Case_Study_Regression.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import datetime\n",
    "import warnings\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import r2_score, mean_squared_error, mean_absolute_error\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from datetime import datetime\n",
    "from geopy.extra.rate_limiter import RateLimiter\n",
    "import folium\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>id</th>\n",
       "      <th>data</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>sqft_lot</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>...</th>\n",
       "      <th>sqft_above</th>\n",
       "      <th>sqft_basement</th>\n",
       "      <th>yr_built</th>\n",
       "      <th>yr_renovated</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>sqft_living15</th>\n",
       "      <th>sqft_lot15</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6243</td>\n",
       "      <td>8850000509</td>\n",
       "      <td>9/23/14</td>\n",
       "      <td>2</td>\n",
       "      <td>1.50</td>\n",
       "      <td>1620</td>\n",
       "      <td>1444</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1080</td>\n",
       "      <td>540</td>\n",
       "      <td>2007</td>\n",
       "      <td>0</td>\n",
       "      <td>98144</td>\n",
       "      <td>47.5892</td>\n",
       "      <td>-122.309</td>\n",
       "      <td>1660</td>\n",
       "      <td>1642</td>\n",
       "      <td>525000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8986</td>\n",
       "      <td>191100410</td>\n",
       "      <td>6/20/14</td>\n",
       "      <td>3</td>\n",
       "      <td>2.75</td>\n",
       "      <td>2470</td>\n",
       "      <td>10125</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2470</td>\n",
       "      <td>0</td>\n",
       "      <td>1960</td>\n",
       "      <td>2012</td>\n",
       "      <td>98040</td>\n",
       "      <td>47.5651</td>\n",
       "      <td>-122.223</td>\n",
       "      <td>2290</td>\n",
       "      <td>10125</td>\n",
       "      <td>970500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9419</td>\n",
       "      <td>6632300040</td>\n",
       "      <td>4/25/15</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1140</td>\n",
       "      <td>7435</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1140</td>\n",
       "      <td>0</td>\n",
       "      <td>1952</td>\n",
       "      <td>1990</td>\n",
       "      <td>98125</td>\n",
       "      <td>47.7300</td>\n",
       "      <td>-122.310</td>\n",
       "      <td>1320</td>\n",
       "      <td>9385</td>\n",
       "      <td>327000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10155</td>\n",
       "      <td>7517500310</td>\n",
       "      <td>5/6/15</td>\n",
       "      <td>3</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1460</td>\n",
       "      <td>6198</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1460</td>\n",
       "      <td>0</td>\n",
       "      <td>1916</td>\n",
       "      <td>0</td>\n",
       "      <td>98107</td>\n",
       "      <td>47.6626</td>\n",
       "      <td>-122.361</td>\n",
       "      <td>2280</td>\n",
       "      <td>5160</td>\n",
       "      <td>775000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10436</td>\n",
       "      <td>7454000990</td>\n",
       "      <td>9/24/14</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>670</td>\n",
       "      <td>6720</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>670</td>\n",
       "      <td>0</td>\n",
       "      <td>1942</td>\n",
       "      <td>0</td>\n",
       "      <td>98126</td>\n",
       "      <td>47.5151</td>\n",
       "      <td>-122.372</td>\n",
       "      <td>710</td>\n",
       "      <td>6720</td>\n",
       "      <td>304950</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   index          id     data   bedrooms   bathrooms   sqft_living   sqft_lot  \\\n",
       "0   6243  8850000509  9/23/14          2        1.50          1620       1444   \n",
       "1   8986   191100410  6/20/14          3        2.75          2470      10125   \n",
       "2   9419  6632300040  4/25/15          2        1.00          1140       7435   \n",
       "3  10155  7517500310   5/6/15          3        1.00          1460       6198   \n",
       "4  10436  7454000990  9/24/14          2        1.00           670       6720   \n",
       "\n",
       "    floors   waterfront   view  ...   sqft_above   sqft_basement   yr_built  \\\n",
       "0      2.0            0      0  ...         1080             540       2007   \n",
       "1      2.0            0      0  ...         2470               0       1960   \n",
       "2      1.0            0      0  ...         1140               0       1952   \n",
       "3      1.5            0      0  ...         1460               0       1916   \n",
       "4      1.0            0      0  ...          670               0       1942   \n",
       "\n",
       "    yr_renovated   zip_code      lat      lon   sqft_living15   sqft_lot15  \\\n",
       "0              0      98144  47.5892 -122.309            1660         1642   \n",
       "1           2012      98040  47.5651 -122.223            2290        10125   \n",
       "2           1990      98125  47.7300 -122.310            1320         9385   \n",
       "3              0      98107  47.6626 -122.361            2280         5160   \n",
       "4              0      98126  47.5151 -122.372             710         6720   \n",
       "\n",
       "    price  \n",
       "0  525000  \n",
       "1  970500  \n",
       "2  327000  \n",
       "3  775000  \n",
       "4  304950  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('train_boston.csv')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. DATA CLEANING AND REVIEWING:\n",
    "\n",
    "#### Results found:\n",
    "#### 1. No nulls/Nans found \n",
    "#### 2. 'data' column typo\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index               int64\n",
       " id                 int64\n",
       " data              object\n",
       " bedrooms           int64\n",
       " bathrooms        float64\n",
       " sqft_living        int64\n",
       " sqft_lot           int64\n",
       " floors           float64\n",
       " waterfront         int64\n",
       " view               int64\n",
       " condition          int64\n",
       " grade              int64\n",
       " sqft_above         int64\n",
       " sqft_basement      int64\n",
       " yr_built           int64\n",
       " yr_renovated       int64\n",
       " zip_code           int64\n",
       " lat              float64\n",
       " lon              float64\n",
       " sqft_living15      int64\n",
       " sqft_lot15         int64\n",
       " price              int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index             0\n",
       " id               0\n",
       " data             0\n",
       " bedrooms         0\n",
       " bathrooms        0\n",
       " sqft_living      0\n",
       " sqft_lot         0\n",
       " floors           0\n",
       " waterfront       0\n",
       " view             0\n",
       " condition        0\n",
       " grade            0\n",
       " sqft_above       0\n",
       " sqft_basement    0\n",
       " yr_built         0\n",
       " yr_renovated     0\n",
       " zip_code         0\n",
       " lat              0\n",
       " lon              0\n",
       " sqft_living15    0\n",
       " sqft_lot15       0\n",
       " price            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isna().sum()\n",
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>id</th>\n",
       "      <th>date</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>sqft_lot</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>...</th>\n",
       "      <th>sqft_above</th>\n",
       "      <th>sqft_basement</th>\n",
       "      <th>yr_built</th>\n",
       "      <th>yr_renovated</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>sqft_living15</th>\n",
       "      <th>sqft_lot15</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6243</td>\n",
       "      <td>8850000509</td>\n",
       "      <td>9/23/14</td>\n",
       "      <td>2</td>\n",
       "      <td>1.50</td>\n",
       "      <td>1620</td>\n",
       "      <td>1444</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1080</td>\n",
       "      <td>540</td>\n",
       "      <td>2007</td>\n",
       "      <td>0</td>\n",
       "      <td>98144</td>\n",
       "      <td>47.5892</td>\n",
       "      <td>-122.309</td>\n",
       "      <td>1660</td>\n",
       "      <td>1642</td>\n",
       "      <td>525000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8986</td>\n",
       "      <td>191100410</td>\n",
       "      <td>6/20/14</td>\n",
       "      <td>3</td>\n",
       "      <td>2.75</td>\n",
       "      <td>2470</td>\n",
       "      <td>10125</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2470</td>\n",
       "      <td>0</td>\n",
       "      <td>1960</td>\n",
       "      <td>2012</td>\n",
       "      <td>98040</td>\n",
       "      <td>47.5651</td>\n",
       "      <td>-122.223</td>\n",
       "      <td>2290</td>\n",
       "      <td>10125</td>\n",
       "      <td>970500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9419</td>\n",
       "      <td>6632300040</td>\n",
       "      <td>4/25/15</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1140</td>\n",
       "      <td>7435</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1140</td>\n",
       "      <td>0</td>\n",
       "      <td>1952</td>\n",
       "      <td>1990</td>\n",
       "      <td>98125</td>\n",
       "      <td>47.7300</td>\n",
       "      <td>-122.310</td>\n",
       "      <td>1320</td>\n",
       "      <td>9385</td>\n",
       "      <td>327000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10155</td>\n",
       "      <td>7517500310</td>\n",
       "      <td>5/6/15</td>\n",
       "      <td>3</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1460</td>\n",
       "      <td>6198</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1460</td>\n",
       "      <td>0</td>\n",
       "      <td>1916</td>\n",
       "      <td>0</td>\n",
       "      <td>98107</td>\n",
       "      <td>47.6626</td>\n",
       "      <td>-122.361</td>\n",
       "      <td>2280</td>\n",
       "      <td>5160</td>\n",
       "      <td>775000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10436</td>\n",
       "      <td>7454000990</td>\n",
       "      <td>9/24/14</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>670</td>\n",
       "      <td>6720</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>670</td>\n",
       "      <td>0</td>\n",
       "      <td>1942</td>\n",
       "      <td>0</td>\n",
       "      <td>98126</td>\n",
       "      <td>47.5151</td>\n",
       "      <td>-122.372</td>\n",
       "      <td>710</td>\n",
       "      <td>6720</td>\n",
       "      <td>304950</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   index          id     date  bedrooms  bathrooms  sqft_living  sqft_lot  \\\n",
       "0   6243  8850000509  9/23/14         2       1.50         1620      1444   \n",
       "1   8986   191100410  6/20/14         3       2.75         2470     10125   \n",
       "2   9419  6632300040  4/25/15         2       1.00         1140      7435   \n",
       "3  10155  7517500310   5/6/15         3       1.00         1460      6198   \n",
       "4  10436  7454000990  9/24/14         2       1.00          670      6720   \n",
       "\n",
       "   floors  waterfront  view  ...  sqft_above  sqft_basement  yr_built  \\\n",
       "0     2.0           0     0  ...        1080            540      2007   \n",
       "1     2.0           0     0  ...        2470              0      1960   \n",
       "2     1.0           0     0  ...        1140              0      1952   \n",
       "3     1.5           0     0  ...        1460              0      1916   \n",
       "4     1.0           0     0  ...         670              0      1942   \n",
       "\n",
       "   yr_renovated  zip_code      lat      lon  sqft_living15  sqft_lot15   price  \n",
       "0             0     98144  47.5892 -122.309           1660        1642  525000  \n",
       "1          2012     98040  47.5651 -122.223           2290       10125  970500  \n",
       "2          1990     98125  47.7300 -122.310           1320        9385  327000  \n",
       "3             0     98107  47.6626 -122.361           2280        5160  775000  \n",
       "4             0     98126  47.5151 -122.372            710        6720  304950  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns = data.columns.str.lower() # lower case\n",
    "data.columns = map(lambda x : x.replace('-', '_').replace(' ', ''), data.columns) \n",
    "data.rename(columns = {'data':'date'}, inplace = True) \n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data1 = data.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "data1.drop('id', axis='columns', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>sqft_lot</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>condition</th>\n",
       "      <th>grade</th>\n",
       "      <th>sqft_above</th>\n",
       "      <th>sqft_basement</th>\n",
       "      <th>yr_built</th>\n",
       "      <th>yr_renovated</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>sqft_living15</th>\n",
       "      <th>sqft_lot15</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9/23/14</td>\n",
       "      <td>2</td>\n",
       "      <td>1.50</td>\n",
       "      <td>1620</td>\n",
       "      <td>1444</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>1080</td>\n",
       "      <td>540</td>\n",
       "      <td>2007</td>\n",
       "      <td>0</td>\n",
       "      <td>98144</td>\n",
       "      <td>47.5892</td>\n",
       "      <td>-122.309</td>\n",
       "      <td>1660</td>\n",
       "      <td>1642</td>\n",
       "      <td>525000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6/20/14</td>\n",
       "      <td>3</td>\n",
       "      <td>2.75</td>\n",
       "      <td>2470</td>\n",
       "      <td>10125</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>8</td>\n",
       "      <td>2470</td>\n",
       "      <td>0</td>\n",
       "      <td>1960</td>\n",
       "      <td>2012</td>\n",
       "      <td>98040</td>\n",
       "      <td>47.5651</td>\n",
       "      <td>-122.223</td>\n",
       "      <td>2290</td>\n",
       "      <td>10125</td>\n",
       "      <td>970500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4/25/15</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1140</td>\n",
       "      <td>7435</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>1140</td>\n",
       "      <td>0</td>\n",
       "      <td>1952</td>\n",
       "      <td>1990</td>\n",
       "      <td>98125</td>\n",
       "      <td>47.7300</td>\n",
       "      <td>-122.310</td>\n",
       "      <td>1320</td>\n",
       "      <td>9385</td>\n",
       "      <td>327000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5/6/15</td>\n",
       "      <td>3</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1460</td>\n",
       "      <td>6198</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>7</td>\n",
       "      <td>1460</td>\n",
       "      <td>0</td>\n",
       "      <td>1916</td>\n",
       "      <td>0</td>\n",
       "      <td>98107</td>\n",
       "      <td>47.6626</td>\n",
       "      <td>-122.361</td>\n",
       "      <td>2280</td>\n",
       "      <td>5160</td>\n",
       "      <td>775000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9/24/14</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>670</td>\n",
       "      <td>6720</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>670</td>\n",
       "      <td>0</td>\n",
       "      <td>1942</td>\n",
       "      <td>0</td>\n",
       "      <td>98126</td>\n",
       "      <td>47.5151</td>\n",
       "      <td>-122.372</td>\n",
       "      <td>710</td>\n",
       "      <td>6720</td>\n",
       "      <td>304950</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      date  bedrooms  bathrooms  sqft_living  sqft_lot  floors  waterfront  \\\n",
       "0  9/23/14         2       1.50         1620      1444     2.0           0   \n",
       "1  6/20/14         3       2.75         2470     10125     2.0           0   \n",
       "2  4/25/15         2       1.00         1140      7435     1.0           0   \n",
       "3   5/6/15         3       1.00         1460      6198     1.5           0   \n",
       "4  9/24/14         2       1.00          670      6720     1.0           0   \n",
       "\n",
       "   view  condition  grade  sqft_above  sqft_basement  yr_built  yr_renovated  \\\n",
       "0     0          3      9        1080            540      2007             0   \n",
       "1     0          3      8        2470              0      1960          2012   \n",
       "2     0          3      7        1140              0      1952          1990   \n",
       "3     0          4      7        1460              0      1916             0   \n",
       "4     0          5      6         670              0      1942             0   \n",
       "\n",
       "   zip_code      lat      lon  sqft_living15  sqft_lot15   price  \n",
       "0     98144  47.5892 -122.309           1660        1642  525000  \n",
       "1     98040  47.5651 -122.223           2290       10125  970500  \n",
       "2     98125  47.7300 -122.310           1320        9385  327000  \n",
       "3     98107  47.6626 -122.361           2280        5160  775000  \n",
       "4     98126  47.5151 -122.372            710        6720  304950  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1.drop('index', axis='columns', inplace=True)\n",
    "data1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date             datetime64[ns]\n",
       "bedrooms                  int64\n",
       "bathrooms               float64\n",
       "sqft_living               int64\n",
       "sqft_lot                  int64\n",
       "floors                  float64\n",
       "waterfront                int64\n",
       "view                      int64\n",
       "condition                 int64\n",
       "grade                     int64\n",
       "sqft_above                int64\n",
       "sqft_basement             int64\n",
       "yr_built                  int64\n",
       "yr_renovated              int64\n",
       "zip_code                  int64\n",
       "lat                     float64\n",
       "lon                     float64\n",
       "sqft_living15             int64\n",
       "sqft_lot15                int64\n",
       "price                     int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1['date'] = pd.to_datetime(data1['date'])\n",
    "data1.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3     7771\n",
       "4     5572\n",
       "2     2212\n",
       "5     1302\n",
       "6      216\n",
       "1      151\n",
       "7       33\n",
       "8       10\n",
       "9        5\n",
       "10       2\n",
       "11       1\n",
       "33       1\n",
       "Name: bedrooms, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['bedrooms'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3    7771\n",
       "4    5572\n",
       "2    2212\n",
       "5    1302\n",
       "6     216\n",
       "1     151\n",
       "7      33\n",
       "Name: bedrooms, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1 = data1[data1['bedrooms'] < 8]\n",
    "data1['bedrooms'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data1['price_per_sqft_living'] = data1['price'] / data1['sqft_living']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data1['price_per_sqft_lot'] = data1['price'] / data1['sqft_lot']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"bin_labels_5 = [0,1,2,3,4]\\ndata1['price_range'] = pd.qcut(data1['price'],\\n                              q=[0, .2, .4, .6, .8, 1],\\n                              labels=bin_labels_5)\\ndata1.isna().sum()\""
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''bin_labels_5 = [0,1,2,3,4]\n",
    "data1['price_range'] = pd.qcut(data1['price'],\n",
    "                              q=[0, .2, .4, .6, .8, 1],\n",
    "                              labels=bin_labels_5)\n",
    "data1.isna().sum()'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### HARDCODE ZIPCODE RANKING "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\" ranking_zipcode = []\\n                                                 \\nfor x in data1['zip_code']:\\n    if x == 98039:\\n        ranking_zipcode.append(5)\\n    elif x == 98004:\\n        ranking_zipcode.append(4)\\n    elif x == 98040:\\n        ranking_zipcode.append(3)\\n    elif x == 98112:\\n        ranking_zipcode.append(2)\\n    elif x == 98102:\\n        ranking_zipcode.append(1)\\n    else:\\n        ranking_zipcode.append(0)\\ndata1['ranking_zipcode'] = ranking_zipcode\\ndata1.head() \""
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' ranking_zipcode = []\n",
    "                                                 \n",
    "for x in data1['zip_code']:\n",
    "    if x == 98039:\n",
    "        ranking_zipcode.append(5)\n",
    "    elif x == 98004:\n",
    "        ranking_zipcode.append(4)\n",
    "    elif x == 98040:\n",
    "        ranking_zipcode.append(3)\n",
    "    elif x == 98112:\n",
    "        ranking_zipcode.append(2)\n",
    "    elif x == 98102:\n",
    "        ranking_zipcode.append(1)\n",
    "    else:\n",
    "        ranking_zipcode.append(0)\n",
    "data1['ranking_zipcode'] = ranking_zipcode\n",
    "data1.head() ''' "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_feature_importance(importance,names,model_type):\n",
    "    #Create arrays from feature importance and feature names\n",
    "    feature_importance = np.array(importance)\n",
    "    feature_names = np.array(names)\n",
    "\n",
    "    #Create a DataFrame using a Dictionary\n",
    "    data={'feature_names':feature_names,'feature_importance':feature_importance}\n",
    "    fi_df = pd.DataFrame(data)\n",
    "\n",
    "    #Sort the DataFrame in order decreasing feature importance\n",
    "    fi_df.sort_values(by=['feature_importance'], ascending=False,inplace=True)\n",
    "\n",
    "    #Define size of bar plot\n",
    "    plt.figure(figsize=(10,8))\n",
    "    #Plot Searborn bar chart\n",
    "    sns.barplot(x=fi_df['feature_importance'], y=fi_df['feature_names'])\n",
    "    #Add chart labels\n",
    "    plt.title(model_type + 'FEATURE IMPORTANCE')\n",
    "    plt.xlabel('FEATURE IMPORTANCE')\n",
    "    plt.ylabel('FEATURE NAMES')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# data1['price_range'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data1['price_range'] = data1['price_range'].astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>98103</th>\n",
       "      <td>497</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98038</th>\n",
       "      <td>473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98052</th>\n",
       "      <td>473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98115</th>\n",
       "      <td>455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98042</th>\n",
       "      <td>454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98102</th>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98010</th>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98024</th>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98148</th>\n",
       "      <td>51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98039</th>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>70 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       counts\n",
       "98103     497\n",
       "98038     473\n",
       "98052     473\n",
       "98115     455\n",
       "98042     454\n",
       "...       ...\n",
       "98102      81\n",
       "98010      81\n",
       "98024      64\n",
       "98148      51\n",
       "98039      38\n",
       "\n",
       "[70 rows x 1 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "categories = pd.DataFrame(data1['zip_code'].value_counts())\n",
    "categories.columns = ['counts']\n",
    "categories"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate Map with Location of all houses "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' map1 = folium.Map(\\n    location=[47.5892,-122.309],\\n    tiles=\\'cartodbpositron\\',\\n    zoom_start=12,\\n)\\ndata1.apply(lambda row:folium.CircleMarker(location=[row[\"lat\"], row[\"lon\"]]).add_to(map1), axis=1)\\nmap1 '"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' map1 = folium.Map(\n",
    "    location=[47.5892,-122.309],\n",
    "    tiles='cartodbpositron',\n",
    "    zoom_start=12,\n",
    ")\n",
    "data1.apply(lambda row:folium.CircleMarker(location=[row[\"lat\"], row[\"lon\"]]).add_to(map1), axis=1)\n",
    "map1 ''' "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         9\n",
       "1         6\n",
       "2         4\n",
       "3         5\n",
       "4         9\n",
       "         ..\n",
       "17271    12\n",
       "17272     4\n",
       "17273     5\n",
       "17274     5\n",
       "17275    12\n",
       "Name: date_month, Length: 17257, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1['date_month']= data1['date'].dt.month\n",
    "data1['date_month']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "data1['date_year']= data1['date'].dt.year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>sqft_lot</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>condition</th>\n",
       "      <th>grade</th>\n",
       "      <th>...</th>\n",
       "      <th>yr_built</th>\n",
       "      <th>yr_renovated</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>sqft_living15</th>\n",
       "      <th>sqft_lot15</th>\n",
       "      <th>price</th>\n",
       "      <th>date_month</th>\n",
       "      <th>date_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2014-09-23</td>\n",
       "      <td>2</td>\n",
       "      <td>1.50</td>\n",
       "      <td>1620</td>\n",
       "      <td>1444</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>...</td>\n",
       "      <td>2007</td>\n",
       "      <td>0</td>\n",
       "      <td>98144</td>\n",
       "      <td>47.5892</td>\n",
       "      <td>-122.309</td>\n",
       "      <td>1660</td>\n",
       "      <td>1642</td>\n",
       "      <td>525000</td>\n",
       "      <td>9</td>\n",
       "      <td>2014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2014-06-20</td>\n",
       "      <td>3</td>\n",
       "      <td>2.75</td>\n",
       "      <td>2470</td>\n",
       "      <td>10125</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>1960</td>\n",
       "      <td>2012</td>\n",
       "      <td>98040</td>\n",
       "      <td>47.5651</td>\n",
       "      <td>-122.223</td>\n",
       "      <td>2290</td>\n",
       "      <td>10125</td>\n",
       "      <td>970500</td>\n",
       "      <td>6</td>\n",
       "      <td>2014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015-04-25</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1140</td>\n",
       "      <td>7435</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>1952</td>\n",
       "      <td>1990</td>\n",
       "      <td>98125</td>\n",
       "      <td>47.7300</td>\n",
       "      <td>-122.310</td>\n",
       "      <td>1320</td>\n",
       "      <td>9385</td>\n",
       "      <td>327000</td>\n",
       "      <td>4</td>\n",
       "      <td>2015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015-05-06</td>\n",
       "      <td>3</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1460</td>\n",
       "      <td>6198</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>1916</td>\n",
       "      <td>0</td>\n",
       "      <td>98107</td>\n",
       "      <td>47.6626</td>\n",
       "      <td>-122.361</td>\n",
       "      <td>2280</td>\n",
       "      <td>5160</td>\n",
       "      <td>775000</td>\n",
       "      <td>5</td>\n",
       "      <td>2015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2014-09-24</td>\n",
       "      <td>2</td>\n",
       "      <td>1.00</td>\n",
       "      <td>670</td>\n",
       "      <td>6720</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>1942</td>\n",
       "      <td>0</td>\n",
       "      <td>98126</td>\n",
       "      <td>47.5151</td>\n",
       "      <td>-122.372</td>\n",
       "      <td>710</td>\n",
       "      <td>6720</td>\n",
       "      <td>304950</td>\n",
       "      <td>9</td>\n",
       "      <td>2014</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        date  bedrooms  bathrooms  sqft_living  sqft_lot  floors  waterfront  \\\n",
       "0 2014-09-23         2       1.50         1620      1444     2.0           0   \n",
       "1 2014-06-20         3       2.75         2470     10125     2.0           0   \n",
       "2 2015-04-25         2       1.00         1140      7435     1.0           0   \n",
       "3 2015-05-06         3       1.00         1460      6198     1.5           0   \n",
       "4 2014-09-24         2       1.00          670      6720     1.0           0   \n",
       "\n",
       "   view  condition  grade  ...  yr_built  yr_renovated  zip_code      lat  \\\n",
       "0     0          3      9  ...      2007             0     98144  47.5892   \n",
       "1     0          3      8  ...      1960          2012     98040  47.5651   \n",
       "2     0          3      7  ...      1952          1990     98125  47.7300   \n",
       "3     0          4      7  ...      1916             0     98107  47.6626   \n",
       "4     0          5      6  ...      1942             0     98126  47.5151   \n",
       "\n",
       "       lon  sqft_living15  sqft_lot15   price  date_month  date_year  \n",
       "0 -122.309           1660        1642  525000           9       2014  \n",
       "1 -122.223           2290       10125  970500           6       2014  \n",
       "2 -122.310           1320        9385  327000           4       2015  \n",
       "3 -122.361           2280        5160  775000           5       2015  \n",
       "4 -122.372            710        6720  304950           9       2014  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " # data1['years_since_renovation'] = data1['date_year'] - data1['yr_renovated']\n",
    "data1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "data1 = data1.reset_index(drop=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\" for row in range(len(data1['years_since_renovation'])):\\n    if data1['years_since_renovation'][row] == data1['date_year'][row]:\\n        data1['years_since_renovation'][row] =  0\\n        \\ndata1.head()\\n\""
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' for row in range(len(data1['years_since_renovation'])):\n",
    "    if data1['years_since_renovation'][row] == data1['date_year'][row]:\n",
    "        data1['years_since_renovation'][row] =  0\n",
    "        \n",
    "data1.head()\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\" def drop_2015_2014(x): \\n    if (x == 2014) | (x == 2015): \\n        return 0\\n    else:\\n        return x\\ndata1['years_since_renovation'] = data1['years_since_renovation'].apply(drop_2015_2014)\\n\\ndata1.head() \""
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' def drop_2015_2014(x): \n",
    "    if (x == 2014) | (x == 2015): \n",
    "        return 0\n",
    "    else:\n",
    "        return x\n",
    "data1['years_since_renovation'] = data1['years_since_renovation'].apply(drop_2015_2014)\n",
    "\n",
    "data1.head() '''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17257, 22)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data2 = data1.copy()\n",
    "data2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['bedrooms', 'bathrooms', 'sqft_living', 'sqft_lot', 'floors',\n",
       "       'waterfront', 'view', 'condition', 'grade', 'sqft_above',\n",
       "       'sqft_basement', 'yr_built', 'yr_renovated', 'zip_code', 'lat', 'lon',\n",
       "       'sqft_living15', 'sqft_lot15', 'price', 'date_month', 'date_year'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_num_exp = data2.select_dtypes(np.number)\n",
    "X_num_exp.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17257, 21)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_num_exp.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4236, 21)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "expensive_houses = X_num_exp.loc[X_num_exp['price'] >= 650000]\n",
    "\n",
    "\n",
    "expensive_houses.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>bedrooms</th>\n",
       "      <td>0.208203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bathrooms</th>\n",
       "      <td>0.467714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_living</th>\n",
       "      <td>0.615326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_lot</th>\n",
       "      <td>0.026336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>floors</th>\n",
       "      <td>0.125553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>waterfront</th>\n",
       "      <td>0.308306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>view</th>\n",
       "      <td>0.335877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>condition</th>\n",
       "      <td>-0.005770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grade</th>\n",
       "      <td>0.510642</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_above</th>\n",
       "      <td>0.471199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_basement</th>\n",
       "      <td>0.298381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yr_built</th>\n",
       "      <td>0.031970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yr_renovated</th>\n",
       "      <td>0.115012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>zip_code</th>\n",
       "      <td>-0.078973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lat</th>\n",
       "      <td>-0.005334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lon</th>\n",
       "      <td>-0.085670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_living15</th>\n",
       "      <td>0.373988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_lot15</th>\n",
       "      <td>0.040822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_month</th>\n",
       "      <td>-0.004040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_year</th>\n",
       "      <td>0.005954</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  price\n",
       "bedrooms       0.208203\n",
       "bathrooms      0.467714\n",
       "sqft_living    0.615326\n",
       "sqft_lot       0.026336\n",
       "floors         0.125553\n",
       "waterfront     0.308306\n",
       "view           0.335877\n",
       "condition     -0.005770\n",
       "grade          0.510642\n",
       "sqft_above     0.471199\n",
       "sqft_basement  0.298381\n",
       "yr_built       0.031970\n",
       "yr_renovated   0.115012\n",
       "zip_code      -0.078973\n",
       "lat           -0.005334\n",
       "lon           -0.085670\n",
       "sqft_living15  0.373988\n",
       "sqft_lot15     0.040822\n",
       "price          1.000000\n",
       "date_month    -0.004040\n",
       "date_year      0.005954"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix = expensive_houses.corr()\n",
    "\n",
    "x = corr_matrix[[\"price\"]]\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(20, 20))\n",
    "ax = sns.heatmap(x[['price']], annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mask = np.zeros_like(expensive_houses.corr())\n",
    "\n",
    "mask[np.triu_indices_from(mask)] = True # optional, to hide repeat half of the matrix\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(20, 20))\n",
    "ax = sns.heatmap(expensive_houses.corr(), mask=mask, annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data1 = data1[data1['price'] < 4000000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['bedrooms', 'bathrooms', 'sqft_living', 'sqft_lot', 'floors',\n",
       "       'waterfront', 'view', 'condition', 'grade', 'sqft_above',\n",
       "       'sqft_basement', 'yr_built', 'yr_renovated', 'zip_code', 'lat', 'lon',\n",
       "       'sqft_living15', 'sqft_lot15', 'price', 'date_month', 'date_year'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_num = data1.select_dtypes(np.number)\n",
    "''' def obj_plotter(data):\n",
    "    obj = list(data.select_dtypes(include=['number']).columns)\n",
    "    for col in obj:\n",
    "        my_data = data[col].value_counts().reset_index()\n",
    "        ax = sns.barplot(x=col, y='index', data = my_data).set_title(col.upper())\n",
    "        plt.figure()\n",
    "obj_plotter(X_num)'''\n",
    "X_num.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hist Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' for column in X_num.columns:\\n    sns.histplot(X_num[column])\\n    plt.show() '"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' for column in X_num.columns:\n",
    "    sns.histplot(X_num[column])\n",
    "    plt.show() ''' "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Box Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function matplotlib.pyplot.show(*args, **kw)>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "    sns.boxplot(X_num['price'])\n",
    "    plt.show"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' mask = np.zeros_like(X_num.corr())\\n\\nmask[np.triu_indices_from(mask)] = True # optional, to hide repeat half of the matrix\\n\\nfig, ax = plt.subplots(figsize=(20, 20))\\nax = sns.heatmap(X_num.corr(), mask=mask, annot=True)\\nplt.show() '"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' mask = np.zeros_like(X_num.corr())\n",
    "\n",
    "mask[np.triu_indices_from(mask)] = True # optional, to hide repeat half of the matrix\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(20, 20))\n",
    "ax = sns.heatmap(X_num.corr(), mask=mask, annot=True)\n",
    "plt.show() ''' "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>bedrooms</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>3.373298</td>\n",
       "      <td>0.889829</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>3.0000</td>\n",
       "      <td>3.0000</td>\n",
       "      <td>4.0000</td>\n",
       "      <td>7.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bathrooms</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>2.116112</td>\n",
       "      <td>0.765535</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>1.7500</td>\n",
       "      <td>2.2500</td>\n",
       "      <td>2.5000</td>\n",
       "      <td>8.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_living</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>2082.342933</td>\n",
       "      <td>922.661663</td>\n",
       "      <td>370.0000</td>\n",
       "      <td>1420.0000</td>\n",
       "      <td>1910.0000</td>\n",
       "      <td>2550.0000</td>\n",
       "      <td>1.354000e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_lot</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>15094.512604</td>\n",
       "      <td>41493.982120</td>\n",
       "      <td>520.0000</td>\n",
       "      <td>5030.0000</td>\n",
       "      <td>7600.0000</td>\n",
       "      <td>10698.0000</td>\n",
       "      <td>1.651359e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>floors</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>1.497363</td>\n",
       "      <td>0.541301</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>2.0000</td>\n",
       "      <td>3.500000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>waterfront</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>0.007649</td>\n",
       "      <td>0.087126</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>view</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>0.233586</td>\n",
       "      <td>0.768506</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>4.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>condition</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>3.406560</td>\n",
       "      <td>0.648434</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>3.0000</td>\n",
       "      <td>3.0000</td>\n",
       "      <td>4.0000</td>\n",
       "      <td>5.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grade</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>7.660312</td>\n",
       "      <td>1.178218</td>\n",
       "      <td>3.0000</td>\n",
       "      <td>7.0000</td>\n",
       "      <td>7.0000</td>\n",
       "      <td>8.0000</td>\n",
       "      <td>1.300000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_above</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>1790.005853</td>\n",
       "      <td>829.866576</td>\n",
       "      <td>370.0000</td>\n",
       "      <td>1190.0000</td>\n",
       "      <td>1570.0000</td>\n",
       "      <td>2210.0000</td>\n",
       "      <td>9.410000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_basement</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>292.337081</td>\n",
       "      <td>445.160346</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>560.0000</td>\n",
       "      <td>4.820000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yr_built</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>1971.222055</td>\n",
       "      <td>29.331745</td>\n",
       "      <td>1900.0000</td>\n",
       "      <td>1952.0000</td>\n",
       "      <td>1975.0000</td>\n",
       "      <td>1997.0000</td>\n",
       "      <td>2.015000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yr_renovated</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>83.741264</td>\n",
       "      <td>400.195448</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.015000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>zip_code</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>98078.064264</td>\n",
       "      <td>53.590905</td>\n",
       "      <td>98001.0000</td>\n",
       "      <td>98033.0000</td>\n",
       "      <td>98065.0000</td>\n",
       "      <td>98118.0000</td>\n",
       "      <td>9.819900e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lat</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>47.559530</td>\n",
       "      <td>0.138670</td>\n",
       "      <td>47.1559</td>\n",
       "      <td>47.4678</td>\n",
       "      <td>47.5718</td>\n",
       "      <td>47.6774</td>\n",
       "      <td>4.777760e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lon</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>-122.214462</td>\n",
       "      <td>0.140516</td>\n",
       "      <td>-122.5190</td>\n",
       "      <td>-122.3290</td>\n",
       "      <td>-122.2300</td>\n",
       "      <td>-122.1250</td>\n",
       "      <td>-1.213150e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_living15</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>1987.953932</td>\n",
       "      <td>687.614353</td>\n",
       "      <td>460.0000</td>\n",
       "      <td>1490.0000</td>\n",
       "      <td>1840.0000</td>\n",
       "      <td>2360.0000</td>\n",
       "      <td>6.210000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqft_lot15</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>12656.209133</td>\n",
       "      <td>25818.872180</td>\n",
       "      <td>651.0000</td>\n",
       "      <td>5100.0000</td>\n",
       "      <td>7615.0000</td>\n",
       "      <td>10095.0000</td>\n",
       "      <td>5.606170e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>540361.923741</td>\n",
       "      <td>372348.377401</td>\n",
       "      <td>80000.0000</td>\n",
       "      <td>321000.0000</td>\n",
       "      <td>450000.0000</td>\n",
       "      <td>643000.0000</td>\n",
       "      <td>7.700000e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_month</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>6.560874</td>\n",
       "      <td>3.116836</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>4.0000</td>\n",
       "      <td>6.0000</td>\n",
       "      <td>9.0000</td>\n",
       "      <td>1.200000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_year</th>\n",
       "      <td>17257.0</td>\n",
       "      <td>2014.322999</td>\n",
       "      <td>0.467636</td>\n",
       "      <td>2014.0000</td>\n",
       "      <td>2014.0000</td>\n",
       "      <td>2014.0000</td>\n",
       "      <td>2015.0000</td>\n",
       "      <td>2.015000e+03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 count           mean            std         min          25%  \\\n",
       "bedrooms       17257.0       3.373298       0.889829      1.0000       3.0000   \n",
       "bathrooms      17257.0       2.116112       0.765535      0.5000       1.7500   \n",
       "sqft_living    17257.0    2082.342933     922.661663    370.0000    1420.0000   \n",
       "sqft_lot       17257.0   15094.512604   41493.982120    520.0000    5030.0000   \n",
       "floors         17257.0       1.497363       0.541301      1.0000       1.0000   \n",
       "waterfront     17257.0       0.007649       0.087126      0.0000       0.0000   \n",
       "view           17257.0       0.233586       0.768506      0.0000       0.0000   \n",
       "condition      17257.0       3.406560       0.648434      1.0000       3.0000   \n",
       "grade          17257.0       7.660312       1.178218      3.0000       7.0000   \n",
       "sqft_above     17257.0    1790.005853     829.866576    370.0000    1190.0000   \n",
       "sqft_basement  17257.0     292.337081     445.160346      0.0000       0.0000   \n",
       "yr_built       17257.0    1971.222055      29.331745   1900.0000    1952.0000   \n",
       "yr_renovated   17257.0      83.741264     400.195448      0.0000       0.0000   \n",
       "zip_code       17257.0   98078.064264      53.590905  98001.0000   98033.0000   \n",
       "lat            17257.0      47.559530       0.138670     47.1559      47.4678   \n",
       "lon            17257.0    -122.214462       0.140516   -122.5190    -122.3290   \n",
       "sqft_living15  17257.0    1987.953932     687.614353    460.0000    1490.0000   \n",
       "sqft_lot15     17257.0   12656.209133   25818.872180    651.0000    5100.0000   \n",
       "price          17257.0  540361.923741  372348.377401  80000.0000  321000.0000   \n",
       "date_month     17257.0       6.560874       3.116836      1.0000       4.0000   \n",
       "date_year      17257.0    2014.322999       0.467636   2014.0000    2014.0000   \n",
       "\n",
       "                       50%          75%           max  \n",
       "bedrooms            3.0000       4.0000  7.000000e+00  \n",
       "bathrooms           2.2500       2.5000  8.000000e+00  \n",
       "sqft_living      1910.0000    2550.0000  1.354000e+04  \n",
       "sqft_lot         7600.0000   10698.0000  1.651359e+06  \n",
       "floors              1.5000       2.0000  3.500000e+00  \n",
       "waterfront          0.0000       0.0000  1.000000e+00  \n",
       "view                0.0000       0.0000  4.000000e+00  \n",
       "condition           3.0000       4.0000  5.000000e+00  \n",
       "grade               7.0000       8.0000  1.300000e+01  \n",
       "sqft_above       1570.0000    2210.0000  9.410000e+03  \n",
       "sqft_basement       0.0000     560.0000  4.820000e+03  \n",
       "yr_built         1975.0000    1997.0000  2.015000e+03  \n",
       "yr_renovated        0.0000       0.0000  2.015000e+03  \n",
       "zip_code        98065.0000   98118.0000  9.819900e+04  \n",
       "lat                47.5718      47.6774  4.777760e+01  \n",
       "lon              -122.2300    -122.1250 -1.213150e+02  \n",
       "sqft_living15    1840.0000    2360.0000  6.210000e+03  \n",
       "sqft_lot15       7615.0000   10095.0000  5.606170e+05  \n",
       "price          450000.0000  643000.0000  7.700000e+06  \n",
       "date_month          6.0000       9.0000  1.200000e+01  \n",
       "date_year        2014.0000    2015.0000  2.015000e+03  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_num.describe().transpose()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17257, 19)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = X_num.drop(['price' , 'zip_code'], axis=1)\n",
    "y = X_num['price']\n",
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4236, 19)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Xe = expensive_houses.drop(['price' , 'zip_code'], axis=1)\n",
    "ye = expensive_houses['price']\n",
    "Xe.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['bedrooms', 'bathrooms', 'sqft_living', 'sqft_lot', 'floors',\n",
       "       'waterfront', 'view', 'condition', 'grade', 'sqft_above',\n",
       "       'sqft_basement', 'yr_built', 'yr_renovated', 'lat', 'lon',\n",
       "       'sqft_living15', 'sqft_lot15', 'date_month', 'date_year'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.6930890567213738,\n",
       " 0.6913899576764164,\n",
       " 125435.99026188182,\n",
       " 212436.03006200964,\n",
       "           true          pred          resid\n",
       " 15526   395000  4.461375e+05   51137.455656\n",
       " 15633   456500  4.999771e+05   43477.054159\n",
       " 10691   387000  3.189750e+05   68024.969565\n",
       " 3124    260000  1.548208e+05  105179.203742\n",
       " 11895   585000  6.680139e+05   83013.886321\n",
       " 13362   919204  9.247477e+05    5543.703005\n",
       " 15892  1700000  1.023466e+06  676534.055185\n",
       " 4980    102500  6.134167e+04   41158.333912\n",
       " 15441   569500  5.362527e+05   33247.345966\n",
       " 16189   380000  4.524644e+05   72464.392736)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "def model_function_linear(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100)\n",
    "    model = LinearRegression()\n",
    "    model.fit(X_train, y_train)\n",
    "    predictions = model.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    return r2_score(y_test, predictions), 1 - (1-r2_score(y_test, predictions))*(len(y_test)-1)/(len(y_test)-X_test.shape[1]-1), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "\n",
    "scores_linear = model_function_linear(X, y)\n",
    "scores_linear"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.510144811692953,\n",
       " 156713.79725119096,\n",
       " 268383.52085114154,\n",
       "           true            pred          resid\n",
       " 15526   395000   598166.666667  203166.666667\n",
       " 15633   456500   437412.222222   19087.777778\n",
       " 10691   387000   551688.888889  164688.888889\n",
       " 3124    260000   283055.555556   23055.555556\n",
       " 11895   585000   542416.666667   42583.333333\n",
       " 13362   919204  1000000.000000   80796.000000\n",
       " 15892  1700000   865814.000000  834186.000000\n",
       " 4980    102500   245344.444444  142844.444444\n",
       " 15441   569500   494544.444444   74955.555556\n",
       " 16189   380000   373544.444444    6455.555556)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "def model_function_knn(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100)\n",
    "    knn = KNeighborsRegressor(n_neighbors=9)\n",
    "    knn.fit(X_train, y_train)\n",
    "    predictions = knn.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    return knn.score(X_test, y_test), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "    \n",
    "scores_knn = model_function_knn(X, y)\n",
    "scores_knn\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### RANDOM FOREST + FEATURE IMPORTANCE\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1. EXPENSIVE HOUSES DATA FRAME (PRICE >= 650,000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.7952818675147095,\n",
       " 139045.58450471697,\n",
       " 240499.99075372246,\n",
       "           true        pred      resid\n",
       " 15085   664000   767872.63  103872.63\n",
       " 1807    905000   935279.50   30279.50\n",
       " 1817    833450  1051394.50  217944.50\n",
       " 6650   1850000  1389550.00  460450.00\n",
       " 7511    980000  1036398.00   56398.00\n",
       " 14984   987000  1072678.00   85678.00\n",
       " 16742   715000   782036.29   67036.29\n",
       " 14453  3300000  2512250.00  787750.00\n",
       " 11188   789000  1061866.12  272866.12\n",
       " 6922   2000000  1339154.50  660845.50)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "def model_function_random_forest(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100)\n",
    "    reg = RandomForestRegressor()\n",
    "    reg.fit(X_train, y_train)\n",
    "    predictions = reg.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    plot_feature_importance(reg.feature_importances_,X_train.columns,'RANDOM FOREST')\n",
    "    return reg.score(X_test, y_test), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "    \n",
    "scores_random_forest = model_function_random_forest(Xe, ye)\n",
    "scores_random_forest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. FULL DATA FRAME: \n",
    "#### a) RANDOM FOREST REGRESSOR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.8854977890506237,\n",
       " 72865.33778099652,\n",
       " 129756.44133510775,\n",
       "           true        pred      resid\n",
       " 15526   395000   356253.04   38746.96\n",
       " 15633   456500   483263.45   26763.45\n",
       " 10691   387000   437726.50   50726.50\n",
       " 3124    260000   317317.50   57317.50\n",
       " 11895   585000   630208.49   45208.49\n",
       " 13362   919204  1078983.07  159779.07\n",
       " 15892  1700000  1409103.04  290896.96\n",
       " 4980    102500   208828.09  106328.09\n",
       " 15441   569500   588695.00   19195.00\n",
       " 16189   380000   380111.50     111.50)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApsAAAHwCAYAAADpSaRgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde5hdZXn///eHk5yPAQGtBtBKBSRI8CcCimItVlSUQ0BaRUtptcpX+0VKRat+rRUK1hYt1thaUFAREIpQQUUCBFBMEEhQC1awKiImIicpINy/P9aKboaZPTPJrJnZk/fruuaatZ/Deu61Z+fi5lnPs3aqCkmSJKkLa0x1AJIkSZq5TDYlSZLUGZNNSZIkdcZkU5IkSZ0x2ZQkSVJnTDYlSZLUGZNNSZIkdcZkU9Kwktye5MEk9ye5M8npSTYc0maDtv4/R+j/syQb9JQdlWRBz+tK8kB7juVJLksyb5hzHZDkurbt8iRnJXlqT/2R7bn+YUi/A9vy00e4xn2TPNaOv+LnSz31z05yYZJ7ktyX5PIkL+ipn92ef0Xf25Mc3+d9XPHzsbZunSQfTvLjtvy2JB9p63rbPzbkHEckeV+SR4a0O67tuyDJ/450XW2b7drzntZTNtqYpyf52yHnWfEerDXM9T7hc9O+fnjIWDf2+fv8uOf1gnasXYe0u6At37d93fve/DLJNUn27Gn/1PYztLz9TF2X5IAh5+z9bP4kyT8kWTPJl3vifmTItfxL23dV/10kyTFJlrYx/DjJOUl2Ge97KE0HJpuS+nllVW0IzAF2A/56SP3BwEPAy5JsM0z/tYD/M8oYu7ZjPAs4HfhYkveuqExyMPBZ4J+AWcBO7ZgLk2zWc57/BuatSHparwduGWX8O6pqw56fV7bj7gBcDSwBtgO2Bc4HvtKbuLQ2ba/hYOA9SX5/SP0rh4zx1rb8r4G5wPOAjYAXA98G6G0P/M+Qc5zV9j97yHn/vmfMtw53XUPem7uBw5I8aRxjjsVon5u/HxLbrk88xYhuaWMHIMkWwPOBnw9pd3Ybw5bAQuCLbRK3efv6YZrP0izgI8Bn289arxWfzRcB84A3VdXLe96js4Zcy5+3/Vb138U/tfXHAJsDvwtcALyip82qvIfSpDLZlDSqqroTuJQmeej1BuBfgJuAI4bpejJwbJJNxzDGsqr6DPBm4K+TbJEkwIeBv62qs6rqwTaWo4D7gXf0nOJOmsTwDwDapOIFwIVjv9LHeR9wbVWdUFW/qKr7qupU4DPASSNcwyLgZp74Po1kD+D8qrqjGrdX1adXMt7xej3wbuARYGgiOiH6fG5WxVk0/1OxZvv6cJr/CXh4hBgeAc4Atga2oPnM3A/8SVXd2X6mPgd8EPhw+5kbeo7v0/yPx1ivY6X/XSR5JvAXwOFV9fWqeqiqftV+/k8c4/jStGKyKWlUaW5Zvxz4fk/Z04B9af7jfxY9s009FgELgGPHMdx/0Mz8PI9mtvNpwDm9DarqMeA8YOgM4qd74jisPddD4xi71+8PHbf1BWCvJOsPrUjyfGBnet6nUXwD+Mskb0myy3CJTheS7AM8Ffg8zfUM97ebiHGe8LmZAHcA3wFe1r5+Pc3ffaQYngQcCfy4qpbR/F3Paz9Dvb5A81n73WHOsSOwD2O4jgn4d7FfG+t1o40lDQqTTUn9XJDkPuBHwF3Ae3vqXg/cVFXfAT4H7JRkt2HO8TfA25JsOZYB25moZTS3D2e1xT8dpulPe+pXOB/YN8kmjJKE9Ni2Xde34ufQtnxWn3HXAHpv4S9L8iBwLXAazS3PXhcMGeNP2/IP0cySHkGTgPwkyRvGEPMKhw4577Y9dacOqftAT90bgC9X1d00SxRenmSrcYw7mn6fG2hm9XpjO2Oc5/808Pokz6JZwnDtMG0OTfLLNobdgQPb8n5/1xX1K1yf5AHguzTJ4WlDOw1jVf9dbDFCfEOt6nsoTRqTTUn9HFhVG9HM1OzI4/9D/HqamRuq6g7gCpok5nGqailwEXD80LrhJFmbZp3dL2iSToDh1r1t01O/YqwHgYtpbg/PqqqrxzDkHVW1ac/PF9ryZX3GfYxmveMKs4ANaWaq9gXWHtLnwCFjfLKN99Gq+ueq2gvYlOZW7qeS/N4Y4gb4wpDz3tFTd8yQuvcAJFkPOITf/u2upVmf+boxjPfrYa5tbZr3o3emsN/nBuCUIbGNJ8EG+CLwEuBtNMsahrPivdmqql5SVYvb8n5/1xX1KzyX5u86D/j/gA2GdhrGqv67WD5CfEOt6nsoTRqTTUmjqqoraDbvnAKQZkf2M2nWVt6Z5E6a/xgfPmSDzgrvBf4UeMoYhns1TVJzHfBfwI9pkqPfSLIGcBBw2TD9Pw38X0ZOQsbqa0PHbR1Ks5bzV72FbeL4YeB/gbeMd7B27eA/0ySxz16JeMfqNcDGwGk9f7unMLZb6f8DzB5Sth3wo2FuSz/hczNR2vf+yzTre8f7d/4acFD7Gep1KM0s6OM2lLVrab9AM2v9N/1OPEH/Li4Dnppk7jiuSZrWTDYljdU/Ar+fZA7NTM1XaZKiOe3PzsD6NGv0HqfdYHE2ze7aYSXZPMkRwD8DJ1XV8qoqmtnCdyd5XZL1kmwN/CtNwvSRYU51Bc26vI+u9JU23g+8IMkH29g2SvI2mqTsr/r0OxE4Lsm6ow2Q5O1pHu+zXpK12lvoG9HuSO/IG4BPAbvw27/dXsCcFY/W6eM84BVJXpbmMUDb0swif75Pn97PzUR6F/Ciqrp9nP0+QvPZ+bckWydZN8nhwAnAO9vP3HBOBI5uP38jWeV/F1V1K83t+s+1n4112hgPy5DHakmDwmRT0phU1c9pZg3fQzML9NF2N++Kn9toZplGup33/xj+NuSNSe6n2XxxFPCOqvrNDFJVnQ38Mc0u4mU0m0PWA/aqquXDxFlVdVlV/WJlr7U9z63A3sCuwO006+gOAv5glNvzF9PMTv5pT9mXhjwT8fy2/EGa3fZ3ttf2F8BBVfWDVYm99bEhYy5O8hSaDSj/OORvtxi4hJH/dgBU1c00u78/RLPM4VrgmzSJ+Uh9ej83Kxw3JLZlI3TvF8sdVbVwJfotp/m7rkvzWVoO/CXwx+1nbaR+S2j+R+adw9W3/3MxUf8ujgE+RvM/Xr+keazXa4DeZ6Wu8nsoTZaM/D9xkiRJ0qpxZlOSJEmdMdmUJElSZ0w2JUmS1BmTTUmSJHXGZFOSJEmdGe4hs5oGZs2aVbNnz57qMCRJkka1ePHiZVU17NcSm2xOU7Nnz2bRokVTHYYkSdKokvxwpDqTzWnq1z//BT//+JlTHYYkSRpQW775j6Y6BMA1m5IkSeqQyaYkSZI6Y7IpSZKkzphsSpIkqTMmm5IkSeqMyWZHktyeZNZUxyFJkjSVTDbHIYmPipIkSRoHk6ceSd4DHAH8CFgGLAYOAK4B9gIuTHIL8G5gHWA5cERV/SzJFsDngC2B64D0nPePgGPaPt8E3lJVj07WdUmSJE0VZzZbSeYCBwG7Aa8F5vZUb1pVL6qqDwMLgedX1W7A54Hj2jbvBRa25RcCT2vP+3vAPGCvqpoDPEqT0A4Xw9FJFiVZtPz+eyf8GiVJkiabM5u/tTfwH1X1IECSL/XUnd1z/FTg7CTb0MxU3taWv5AmSaWqLk5yd1u+H7A78K0kAOsBdw0XQFXNB+YDzHn69jUB1yRJkjSlTDZ/K33qHug5/ijwD1V1YZJ9gff11A2XIAY4o6r+epUjlCRJGjDeRv+thcArk6ybZEPgFSO02wT4SXv8hp7yK2lvjyd5ObBZW34ZcHCSrdq6zZM8faKDlyRJmo5MNltV9S2atZY3Al8EFgH3DNP0fcA5Sa6i2US0wvuBFya5HngZ8D/teb9Ds6HoK0luAr4KbNPRZUiSJE0rqXJp4ApJNqyq+5OsTzNTeXRVXT8Vscx5+vb11eP/31QMLUmSZoAt3/xHkzZWksVVNXe4OtdsPt78JM8G1qVZZzkliaYkSdJMYbLZo6peN9UxSJIkzSSu2ZQkSVJnTDYlSZLUGW+jT1Nrbbn5pC7slSRJ6oIzm5IkSeqMyaYkSZI6Y7IpSZKkzrhmc5p65Od3cOfH3z/VYUiSpsjWb37vVIcgTQhnNiVJktQZk01JkiR1xmRTkiRJnTHZlCRJUmdMNiVJktSZGZ1sJnlSkq8luSHJvCTvGkOf+9vf2yY5d5S2r0py/ETFK0mSNNPM9Ecf7QasXVVz4DeJ5N+NpWNV3QEcPEqbC4ELVzVISZKkmWrgZjaTbJDk4iQ3Jlnazljun+R7SRYmOTXJRUm2As4E5rQzm+cA67XHZ41hnNlJlrbH30yyU0/dgiS7JzkyycfastPbsa9J8oMkB7flayQ5LcnNbVz/uaJOkiRpphu4ZBPYH7ijqnatqp2BS4BPAq8E9gG2Bqiqu4CjgKuqak5VHQI82B4fMc4xPw8cCpBkG2Dbqlo8TLttgL2BA4AT27LXArOBXdp49hxpkCRHJ1mUZNHy+381zhAlSZKmn0FMNpcAL01yUpJ9gO2A26rq1qoqmtnMifYF4JD2+FDgnBHaXVBVj1XVd4Ant2V7A+e05XcCl480SFXNr6q5VTV3iw3Xn6jYJUmSpszAJZtVdQuwO03S+SHgVUB1POZPgOVJngPMo5npHM5DPccZ8luSJGm1M3DJZpJtgV9V1ZnAKcALgO2S7NA2ObxP90eSrL2SQ38eOA7YpKqWjKPfQuCgdu3mk4F9V3J8SZKkgTOIu9F3AU5O8hjwCPBmYBZwcZJlNMndziP0nQ/clOT6lVi3eS7wT8AHxtnvPGA/YClwC/BN4J5xnkOSJGkgpVnmOHMk2Rc4tqoOmOpYVkiyYVXdn2QL4Dpgr3b95oh2ffq2denxfzY5AUqSpp2t3/zeqQ5BGrMki6tq7nB1gzizOYguSrIpsA7wgdESTUmSpJlixiWbVbUAWNCvTTvDeNkwVftV1fIOYtp3os8pSZI0CGZcsjkWbUI5Z6rjkCRJmulWy2RzEKy95bau15EkSQNv4B59JEmSpMFhsilJkqTOmGxKkiSpMyabkiRJ6owbhKapB3/+fW4+7VVTHYY0Y+30lgunOgRJWi04sylJkqTOmGxKkiSpMyabkiRJ6ozJpiRJkjpjsjmBktw/Sv2mSd4yWfFIkiRNNZPNybUpYLIpSZJWGyabHUiyYZLLklyfZEmSV7dVJwI7JLkhyclTGaMkSdJk8Dmb3fhf4DVVdW+SWcA3klwIHA/sXFVzpjY8SZKkyWGy2Y0Af5fkhcBjwFOAJ4/aKTkaOBpgm83X6zRASZKkyeBt9G4cAWwJ7N7OYv4MWHe0TlU1v6rmVtXczTZcp+sYJUmSOmey2Y1NgLuq6pEkLwae3pbfB2w0dWFJkiRNLpPNbpwFzE2yiGaW83sAVbUcuDrJUjcISZKk1YFrNidQVW3Y/l4G7DlCm9dNalCSJElTyJlNSZIkdcZkU5IkSZ0x2ZQkSVJnTDYlSZLUGTcITVPrbfkMdnrLhVMdhiRJ0ipxZlOSJEmdMdmUJElSZ0w2JUmS1BmTTUmSJHXGDULT1H3LbuWyf33FVIehIfY76uKpDkGSpIHizKYkSZI6Y7IpSZKkzphsSpIkqTMmm5IkSeqMyeYESHL/VMcgSZI0HZlsSpIkqTMmmxMojZOTLE2yJMm8tnzfJAuSnJvke0nOSpKpjleSJKlrPmdzYr0WmAPsCswCvpXkyrZuN2An4A7gamAvYOFUBClJkjRZnNmcWHsDn6uqR6vqZ8AVwB5t3XVV9eOqegy4AZg9tHOSo5MsSrLol/c9PGlBS5IkdcVkc2L1uzX+UM/xowwzq1xV86tqblXN3XSjdSY8OEmSpMlmsjmxrgTmJVkzyZbAC4HrpjgmSZKkKeOazYl1PrAncCNQwHFVdWeSHac2LEmSpKlhsjkBqmrD9ncB72x/eusXAAt6Xr91EsOTJEmaMt5GlyRJUmdMNiVJktQZk01JkiR1xmRTkiRJnXGD0DS10axnst9RF091GJIkSavEmU1JkiR1xmRTkiRJnTHZlCRJUmdMNiVJktQZNwhNU79Ydiuf//c/mOowJtVhb7x0qkOQJEkTzJlNSZIkdcZkU5IkSZ0x2ZQkSVJnTDYlSZLUGZNNSZIkdcZkcxRJZidZOs4+r0pyfHv8viTHtsdHJtm2izglSZKmI5PNVpI1J+pcVXVhVZ04TNWRgMmmJElabaw2yWaSDyT5Pz2vP5jkmCSXJ/kssKRP97WSnJHkpiTnJlm/PcftSWa1x3OTLGiPj0zysSHjHwzMBc5KckOS9Sb4EiVJkqad1SbZBP4NeANAkjWAw4CfAM8DTqiqZ/fp+yxgflU9B7gXeMt4B6+qc4FFwBFVNaeqHhzaJsnRSRYlWXTf/Q+PdwhJkqRpZ7VJNqvqdmB5kt2AlwHfBpYD11XVbaN0/1FVXd0enwns3VGM86tqblXN3WjDdboYQpIkaVKtbl9X+a806ya3Bj7Vlj0whn41wutf89uEfd1VDU6SJGmmWW1mNlvnA/sDewDj+SLupyXZsz0+HFjYHt8O7N4eHzSG89wHbDSOcSVJkgbaapVsVtXDwOXAF6rq0XF0/S7whiQ3AZsDH2/L3w/8U5KrgLGc73TgX9wgJEmSVhepGnqHeOZqNwZdDxxSVbdOdTz9bD97k/q79z5/qsOYVIe9cTyTzZIkabpIsriq5g5Xt9rMbCZ5NvB94LLpnmhKkiTNFKvNBqGq+g6w/Uj1SbYALhumar+qWt5ZYJIkSTPYapNsjqZNKOdMdRySJEkzicnmNLX5rGe6hlGSJA281WbNpiRJkiafyaYkSZI6Y7IpSZKkzphsSpIkqTNuEJqm7vrFrXz0rD+Y6jBW2duOcJOTJEmrM2c2JUmS1BmTTUmSJHXGZFOSJEmdMdmUJElSZ1aLZDPJ25OsvxL9dkxyQ5JvJ9lhquKQJEkaVKtFsgm8HRhXkpdkTeBA4D+qareq+u+euiRZmfdu3HFIkiQNsoFKNpMcl+SY9vgjSb7eHu+X5MwkH0+yKMnNSd7f1h0DbAtcnuTytuxlSa5Ncn2Sc5Js2JbfnuRvkiwE5tEkh0cluTzJ7CTfTXIacD3wO0lOTrI0yZIk89pz7JtkQZJzk3wvyVltcvqEOCRJkma6gUo2gSuBfdrjucCGSdYG9gauAk6oqrnAc4AXJXlOVZ0K3AG8uKpenGQW8G7gpVX1XGAR8Jc9Y/xvVe1dVZ8F/gX4SFW9uK17FvDpqtqtHX8OsCvwUuDkJNu07XajSVSfDWwP7DU0jgl+XyRJkqalQUs2FwO7J9kIeAi4libp24cm2Tw0yfXAt4GdaJK9oZ7fll+d5AbgDcDTe+rP7jP+D6vqG+3x3sDnqurRqvoZcAWwR1t3XVX9uKoeA24AZo/l4pIc3c7MLrr/3ofH0kWSJGlaG6hvEKqqR5LcDrwRuAa4CXgxsAPwIHAssEdV3Z3kdGDdYU4T4KtVdfgIwzzQJ4TeuvRp91DP8aOM8X2uqvnAfICnbb9JjaWPJEnSdDZoM5vQ3Eo/tv19FfDnNLOHG9Mkg/ckeTLw8p4+9wEbtcffAPZK8gyAJOsn+d2VjGNekjWTbAm8ELhulD69cUiSJM14g5hsXgVsA1zb3r7+X+CqqrqR5vb5zcCngKt7+swHvpzk8qr6OXAk8LkkN9EknzuuRBzn08ys3gh8HTiuqu4cpc9v4liJ8SRJkgZOqrxbOx09bftN6p0feP5Uh7HK3nbEpVMdgiRJ6liSxe0m7ScYxJlNSZIkDQiTTUmSJHXGZFOSJEmdMdmUJElSZwbqOZurk602f6abayRJ0sBzZlOSJEmdMdmUJElSZ0w2JUmS1BmTTUmSJHXGDULT1I/vvpVjz91/qsNYJaccfMlUhyBJkqaYM5uSJEnqjMmmJEmSOmOyKUmSpM6YbEqSJKkzJpuSJEnqzEAlm0melORrSW5IMi/Ju8bQ5/7297ZJzh2l7auSHL8K8b01yfeTVJJZPeX7JrmnjfuGJH+zsmNIkiQNkkF79NFuwNpVNQd+k0j+3Vg6VtUdwMGjtLkQuHAV4rsauAhYMEzdVVV1wCqcW5IkaeBM+cxmkg2SXJzkxiRL2xnL/ZN8L8nCJKcmuSjJVsCZwJx2dvAcYL32+KwxjDM7ydL2+JtJduqpW5Bk9yRHJvlYW3Z6O/Y1SX6Q5OC2fI0kpyW5uY3rP1fUVdW3q+r2iX+XJEmSBtOUJ5vA/sAdVbVrVe0MXAJ8EnglsA+wNUBV3QUcRTNDOKeqDgEebI+PGOeYnwcOBUiyDbBtVS0ept02wN7AAcCJbdlrgdnALm08e45xzD3bhPrLvYluryRHJ1mUZNGv7n14zBcjSZI0XU2HZHMJ8NIkJyXZB9gOuK2qbq2qopnNnGhfAA5pjw8Fzhmh3QVV9VhVfQd4clu2N3BOW34ncPkYxrseeHpV7Qp8FLhguEZVNb+q5lbV3PU3Xmes1yJJkjRtTXmyWVW3ALvTJJ0fAl4FVMdj/gRYnuQ5wDyamc7hPNRznCG/xzPevVV1f3v8n8DavRuIJEmSZqopTzaTbAv8qqrOBE4BXgBsl2SHtsnhfbo/kmTtlRz688BxwCZVtWQc/RYCB7VrN58M7DtahyRbJ0l7/Dya9335+EOWJEkaLNNhN/ouwMlJHgMeAd4MzAIuTrKMJrnbeYS+84Gbkly/Eus2zwX+CfjAOPudB+wHLAVuAb4J3AOQ5BiaBHbrNq7/rKqjaHbBvznJr4EHgcPaJQKSJEkzWqZ7zpNkX+DY6fTYoCQbVtX9SbYArgP2atdvTpitd9ik/uikse49mp5OOfiSqQ5BkiRNgiSLq2rucHXTYWZzEF2UZFNgHeADE51oSpIkzRTTPtmsqgUM/5D032hnGC8bpmq/qprwtZFVte9En1OSJGkmmvbJ5li0CeWcqY5DkiRJjzcjks2Z6KmbPdM1j5IkaeBN+aOPJEmSNHOZbEqSJKkzJpuSJEnqjGs2p6lbf/k/vPyCt011GI/z5QM/OtUhSJKkAePMpiRJkjpjsilJkqTOmGxKkiSpMyabkiRJ6ozJpiRJkjpjsilJkqTOzLhkM8mTknwtyQ1J5iV51yqc6/6JjE2SJGl1MxOfs7kbsHZVzYHfJIx/N7UhSZIkrZ4GYmYzyQZJLk5yY5Kl7Yzl/km+l2RhklOTXJRkK+BMYE47s3kOsF57fFaf81+QZHGSm5McPaTuw0muT3JZki3bsjlJvpHkpiTnJ9ksye8lua6n3+wkN7XHuye5oh3j0iTbdPJGSZIkTTMDkWwC+wN3VNWuVbUzcAnwSeCVwD7A1gBVdRdwFHBVVc2pqkOAB9vjI/qc/01VtTswFzgmyRZt+QbA9VX1XOAK4L1t+aeBv6qq5wBLgPdW1XeBdZJs37aZB3whydrAR4GD2zE+BXxwuCCSHJ1kUZJFD9/74DjfIkmSpOlnUJLNJcBLk5yUZB9gO+C2qrq1qopmNnNVHJPkRuAbwO8Az2zLHwPObo/PBPZOsgmwaVVd0ZafAbywPf4CcGh7PK/t+yxgZ+CrSW4A3g08dbggqmp+Vc2tqrnrbLzeKl6SJEnS1BuINZtVdUuS3YE/BD4EfAWoiTh3kn2BlwJ7VtWvkiwA1h0plFFOdzZwTpIvNmHXrUl2AW6uqj0nIl5JkqRBMhAzm0m2BX5VVWcCpwAvALZLskPb5PA+3R9pb2WPZBPg7jbR3BF4fk/dGsDB7fHrgIVVdQ9wdzvDCvDHNLfYqar/Bh4F3sNvZ0T/C9gyyZ7ttaydZKdRL1qSJGkGGIiZTWAX4OQkjwGPAG8GZgEXJ1kGLKS5VT2c+cBNSa4fYd3mJcCft5t5/ovmVvoKDwA7JVkM3ENzaxzgDcC/JFkf+AHwxp4+ZwMn09zqp6oeTnIwcGp7C34t4B+Bm8fzBkiSJA2iNEseh6lI9gB+VFV3tq9fDxwE/BB4X1X9YtKiHEV7K/zYqjpgqmOZKJs8Y6t6wSnzRm84ib584EenOgRJkjQNJVlcVXOHq+t3G/0TwMPtCV4InEizC/semtlCSZIkqa9+t9HX7Jm9nAfMr6rzgPPaXdXTRlUtABb0a9M+zuiyYar2q6rlHYQlSZK02uubbCZZq6p+DewH9D7sfFDWev5Gm1DOmeo4JEmSVif9ksbPAVe0G3AeBK4CSPIMmlvp6tAzN32aayQlSdLAGzHZrKoPJrkM2Ab4Sv12J9EawNsmIzhJkiQNthE3CCV5SVV9o6rOB7ZaUV5VtwCzJyE2SZIkDbh+u9FP6Tk+b0jduzuIRZIkSTNMvzWbGeF4uNeaYLf+8me84osfnvDzXvza/zvh55QkSRpJv5nNGuF4uNeSJEnSE/Sb2dw+yYU0s5grjmlfb9d5ZJIkSRp4/ZLNV/ccnzKkbuhrSZIk6Qn6PfroiskMRJIkSTPPiMlmkpv6dayq50x8OJIkSZpJ+t1Gf4xmI9BngS/RfIuQJEmSNGYj7kavqjnA4cCGNAnnB4GdgJ9U1Q8nJ7ypk+RJSb6W5IYk85K8awx97h+lfnaS101clJIkSdNbv0cfUVXfq6r3VtVzaWY3Pw28Y1Iim3q7AWtX1ZyqOhsYNdkcg9mAyaYkSVpt9LuNTpKnAIcBrwHupkk0z5+EuDqRZAPgC8BTgTWBDwD3AP8ILAOuB7YH3gScCWyZ5AbgVmC99vjmqjpilHEC/D3wcpqlCH/bJqwnAr/XnueMqvrIkH5HA0cDrDtrswm5ZkmSpKnUb4PQFcBGNMnZkcAv2qp1kmxeVb8Yqe80tj9wR1W9AiDJJsBS4CXA94GzAarqriRHAcdW1QFt2/vbpQVj8VpgDrArMAv4VpIrgeN7zzlUVc0H5gNs8ozf8cH5kiRp4PW7jf50YDPgz4CvAIvan8Xt70G0BHhpkpOS7EPzcPrbqurWqiqa2cyJsDfwuap6tKp+BlwB7DFB55YkSRoY/Z6zOXsS45gUVXVLkt2BPwQ+RJNEdzGD6HfHS5IkMcoGoaGS7JDkhCRLuwqoS0m2BX5VVWfSfAvSC4DtkuzQNjm8T/dHkqw9xqGuBOYlWTPJlsALgeuA++xTp/QAACAASURBVGiWJkiSJK0WRk02k2yT5O1JrgNuppkN7ZeUTWe7ANe1G3ROAN5NsyHn4iQLgX6PdJoP3JTkrDGMcz5wE3Aj8HXguKq6sy37dZIbk6wuu/olSdJqLM1SxWEqkj+lSSqfSrNJ6AvAf1TVdpMX3uRKsi99NvBMpk2e8Tu199+/fcLPe/Fr/++En1OSJK3ekiyuqrnD1fV79NE/A9cCr6uqRe2J3CEtSZKkMeuXbG4LHAL8Q5In08xsjnXN4kCqqgXAgn5tkmwBXDZM1X5VtbyDsCRJkgZWv93oy4CPAx9P8lSah7vfleS7wPlVNRHfqDNw2oRyrM/blCRJWq2NuGZzxA7Js4DDqur93YQkgLlz59aiRYP6OFNJkrQ6Wak1m0le2Oecl69yVJIkSZrx+q3ZfOcwZUXzFYwrvltckiRJGlG/NZuv7H2dZG+aZ1P+FHhrx3FJkiRpBug3swlAkv2A99DMav5dVX2186jE9+9exgHnfWpCznXRQW+akPNIkiSNV781m6+gmcm8Bzihqq6etKgkSZI0I/Sb2fwS8GNgOfBXSR5XWVWv6jAuSZIkzQD9ks0XT1oUkiRJmpH6bRC6YjIDkSRJ0syzxlQHIEmSpJnLZFOSJEmdWalkM8moj0yaTpI8KcnXktyQZF6SUb/XPcn9o9TPTvK6ntdbJLk8yf1JPjak7YIk/9WOf0OSrVb+aiRJkgbHiMlmkoU9x58ZUn1dZxF1Yzdg7aqaU1VnA6Mmm2MwG3hdz+v/pXke6bEjtD+iHX9OVd01AeNLkiRNe/1mNjfoOd5pSF2YYkk2SHJxkhuTLG1nLPdP8r0kC5OcmuSidhbxTGBOO6t4DrBee3zWGMZJkpPbMZYkmddWnQjs057nHVX1QFUtpEk6V/aajk6yKMmih+/tO7EqSZI0EPrdDq+VrJss+wN3VNUrAJJsAiwFXgJ8HzgboKruSnIUcGxVHdC2vb+q5oxxnNcCc2i+E34W8K0kVwLH955zDP49yaPAecDfVtUT3sOqmg/MB9h0h9nT4T2WJElaJf2SzU2TvIZm9nPTJK9tywNs0nlko1sCnJLkJOAi4D7gtqq6FSDJmcDREzDO3sDnqupR4GdJrgD2AO4dxzmOqKqfJNmIJtn8Y+DTExCbJEnStNYv2bwCeFXP8St76q7sLKIxqqpbkuwO/CHwIeArdDPjuspLBqrqJ+3v+5J8FngeJpuSJGk10O+h7m+czEDGK8m2wC+q6sx25/ifA9sl2aGq/hs4vE/3R5KsXVWPjGGoK4E/S3IGsDnwQuCdwFOAjcYQ51rAplW1LMnawAHA18YwriRJ0sDr+wijJGsCm1XVsvb1OsCRwDuq6ve6D6+vXYCTkzwGPAK8mWZN5cVJlgELgZ1H6DsfuCnJ9VV1xCjjnA/sCdxIM3N6XFXdmWQ58OskNwKnV9VHktwObAysk+RA4GXAD4FL20RzTZpE85MrfdWSJEkDZMRkM8lhwCeAB5LcCrwP+AzwLWC0BK1zVXUpcOkwVTsCJNmXNtmsqgXAgp6+fwX81Sjn37D9XTQzme8cUv8IsN+QstkjnG73fmNJkiTNVP1mNt8N7F5V30/yXOBa4LCqOn9yQpMkSdKg65dsPlxV3weoquuT3DZIiebQ2czhJNkCuGyYqv2qankHYUmSJK1W+iWbWyX5y57XG/a+rqp/6C6sydEmlGN93qYkSZLGqV+y+Ukev9t66Gt16BmbzeKig9401WFIkiStkn6PPnr/ZAYiSZKkmaffbvRThxQVsAy4vP0OcEmSJKmvfrfRFw9TtjnNsy3Prqp/7CgmSZIkzRBpHiM5jg7JesA1VbVbNyEJYNMddqi9Tzxplc5x0SEHT1A0kiRJI0uyuKrmDle3xnhPVlUPrnpIkiRJWh30/brKodrv+f5j4MfdhCNJkqSZpN8GoftoNgX1ehC4AvizLoOSJEnSzNBvZnPnqvrhpEUiSZKkGaffms2B+WpKSZIkTU/9ks1MWhQDJsm2Sc6d6jgkSZKmu3630Z8yzIPdf6OqjukgnoFQVXcAPldIkiRpFP2SzQcZ/sHuq5UkJwE/rKrT2tfvA+4D3lhVOydZEzgR2Bd4EvDPVfWJJKcBl1TVhUnOB+6uqjcl+RNgu6p691RcjyRJ0mTql2wur6ozJi2S6evzwD8Cp7WvDwX+HHhj+/pPgHuqao8kTwKuTvIV4EpgH+BC4CnANm37vdtzPkGSo4GjAdabNWvir0SSJGmS9Vuz+fBwhUn2SvLPHcUz7VTVt4Gt2nWauwJ3A//T0+RlwOuT3AB8E9gCeCZwFbBPkmcD3wF+lmQbYE/gmhHGml9Vc6tq7jobb9zdRUmSJE2SEWc2q+r5K46TzAFeRzOrdxvwxe5Dm1bOpVmjuTVPnJUM8LaqunRopySbAfvTzHJuTvP+3V9V93UbriRJ0vTQ76HuvwscBhwOLAfOpvku9RdPUmzTyeeBTwKzgBfRrM1c4VLgzUm+XlWPtO/bT6rqAeBa4O3AS2hmPM9tfyRJklYL/W6jfw/YD3hlVe1dVR8FHp2csKaXqroZ2IgmifzpkOp/pblNfn2SpcAn+G0SfxWwVlV9H7ieZnbzqsmJWpIkaer12yB0EM3M5uVJLqGZ3Vttn71ZVbv0HN8O7NwePwa8q/0Z2uffgH9rjx8BNpiMWCVJkqaLEWc2q+r8qpoH7AgsAN4BPDnJx5O8bJLikyRJ0gDrdxsdgKp6oKrOqqoDgKcCNwDHdx6ZJEmSBt6IyWaSl/QcbwdQVb+oqk8Aq82jjyRJkrTyUlXDVyTXV9Vzhx4P91oTb+7cubVo0aKpDkOSJGlUSRZX1dzh6vrdRs8Ix8O9liRJkp6gX7JZIxwP91qSJEl6gn6PPto+yYU0s5grjmlfb9d5ZJIkSRp4/ZLNV/ccnzKkbuhrSZIk6Qn6JZu3VdX/TFokepzv330vB577tXH1ueDgl3YUjSRJ0srpt2bzghUHSc6bhFgkSZI0w4x1N/r2XQciSZKkmWdld6NLkiRJo+q3ZnPXJPfSzHCu1x7Tvq6q2rjz6CRJkjTQRpzZrKo1q2rjqtqoqtZqj1e8nvREM8nsJEvH0f7IJNv2vL49yaxuopMkSdJw+t1GH3RHAtuO1qhXkn4zvZIkSRqnQUs210pyRpKbkpybZP0kf5PkW0mWJpmfxsHAXOCsJDckWa/t/7Yk1ydZkmRHgCTva/t9Bfh0kqcnuawd47IkT2vbjVR+epKPJ7k8yQ+SvCjJp5J8N8npbZs123ZL27HfMenvnCRJ0hQYtGTzWcD8qnoOcC/wFuBjVbVHVe0MrAccUFXnAouAI6pqTlU92PZfVlXPBT4OHNtz3t2BV1fV64CPAZ9uxzgLOLVtM1I5wGbAS4B3AF8CPgLsBOySZA4wB3hKVe1cVbsA/z6B74kkSdK0NWjJ5o+q6ur2+Exgb+DFSb6ZZAlNwrdTn/5fbH8vBmb3lF/Yk5DuCXy2Pf5MO0a/coAvVVUBS4CfVdWSqnoMuLkd5wc0X/n50ST70yTKT5Dk6CSLkix6+N57+lyGJEnSYBi0ZHPoI5gKOA04uJ0x/CSwbp/+D7W/H+XxO/EfGMeYw5WvOO9jPccrXq9VVXcDuwILgL8A/nXYE1bNr6q5VTV3nY036ROSJEnSYBi0ZPNpSfZsjw8HFrbHy5JsCBzc0/Y+YKOVGOMa4LD2+IieMUYqH1W7C36NqjoPeA/w3JWIS5IkaeAM2u7r7wJvSPIJ4FaatZeb0dy+vh34Vk/b04F/SfIgzS3wsToG+FSSdwI/B944SvlYPAX49yQrkvu/HkdfSZKkgZVmqaGmm013+N3a96TTxtXngoNf2lE0kiRJI0uyuKrmDlc3aLfRJUmSNEBMNiVJktQZk01JkiR1xmRTkiRJnRm03eirjWdstrEbfiRJ0sBzZlOSJEmdMdmUJElSZ0w2JUmS1BmTTUmSJHXGDULT1A/ufpBDzrtp1HbnHPScSYhGkiRp5TizKUmSpM6YbEqSJKkzJpuSJEnqjMmmJEmSOmOyKUmSpM5Mi2QzyZOSfC3JDUnmJXnXKO1nJ1k6WfFNlDbu1011HJIkSZNlWiSbwG7A2lU1p6rOBvommwNsNmCyKUmSVhudJZtJNkhycZIbkyxtZyz3T/K9JAuTnJrkoiRbAWcCc9qZzXOA9drjs/oMsVaSM5LclOTcJOu34/5Nkm+1Y85Pkrb8mCTfadt/vifGT7Xtv53k1W35kUkuSPKlJLcleWuSv2zbfCPJ5m27HZJckmRxkquS7NiWn95e3zVJfpDk4DbmE4F92mt7RxfvuyRJ0nTS5czm/sAdVbVrVe0MXAJ8EnglsA+wNUBV3QUcBVzVzmweAjzYHh/R5/zPAuZX1XOAe4G3tOUfq6o92jHXAw5oy48Hdmvb/3lbdgLw9araA3gxcHKSDdq6nWlmIZ8HfBD4VVXtBlwLvL5tMx94W1XtDhwLnNYT3zbA3u34J/bEsOI6PzL0gpIcnWRRkkUP3Xt3n0uXJEkaDF0mm0uAlyY5Kck+wHbAbVV1a1UVzWzmqvhRVV3dHp9Jk9gBvDjJN5MsAV4C7NSW3wScleSPgF+3ZS8Djk9yA7AAWBd4Wlt3eVXdV1U/B+4BvtRzXbOTbAi8ADin7f8JmgRzhQuq6rGq+g7w5LFcUFXNr6q5VTX3SRtvNsa3QZIkafrq7Osqq+qWJLsDfwh8CPgKUBM5xNDXSdalmV2cW1U/SvI+mgQS4BXAC4FXAe9JshMQ4KCq+q/eEyX5/4CHeooe63n9GM37tgbwy6qaM0J8vf0znguTJEmaKbpcs7ktza3nM4FTaGYBt0uyQ9vk8D7dH0my9ihDPC3Jnj3nWshvE8tl7czjwW0sawC/U1WXA8cBmwIbApcCb+tZ17nbWK+vqu4FbktySNs3SXYdpdt9wEZjHUOSJGnQdXkbfRfguvYW8wnAu4GjgYuTLAR+2KfvfOCmUTYIfRd4Q5KbgM2Bj1fVL2nWhS4BLgC+1bZdEzizvbX+beAjbdsPAGu3Yy1tX4/HEcCfJLkRuBl49SjtbwJ+3W6acoOQJEma8dIsn5yCgZN9gWOr6oDR2q6ONt9hp9rv7z83artzDnrOJEQjSZI0siSLq2rucHXT5TmbkiRJmoE62yA0mqpaQLMDfERJtgAuG6Zqv6pa3kFYkiRJmkBTlmyORZtQjrTbW5IkSdPctE42V2fbb7ae6zElSdLAc82mJEmSOmOyKUmSpM6YbEqSJKkzJpuSJEnqjBuEpqmf/vIRPnj+T/u2OeE120xSNJIkSSvHmU1JkiR1xmRTkiRJnTHZlCRJUmdMNiVJktQZk01JkiR1Ztonm0nel+TYPvUHJnn2ZMY0mqExJVmQZO5UxiRJkjQVpn2yOQYHAtMq2WR6xiRJkjTppmWymeSEJP+V5GvAs9qyP03yrSQ3JjkvyfpJXgC8Cjg5yQ1Jdmh/LkmyOMlVSXbsM87pST6e5PIkP0jyoiSfSvLdJKf3tDs8yZIkS5Oc1FN+f5IPtjF9I8mTh4upbX5IkuuS3JJknw7eNkmSpGln2iWbSXYHDgN2A14L7NFWfbGq9qiqXYHvAn9SVdcAFwLvrKo5VfXfwHzgbVW1O3AscNooQ24GvAR4B/Al4CPATsAuSeYk2RY4qW0zB9gjyYFt3w2Ab7QxXQn86QgxAaxVVc8D3g68d4RrPzrJoiSLHrh3+RjfMUmSpOlrOn6D0D7A+VX1K4AkF7blOyf5W2BTYEPg0qEdk2wIvAA4J8mK4ieNMt6XqqqSLAF+VlVL2nPdDMwGng4sqKqft+VnAS8ELgAeBi5qz7MY+P0+43yxp93s4RpU1XyaZJmnPGPXGiVuSZKkaW86JpsAwyVapwMHVtWNSY4E9h2mzRrAL6tqzjjGeqj9/VjP8YrXawG/7tP3kapaEeuj9H8/HxpjO0mSpBlj2t1Gp7kd/Zok6yXZCHhlW74R8NMkawNH9LS/r62jqu4FbktyCEAau65iPN8EXpRkVpI1gcOBK0bp85uYJEmSVmfTLtmsquuBs4EbgPOAq9qq99Akfl8FvtfT5fPAO5N8u92McwTwJ0luBG4GXr2K8fwU+GvgcuBG4Pqq+o9Rug2NSZIkabWU394F1nTylGfsWm85+ZK+bU54zTaTFI0kSdLIkiyuqmGfKT7tZjYlSZI0c6wWG1WSnAAcMqT4nKr64FTEI0mStLpYLZLNNqk0sZQkSZpkq0WyOYi22XRt12RKkqSB55pNSZIkdcZkU5IkSZ0x2ZQkSVJnTDYlSZLUGTcITVO/vPvXXHDOsseVHXjIrCmKRpIkaeU4sylJkqTOmGxKkiSpMyabkiRJ6ozJpiRJkjpjsjmCJKcnObg9/tckz26P3zWk3TVTEZ8kSdIgMNkcg6o6qqq+075815C6F0xBSJIkSQNhxiSbSV6f5KYkNyb5TJKnJ7msLbssydPadqcnOTXJNUl+0DN7mSQfS/KdJBcDW/Wce0GSuUlOBNZLckOSs9q6+3v6n5xkaZIlSea15fu2/c9N8r0kZyXJZL8/kiRJU2FGPGczyU7ACcBeVbUsyebAGcCnq+qMJG8CTgUObLtsA+wN7AhcCJwLvAZ4FrAL8GTgO8CnesepquOTvLWq5gwTxmuBOcCuwCzgW0mubOt2A3YC7gCuBvYCFk7EtUuSJE1nM2Vm8yXAuVW1DKCqfgHsCXy2rf8MTXK5wgVV9Vh7a/zJbdkLgc9V1aNVdQfw9XHGsHdP/58BVwB7tHXXVdWPq+ox4AZg9nAnSHJ0kkVJFt177/JxDi9JkjT9zJRkM0CN0qa3/qEhfYdrszIxjKR3vEcZYUa5quZX1dyqmrvxxlusQiiSJEnTw0xJNi8DDk2yBUB7G/0a4LC2/ghGv219JXBYkjWTbAO8eIR2jyRZe4T+89r+W9LMlF43zuuQJEmaUWbEms2qujnJB4ErkjwKfBs4BvhUkncCPwfeOMppzqe5Hb8EuIXmNvhw5gM3Jbm+qo4Y0n9P4EaaGdLjqurOJDuu7HVJkiQNulStyp1jdeUZO8ypU0782uPKDjxk1hRFI0mSNLIki6tq7nB1M+U2uiRJkqYhk01JkiR1xmRTkiRJnTHZlCRJUmdmxG70mWjTzdZyQ5AkSRp4zmxKkiSpMyabkiRJ6ozJpiRJkjpjsilJkqTOuEFomnpg2a+57t/velzZ89641RRFI0mStHKc2ZQkSVJnTDYlSZLUGZNNSZIkdcZkU5IkSZ0Z2GQzyewkSye7ryRJksZuYJPNLiRxd74kSdIEGvRkc60kZyS5Kcm5SdZPsnuSK5IsTnJpkm0A2vIbk1wL/MWKEyQ5Mv9/e/ceZVlZ3nn8+xvoDg3d0FwEjAgdUUEg2NoFioEI6ph45SJKk9YAccLKGiNK0mQmapLORCbJ4FIjxGU6hEsbDASBSDLKRUVAuUi1NHcIEBhvCQrKvYOAz/xxdsHhUJdTdfpU1an+ftY6q/d+9/u++91PvVXr6X05Ozk3yT8DlyTZJsk/NX1ek2Tvpt5Y5auaMVyS5N4khyX5P0luSnJRknlNvb9IcmvT/hPTHypJkqTpN+jJ5m7A6qraG3iYVhJ5MnB4VS0DTgNObOqeDhxXVfuN0s9+wFFV9QbgT4Hrmz4/Aqxp6oxVDrAr8DbgYODvgcuq6peB9cDbkmwDHArs2bT/+AY5ekmSpFlu0JPN71XVt5rlvwd+DdgLuDTJOuBjwE5JtgIWV9XlTd3Pd/RzaVX9pFnef2R7VX0d2LZpP1Y5wFeq6kngJmAT4KKm/CZgCa1E+D+BU5McBjw+2sEkOTbJcJLhBx99YPLRkCRJmmUGPdmsjvVHgFuqamnz+eWqejOQUeq2e6xtOWPsZ6xygCcAqurnwJNVNVL+c2DTqnoK2Bc4DziEZ5PR53ZWtbqqhqpqaPHCbccZriRJ0mAY9GRz5yQjl8WPBK4BXjBSlmRekj2r6kHgoST7N3VXjNPnFSPbkxwI3F9VD49TPqEkC4GtqurLwIeBpd0foiRJ0uAa9KevbwOOSvI3wJ207te8GPhMc4l7U+DTwC3AMcBpSR5v6oxlFXB6khtpXe4+aoLybiwCvpRkM1pnSI+fRFtJkqSBlWev+Go2ecWSpXXmn1zynLJ9j9l+hkYjSZI0tiRrq2potG2DfhldkiRJs5jJpiRJkvrGZFOSJEl9Y7IpSZKkvhn0p9HnrC2229QHgiRJ0sDzzKYkSZL6xmRTkiRJfWOyKUmSpL7xns1Z6mf3Pcn3P/Efz6zvtHLHGRyNJEnS1HhmU5IkSX1jsilJkqS+MdmUJElS35hsSpIkqW9MNiVJktQ3JpuSJEnqG5PNKUpydJJfnGSbJUlu7teYJEmSZptZmWwm2aSLOkkyk+M/GphUsilJkrSxmZFkLcmfJflQ2/qJSY5LclmSLwA3jdFuSZLbknwW+A7w4iQnJLkuyY1J/rSj3t8muSXJJUkWNNuWJrmmqX9Bkq2TvCLJtzv2c2Oz/MdN/zcnWd0kuYcDQ8BZSdYlWZBkWZLLk6xNcnGSFzbtlyW5IcnVwAf6E1FJkqTZaabODP4dcBRAc3ZyOfADYF/go1W1xzhtdwPWVNWrmuWXNe2WAsuS/GpT72XAX1fVnsCDwLua8jXA/6iqvWkltX9SVbcB85O8pKlzBPCPzfIpVbVPVe0FLADeXlVfBIaBFVW1FHgKOBk4vKqWAacBJzbtTweOq6r9JgpKkmOTDCcZ/smjD0xUXZIkadabkWSzqu4FHkjyKuDNwPXAA8C3q+qeCZr/v6q6pll+c1v77wC700oyAe6pqnXN8lpgSZKtgMVVdXlTfiYwkpz+I/CeZvkI4Jxm+aAk1ya5CXgDsOcoY9oN2Au4NMk64GPATqPs7/PjHVhVra6qoaoa2mbhtuNHQZIkaQDM5LvRT6V13+OOtM4EAjzWRbv2OgH+vKr+pr1CkiXAE21FT9M6Kzmec4Bzk5wPVFXdmWQz4LPAUFV9L8kqYLNR2ga4pfPsZZLFQE14RJIkSXPUTD5gcwHw68A+wMVT7ONi4LeSLARI8qIk249VuaoeAn6a5ICm6H3A5c22u2klpX/Es2c1RxLL+5t9HN7W3SPAomb5DuAFSfZrxjEvyZ5V9SDwUJL9m3orpnickiRJA2nGzmxW1c+SXAY8WFVPJ5lKH5ckeQVwddP+UeC9tJLGsRwFfC7J5sC/Ace0bTsHOAn4pab/B5P8La17O+8Frmure0bTz3pgP1qJ6GeaS+ebAp8Gbmn6Py3J40w9qZYkSRpIqZqZq7zNg0HfAd5dVXfOyCBmsb1f/Mr68oeezU13WrnjDI5GkiRpbEnWVtXQaNtm6quP9gDuAr5moilJkjR3zchl9Kq6FXjJWNuTbAt8bZRNb6wqvxNIkiRpQMzk0+hjahLKpTM9DkmSJPVmViabgvk7zPM+TUmSNPBm5bvRJUmSNDeYbEqSJKlvTDYlSZLUN96zOUs9ed/j3Pepdc+s73C8z0tJkqTB45lNSZIk9Y3JpiRJkvrGZFOSJEl9Y7IpSZKkvjHZlCRJUt+YbHZIclyS25L8IMkpMz0eSZKkQeZXHz3ffwfeArweGOq1sySbVtVTPY9KkiRpAHlms02SzwEvAS4Etm4r3yXJ15Lc2Py78wTlZyT5ZJLLgL9M8vok65rP9UkWzcTxSZIkTTeTzTZV9TvAD4GDgJ+2bToFWFNVewNnAZ+ZoBzg5cCbqur3gZXAB6pqKXAAsH60/Sc5NslwkuGfPPbgBjwySZKkmWGy2Z39gC80y58H9p+gHODcqnq6Wf4W8MkkxwGLx7qsXlWrq2qoqoa22WLxBj0ASZKkmWCyOTXVRfljzxRW/QXw34AFwDVJdu/j2CRJkmYNk83uXAUsb5ZXAN+coPw5kuxaVTdV1V8Cw4DJpiRJ2ij4NHp3jgNOS3IC8GPgmAnKO304yUHA08CtwFf6PF5JkqRZwWSzQ1UtaRbPaD5U1b3AG0apO1b50R3rH9yQY5QkSRoUXkaXJElS35hsSpIkqW9MNiVJktQ3JpuSJEnqGx8QmqXm7bA5Oxy/dKaHIUmS1BPPbEqSJKlvTDYlSZLUNyabkiRJ6huTzVnqqR89zI9OvnSmhyFJktQTk01JkiT1jcmmJEmS+sZkU5IkSX1jsilJkqS+MdmUJElS35hsSpIkqW/mTLKZZFWSleNsPyTJHtM5JkmSpI3dnEk2u3AIMGuSzSS+l16SJM15A51sJvlokjuSfBXYrSn77STXJbkhyXlJNk/yOuCdwElJ1iXZtflclGRtkiuT7D7GPhYluSfJvGZ9yyT3Jpk3Vh9J3pHk2iTXJ/lqkh2a8lVJVie5BFgzyr6OTTKcZPiBRx/qT9AkSZKm0cAmm0mWAcuBVwGHAfs0m86vqn2q6pXAbcD7q+oq4ELghKpaWlV3A6uBD1bVMmAl8NnR9lNVjwDfAN7WFC0HzquqJ8fp45vAa6vqVcDZwB+0dbkMOLiqfmOUfa2uqqGqGtp24VaTD4okSdIsM8iXcg8ALqiqxwGSXNiU75Xk48BiYCFwcWfDJAuB1wHnJhkp/oVx9nUqrYTxn4BjgN+eoI+dgHOSvBCYD9zT1teFVbV+EscpSZI0sAY52QSoUcrOAA6pqhuSHA0cOEqd/wI8WFVLu9pJ1beSLEnyemCTqro5yZbj9HEy8MmqujDJgcCqtm2PdbNPSZKkuWBgL6MDVwCHJlmQZBHwjqZ8EfDvzT2WK9rqP9Jso6oeBu5J8m6AtLxygv2tAf4BOL2LPrYCftAsH9XDMUqSJA20gU02q+o7wDnAOuA84Mpm0x8B1wKXAre3NTkbOKF5aGdXWono+5PcANwCHDzBLs8CtqaVcI4Yq49VtC6vXwncP6UDlCRJmgNSNdqVaHVKcjitB3veNx37W7rzy+uSE/6a7T/4X6djb+Ak3gAACKNJREFUd5IkSVOWZG1VDY22bdDv2ZwWSU4G3gK8dabHIkmSNEhMNtsk+Sjw7o7ic6vqgzMxHkmSpEFnstmmqk4ETpzpcUiSJM0VA/uA0Fy36fZber+mJEkaeCabkiRJ6hufRp+lkjwC3DHT4xgw2+FXTU2WMZs8YzZ5xmxqjNvkGbPJ21Ax26WqXjDaBu/ZnL3uGOsrBDS6JMPGbHKM2eQZs8kzZlNj3CbPmE3edMTMy+iSJEnqG5NNSZIk9Y3J5uy1eqYHMICM2eQZs8kzZpNnzKbGuE2eMZu8vsfMB4QkSZLUN57ZlCRJUt+YbE6zJL+e5I4kdyX5n6NsT5LPNNtvTPLqbtvOZT3G7d4kNyVZl2R4ekc+c7qI2e5Jrk7yRJKVk2k7V/UYM+fZ6DFb0fxO3pjkqiSv7LbtXNVjzJxno8fs4CZe65IMJ9m/27ZzVY8x27DzrKr8TNMH2AS4G3gJMB+4Adijo85bga8AAV4LXNtt27n66SVuzbZ7ge1m+jhmYcy2B/ah9YrWlZNpOxc/vcTMeTZuzF4HbN0sv2Vj/5vWS8ycZ+PGbCHP3hq4N3C782xqMevHPPPM5vTaF7irqv6tqn4GnA0c3FHnYGBNtVwDLE7ywi7bzlW9xG1jNWHMqupHVXUd8ORk285RvcRsY9VNzK6qqp82q9cAO3Xbdo7qJWYbq25i9mg1WRKwBVDdtp2jeonZBmeyOb1eBHyvbf37TVk3dbppO1f1Ejdo/QJdkmRtkmP7NsrZpZf5srHOtV6P23k2cczeT+sKxFTazhW9xAycZzBGzJIcmuR24P8CvzWZtnNQLzGDDTzPfIPQ9MooZZ3/kxirTjdt56pe4gbwK1X1wyTbA5cmub2qrtigI5x9epkvG+tc6/W4nWcto8YsyUG0EqeR+8KcZ8/qNmbgPBvxvJhV1QXABUl+Ffgz4E3dtp2DeokZbOB55pnN6fV94MVt6zsBP+yyTjdt56pe4kZVjfz7I+ACWpcX5rpe5svGOtd6Om7nGTBGzJLsDZwKHFxVD0ym7RzUS8ycZy3jzpUmKdo1yXaTbTuH9BKzDT7PTDan13XAy5L8UpL5wHLgwo46FwK/2Txd/Vrgoar69y7bzlVTjluSLZIsAkiyBfBm4ObpHPwM6WW+bKxzbcrH7TwbO2ZJdgbOB95XVf86mbZz1JRj5jwbN2YvTZJm+dW0Hop5oJu2c9SUY9aPeeZl9GlUVU8l+V3gYlpPip1WVbck+Z1m++eAL9N6svou4HHgmPHazsBhTLte4gbsQOsSAbTm+xeq6qJpPoRp103MkuwIDANbAj9P8mFaTys+vDHOtV5iBmyH82ys380/BrYFPtvE56mqGtpY/6b1EjP8ezZezN5F64TDk8B64Ijm4Rfn2SRjlmSDzzPfICRJkqS+8TK6JEmS+sZkU5IkSX1jsilJkqS+MdmUJElS35hsSpIkqW9MNiWpQ5Knk6xr+yxJcmCShzrK39TW5tAklWT3Zv3aps53k/y4o69HO/Z3dJJTmuVVSX7Q1L01yZFt9c5Ick9bX1eNMvYDk/xLW7+V5I2jjPPwZv0bSe5IckOSbyXZrSmfn+TTSe5OcmeSLyXZqa2fkRjdnOSfkyye4Jg3TXJ/kj/vGO83kgy3rQ8l+Ubb+r5JrmjGeHuSU5Ns3hxb+z7WJdlj0j9sSX3n92xK0vOtr6ql7QVJlgBXVtXbx2hzJPBNWl+evKqqXtO0OxoYqqrfbetrov1/qqo+keRlwNokX6yqJ5ttJ1TVFydxLDc1Y/tas74cuKGjzoqqGk7rHcgnAe8E/jewCHh5VT2d5Bjg/CSvab6/8JkYJTkT+MAEx/xW4A7gPUk+Us/93r3tk7ylqtrfAU7zfX/nAsur6uq0AveuZlwA57TvQ9Ls5JlNSepRkoXAr9B6j/XyDdVvVd1J6yUFW/fQzZXAvknmNeN8KbBujLpXAC9NsjmtFyMcX1VPN2M5HXgCeMMo7a4GXjTBOI4E/gr4LvDajm0nAR8bpc0HgDOr6upmDFVVX6yq+ybYl6RZxGRTkp5vQdul2Qvayg/ouGy7a1N+CHBR82rBn6T16reeNf3c2byfeMRJbfs/q4tuCvgq8GvAwYz/qr530DoT+lLgu1X1cMf2YWDPjjFuArxxvH6TLGjq/AvwD7QSz3ZXA08kOaijfC9g7TjjPaLj57FgnLqSZojJpiQ93/qqWtp8Dm0rv7KtfGlV3d2UHwmc3SyfzfOTqW60X1Y+PskdwLXAqo56J7Ttf0WXfZ9N64zrclrJXqezkqyjdXZ2JZCO8YxoL1/QtHkA2Aa4dJz9vx24rKoeB84DDm2S1HYfZ/Szm+M5p+PnsX6S7SVNA5NNSepBkm1pXVo+Ncm9wAm0zriNd2Pm+iTz29a3Ae5vW/9UVe0GHAGsSbJZL2Osqm/TOku4XXP2tdOKJlk7pKq+B9wF7JJkUUe9VwO3jhxDc8/mLsB8Wpe8x3Ik8KYmPmtpvff7OWcxq+rrwGY89xL7LcCyLg5R0ixmsilJvTkcWFNVu1TVkqp6MXAPsP84bS4H3gvPXGJ+D3BZZ6WqOp/WpeujNsA4/xD4SDcVq+ox4EzgkyNnIJP8JrA58PWOug8BxwErk8zr7CvJlrRisXMTnyW0EtPRzv6eCPxB2/opwFFJXtPW33uT7NjNcUiaHUw2Jal7nfdsHk4rabqgo955wG+M08+HgMOay9DXAOdW1RVj1P1fwO8lGfl7fVLHGOaP0e45quorVfW8hHYcfwj8J/CvSe4E3g0c2vEU+Ujf19N6wn20h6MOA75eVU+0lX0JeGeSX+jo58vAj9vW72v6/ETz1Ue3AQcAI/eSdt6z+bpJHJ+kaZJR/m5IkiRJG4RnNiVJktQ3JpuSJEnqG5NNSZIk9Y3JpiRJkvrGZFOSJEl9Y7IpSZKkvjHZlCRJUt+YbEqSJKlv/j875jVEGmvTxQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "def model_function_random_forest(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100)\n",
    "    reg = RandomForestRegressor()\n",
    "    reg.fit(X_train, y_train)\n",
    "    predictions = reg.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    plot_feature_importance(reg.feature_importances_,X_train.columns,'RANDOM FOREST')\n",
    "    return reg.score(X_test, y_test), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "    \n",
    "scores_random_forest = model_function_random_forest(X, y)\n",
    "scores_random_forest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### b) MLPRegressor "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9791     4.191992e+06\n",
       "15192    3.048654e+06\n",
       "16782    2.200909e+06\n",
       "2827     2.186043e+06\n",
       "7624     1.988783e+06\n",
       "Name: resid, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.5445703156592312,\n",
       " 166290.21525634246,\n",
       " 258781.16093573446,\n",
       "           true           pred          resid\n",
       " 15526   395000  730309.748038  335309.748038\n",
       " 15633   456500  310617.390071  145882.609929\n",
       " 10691   387000  372942.588912   14057.411088\n",
       " 3124    260000  224633.216931   35366.783069\n",
       " 11895   585000  468908.108213  116091.891787\n",
       " 13362   919204  881528.305896   37675.694104\n",
       " 15892  1700000  967959.579060  732040.420940\n",
       " 4980    102500  152152.516411   49652.516411\n",
       " 15441   569500  505995.545787   63504.454213\n",
       " 16189   380000  404855.370774   24855.370774)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neural_network import MLPRegressor\n",
    "\n",
    "def model_function_MPL(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100)\n",
    "    mpl = MLPRegressor()\n",
    "    mpl.fit(X_train, y_train)\n",
    "    predictions = mpl.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    x = results['resid'].sort_values(ascending=False )\n",
    "    display(x.head())\n",
    "    return mpl.score(X_test, y_test), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "    \n",
    "scores_MPL = model_function_MPL(X, y)\n",
    "scores_MPL"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### c) XGBOOST REGRESSOR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17205    2226311.0\n",
       "4051     1802911.0\n",
       "9791     1770026.0\n",
       "8874     1283481.0\n",
       "2827     1099126.5\n",
       "Name: resid, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(0.8933149352553882,\n",
       " 68734.33056470887,\n",
       " 125248.86598685541,\n",
       "           true          pred         resid\n",
       " 15526   395000  3.188074e+05   76192.56250\n",
       " 15633   456500  4.829555e+05   26455.46875\n",
       " 10691   387000  3.903453e+05    3345.28125\n",
       " 3124    260000  2.743578e+05   14357.78125\n",
       " 11895   585000  6.772319e+05   92231.93750\n",
       " 13362   919204  1.002423e+06   83219.18750\n",
       " 15892  1700000  1.534657e+06  165343.12500\n",
       " 4980    102500  1.768425e+05   74342.53125\n",
       " 15441   569500  4.839170e+05   85583.03125\n",
       " 16189   380000  3.779468e+05    2053.15625)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from xgboost import XGBRegressor\n",
    "\n",
    "def model_function_xgb(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100)\n",
    "    xgb = XGBRegressor()\n",
    "    xgb.fit(X_train, y_train)\n",
    "    predictions = xgb.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    x = results['resid'].sort_values(ascending=False )\n",
    "    display(x.head())\n",
    "    return xgb.score(X_test, y_test), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "    \n",
    "scores_xgb = model_function_xgb(X, y)\n",
    "scores_xgb"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. SCALE DATA AND TEST MODEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' from sklearn.preprocessing import Normalizer\\ndef normalizer_transform(data): \\n    transformer = Normalizer().fit(data)\\n    nums_norm = pd.DataFrame(transformer.transform(data), columns=data.columns)\\n    return nums_norm '"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' from sklearn.preprocessing import Normalizer\n",
    "def normalizer_transform(data): \n",
    "    transformer = Normalizer().fit(data)\n",
    "    nums_norm = pd.DataFrame(transformer.transform(data), columns=data.columns)\n",
    "    return nums_norm '''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "def stdscaler_transform(data): \n",
    "    transformer = StandardScaler().fit(data)\n",
    "    nums_scaled = pd.DataFrame(transformer.transform(data), columns=data.columns)\n",
    "    return nums_scaled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEHCAYAAACjh0HiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de3yc1X3n8c9vRve7L7JlyXZ8wWCMwy0OkECTtkm25AZNkzRA2VzaLKVpet3slmS7TbvZfbXZdvtqm01DWJo2N0oDoRQSmjSQBJICDuYWDL5gG2zLtixZki1pZM1oZn77x8wYWYzu88wzo/m+Xy+/pHnmmZlfjPN855zznHPM3RERkcoVCbsAEREJl4JARKTCKQhERCqcgkBEpMIpCEREKlxV2AXM1fLly33dunVhlyEiUlaefPLJE+7enu+5sguCdevWsWPHjrDLEBEpK2Z2cKrn1DUkIlLhFAQiIhVOQSAiUuEUBCIiFU5BICJS4RQEIiIVTkEgIlLhFAQiIhVOQRCSeDIVdgkiIkAZziwuZ3dsP8TOI6f4/u5eeobG2NrVyrsvXMWvv3lj2KWJSAVTi6CIBmIJ7nryMGl3Ll+/lF3Hhrj14f2MJpJhlyYiFUxBUCTuzj1PdxMx4yNXrufai7v4yJXrGBwd53Pf3xd2eSJSwRQERfLYgX4O9MX4hQs6aK2vBmDD8iYuXdvG7T86wL7ekZArFJFKpSAokod29VIVMS5du+Ss41dvXUXEjK889nIodYmIKAiK5Pu7e9nQ3khN1dl/5U21Vbxty0ruf/YoiWQ6pOpEpJLprqEiONA3wksnYlxzUWfe55c21jA4Os7/uP8FtnS2AHDD5WuLWaKIVDC1CIrg+7t7ATivoznv85tWNNNYE+Xpw4PFLEtEBFAQFMXDe/s4d2UTSxpq8j4fjRgXrm5jT8+wuodEpOgUBAFzd54/OsQla5ZMe955Hc0k087L/bEiVSYikqEgCFjfSJyBWILNq/J3C+WsX95IVcR48fhwkSoTEclQEARs97HMhX1zR8u051VHI6xf3sje45pPICLFpSAI2J6eXBBM3yIA2LSiib6ROCdHE0GXJSJyhoIgYLt6hljZUsuSxvwDxRNtWpkJixfVKhCRIlIQBGz3seEZu4VyVjTX0lpfzd5ejROISPEoCAKUTKXZ1zsy40BxjpmxaUUT+/tGSKZ0G6mIFIeCIEAvnYiRSKVnNT6Qs2llM2PjaZ45fDLAykREXqEgCND+vkxf/6YVsw+Cc9qbMOCRvX0BVSUicjYFQYAOD5wGYM3Shlm/pr4mypqlDTysIBCRItGicwG4Y/shAB7afZy66gjf/umxOb1+04omvr+nl4FYgqWzuNtIRGQh1CII0GBsfMr1haZz7spm3OHH+04EUJWIyNkUBAEaGE3MKwi6ltTT1lCtcQIRKQoFQUDcnZOj8+vaiZhx5TnLeWRvH+4eQHUiIq9QEARkJJ5kPOUsaaie1+vffG47vcNxdvdocpmIBEtBEJDB0XGAeXUNAbxpUzug20hFJHgKgoAMxjILx81mjaF8OlrrOGdFE4/u7y9kWSIir6IgCMhgdgXR+bYIAN6wYRlPvDzAuJabEJEAKQgCMhBL0FhbRU3V/P+Kr9iwjNFEiueOnCpgZSIiZ1MQBGRwNMHSeQ4U51yxYSkAj6l7SEQCFGgQmNnVZrbHzPaZ2S3TnPd6M0uZ2fuCrKeYTo6O07qAbiGAZU21nLeymccPKAhEJDiBLTFhZlHg88DbgG7gCTO7z91fyHPeZ4HvBlVLsbk7Q2PjbK6b/WJzk+WWqVjSWMPjB/r5ymMvUxWJcMPlawtUpYhIRpAtgsuAfe5+wN0TwJ3AtXnO+y3gm0BvgLUU1dh4mvGU01K/sK4hgHXLGhhPOceH4gWoTETk1YIMgi7g8ITH3dljZ5hZF/Ae4Nbp3sjMbjKzHWa2o6+v9O+rHxrLzCEoRBB0tdUD0D04uuD3EhHJJ8ggsDzHJq+X8FfAH7h7aro3cvfb3H2bu29rb28vWIFBGTqdDYK6hQfB0sYa6qujHBk8veD3EhHJJ8hlqLuBNRMerwaOTjpnG3CnmQEsB95hZkl3vzfAugKXaxG0FqBFYGasXlLPkZMKAhEJRpAtgieATWa23sxqgOuA+yae4O7r3X2du68D7gY+Vu4hAHDqdBKA5rrC5GzXknqOD42RSGpimYgUXmAtAndPmtnHydwNFAW+5O7Pm9nN2eenHRcoZ0Nj4zTURKmOFiZnV7c1kHboOaVWgYgUXqA7lLn7A8ADk47lDQB3/3CQtRTT0OnxgowP5HQtyQ4Yq3tIRAKgmcUBGBobp6W+cBnbWl9Nc20VRxUEIhIABUEAhk4nC9oigMxqpD1DYwV9TxERUBAU3HgqTSyeLMgcgolWttTROxQnqZVIRaTAFAQF1jscx4HWQrcIWupIpp2X+zWxTEQKS0FQYD2nMt03hRwjAFjZWgfAHm1dKSIFpiAosN5sP35zgVsEK5prMWBPz1BB31dEREFQYL3DmcXhCj1GUB2NsLypVpvZi0jBKQgKrG84TsSgoSZa8Pde2VrHnuMKAhEpLAVBgfUNx2msrSJi+dbcW5iOlloO9o8SiycL/t4iUrkUBAXWOzxWsDWGJutoyQwY71WrQEQKSEFQYH0jcZprCzs+kLOyRXcOiUjhKQgKrG84TlNALYIljTU01EQ1YCwiBaUgKKB02jkxkqC5NpggiJhx7spmtQhEpKAUBAU0MJoglfbAWgQAmzua2d0zhPvkzd5EROZHQVBAfdk5BIWeTDbReR3NDI6On/ksEZGFUhAU0JkgCKhrCDJBAGicQEQKRkFQQK+0CILsGmoBdOeQiBSOgqCAcstLBDlGsLSxhvZmLTUhIoWjICigvuE4jTVRaqsKv7zERJs7mtlzXIvPiUhhKAgKqG8kTntzbeCfs7mjmRePj2iTGhEpCAVBAfUNjxUlCM7raCGeTGuTGhEpCAVBAfUOx1neVJwWAWjAWEQKQ0FQQAOxRFGC4JwVTURMm9SISGEoCAokmUpzcnScpY01gX9WXXWUdcsbdeeQiBSEgqBABkfHAVjWFHwQQO7OIQWBiCycgqBABmIJAJY0FCcIzlvZok1qRKQgFAQF0h/LTCZbVoSuIYDNqzIDxtqkRkQWSkFQILkWwdIidg2B7hwSkYVTEBTImSAoUotgzZIGbVIjIgUR3KI4FaZ/pDhjBHdsP3Tm96WNNTyyt487th/ihsvXBvq5IrJ4qUVQIIOjCVrrq6mOFu+vtKOljp6hMW1SIyILoiAokP5YomgDxTkrW+oYTaQY1p1DIrIACoICGRhJFG18IKejtQ6A46fGivq5IrK4KAgKZCAWQhC0ZIKgZ0hBICLzp8HiBZg4cHvk5Gma66rOOha0xtoqmmurOK4gEJEFCLRFYGZXm9keM9tnZrfkef5aM/upmT1jZjvM7Kog6wlK2p3RRJLGAPcqnkpHax3H1DUkIgsQWBCYWRT4PPB2YAtwvZltmXTaQ8BF7n4x8KvA7UHVE6Sx8RRpJ5Qg6Gyr5/jQGGPjqaJ/togsDkG2CC4D9rn7AXdPAHcC1048wd1H/JV7HxuBsrwPMhbPXIQba4LdojKfzrZ60q6lJkRk/oIMgi7g8ITH3dljZzGz95jZbuDbZFoFr2JmN2W7jnb09fUFUuxCjCYyt2+G0SLoaqsHYOcR7U0gIvMTZBBYnmOv+sbv7v/s7puBXwQ+k++N3P02d9/m7tva29sLXObC5VYADSMIljRUU1cd4fmjp4r+2SKyOAQZBN3AmgmPVwNHpzrZ3R8BNprZ8gBrCkSYXUNmRmdrPTuPqkUgIvMTZBA8AWwys/VmVgNcB9w38QQzO8fMLPv7pUAN0B9gTYGIhdg1BJlxgl3HhhhPpUP5fBEpb4Fdudw9aWYfB74LRIEvufvzZnZz9vlbgfcCHzSzceA08AEvw4VzYvEkNdFIUdcZmqizrY5EMs3+vhE2d7SEUoOIlK9Av8K6+wPAA5OO3Trh988Cnw2yhmKIJVI01ha/Wyins/WVAWMFgYjM1ay+wprZN83snWamJSnyiMXDmUyWs7y5lvrqKDuPaMBYROZuthf2LwA3AC+a2Z+Z2eYAayo7sXiSxprwgiBixpbOFt05JCLzMqsgcPcH3f1XgEuBl4HvmdmjZvYRM6sOssByEHbXEMDWzhZeODpEOl12QywiErJZd/WY2TLgw8BHgaeBvyYTDN8LpLIy4e6htwgALuhqJZZI8XJ/LNQ6RKT8zHaM4B7gR0AD8G53v8bd/8ndfwtoCrLAUpdIpUmmPdQxAoCtna0Amk8gInM22xbB7e6+xd3/1N2PAZhZLYC7bwusujJwZjJZyF1Dm1Y2URON8LwGjEVkjmYbBP8zz7HHCllIuTqzvETIXUPV0QibVzWzUwPGIjJH0169zKyDzEJx9WZ2Ca+sH9RCppuo4uVmFTeE3DV0x/ZD1FZFeergSb7++EGyE7a54fK1odYlIqVvpqvXL5AZIF4N/OWE48PApwKqqayEuc7QZJ1tdTzx8gAnR8dZUuRtM0WkfE0bBO7+ZeDLZvZed/9mkWoqK2EuQT1ZbobxkZOnFQQiMmszdQ3d6O5fA9aZ2e9Pft7d/zLPyypKLJ4kGjFqq8KfdN3RWkfE4Oip02ztag27HBEpEzN9jW3M/qzoW0SnE4unaKyJnumTD1N1NMKK5jqOndQexiIyezN1DX0x+/NPilNO+YmFtGn9VDrb6njx+EjYZYhIGZnthLL/bWYtZlZtZg+Z2QkzuzHo4spB2AvOTdbZVs9wPMnQ2HjYpYhImZhtx/Z/cPch4F1kdh47F/gvgVVVRmKJVEncMZSTGzA+evJ0yJWISLmYbRDkFpZ7B/CP7j4QUD1lJxZPhj6HYKKO1joAek5pnEBEZme2V7D7zWw3mV3EPmZm7UDFX2mSqTTxZDr0WcUT1VVHWdpYw1EFgYjM0myXob4FeAOwzd3HgRhwbZCFlYNYojTWGZpsVWsdx9Q1JCKzNJevsueTmU8w8TVfKXA9ZaVU1hmarKO1jheODhFPpsIuRUTKwKyuYGb2VWAj8AyQu7o4lR4EJTSreKLO1nocOK7uIRGZhdlewbYBW9xd219NUCpLUE+WGzA+NqQgEJGZzfauoZ1AR5CFlKNc11BTiXUNtdVXU1cd0QxjEZmV2V7BlgMvmNlPgHjuoLtfE0hVZSKWSBIxqCuheQQAZsaq1nqOndKAsYjMbLZB8MdBFlGuYvEU9TVVREpgnaHJVrVmlqROpZ1opPTqE5HSMdvbRx8GXgaqs78/ATwVYF1lIbNpfWm1BnJWtdYxnnIOajN7EZnBbNca+k/A3cAXs4e6gHuDKqpcjJbYgnMTrcouNfHCMW1mLyLTm+1g8W8CVwJDAO7+IrAiqKLKRW4J6lK0ormWiMEuBYGIzGC2QRB390TuQXZSWcXfSlpqS1BPVJXdm+CFowoCEZnebIPgYTP7FJlN7N8G3AXcH1xZpS+Vdk4nUiUbBJCdYawWgYjMYLZBcAvQBzwH/DrwAPCHQRVVDgZHEzilsWn9VFa11nF8KE7/SHzmk0WkYs3q66y7p83sXuBed+8LuKayMBDL9JSVcosgN2C869gwV22qDbkaESlV07YILOOPzewEsBvYY2Z9ZvZHxSmvdPWPlH4QdGaXmnjuyKmQKxGRUjZT19Dvkrlb6PXuvszdlwKXA1ea2e8FXl0JGxzNBEFDCXcNNdRW8ZplDTx7+GTYpYhICZspCD4IXO/uL+UOuPsB4MbscxWrvwy6hgAuWt3Gs90KAhGZ2kxBUO3uJyYfzI4TVOc5/yxmdrWZ7TGzfWZ2S57nf8XMfpr986iZXTT70sM1kOsaKrEF5ya7eE0bx06NcVwrkYrIFGYKgsQ8n8PMosDngbcDW4DrzWzLpNNeAt7s7hcCnwFum6GekjEQi1NXHSn5dXwuWtMGwDPqHhKRKcwUBBeZ2VCeP8PAa2d47WXAPnc/kJ2MdieTtrd090fdfTD78HFg9Xz+R4ShP5Yo+dYAwAWdLVRFTOMEIjKlaa9k7r6QkdAu4PCEx91kBpqn8mvAvy7g84pqIJYo+fEByGxmv3lVs8YJRGRKs51QNh/5+kzyLkthZj9HJgj+YIrnbzKzHWa2o6+vNKYxlEsQAFyyZgnPHDrJeCoddikiUoKCDIJuYM2Ex6uBo5NPMrMLgduBa929P98buftt7r7N3be1t7cHUuxcZbqGSvfW0YnesHEZsUSKn3ZrPoGIvFqQQfAEsMnM1ptZDXAdcN/EE8xsLXAP8B/dfW+AtRSUuzNYRi2CKzYsA+Cx/a+6AUxEJLggcPck8HHgu8Au4Bvu/ryZ3WxmN2dP+yNgGfC3ZvaMme0Iqp5CGhpLkkx72bQIljbWcP6qFh47kLfBJSIVLtCvtO7+AJkF6iYeu3XC7x8FPhpkDUEoh3WGJnvjxmV87fGDjI2nqKsujwATkeIIsmto0RqIZVbzLLcgiCfTPHVocOaTRaSilM+VrIT0l8msYoA7th8CYGw8RdSML/xwP2/cuDzkqkSklKhFMA+vdA2VTxdLXXWUDe2NPH90CPeK31xORCZQEMxDuSw4N9mWzhYGYgn2HB8OuxQRKSEKgnkYiCVoqIlSHS2vv74tq1ow4Ds7e8IuRURKSHldyUrEYCzB0saasMuYs+a6atYubVAQiMhZFATz0F+mQQBwQVcru3uG2d83EnYpIlIiFATzMFDGQXBhVytmcN8zr1rtQ0QqlIJgHso5CFrqq7l8/VLuf/ao7h4SEUBBMC/9sTjLyjQIAK65qIsDJ2I8f3Qo7FJEpAQoCOZoNJFkbDzN0sbasEuZt7dv7aAqYtz3rLqHRERBMGe5WcXl3CJY0ljDm85t5/5nj5JOq3tIpNIpCOZocDQTBOU6RpBzzUWdHDs1xo6DWntIpNIpCOYoN6t4aVN5B8HbtqykrjrC/eoeEql45bVGQgkYWARdQ7mF6DataOaep7o5d2Uz0Yhxw+VrQ65MRMKgFsEc5RacW1LGQZCztauVWCLFoYHRsEsRkRApCOaoP5agOmo0l9mCc/lsWtFE1IzdPbqNVKSSKQjmaCAWZ2ljDWYWdikLVlcdZX17I7uOaTVSkUqmIJijzKzi8p1DMNn5Hc2cGIlzYjgedikiEhIFwRz1xxJlPVA82eZVLQDsUveQSMVSEMxRuS5BPZUlDTV0tNSxu0fdQyKVSkEwR+W8BPVUzl/VzMH+GIPZO6JEpLIoCOYgkUwzPJZcVF1DAJs7Wkg7/HBvb9iliEgIFARzkFteYjHMIZioa0k9zbVVPLhLQSBSiRQEc7AYFpzLJ2LGeR3NPLynj0QyHXY5IlJkCoI5yM0qXmxjBADnr2phJJ7kJy8NhF2KiBSZgmAO+mOZe+2XlfmCc/lsbG+irjrCg7uOh12KiBSZgmAOXmkRLJ4JZTk1VRGuOqed771wXFtYilSY8l8wp0ju2H6IR/b2YcADzx0jsgiWmJjsbVtW8OCu4+zuGeb87EQzEVn81CKYg5F4ksbaqkUZAgA/v3klZvDgC+oeEqkkCoI5GBlL0rQIVh2dSntzLRevadM4gUiFURDMwUg8SVPd4g0CgLeev5Jnu09xfGgs7FJEpEgUBHMwEl/cLQLIbGEJ8JAml4lUjMV9VSsgd1/0QXDH9kO4O0sba/jyoy+fOa4tLEUWN7UIZimRTDOe8kUdBABmxvkdzezvGyGeTIVdjogUQaBBYGZXm9keM9tnZrfkeX6zmT1mZnEz+0SQtSzUSDwJsOiDAOD8zhaSaWfv8ZGwSxGRIggsCMwsCnweeDuwBbjezLZMOm0A+G3gL4Kqo1DOBMEiHywGWLeskabaKp47cirsUkSkCIJsEVwG7HP3A+6eAO4Erp14grv3uvsTwHiAdRTE8FjltAgiZlzQ2cKeniEtQidSAYIMgi7g8ITH3dljc2ZmN5nZDjPb0dfXV5Di5qqSuoYAtna1Mp5y9h7XzmUii12QQZBv+u28FrFx99vcfZu7b2tvb19gWfOTC4LGCgmCdcsaaayJqntIpAIEGQTdwJoJj1cDRwP8vECNxJM01ESJRhbn8hKTRSPGBZ2t7OkZZmxcdw+JLGZBBsETwCYzW29mNcB1wH0Bfl6gFvvyEvls7WolkUrzwz3hdMeJSHEEFgTungQ+DnwX2AV8w92fN7ObzexmADPrMLNu4PeBPzSzbjMryWUvF/tksnzWL2+koSbKA88dC7sUEQlQoFc2d38AeGDSsVsn/N5Dpsuo5I3Ek6xeUh92GUWV6R5q4aFdxxkbT1FXHQ27JBEJgGYWz1IltggAXtvVRiyR0tpDIouYgmAWRhNJEsl0RQbBhvZGVrXWcfeTh2c+WUTKkoJgFo6dyizJ3FpfHXIlxRcx45cu7eLhvX30amlqkUVJQTALPRUcBADvvXQ1aYd7nj4SdikiEgAFwSwcPXkaqNwg2NDexOvXLeGO7YdIpbWxvchioyCYhVyLoKVCgwDg165az6GBUb6zsyfsUkSkwBQEs3BsaIyGmijV0cr963rblg7WLWvgtkf2465WgchiUrlXtjk4dvI0bRXcGoDMnIJf+5kNPNt9ih+9eCLsckSkgBQEs3Ds1FhFdwvdsf1QZhvLdGYby0/c9Sxffexg2GWJSIEoCGbh2Kmxih0onqgqGuEdWzvoHY6z/aX+sMsRkQJREMxgNJHk1OlxBUHW+ata2LSiiX97/jgvn4iFXY6IFICCYAaVPodgMjPjPZd0EYnA7/zTM4yntIOZSLlTEMygkmcVT6WtoYb3XLKaZw+f5G8eejHsckRkgRQEM1AQ5Pfarlbe97rVfP4H+3ji5YGwyxGRBVAQzOBYdlZxJd81NJU/vuYC1ixt4HfvfIaTo4mwyxGReVIQzKB78DTLm2oqejLZVJpqq/jc9ZfQOzzGJ+56VhPNRMqUrm4z2N83wob2prDLKEl3bD/EziND/MIFHTy4q5ePff2psEsSkXlQEMxgf98I56xQEEznDRuWcUFnC999vocnD2q8QKTcKAimMRBLMDg6zka1CKZlZrz30tW0NdTwsa8/dWa1VhEpDwqCaezvGwFgY3tjyJWUvrrqKDde/hpG4yk+9KWfaPBYpIwoCKaxrzcXBGoRzEZHax1f/ODrONg/yge++PiZyXgiUtoUBNPY3ztCbVWErrb6sEspG2/cuJy//8jrOXLyNO/523/nqUODYZckIjNQEEwjd8dQJGJhl1I27th+iIP9o3z4jesYG0/xvi88ym9+/SndWipSwhQE09jfF9MdQ/PU2VbPx39uE+d1tPDt547xG197iqGx8bDLEpE8FARTGBtPcXhwVAPFC1BfE+XGy9fyjq0dPLjrOO/+3I/Z3TMUdlkiMomCYAo/7T6Fe2bZZZk/M+OqTe3cedMVjI2neP+tj2ltIpESoyCYwuMH+jGDy9cvDbuURWHv8RE+9IZ11FZFuOH/Pc6n/2Und2w/FHZZIoKCYEqP7e/n/I4W2hpqwi5l0WhrqOGmN22kvbmWrz5+kGcO644ikVKgIMhjbDzFU4cGuWLDsrBLWXSaaqv46FUbeM2yRr6xo5svP/py2CWJVDwFQR7PHD5JPJnmDRsVBEGoq47y4Teu4/xVLXz6vuf57/fuJJHUTmciYakKu4BS9Nj+fiIGl2l8IDDV0Qg3XLaW7sFRvvjIAZ7tPsn/ef9FbFrZHHZpIhVHLYJJUmnnn58+wutes0S7kgUsGjE++Y7zufXG19E9eJp3fu7HfPY7uzXfQKTIrNxmfG7bts137NgR2Pt/6p7nuOMnh7jhsrVs7WoN7HPkbMNj43xnZw9PHz6ZubPo8rW868JOLl7TRlQzu0UWzMyedPdt+Z5T19AkP953gqWNNWzp1PyBYmquq+b929Zw5TnL+fG+E3z1sYP8/b+/zLLGGt58XjsXrW5j3fJG2uqrMYO0Q9odd6e2KsrqJfW01ldjptAQmSsFwQTf2HGYQwOjvPvCVUR0QQlFZ1s9v7xtDe++sJO9vcPsPjbEd3b2cM9TR2Z8bVNtFWuWNnDeyibO7WhmY3sTq1rr6GipY1lTrVoWIlMINAjM7Grgr4EocLu7/9mk5y37/DuAUeDD7h7KfocP7+3jU/c8xzntTVy2XncLha2+JspFq9u4aHUb7s7QWJKTowlOJ1I4ELHMrGUDEqk0g6PjDI4m6B+J84M9fdz7zNGz3i8aMVY017J6ST3rljWybnlj9mcD65Y10lir70RSuQL7129mUeDzwNuAbuAJM7vP3V+YcNrbgU3ZP5cDX8j+LIrhsXGePXyKe57q5p6nj7BpRRPXvX6tvjmWGDOjtb56ToP3Y+MpTozEGTo9ztBYMvtznL7hOAf7R7nrye6zzm+uq6K5torG2ioaaqLUVEWoqYpQHY1QE838XlcdpbEmSkNtFU3Z86py/1ayLcjcv5zaqgjNdZn3a8q+b232/XLvWV1lVEcjVEUsb5dWOu2k3EmlPdsNBjVVU58vMl9Bfg26DNjn7gcAzOxO4FpgYhBcC3zFMyPWj5tZm5mtcvdjhS7mB7t7+aP7dpJMOeOpNKcTKWKJFJD5P9dv/OxGPv5z5/Avk75JSnmqq46yekkDLMn/fDyZon8kQX8s04oYjidJjKcZS6Y4PZ5iOJ4klXKS6cyFOJlOM55yEsk08WSKdIHvsaiKGNGI4c6Zi/9UIga1VZmwyoXLRLkbQBxwByf72F85lj3zzO+efZ2feY+z34fsa8lz3sS6cq00y/1uEDE781zkzGMjEsn8rkibvesvW8uvv3ljwd83yCDoAg5PeNzNq7/t5zunCzgrCMzsJuCm7MMRM9tT2FJZfgucuKXAb1pAy4ETYRcxg1KvUfUtTKnXB6Vf44Lrexi4ef4vf81UTwQZBPmCfvLXnNmcg7vfBtxWiKLyMbMdU91WVQpKvT4o/RpV38KUen1Q+jWWcn1BTijrBtZMeLwamNzvMptzREQkQEEGwRPAJjNbb2Y1wHXAfZPOuQ/4oGVcAZwKYnxARESmFljXkLsnzezjwHfJ3D76JXd/3sxuzj5/K/AAmVtH95G5ffQjQWqGhYgAAAXWSURBVNUzg8C6nQqk1OuD0q9R9S1MqdcHpV9jydZXdktMiIhIYWnRORGRCqcgEBGpcAqCLDP7czPbbWY/NbN/NrO2sGuCzDIdZrbHzPaZWUlNdTCzNWb2AzPbZWbPm9nvhF1TPmYWNbOnzexbYdeST3Yi5d3Zf3+7zOwNYdc0kZn9Xva/704z+0czqwu5ni+ZWa+Z7ZxwbKmZfc/MXsz+nGIqYag1luQ1BhQEE30P2OruFwJ7gU+GXM/EZTreDmwBrjezLeFWdZYk8J/d/XzgCuA3S6y+nN8BdoVdxDT+GviOu28GLqKEajWzLuC3gW3uvpXMjR/XhVsV/wBcPenYLcBD7r4JeCj7OEz/wKtrLLlrTI6CIMvd/83dk9mHj5OZ0xC2M8t0uHsCyC3TURLc/VhukUB3HyZzAesKt6qzmdlq4J3A7WHXko+ZtQBvAv4OwN0T7n4y3KpepQqoN7MqoIGQ5/q4+yPAwKTD1wJfzv7+ZeAXi1rUJPlqLNFrDKAgmMqvAv8adhFMvQRHyTGzdcAlwPZwK3mVvwL+K1CqmyJvAPqAv892X91uZo1hF5Xj7keAvwAOkVn65ZS7/1u4VeW1MjcHKftzRcj1zKRUrjFAhQWBmT2Y7eec/OfaCef8NzJdHl8Pr9IzZrUER9jMrAn4JvC77j4Udj05ZvYuoNfdnwy7lmlUAZcCX3D3S4AY4XdrnJHta78WWA90Ao1mdmO4VZW3ErvGABW2MY27v3W6583sQ8C7gLd4aUywKPklOMysmkwIfN3d7wm7nkmuBK4xs3cAdUCLmX3N3UvpQtYNdLt7riV1NyUUBMBbgZfcvQ/AzO4B3gh8LdSqXu14buViM1sF9IZdUD4leI0BKqxFMJ3sJjp/AFzj7qNh15M1m2U6QpPdWOjvgF3u/pdh1zOZu3/S3Ve7+zoyf3ffL7EQwN17gMNmdl720Fs4e6n2sB0CrjCzhux/77dQQoPZE9wHfCj7+4eAfwmxlrxK9BoDaGbxGWa2D6gF+rOHHnf3Baz4WhjZb7N/xSvLdPyvkEs6w8yuAn4EPMcrffCfcvcHwqsqPzP7WeAT7v6usGuZzMwuJjOYXQMcAD7i7oPhVvUKM/sT4ANkujOeBj7q7vEQ6/lH4GfJLOt8HPg0cC/wDWAtmfB6v7tPHlAOu8ZPUoLXGFAQiIhUPHUNiYhUOAWBiEiFUxCIiFQ4BYGISIVTEIiIVDgFgYhIhVMQiORhZrXZJUmeMbMPmNmnZvGakezPTjO7e4Zzrym1ZcWlcmkegUgeZnYF8Fl3f3P28Yi7N83wmhnPESlFahFIxTCzRjP7tpk9m11s8APZjX92m9mPzexvzOxbZraCzFo6F2dbBHeRWYb5GTObcaEwM1uX25DEzLab2QUTnvuhmb3OzD5sZv83e+wfsp/9qJkdMLP3ZY9HzOxvs5vCfMvMHsg9J1JIFbXonFS8q4Gj7v5OADNrBXYCPw/sA/4JwN17zeyjTFiSIvtt/+J5fOadwC8Dn84uhtbp7k+a2WsnnbcKuArYTGbdnLuBXwLWAa8ls6zyLuBL86hBZFpqEUgleQ54q5l91sx+hszSyi+5+4vZlSCDWFHzG8D7s7//MnDXFOfd6+5pd38BWJk9dhVwV/Z4D/CDAOoTURBI5XD3vcDryATCnwLXEPD+DtmNXfrN7EIyC7fdOcWpExdxs0k/RQKlIJCKYWadwKi7f43MrltvBNab2cbsKddP8/Lx7N4L83EnmV3SWt39uTm87sfAe7NjBSvJrGYpUnAaI5BK8lrgz80sDYwDv0FmmeBvm9kJMhferVO89jbgp2b2lLv/yhw/924yG9R/Zo6v+yaZ9f93ktnsfDtwao7vITIj3T4qklWKexaYWZO7j5jZMuAnwJXZ8QKRglGLQKS0fcvM2shsWvMZhYAEQS0CkTnIfjN/KM9Tb3H3/jzHRUqegkBEpMLpriERkQqnIBARqXAKAhGRCqcgEBGpcP8fdOtvHuvw0RgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEHCAYAAAC0pdErAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAeg0lEQVR4nO3dfZRcd33f8fdnZleyLQs/SGswkowUV21wKDbuIhtMwE7BkY1BJckpEk+BA1Uh9jlAWhJDe3hIe04ppDQlOCgq6Jgn25zUFihEYDuU4gAxSDayJflRESJey7HWlh8kW0iamW//uL/RXo3u7I60ujtr38/rnDk7cx92vtyD96Pfw/1dRQRmZmadav0uwMzMpicHhJmZFXJAmJlZIQeEmZkVckCYmVmhgX4XcDzNnTs3Fi5c2O8yzMyeM+64447HImKoaN/zKiAWLlzIxo0b+12GmdlzhqRfdtvnLiYzMyvkgDAzs0IOCDMzK+SAMDOzQg4IMzMr5IAwM7NCDggzMyvkgDAzs0IOiAJbHn6KN3/hRzx7oNHvUszM+sYBUeCenU9z98hT7Hp6f79LMTPrGwdEgUYre8pe00/bM7MKc0AUaLRaADRbDggzqy4HRIFGM7UgHBBmVmEOiALtYHBAmFmVOSAKHExdTC2PQZhZhTkgCjRTF1PDLQgzq7DSAkLSAkk/kHSvpK2SPlhwjCR9XtI2SXdLOj+3b6mk+9O+q8uqs0g7GFoOCDOrsDJbEA3gP0TES4ELgSslndNxzGXA4vRaCXwRQFIduCbtPwdYUXBueYV7FpOZWXkBERGPRMSd6f0e4F5gXsdhy4CvRuZ24FRJZwJLgG0RsT0iDgA3pGOnhO+DMDObojEISQuBVwA/7dg1D3go93kkbeu2veh3r5S0UdLG0dHR41Jv09NczczKDwhJJwM3Ah+KiKc7dxecEuNsP3JjxOqIGI6I4aGhockVmzQ8zdXMjIEyf7mkQbJw+EZE3FRwyAiwIPd5PrATmNFl+5RoeJqrmVmps5gEfBm4NyI+1+WwdcC70mymC4GnIuIRYAOwWNIiSTOA5enYKTF2J/VUfaOZ2fRTZgviIuCdwGZJm9K2jwFnAUTEKmA9cDmwDXgWeE/a15B0FXAzUAfWRMTWEms9zFgXkxPCzKqrtICIiB9RPJaQPyaAK7vsW08WIFNubKmNfny7mdn04DupCxxMyeBprmZWZQ6IAk3fSW1m5oAo0h6D8FpMZlZlDogCjdTF5BaEmVWZA6KAl9owM3NAFPIT5czMHBCFDg1SuwVhZhXmgCjQXmqj3ZIwM6siB0SBhlsQZmYOiCIegzAzc0AUanoWk5mZA6LIwfYjRz0GYWYV5oAo4BaEmZkDolB7DMJ3UptZlTkgCrSnuboFYWZV5oAo0PRifWZm5T0wSNIa4ApgV0S8rGD/R4C35+p4KTAUEbsl7QD2AE2gERHDZdVZ5KC7mMzMSm1BXAss7bYzIj4bEedFxHnAR4EfRsTu3CGXpP1TGg7gJ8qZmUGJARERtwG7JzwwswK4vqxajlZ7DMJ3UptZlfV9DELSSWQtjRtzmwO4RdIdklZOcP5KSRslbRwdHT0uNbVnMbWDwsysivoeEMCbgB93dC9dFBHnA5cBV0p6bbeTI2J1RAxHxPDQ0NCki4mIsedBOB/MrMKmQ0Asp6N7KSJ2pp+7gLXAkqkqJr/+kgepzazK+hoQkk4BXgd8O7dtlqTZ7ffApcCWqaopP7XV90GYWZWVOc31euBiYK6kEeATwCBARKxKh70FuCUinsmd+kJgraR2fddFxPfKqrNTvgXh1VzNrMpKC4iIWNHDMdeSTYfNb9sOnFtOVRPLPyTIAWFmVTYdxiCmlfzMJXcxmVmVOSA6eJDazCzjgOhwMBcKXovJzKrMAdEh/5AgtyDMrMocEB0OegzCzAxwQBzB01zNzDIOiA75aa5erM/MqswB0SE/zTUfFmZmVeOA6JCfueQWhJlVmQOiQ3vcQfIYhJlVmwOiw8G0xveMeg33MJlZlTkgOrRbDTMGajT9wCAzqzAHRIf2GMTMgbofGGRmleaA6NCeuTRzoOY7qc2s0hwQHdrdSjMHa76T2swqrbSAkLRG0i5JhU+Dk3SxpKckbUqvj+f2LZV0v6Rtkq4uq8YiB1MLYka95llMZlZpZbYgrgWWTnDM30XEeen1JwCS6sA1wGXAOcAKSeeUWOdh2qEwc7DugDCzSistICLiNmD3MZy6BNgWEdsj4gBwA7DsuBY3jrFBarcgzKza+j0G8SpJd0n6rqTfSNvmAQ/ljhlJ2wpJWilpo6SNo6Ojky6okaYuzRyo+U5qM6u0fgbEncBLIuJc4M+Bb6XtKji261/qiFgdEcMRMTw0NDTpototiBn1mh8YZGaV1reAiIinI2Jver8eGJQ0l6zFsCB36Hxg51TVNTYG4WmuZlZtfQsISS+SpPR+SarlcWADsFjSIkkzgOXAuqmq6+ChLqa6p7maWaUNlPWLJV0PXAzMlTQCfAIYBIiIVcDvAR+Q1AD2AcsjIoCGpKuAm4E6sCYitpZVZ6dmy9NczcygxICIiBUT7P8C8IUu+9YD68uoayKNXBeTA8LMqqzfs5imnfxSGw4IM6syB0SHZquFBAN1T3M1s2pzQHQ42AoGazXqklsQZlZpDogOzVZQr4laTbQCwq0IM6soB0SHRjMYqImBWna/nlsRZlZVDogOjVaLgbqotwPCLQgzqygHRIdGK6jXatSye/jwU0fNrKocEB0azRYDNVFPV6bhhDCzinJAdGi0InUxZZfG+WBmVeWA6NBsZYPU9bSmrMcgzKyqHBAdGs1smmvds5jMrOIcEB0arRaD9Rq1FBC+m9rMqsoB0aF9o1z7Pgg/NMjMqsoB0eFgulFubJqrA8LMqskB0aHZCgbqNY9BmFnlOSA6HGy2Dh+k9hiEmVVUaQEhaY2kXZK2dNn/dkl3p9dPJJ2b27dD0mZJmyRtLKvGIs1WMFj3LCYzs54CQtKNkt4o6WgC5Vpg6Tj7fwG8LiJeDvwXYHXH/ksi4ryIGD6K75y09lIbdTkgzKzaev2D/0XgbcCDkj4t6dcnOiEibgN2j7P/JxHxRPp4OzC/x1pK1WhlS23U3IIws4rrKSAi4m8j4u3A+cAO4NbULfQeSYPHoY73At/NfyVwi6Q7JK0c70RJKyVtlLRxdHR00oW0l/tutyB8H4SZVVXPXUaS5gDvBt4H/Bz4X2SBcetkCpB0CVlA/HFu80URcT5wGXClpNd2Oz8iVkfEcEQMDw0NTaYUoD2LSdTrvg/CzKptoJeDJN0E/DrwNeBNEfFI2vXNyQwiS3o58CXgsoh4vL09Inamn7skrQWWALcd6/ccjc4xCN8HYWZV1VNAAF+KiPX5DZJmRsT+Yx1ElnQWcBPwzoh4ILd9FlCLiD3p/aXAnxzLdxyLRqvFoNdiMjPrOSD+K7C+Y9vfk3UxFZJ0PXAxMFfSCPAJYBAgIlYBHwfmAH+h7F/rjRQ2LwTWpm0DwHUR8b0e65y09mJ97TupfR+EmVXVuAEh6UXAPOBESa8A0iLYvAA4abxzI2LFBPvfRzae0bl9O3DukWdMjfbzIAbqbkGYWbVN1IL4bbKB6fnA53Lb9wAfK6mmvmov1lfzfRBmVnHjBkREfAX4iqTfjYgbp6imvsoeGDS2FpOnuZpZVU3UxfSOiPg6sFDSH3buj4jPFZz2nNZsBQ8+uodbtz4KwA/uG+WfntoPwNsuOKufpZmZTamJuphmpZ8nl13IdNFotahJpEdSu4vJzCproi6mv0w/PzU15fRfsxXUaiLNosLxYGZV1etifZ+R9AJJg5K+L+kxSe8ou7h+aLaCmsYujMcgzKyqel1q49KIeBq4AhgB/jnwkdKq6pNWK2gFWRdTuwXhgDCziuo1INoL8l0OXB8RXVdpfS5r3xQnja3m2mr1syIzs/7p9U7qv5Z0H7AP+ANJQ8CvyiurP9oD0nVBakC4i8nMKqvX5b6vBl4FDEfEQeAZYFmZhfVDOyBqtXwXUz8rMjPrn15bEAAvJbsfIn/OV49zPX3VXtq7Jo21IDyPycwqqtflvr8GnA1sApppc/A8C4h2C0LiUAvCy32bWVX12oIYBs6J5/mUnmauBVE7NAbRx4LMzPqo11lMW4AXlVnIdDA2SO1prmZmvbYg5gL3SPoZsL+9MSLeXEpVfdJIc1prtfwspj4WZGbWR70GxCeP9hdLWkN2Y92uiHhZwX6RPdf6cuBZ4N0RcWfatzTtq5M9ze7TR/v9x2JsDMItCDOzXqe5/hDYAQym9xuAOyc47Vpg6Tj7LwMWp9dK4IsAkurANWn/OcAKSef0UudkHT4G0X6i3FR8s5nZ9NPrWkz/Dvg/wF+mTfOAb413TkTcBox3x/Uy4KuRuR04VdKZwBJgW0Rsj4gDwA1M0T0XYwEx1sXkFoSZVVWvg9RXAhcBTwNExIPAGZP87nnAQ7nPI2lbt+2laxS0IDwGYWZV1WtA7E//mgcg3Sw32T+dKtgW42wv/iXSSkkbJW0cHR2dVEH5Lqbsp1sQZlZdvQbEDyV9DDhR0huAvwL+epLfPQIsyH2eD+wcZ3uhiFgdEcMRMTw0NDSpgsaW2sg+1ySvxWRmldVrQFwNjAKbgX8PrAf+8yS/ex3wLmUuBJ6KiEfIBsAXS1okaQawPB1bukZHC0JyF5OZVVdP01wjoiXpW8C3IqKnfhxJ1wMXA3MljQCfIC0bHhGryELmcmAb2TTX96R9DUlXATeTTXNdExFbj+Z/1LE6sotJ7mIys8oaNyDSvQqfAK4iGxuQpCbw5xHxJ+OdGxErJtgfZIPfRfvWkwXIlDp0o1waBcm6mKa6CjOz6WGiLqYPkc1eemVEzImI04ELgIskfbj06qZY++FAh3cxOSHMrJomCoh3ASsi4hftDRGxHXhH2ve8MrbUxlgXk1sQZlZVEwXEYEQ81rkxjUMMFhz/nJa/Ua7902MQZlZVEwXEgWPc95zUOYvJLQgzq7KJZjGdK+npgu0CTiihnr5qFU5zdUKYWTWNGxARUZ+qQqaDxhFdTL5Rzsyqq9cb5Sqh6D4IdzGZWVU5IHIOtSBSE6Jek59JbWaV5YDIaXV0MdVrOtSqMDOrGgdETucspnpNND0GYWYV5YDIaaYb5eQWhJmZAyKv3YKo51sQDggzqygHRE6zc5BaDggzqy4HRE7nNFe3IMysyhwQOe0uJo9BmJk5IA7TudSGZzGZWZWVGhCSlkq6X9I2SVcX7P+IpE3ptUVSU9Lpad8OSZvTvo1l1tnWudSGxyDMrMp6euTosZBUB64B3gCMABskrYuIe9rHRMRngc+m498EfDgidud+zSVFy42XpdkK6jWhXAui4YAws4oqswWxBNgWEdsj4gBwA7BsnONXANeXWM+EGq04NMUVvNSGmVVbmQExD3go93kkbTuCpJOApcCNuc0B3CLpDkkru32JpJWSNkraODo6OqmCW5G1INo8SG1mVVZmQKhgW7e/tm8CftzRvXRRRJwPXAZcKem1RSdGxOqIGI6I4aGhoUkV3GgGAw4IMzOg3IAYARbkPs8HdnY5djkd3UsRsTP93AWsJeuyKlWz1aJe7wiICD921MwqqcyA2AAslrRI0gyyEFjXeZCkU4DXAd/ObZslaXb7PXApsKXEWoHiMQjAz4Qws0oqbRZTRDQkXQXcDNSBNRGxVdL70/5V6dC3ALdExDO5018IrE2ziQaA6yLie2XV2nbEGEQKi/bsJjOzKiktIAAiYj2wvmPbqo7P1wLXdmzbDpxbZm1FisYgAI9DmFkl+U7qnGYrDi3UB2MB0UjLgJuZVYkDIqfRKm5BuAFhZlXkgMhpFtwHAe5iMrNqckDkNJvBQG3skriLycyqzAGR0+gcg5BbEGZWXQ6InGar5VlMZmaJAyKnGRSOQXjBPjOrIgdETrcWRMNLbZhZBTkgchrNw8cgBtzFZGYV5oDIaXa5D8IBYWZV5IDI6bwPov1sao9BmFkVOSByurUg/NhRM6siB0ROo3l4C8JjEGZWZQ6InM5lvT0GYWZV5oDIacbhS220ZzQ1Pc3VzCqo1ICQtFTS/ZK2Sbq6YP/Fkp6StCm9Pt7ruWXobEEMeKkNM6uw0h4YJKkOXAO8gez51BskrYuIezoO/buIuOIYzz2uGq2Wu5jMzJIyWxBLgG0RsT0iDgA3AMum4Nxj1mx6DMLMrK3MgJgHPJT7PJK2dXqVpLskfVfSbxzlucdVNgaRuw/CAWFmFVbmM6lVsK3zL+2dwEsiYq+ky4FvAYt7PDf7EmklsBLgrLPOOvZqOXIMoiZRkwepzayaymxBjAALcp/nAzvzB0TE0xGxN71fDwxKmtvLubnfsToihiNieGhoaFIFNzoCArJuJrcgzKyKygyIDcBiSYskzQCWA+vyB0h6kZRNFZK0JNXzeC/nlqGzBQEOCDOrrtK6mCKiIekq4GagDqyJiK2S3p/2rwJ+D/iApAawD1geEQEUnltWrW2dS21A9lQ5B4SZVVGZYxDtbqP1HdtW5d5/AfhCr+eWLetiOrxR5RaEmVWV76TOybqYDt/mgDCzqnJAJBGRAqKgBeFZTGZWQQ6IpN1IOGIMwi0IM6soB0TSaLUAPIvJzCxxQCTtEDgiIDyLycwqygGRtEPAXUxmZhkHRNK1BeGAMLOKckAkjfECwrOYzKyCHBBJ9xZEzS0IM6skB0TSdQxCXu7bzKrJAZGMtSC81IaZGTggDhkbgzh8u7uYzKyqHBBJ89CNcl5qw8wMHBCHNLN8KLgPwmMQZlZNDoik61IbvpPazCrKAZEcGqSWp7mamUHJASFpqaT7JW2TdHXB/rdLuju9fiLp3Ny+HZI2S9okaWOZdUJukLruO6nNzKDEJ8pJqgPXAG8ARoANktZFxD25w34BvC4inpB0GbAauCC3/5KIeKysGvNa46zFFEDLA9VmVjFltiCWANsiYntEHABuAJblD4iIn0TEE+nj7cD8EusZ13hLbYAHqs2sesoMiHnAQ7nPI2lbN+8Fvpv7HMAtku6QtLLbSZJWStooaePo6OgxF9t9DMIBYWbVVFoXE6CCbYV/ZSVdQhYQr8ltvigidko6A7hV0n0RcdsRvzBiNVnXFMPDw8f8V7zdghgoGIPI7zczq4oyWxAjwILc5/nAzs6DJL0c+BKwLCIeb2+PiJ3p5y5gLVmXVWlaXZbaGEgtipYDwswqpsyA2AAslrRI0gxgObAuf4Cks4CbgHdGxAO57bMkzW6/By4FtpRY61gLomMMouYuJjOrqNK6mCKiIekq4GagDqyJiK2S3p/2rwI+DswB/kLZv9QbETEMvBBYm7YNANdFxPfKqhXGltqoeQzCzAwodwyCiFgPrO/Ytir3/n3A+wrO2w6c27m9TBOOQXiaq5lVjO+kTro9MOiEgewS7TvQnPKazMz6yQGRdHtg0NzZMwEY3bN/ymsyM+snB0TS7mLqHIM45cRBButi155f9aMsM7O+cUAkrS5jEDWJodkz3YIws8pxQCTdltoAOGP2CexyQJhZxTggkm5LbQAMzZ7JU/sO8sz+xlSXZWbWNw6I5IlnDyDBySccOfN36ORsoPofRvdOdVlmZn3jgEgefmIfZ8yeycyB+hH7zkgzmbbtckCYWXU4IJKHn9zHvFNPLNw35+SZ1OSAMLNqcUAkDz+5j3mnnVS4r14Tc06eyYYduwnfUW1mFeGAIJvi+siTv+raggC4cNHpbNjxBF/+0S+msDIzs/4pdS2m54rRvfs50Gwx79QTuh5z4a/N4UCzxae/ex+vPnsu57z4BVNYoZnZ1HMLAhh5Yh8A807r3oKQxGd+91xOOXGQT67b6q4mM3vec0CQjT8AzDu1eAyi7ZSTBvnIb/8LfrZjN+vuOuLZR2ZmzyvuYiKb4grjtyAArvvpP9KKYN6pJ/LRmzbz0O59nD5rBm+74KypKNPMbEq5BQE8/OSznHLiICfPnDgvaxLLX7mACPja7Tt41ndXm9nzVKkBIWmppPslbZN0dcF+Sfp82n+3pPN7Pfd4eviJ7vdAFJlz8kyWL1nAY3sO8Gfff5DrfvqPPPDoHvb86qDHJszseaO0LiZJdeAa4A3ACLBB0rqIuCd32GXA4vS6APgicEGP5x43Dz+5j5fMmXVU5yw+YzYfuPhsbrxzhI+t3Xxo+4x6jdNnzeC0WTOYk/t5esf702fNYEa9RiuCVkCkn9nnICJbH+qJZw+wbddefvjAKA/tfpYTBuu8+uy5vP6cM1g0dxYnzRhgoCbq7Zd06DnaZmaTUeYYxBJgW3p8KJJuAJYB+T/yy4CvRvbP7tslnSrpTGBhD+ceFxHBw0/s49Vnzz3qc1986olceck/Y3TPfnY+uY+9+xs8s7/BMweaPLO/wS8ff4Z7H3maZw40+NXB1qTqHDp5JmeeegL7DjT5yt/vYM2Pu9+PIWWLDtYkpOxzLfc5e0/6nL1vb58KcSgIs+ufr6+9mu5kSjnac0XvJxz97zYr32mzZrD2Dy467r+3zICYBzyU+zxC1kqY6Jh5PZ4LgKSVwMr0ca+k+4+l2E+lV4G5wGPH8juPp18e/nFa1FRgOtblmno3HeuajjXBNKxLVx5zTS/ptqPMgCj6x1NnB323Y3o5N9sYsRpYfXSl9U7SxogYLuv3H4vpWBNMz7pcU++mY13TsSaYnnWVUVOZATECLMh9ng903jzQ7ZgZPZxrZmYlKnMW0wZgsaRFkmYAy4F1HcesA96VZjNdCDwVEY/0eK6ZmZWotBZERDQkXQXcDNSBNRGxVdL70/5VwHrgcmAb8CzwnvHOLavWCZTWfTUJ07EmmJ51uabeTce6pmNNMD3rOu41yfP2zcysiO+kNjOzQg4IMzMr5IDoYiqX+jgaknZI2ixpk6SNfaphjaRdkrbktp0u6VZJD6afp02Tuj4p6eF0vTZJunyKa1og6QeS7pW0VdIH0/a+Xa9xaur3tTpB0s8k3ZXq+lTa3s9r1a2mvl6rVENd0s8lfSd9Pu7XyWMQBdJSHw+QW+oDWFHWUh9HQ9IOYDgi+naTjqTXAnvJ7oJ/Wdr2GWB3RHw6BeppEfHH06CuTwJ7I+JPp7KWXE1nAmdGxJ2SZgN3AP8GeDd9ul7j1PRv6e+1EjArIvZKGgR+BHwQ+B36d6261bSUPl6rVNsfAsPACyLiijL+G3QLotihZUIi4gDQXurDgIi4DdjdsXkZ8JX0/itkf3CmVJe6+ioiHomIO9P7PcC9ZCsF9O16jVNTX0Vmb/o4mF5Bf69Vt5r6StJ84I3Al3Kbj/t1ckAU67YEyHQQwC2S7kjLjEwXL0z3sJB+ntHnevKuUrZa8Jp+dH21SVoIvAL4KdPkenXUBH2+VqnbZBOwC7g1Ivp+rbrUBP29Vn8G/BGQX+TtuF8nB0Sxnpf66IOLIuJ8spVwr0zdKtbdF4GzgfOAR4D/0Y8iJJ0M3Ah8KCKe7kcNnQpq6vu1iohmRJxHtnrCEkkvm+oaeqypb9dK0hXAroi4o+zvckAU62WZkL6IiJ3p5y5gLVl32HTwaOrbbvdx7+pzPQBExKPpP/AW8L/pw/VKfdc3At+IiJvS5r5er6KapsO1aouIJ4H/R9bXPy3+v5Wvqc/X6iLgzWk88gbgtyR9nRKukwOi2LRc6kPSrDSoiKRZwKXAlvHPmjLrgN9P738f+HYfazmk/R9M8ham+HqlQc4vA/dGxOdyu/p2vbrVNA2u1ZCkU9P7E4HXA/fR32tVWFM/r1VEfDQi5kfEQrK/Tf83It5BGdcpIvwqeJEtAfIA8A/Af+p3PammXwPuSq+t/aoLuJ6sWX2QrLX1XmAO8H3gwfTz9GlS19eAzcDd6T+gM6e4pteQdU/eDWxKr8v7eb3Gqanf1+rlwM/T928BPp629/Nadaupr9cqV9/FwHfKuk6e5mpmZoXcxWRmZoUcEGZmVsgBYWZmhRwQZmZWyAFhZmaFHBBmZlbIAWF2nEiaKelv0/LPb5X0sR7O2TvB/oWS3nb8qjTrnQPC7Ph5BTAYEedFxDeBCQOiBwsBB4T1hQPCbBxpeZO/SQ+M2ZJaBksl3SfpR5I+L+k7ks4Avg6cl1oQfwWcmN5/o4fvkaTPpu/YLOmtadengd9Mv+fDJf5PNTvCQL8LMJvmlgI7I+KNAJJOIVty4beAbcA3IVs8UdL7gP8YEVekY/dGtgpoL36HbGXQc4G5wAZJtwFX53+n2VRyC8JsfJuB10v675J+E1gE/CIiHoxsnZqvH6fveQ1wfWQrhD4K/BB45XH63WbHxAFhNo6IeAD4V2RB8d+AN1POs0GKnkFi1lcOCLNxSHox8GxEfB34U+DVwCJJZ6dDVoxz+sH03IVe3Aa8NT29bAh4LfAzYA8w+9iqN5scj0GYje9fAp+V1CJbRvwDZGMEfyPpMbKH2Hd76tlq4G5Jd0bE2yf4nrXAq8iWcg/gjyLinyQ9DjQk3QVcGxH/c/L/k8x64+W+zSZB0sV4ENmep9zFZGZmhdyCMCuZpPaTvjr964h4fKrrMeuVA8LMzAq5i8nMzAo5IMzMrJADwszMCjkgzMys0P8H1FOw8r3gDOYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEJCAYAAACZjSCSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3hc9ZXw8e+ZUZesLjcVS7ZlbHC3XMAQCITEtBAILC0hQALLZkk2m+xuSNhN3n33zSa7bEk2IXHYxECyMb15iYkJhGJcJRfciyzJlizbqpasXua8f2hshFC1dDWauefzPHo8t8ydM7Y8Z+6vnJ+oKsYYY9zLE+gAjDHGBJYlAmOMcTlLBMYY43KWCIwxxuUsERhjjMtZIjDGGJdzNBGIyAoROSgihSLycC/H/1ZEdvp/9ohIp4gkOxmTMcaYjxKn5hGIiBc4BFwNlAH5wB2quq+P828A/lpVr+zvuqmpqZqdnT3C0RpjTGjbtm1blaqm9XYszMHXXQIUqmoRgIg8A9wI9JoIgDuApwe6aHZ2NgUFBSMWpDHGuIGIHO3rmJNNQ+lAabftMv++jxGRGGAF8GIfxx8QkQIRKaisrBzxQI0xxs2cTATSy76+2qFuADaoak1vB1X1cVXNU9W8tLRe72yMMcacJycTQRmQ2W07Ayjv49zbGUSzkDHGmJHnZCLIB3JFJEdEIuj6sF/T8yQRSQAuB151MBZjjDF9cKyzWFU7ROQhYB3gBVap6l4RedB/fKX/1JuAN1S10alYjDHG9M2x4aNOycvLUxs1ZIwxQyMi21Q1r7djNrPYGGNczhKBMca4nCUCY4xxOSdnFhtjzJixesuxPo/duTRrFCMZe+yOwBhjXM4SgTHGuJwlAmOMcTlLBMYY43KWCIwxxuUsERhjjMtZIjDGGJezRGCMMS5nicAYY1zOEoExxricJQJjjHE5SwTGGONylgiMMcblLBEYY4zLWSIwxhiXs0RgjDEuZ4nAGGNczhKBMca4nKOJQERWiMhBESkUkYf7OOcKEdkpIntF5F0n4zHGGPNxjq1ZLCJe4DHgaqAMyBeRNaq6r9s5icDPgRWqekxExjsVjzHGmN45eUewBChU1SJVbQOeAW7scc6dwEuqegxAVSscjMcYY0wvnEwE6UBpt+0y/77uZgBJIvKOiGwTkbt7u5CIPCAiBSJSUFlZ6VC4xhjjTk4mAulln/bYDgMWAdcBnwH+QURmfOxJqo+rap6q5qWlpY18pMYY42KO9RHQdQeQ2W07Ayjv5ZwqVW0EGkXkPWAecMjBuIwxxnTj5B1BPpArIjkiEgHcDqzpcc6rwGUiEiYiMcBSYL+DMRljjOnBsTsCVe0QkYeAdYAXWKWqe0XkQf/xlaq6X0T+AOwCfMCvVHWPUzEZY4z5OCebhlDVtcDaHvtW9th+FHjUyTiMMcb0zWYWG2OMy1kiMMYYl7NEYIwxLmeJwBhjXM4SgTHGuJwlAmOMcTlLBMYY43KWCIwxxuUsERhjjMtZIjDGGJezRGCMMS5nicAYY1zOEoExxricJQJjjHE5SwTGGONylgiMMcblLBEYY4zLWSIwxhiXs0RgjDEuZ4nAGGNczhKBMca4nCUCY4xxOUcTgYisEJGDIlIoIg/3cvwKEakTkZ3+n+85GY8xxpiPC3PqwiLiBR4DrgbKgHwRWaOq+3qcul5Vr3cqDmOMMf1z8o5gCVCoqkWq2gY8A9zo4OsZY4w5D04mgnSgtNt2mX9fTxeLyAci8rqIXNTbhUTkAREpEJGCyspKJ2I1xhjXcjIRSC/7tMf2dmCKqs4Dfgq80tuFVPVxVc1T1by0tLQRDtMYY9zNyURQBmR2284AyrufoKr1qtrgf7wWCBeRVAdjMsYY04OTiSAfyBWRHBGJAG4H1nQ/QUQmioj4Hy/xx1PtYEzGGGN6cGzUkKp2iMhDwDrAC6xS1b0i8qD/+ErgFuAvRKQDaAZuV9WezUfGGGMc5FgigHPNPWt77FvZ7fHPgJ85GYMxxpj+2cxiY4xxOUsExhjjcpYIjDHG5RztIzDGjG2rtxzr89idS7NGMRITSHZHYIwxLmeJwBhjXM4SgTHGuJwlAmOMcTlLBMYY43I2asgYE7Je3lHGc/llhHmFifFRLMhKCnRIY5IlAmPMiAv0sFSfT3n4pV08V1DGtLRYFFh/uIoTdS1cM3si/lqXxs8SgTEm5Dy1qYTnCsp48PJp/M2nZwBw16+28H5hFZMS7M6gJ+sjMMaElIMnz/DD1w9w1czxfHvFBYR5PYR5PdwwbzKZSdGs3X2C5rbOQIc5ptgdgTEmJJxtjnpiQzFhHmHp1BSe3vrharkeET63IJ3H3i7kzQOnuGHu5ECFOubYHYExJmQcrW7kcEUDn8hNIy7y499zJyVEMz8ziW0ltbS0213BWZYIjDEh4639FcRGhrFsakqf51w8LYW2Th/bjtaOYmRjmyUCY0xIKD/dTGFlA5/ITSUirO+PtvTEaLKSY9hcVI3PFkQELBEYY0LE5qJqwr1C3pTkAc9dNjWF6sY2SqoaRyGysc8SgTEm6NU1tfNB2WnmZSQSHeEd8PwLJ8UT7hV2H68bhejGPksExpig98L2Mto7td++ge4iwjxcMGEce8vrrXkISwTGmBDwwrYyMpKimZwYPejnzE5PoKG1w5qHsERgjAly+0/Us/9EPQsyE4f0vJkTrXnoLEcTgYisEJGDIlIoIg/3c95iEekUkVucjMcYE3pe3nGcMI8wN2NoiSAizMP08eM4ePIM6vLmoUElAhF5UUSuE5FBJw4R8QKPAdcAFwJ3iMiFfZz3L8C6wV7bGGMAOn3KKzuO88mZ44ntZQLZQGZMiON0cztHKhsciC54DPaD/RfAncBhEfmRiMwcxHOWAIWqWqSqbcAzwI29nPc14EWgYpCxGGMMABsKq6g408rNC9LP6/kzxo8D4J2DlSMZVtAZVCJQ1TdV9S5gIVAC/FFENorIvSIS3sfT0oHSbttl/n3niEg6cBOwsr/XF5EHRKRARAoqK939D2aM+dBL28uIjwrjylnjz+v5SbERpMVF8u4hd3+uDKWpJwW4B/gKsAP4CV2J4Y99PaWXfT0b4n4MfFtV+y36oaqPq2qequalpaUNNmRjTAhraO3gD3tPcsO8yUSGDTx3oC8zJsSxpbjG1RVJB9tH8BKwHogBblDVz6rqs6r6NSCuj6eVAZndtjOA8h7n5AHPiEgJcAvwcxH53BDiN8a41B/2nKSl3cfNC8+vWeis3AnjaOvwsbWkZoQiCz6D7V35laqu7b5DRCJVtVVV8/p4Tj6QKyI5wHHgdrr6Gc5R1Zxu13sSeE1VXxls8MYY93pxWxlTUmJYOMxFZrJTYgnzCJuLqrl8hjtbHAabCP4fsLbHvk10NQ31SlU7ROQhukYDeYFVqrpXRB70H++3X8CYQAn0MotmYKU1TWwqquZbV88Y9rKTEWEe5mUmsrmoeoSiCz79JgIRmUhXB2+0iCzgw3b/eLqaifrlv4tY22NfrwlAVe8ZRLzGGMPLO44DcNMwm4XOWjY1mZXvFtHQ2tHrOgahbqB3/Bm6OogzgP/otv8M8F2HYjImJNmdxshQVV7YVsYl01LISBrw++igLJuawmNvH6GgpIYrLvjoCKT+/t0gNP7t+k0EqvoU8JSIfF5VXxylmIwxpk/5JbUcq2niG5/KHbFrLpqSRLhX2Fz08UTgBgM1DX1BVf8HyBaRb/Y8rqr/0cvTjDHGMS9uKyM2wsuK2RNH7JoxEWHMzXBvP8FAw0dj/X/GAeN6+THGmFHT1NbB73ef4No5k4iJGNm2/LzsJPaW17lyLeOBmoZ+6f/zH0cnHGNMb6x/ocu6vSdpaO3glkUZI37tRVlJ/LKziD3H68jLHniVs1Ay2All/yoi8SISLiJviUiViHzB6eCMMaa7Z7aWkpUcw2IHPqgXTumaj+DGRe0HW2Li06paD1xP14zhGcDfOhaVMcb0sP9EPVuKa/jCsiw8nuHNHehNalwk2Skxlgj6cbaw3LXA06rq3rnYxpiAeHJDCdHhXm7Lc64pbGFWEtuP1bpufYLBJoL/FZEDdNUGektE0oAW58IyxpgP1TS28crO49y0MJ2EmL4KHg/fwilJVDW0caymybHXGIsGW4b6YeBiIE9V24FGel9bwBhjRtx/ry+irdPHfcuzHX2dRS7tJxjK+KtZdM0n6P6c34xwPMYY8xHVDa08tbGE6+dOZvp4Z0etz5gwjrjIMLYfq+XmhSM/MmmsGlQiEJHfAtOAncDZQbaKJQJjxrT+hp2qKoUVDRytaaKuuZ3YCC/pSTHkjo8jKvz86/uPtMffK6KlvZO/umrkZhL3xesRFmQlsu3oacdfaywZ7B1BHnChuq0HxZgRVNfcztHqRspqm6lqaOVMSwftnT4iwjysP1zJwqwkVsyeSGbyyNTP6U9JVSOv7SqnvK4FAeIiw2hq76TTV0WYR5ifmcj8zEQunBzveCz9OXzqDKs2FHPTggymj+9r6ZORtTAriZ/+6TBnWtoZF+Vcf8RYMthEsAeYCJxwMBZjQkp7p4/NRdW8sfcUfzpQwfHTzQCEeYTUuEjio8MI94bT1uFjb3k9r+85yQ/W7ufKmeP57rUzHWsG2VxUzWu7ykmIDueWRRnMnpxARJiHTp9yrKaJnaWn2Vlay7X/tZ4l2cl86ZJsPn3RBMK9vXcpNrR2sKv0NDtKT7O7rI5TZ1o4WddCVLiX8eMiyU6NZdbEeCLCBr0gIgA+n/Kdl3YTGxnGd68dzDLpI2PRlCR8Ch+U1nFpbuqovW4gDTYRpAL7RGQr0Hp2p6p+1pGojAlSbR0+3j5Yweu7T/DWgQrOtHQQFe7hstw05mcmMiUlhkkJ0Xh7jIO/c2kWx08383xBKaveL2bFj9fz9atyeeiT00d0zPz6w5W8vuckF0wYx22LMz/SBOT1CDmpseSkxrLioon4VHlqUwl/uXo746LCuGRaCtPHxzEuKpzmtk6O1TSxr7yeQxVnONtWkJMaS3piNAnRXefsLD3NluIawr3CoinJQ1r45SdvHabgaC3/estcUuIiR+zvYCDzsxIR6eowtkTwUf/HySCMCXYHTtbzfEEZL+84Tk1jG0kx4ay4aCKfvmgil05PJTrCO2A54/TEaL7xqRl8cdkU/vF/9/EffzzEjmO1PHZXn+s/DUl+SQ2v7znJnPQEbluciaefBV2iI7zcuTSL+y7N4Z2DFbyx9xSbi6t5c38Fnb6uT/3JCVHMmDiOa+ZMZEFWEvMyEkiMiQA+7JvwqXK0uoltR2vZWlxNfkkNx0838RdXTCc9MbrP13++oJSfvHWYWxZlcKsD5ST6Ex8Vzozx4yg46p7pUoNKBKr6rohMAXJV9U0RiaFr1TFjXKutw8eaD8r5zaYSdpXVEe4VPjVrAn+Wl8lluamE9dGUMpCUuEh+cvt8Fmcn8f01e7nniXyuuWgikcPowD1S2cCrO4+TOz6OW/My+k0C3Xk9wlWzJnDVrAkAdHT6aO9UwrzSZ1NRdx758C7jqpnjefdQJc/ml/Jsfim35mVy98VTmDnxw36IhtYOfvLmIf57fTGXTEvhn2+a85EVyAZKpiNl4ZRE1u4+6ZqJZYMdNXQ/8ACQTNfooXRgJXCVc6EZMzY1tXWwessxfv1+MSfqWpgxIY7vXX8hn1uQTnJsxHlds7cPOK/Hw615mTxfUMrJuhbuuST7vEbzVDe0snrLMVLjIrljSRZhnvNLUABhXg9h55mPkmIj+NyCdP7z9vk89nYhLxSUsXrLMbJTYpgxYRytHT62FtfQ3N7J3RdP4ZHrZg25X2GkzM1I5OmtpRytdsfEssE2Df0lsATYAqCqh0XEfas3GFfzqfLyjjJ+9PoBTtW3smxqMj+8eQ6Xz0g79611pL+xzstIxCvCs/mlrNpQzL2X5BAdMfhP4ua2Tp7adBQRuPvi80skIy09MZp/vmkOf/PpC3h153E2F1VzpLKRqHAPs9PjWTQlmazkGF7cdjxgMc7NSADggzJ3DCMdbCJoVdW2s7/s/kll7rhnMgYoP93MqzuPU1rbzNyMBH5250JHKmD2ZnZ6Al6PsHrrMVZtKOa+5YNLBu2dPlZvPUptYxv3XZpz3ncrTkmOjeDe5Tncuzzn3L7RavoZyIwJ44gK9/BBad2oDVsNpMHed70rIt+laxH7q4Hngf91LixjxgafKu8dquQX7xyhtqmdR2+ZyytfXT5qSeCsWZPiuWtJFifrW1i1oZjmtv4XT1FVvr9mL0cqG7lpQTo5qbH9nm8+Ktzr4aLJCexyyR3BYBPBw0AlsBv4c2At8PcDPUlEVojIQREpFJGHezl+o4jsEpGdIlIgIpcOJXhjnNTa0cnTW4/xh70nmTlpHN+4Kpdb8zIdKYE8GDMnxXPX0q5k8OsNRdQ1t/d6ns+n/N/X9rF6yzE+kZt2rs6+GZq5GQnsKa87N0oqlA121JBPRF4BXlHVysE8R0S8wGPA1XStYZAvImtUdV+3094C1qiqishc4Dlg9GaOGNOHhtYOntxYzInTLVw7ZxLLp6V8ZPRKoMycGM8Xlmbx9NZSfvZ2IZ+bPxlVPRfbyboWvvfqHt7Yd4r7lucwNc3uBM7X/MxEnthQQsWZFiYl9D3UNRQMtHi9AN8HHgLEv6sT+Kmq/t8Brr0EKFTVIv+1nqGrYum5RKCqDd3Oj8X6HcwYUN/Szq/WF1PX3MYXewxvHAsumBjPV6+Yxuqtx/jdlmNsLa5hXmYiNY1tbDpSjU+Vv79uFl++NIent5YGOtygNTcjEYCy2mZ3JwLgG8ByYLGqFgOIyFTgFyLy16r6n/08Nx3o/ltYBizteZKI3AT8EBgPXNfbhUTkAbqGr5KV5Z71Wc3oq2lsY9X7xdQ3t3PvJTlkj9G29fHxUXztylw+KDtNWW0T7x2qJC4qjM8tSOcvLp9GVoqz9YrGSqeuk7JTYoiPCqOstmnU+4RG20CJ4G7galWtOrtDVYv86xW/AfSXCHq7j/7YN35VfRl4WUQ+AfwT8KleznkceBwgLy/P7hqMI1raO7nvyXxqGtv40iXZYzYJnOX1CAuzkvi3W+cFOpSQJCLMy0yksKJh4JOD3ECdxeHdk8BZ/n6CgcrylQGZ3bYzgPK+TlbV94BpIuKO4h5mTPH5lG899wEflJ3mtsWZTEsL/SGDZmBzMxI4Vd9Ce6cv0KE4aqBE0HaexwDygVwRyRGRCOB2YE33E0Rkur8fAhFZCEQA1QNc15gR99/ri/j97hM8vGImF01OCHQ4ZoyYl5GIT7vmkYSygZqG5olIfS/7BYjq74mq2iEiDwHr6KpLtEpV94rIg/7jK4HPA3eLSDvQDNxmax6Y0bbtaC2PrjvItXMm8sAnploHqzlnXuaHHcZTUsZ2U+Fw9JsIVHVY89FVdS1dcw6671vZ7fG/AP8ynNcwZjjqmtr5+tM7mJQYxQ9vnjsmhoiasWNCfBTxUWHn1pIIVUNZs9iYkKKq/N2LH1BxpoUXHryEhGh3rEZlhiYjKYbSmtAuPheY0n7GjAHPFZSybu8pvr1i5rkmAGN6ykiKprqxbcCyHsHMEoFxpZN1Lfy/1/azbGoy93UremZMT+lJXZPJQrl5yBKBcR1V5e9f2U27z8ePbp4bsNpBJjhkJHZNziurDd3mIesjMK6z5oNy3txfwTWzJ7LxSDUbj9iIZdO36AgvKbERlNWG7h2BJQLjKtUNrfzj/+4jIyma5dNt7mIgBGN5ioykaIqrGgMdhmOsaci4yo9eP0B9czs3Lxz8ur3GZCTFUN/SQX0fpb+DnSUC4xrbjtby/LYyvnxZDhPj+50PacxHZPg7jEO1eciahowrdPqU7726h4nxUXz9ylxe3dln2at+BWOzhhm+yYnReATKTjdx4eSxVZZ8JFgiMCGp5wf25qJq9pbXc/vizPNOAsa9wr0eJsRHcTxE7wisaciEvIbWDt7Yd5JpabHMSbeCcub8pCdGU1bbTCiWQ7NEYELem/tO0dbh44a5k62WkDlv6UnRNLd3UtsUeh3GlghMSDtZ10J+SQ3LpqYw3jqIzTCE8sQySwQmZKkqa/ecICrcy5Uzxwc6HBPkJiRE4vVISJaasM5iMyj9jZa5c+nYXEf60KkGCisauG7OJGIi7FfdDE+Yx8OkhKiQHEJqdwQmJHX6uu4GUmIjWDo1tBceN6MnPTGa8tPN+EKsw9gSgQlJBUdrqDzTyjWzJxLmsV9zMzIykqJp7fBR1dAa6FBGlP0PMSGnpb2Ttw9UkJUcw6xJoTf5xwROelJXh3GozSewhlMTtPrqt1h/uJL6lg7+bHGmDRcdBptF/XFpcZGEe4Wy080syEoKdDgjxu4ITEhpae/k3UOV5I6PY2pqXKDDMSHG6xEmJ0aH3B2BJQITUjYUVtHU1snVF04IdCgmRGX4O4w7faHTYWyJwISMptYO3i+s4sJJ8WT423KNGWnpSTF0+JSKMy2BDmXEOJoIRGSFiBwUkUIRebiX43eJyC7/z0YRmedkPCa0rS+soq3DZ3cDxlEZif41jEOoecixzmIR8QKPAVcDZUC+iKxR1X3dTisGLlfVWhG5BngcWOpUTCZ0Nbd1srmomtnpCUywUhJmiIbSMZ4cF0FUuIey2mbysp2LaTQ5eUewBChU1SJVbQOeAW7sfoKqblTVWv/mZiDDwXhMCNtcXE1rh4/LZ6QFOhQT4jzi7zAOoVITTiaCdKC023aZf19fvgy83tsBEXlARApEpKCysnIEQzShoK3Dx4bCKi6YMI7J/tt2Y5yUkRjDyboWOjp9gQ5lRDiZCHobwN1rN7uIfJKuRPDt3o6r6uOqmqeqeWlp9o3PfFTB0Rqa2jq54gL73TCjIz0pmk5VTtaHRoexkxPKyoDMbtsZwMeWhhKRucCvgGtUtdrBeEwI6vD5WH+4iuyUGKakxAY6HOMSobaGsZN3BPlArojkiEgEcDuwpvsJIpIFvAR8UVUPORiLCVE7j52mrrmdKy6wMtNm9CRGhxMT4Q2ZkUOO3RGoaoeIPASsA7zAKlXdKyIP+o+vBL4HpAA/95cC6FDVPKdiMqFFVdlwpIpJCVHkjrdZxGb0iAgZSdGUnQ6NRWocrTWkqmuBtT32rez2+CvAV5yMwYSuoqpGTtW38vmF6VZTyIy69MQYDp+qoKmtI+jXuwju6I2rbSysIjbCy9yMxECHElBWHC4wMpKiUWBveT2Ls4N7zQsrMWGC0tHqRg6cPMOSnGTCvfZrbEZfur/DeFdZXYAjGT77H2SC0lMbjyICS3NSAh2Kcan4qHDio8LYVXY60KEMmyUCE3QaWjt4vqCU2ekJxEeHBzoc42LpSTHstjsCY0bfi9vKONPawfJpqYEOxbhcRlI0RVWN1DW3BzqUYbFEYIKKz6c8ubGE+ZmJZCZbqWkTWOn+kiZ7jwf3XYElAhNU3j1USXFVI/cuzw50KMacK0n9QZA3D1kiMEFl1YZiJsRHcu2cSYEOxRhiIsOYkhLDjmO1A588hlkiMEGjsOIM6w9X8cVlU2zIqBkzFmYlsf3YaVSDd+lKm1BmBrS3vI7XdpVTVttMfXM7cVFhZCbFsDg7mYkJo7cIzJMbS4gI83DHkqxRe01jBrIwK5GXdxynrLY5aPutLBGYPhVXNfKD3+/jzf0VeD1CVnIM2amxNLR0kF9Sw6aiahZnJ/PZ+ZOJi3T2V6muqZ0Xtx3nxnmTSYmLdPS1jBmKhVOSANh+rNYSgRk5A5UMuHOps9+IVZXVW4/xT6/tI8zj4e9WXECE1/OReipNrR28e7iS9w9XcePP3ud/vrKUSQnOLQrzbMExmts7uXd5jmOvYcz5uGDCOGIivGw7WsuN8/tbe2vsskTgIoNJMG0dPr778m5e2FbGZbmp/Put8xgfH/Wx58ZEhnHN7EnMmDCO5/JLuXXlJp6+f5kj34jaOnw8saGEZVOTuXBy/Ihf35jhCPN6mJeRyPYg7jC2HjdzTl1zO/c8sZUXtpXxV1fl8tS9Sxg/wELw09LiWH3/Muqb27nvyXxHJtas+aCcE3Ut/Pnl00b82saMhIVTEtl/4gxNbR2BDuW8WCIwANQ2tnHLLzaSX1LDv986j7++egYez+BKO8/JSGDlFxdRUt3IQ6u30+kbudETPp/yy3ePMHPiOK6whenNGLUwK4lOnwZtATprGjKU1Tbxm01H8Qj85r6lXDxtaIXczjYbXT9nMi/vPM6D/7ONT3ZbMWw4fRp/OlDB4YoGfnzbfFtzwIxZC7I+7DBeNjX4CiFaInC5feX1PFtwjLjIMJ5/8GKmjx933tfKy07iSFUDb+0/RU5KLNmpw19DeOW7R0hPjOb6uTaBzIxdybERTE2NZfvR4KxEak1DLqWqvHeokt9tOcqE+CgevHzasJIAdC3f97n56STGRPBsQemw20sLSmooOFrL/ZflEGYTyMwYtyAriR3HaoNyYpndEbhQa3snL2wvY295PbMnx3PLokwiwkbmgzYq3MvtizP55btFvLT9OHcNo1nokZf3EBPhBcRW4TJj3sIpiby4vYxjNU1MSRn+3fBosq9ZLnOyroWfv3OE/SfquWb2RO5YkjViSeCsjKQYPjN7IvtO1LOpqPq8rrHpSDUHT53hsumpIx6fMU5Y6O8n2HY0+IaR2v8wl2hu6+TN/ad47O1Cmto7uW95DpflpjnWAbt8WgozJ47j9T0n2TPEEr0+n/LD1/eTEB3OJdNtzQETHGZMGEdcZJglAjP2NLR28NTGEj75b+/wpwMVzE6P5xtX5TI1Lc7R1xURPr8wg9gILw+t3k5D6+D7C17YXsausjo+NWuCFZczQcPrERZNSWJrcU2gQxkyR/+XicgKETkoIoUi8nAvx2eKyCYRaRWRv3EyFrdoaO1g/4l6nssv5aHV21n6gzf5/pq9pCdFc/9lU7ltcRaxDtcFOis2MozbFmdxrKaJ7760e1CdaCfrWvin1/axJDuZBVmJoxClMSNn2dQUDlc0UNXQGuhQhsSxTwQR8QKPAVcDZUC+iKxR1X3dTqsBvg58zqk4gp1PldKaJgorGjhR10JtUxu/fO8ITW2dtLR30ulTfMCqWTwAAAzHSURBVKr4fNCp+pHJXKlxEdwwbzK3Lc5kQVZSQDpcc1Jj+danL+DRdQfJTo3lm1fP6PPcjk4ff/vCB7R3+vjXW+ay8cj59S8YEyhLpyYDsLW4JqjWzHDyq+ESoFBViwBE5BngRuBcIlDVCqBCRK5zMI6g1NreyaaiarYW13C6uR0BUuMiSYmL4IKJXUWuIsO8hHkEr0fweASvCHFRYUxOjOaiyfHkpMQOenYwDFyL6Hx99YpplFQ18l9vHSYu0sv9l039WN+EqvIPr+5l/eEqfnTzHLJTYy0RmKAzJz2BmAgvm4uqLRH4pQOl3bbLgKXncyEReQB4ACArK7Rr0asq247W8oc9J2hs62R6WhyfmT2RGePHER3hBZyvPjrSRIR/vnkOjW0d/PPaAxRVNvKda2aREBMOwJmWdv7hlT28srOcr14xjdttvQETpMK9HhZNSWJLUXD1EziZCHr7KnpeMy1U9XHgcYC8vLzgm60xSKeb2vjWcx/w1oEKpiTHcPecSUFb37yncK+Hn92xkH9LOcjP3znC73ed4LIZqXg9HtYfrqS+uZ1vXj2Dr105PdChGjMsy6am8Oi6g1Q3tAbN2hlOJoIyILPbdgZQ7uDrBbWDJ89w/28KOFnXwnVzJnHxtBQ8IVZbx+MR/m7FTG6YN5lfvHOEPeV1NLV2cuUF47lr2RQW+Rf4MCaYLZ+eyqPrDrLhSDWfnTc50OEMipOJIB/IFZEc4DhwO3Cng68XtPJLavjyk/lEhXt55s+XceDEmUCH5KhZk+L5rzsWBDoMYxwxJz2BxJhw1h+qtESgqh0i8hCwDvACq1R1r4g86D++UkQmAgVAPOATkW8AF6pqvVNxjRVnO2aLqxp5cmMxCdHhfOmS7JBPAsaEOq9HWD49lfWHq1DVoKia6+iAclVdC6ztsW9lt8cn6WoycqWu8s8lJMZEcP9lUx1f99cYMzo+kZvK73edoLCigdwJwyvmOBrsk8dB/Q3HPFnfwhMbSoiJ8HLf8pwhJYH+rhtsI4qMCUWX5nYtovTuocqgSAQ2fz8AapvaeOL9YsK8wpcvnUpCdHigQzLGjKD0xGimj4/jnYOVgQ5lUOyOYJS1tnfy201Haff5+PNPTCM5NmJEr2/lmo0ZGz41awK/Wl9EXXP7mP+yZ3cEo8inyrMFpVScaeGOJVlMGGBheGNM8Pr0RRPo8CnvHKwIdCgDskQwitbtPcmBk2e4bu5kcoe5GpgxZmybn5FI2rhI3th7KtChDMgSwSjZdrSG9YerWJqTzMVBuLi1MWZoPB7hU7Mm8M7BClo7OgMdTr8sEYyC4qpGXtlRzvS0OK6fGxwTTIwxw7di9kQa2zrHfKexJQKH1TS28bstR0mKDeeOJVl4h1AN1BgT3JZPSyE1LoKXtx8PdCj9slFDDmpt7+S3m0vwqXL3suxz1UPdxkYyGbcK83q4Yd5kfrf5GHVN7ecq7o41dkfgkE5f1wihyjOt3LlkCqnjgqMKoTFmZN28IIO2Th+/330i0KH0yRKBQx5dd/DcCKHp451dH9gYM3bNTo8nd3wczxaUDnxygFgicMBL28tY+e4RluQksywnOdDhGGMCSET4wrIpfFB6mu3HagMdTq8sEYyw/JIaHn5xNxdPTeGGuZODovKgMcZZn1+UwbjIMJ7YUBLoUHpliWAE7T9Rz31P5pORFM3P71poI4SMMQDERYbxZ4szeX33CcpPNwc6nI+xRDBCjlY3cveqrcRGhPHbrywlaYRrCBljgtu9y7MRgZ/+6XCgQ/kYSwQj4FR9C1/89VbaO3389stLSE+MDnRIxpgxJiMphruWTuG5gjKOVDYEOpyPsEQwTCVVjdyyciNVDa08cc/ioKg9bowJjIeunE5UmIcfrt2PqgY6nHMsEQzD3vI6blm5iYaWDlbfv4wFWbb4ujGmb6lxkXztqlze3F/BqzvLAx3OOZYIztP7h6u4/ZebCfcKzz94MfMzEwMdkjEmCNx/2VQWZiXyvVf3cHyMdBxbIhiijk4fP3nzMF9ctYVJiVG88BeXMN1KShtjBsnrEf79z+ajCvc+sZW65vZAh2SJYCj2ltdx8y828p9vHuLGeZN55S+XW8ewMWbIclJj+eUXF1Fc1ci9T2ylprEtoPFYIhiEwoozfPO5nVz/0/cpq23msTsX8uPbFxATYTX7jDHn55Lpqfz0jgXsKa/nxsfeD+isY0c/yURkBfATwAv8SlV/1OO4+I9fCzQB96jqdidjGqzKM628fbCCV3YcZ+ORaqLCPTxw2VS++snpY379UWNMcFgxexLPPhDFV3+3nZt/vpGbF6Rz36U5zE5PGNU4HEsEIuIFHgOuBsqAfBFZo6r7up12DZDr/1kK/ML/p6NUlbZOHy3tPuqa2qk400LlmVaKqxvZV17PvhP1FFU2AjAlJYZvXj2DLyybMuILzRtjzIKsJP74zcv56VuH+c2mo7y04zhT02JZPi2V3AlxTEuLIys5huTYCGIivI6UrXHyjmAJUKiqRQAi8gxwI9A9EdwI/Ea7BtRuFpFEEZmkqiNer/WNvSf5zku7aW7vpKW9E18fQ3gzkqK5cFI8n1+YweUz0rhocrzVCzLGOCouMozvXDuLr35yOq/uPM4f953i5R3HaWjt+Mh5D3xiKt+9dtaIv76TiSAd6F53tYyPf9vv7Zx04COJQEQeAB7wbzaIyMFeXi8VqBpOwABHgQ3DvUhgjMj7D2Jufv9ufu8Q4Pd/1yi+1iP/Ao98fPdg3/+Uvg44mQh6+xrd83v4YM5BVR8HHu/3xUQKVDVv8OGFFnv/7n3/bn7vYO9/JN6/k6OGyoDMbtsZQM+pdIM5xxhjjIOcTAT5QK6I5IhIBHA7sKbHOWuAu6XLMqDOif4BY4wxfXOsaUhVO0TkIWAdXcNHV6nqXhF50H98JbCWrqGjhXQNH713GC/Zb9ORC9j7dy83v3ew9z/s9y9jqQKeMcaY0Wczi40xxuUsERhjjMuFVCIQkX8SkV0islNE3hCRyYGOaTSJyKMicsD/d/CyiLimNraI3Coie0XEJyKuGUooIitE5KCIFIrIw4GOZzSJyCoRqRCRPYGOJRBEJFNE3haR/f7f/b8632uFVCIAHlXVuao6H3gN+F6gAxplfwRmq+pc4BDwnQDHM5r2ADcD7wU6kNHSrYzLNcCFwB0icmFgoxpVTwIrAh1EAHUA31LVWcAy4C/P998/pBKBqtZ324yll8lpoUxV31DVs3PSN9M1L8MVVHW/qvY24zyUnSvjoqptwNkyLq6gqu8BNYGOI1BU9cTZIp2qegbYT1dlhiELuTrKIvID4G6gDvhkgMMJpPuAZwMdhHHUYMq4GBcQkWxgAbDlfJ4fdIlARN4EJvZy6BFVfVVVHwEeEZHvAA8B3x/VAB020Pv3n/MIXbeNvxvN2Jw2mPfuMoMq0WJCm4jEAS8C3+jRKjJoQZcIVPVTgzx1NfB7QiwRDPT+ReRLwPXAVRpik0SG8G/vFlaixeVEJJyuJPA7VX3pfK8TUn0EIpLbbfOzwIFAxRII/oWAvg18VlWbAh2PcdxgyriYEOVf2OvXwH5V/Y9hXSuUvjSKyIvABYCProrSD6rq8cBGNXpEpBCIBKr9uzar6oMBDGnUiMhNwE+BNOA0sFNVPxPYqJwnItcCP+bDMi4/CHBIo0ZEngauoKsM8yng+6r664AGNYpE5FJgPbCbrs88gO+q6tohXyuUEoExxpihC6mmIWOMMUNnicAYY1zOEoExxricJQJjjHE5SwTGGONylgiMMcblLBEYc55EJHuoJZBF5LNny0WLyP8Rkb/xP77HbWXTzdhhicCYAfjLPY8IVV2jqj/q5dA9gCUCExCWCIzr+Rc0+qtu2z8Qka/7F/1YTdfMzb6EichT/sWAXhCRGP81SkQk1f84T0Te8T++R0R+1uP1bwHygN/5F1WKHuG3aEy/LBEY01Wv5UsAIuKhq2bPcbrq/T+iqv0t9nEB8Lh/MaB64KtDfXFVfQEoAO5S1fmq2jzUaxgzHJYIjOupaglQLSILgE8DO+iq17RVVYsHeHqpqm7wP/4f4FLHAjXGIUFXhtoYh/yKrnb6icAq/77GQTyvZ7Gus9sdfPhFK2q4wRnjJLsjMKbLy3Stf7sYWDeE52WJyMX+x3cA7/sflwCL/I8/P4jrnAHGDeF1jRkxlgiMAfxr/r4NPKeqnUN46n7gSyKyC0gGfuHf/4/AT0RkPTCY6z0JrLTOYhMIVobaGM51Em8HblXVw4GOx5jRZHcExvVE5EKgEHjLkoBxI7sjMGYAIpICvNXLoatUtbqX/cYEFUsExhjjctY0ZIwxLmeJwBhjXM4SgTHGuJwlAmOMcbn/D4XWmddAh8GwAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEHCAYAAACk6V2yAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAX0ElEQVR4nO3de3Rd5X3m8e+jiy3Zsi3bkrEtywgM2FwCNojbkKYTaDIJIZBOyIQkZJq0CbPWkMltVjtJpk3Imnama6aLNu0knRBCCwm5FcgFmikQIIRMiMH4AhgbDLbx3ZYv8kW2df3NH2cLhCysI52zdY62ns9aWj5nn733+9sCP+f1u/d+tyICMzPLnopSF2BmZulwwJuZZZQD3swsoxzwZmYZ5YA3M8uoqlIXMFBDQ0O0tLSUugwzs3HjmWee2RsRjUN9VlYB39LSwooVK0pdhpnZuCHp1Tf7zEM0ZmYZ5YA3M8soB7yZWUY54M3MMsoBb2aWUQ54M7OMcsCbmWWUA97MLKMc8GZmGVVWd7KWo+8t33LCsg9furAElZiZjYx78GZmGeWANzPLKAe8mVlGOeDNzDLKAW9mllEOeDOzjHLAm5lllAPezCyjHPBmZhnlgDczyygHvJlZRjngzcwyygFvZpZRDngzs4xywJuZZZQD3swsoxzwZmYZ5YA3M8soB7yZWUY54M3MMsoBb2aWUQ54M7OMcsCbmWWUA97MLKMc8GZmGeWANzPLKAe8mVlGOeDNzDIq1YCX9DlJayU9L+n7kmrSbM/MzF6XWsBLagI+DbRGxHlAJXBDWu2ZmdkbpT1EUwXUSqoCpgA7Um7PzMwSqQV8RGwH/grYAuwEDkbEQ2m1Z2Zmb5TmEM1M4DrgNGA+MFXSjUOsd5OkFZJWtLW1pVWOmdmEk+YQze8BmyKiLSK6gfuAfzV4pYi4LSJaI6K1sbExxXLMzCaWNAN+C3CZpCmSBFwFrEuxPTMzGyDNMfjlwD3ASuC5pK3b0mrPzMzeqCrNnUfEV4CvpNmGmZkNzXeymplllAPezCyjHPBmZhnlgDczyygHvJlZRjngzcwyygFvZpZRDngzs4xywJuZZZQD3swsoxzwZmYZ5YA3M8soB7yZWUY54M3MMsoBb2aWUQ54M7OMcsCbmWWUA97MLKMc8GZmGeWANzPLKAe8mVlGOeDNzDLKAW9mllEOeDOzjHLAm5lllAPezCyjHPBmZhnlgDczyygHvJlZRjngzcwyygFvZpZRDngzs4xywJuZZZQD3swsoxzwZmYZlWrAS6qXdI+k9ZLWSbo8zfbMzOx1VSnv/2vAv0TE9ZImAVNSbs/MzBKpBbyk6cDbgI8BREQX0JVWe2Zm9kZpDtGcDrQB/yBplaTbJU0dvJKkmyStkLSira0txXLMzCaWNAO+CrgQ+PuIWAZ0AF8YvFJE3BYRrRHR2tjYmGI5ZmYTS5oBvw3YFhHLk/f3kAt8MzMbA6kFfETsArZKWpwsugp4Ia32zMzsjdK+iuY/AXcnV9BsBD6ecntmZpZINeAjYjXQmmYbZmY2tLyGaCTdK+k9knznq5nZOJFvYP898GFgg6S/lLQkxZrMzKwI8gr4iPhFRHyE3FUwm4GHJf1G0sclVadZoJmZjU7eQy6SZpO7K/UTwCpy0xBcCDycSmVmZlaQvE6ySroPWAJ8B3hvROxMPvqhpBVpFWdmZqOX71U0t0fEzwcukDQ5IjojwlfJmJmVoXyHaP58iGVPFrMQMzMrrpP24CXNBZqAWknLACUfTcdT/5qZlbXhhmj+DbkTqwuAWwcsPwx8KaWazMysCE4a8BFxJ3CnpPdHxL1jVJOZmRXBcEM0N0bEd4EWSZ8f/HlE3DrEZmZmVgaGG6Lpf0BHXdqFmJlZcQ03RPPN5M+vjk05ZmZWLPlONvY/JU2XVC3pEUl7Jd2YdnFmZjZ6+V4H/86IOARcQ+5JTWcBf5xaVWZmVrB8A75/QrGrge9HxP6U6jEzsyLJd6qC+yWtB44B/1FSI3A8vbLMzKxQ+U4X/AXgcqA1IrqBDuC6NAszM7PCjOSRfWeTux5+4DZ3FbkeMzMrknynC/4OsAhYDfQmiwMHvJlZ2cq3B98KnBMRkWYxZmZWPPleRfM8MDfNQszMrLjy7cE3AC9Iegro7F8YEdemUpWZmRUs34C/Jc0izMys+PIK+Ih4XNKpwJkR8QtJU4DKdEszM7NC5DsXzSeBe4BvJouagJ+kVZSZmRUu35OsNwNXAIcAImIDMCetoszMrHD5BnxnRHT1v0ludvIlk2ZmZSzfgH9c0pfIPXz7HcA/AfenV5aZmRUq34D/AtAGPAf8B+DnwJ+mVZSZmRUu36to+iT9BPhJRLSlXJOZmRXBSXvwyrlF0l5gPfCipDZJXx6b8szMbLSGG6L5LLmrZy6OiNkRMQu4FLhC0udSr87MzEZtuID/98CHImJT/4KI2AjcmHxmZmZlariAr46IvYMXJuPw1UOsb2ZmZWK4gO8a5WevkVQpaZWkB/Ivy8zMCjXcVTQXSDo0xHIBNXm28RlgHTB9JIWZmVlhTtqDj4jKiJg+xM+0iBh2iEbSAuA9wO3FKtjMzPKT741Oo/U3wJ8AfW+2gqSbJK2QtKKtzZfYm5kVS2oBL+kaYE9EPHOy9SLitohojYjWxsbGtMoxM5tw0uzBXwFcK2kz8APgSknfTbE9MzMbILWAj4gvRsSCiGgBbgAejYgb02rPzMzeKO0xeDMzK5F8n8lakIj4JfDLsWjLzMxy3IM3M8soB7yZWUY54M3MMsoBb2aWUQ54M7OMcsCbmWWUA97MLKMc8GZmGeWANzPLKAe8mVlGOeDNzDLKAW9mllEOeDOzjHLAm5lllAPezCyjHPBmZhnlgDczyygHvJlZRjngzcwyygFvZpZRDngzs4yqKnUB5aqrp48d7cfo6eujqsLfg2Y2/jjg38Tnf7SaB57dSaXE+5Y1cdGpM0tdkpnZiLhrOoT71+zggWd3csPFzcyvr+Gfn9vB0c6eUpdlZjYiDvhBOjp7+LOfPs/S5nr+/H3n8fsXLqCzu49frN9T6tLMzEbEAT/IL9btpv1oN1+6+myqKiuYO72Gi1tm8dSmfXS4F29m44gDfpD71+xg/owaWgeMuV/cMou+gPW7DpWwMjOzkXHAD3DwaDePv9TGe86fR0WFXls+v76G+tpq1u5wwJvZ+OGAH+DBtbvo7g3ee8H8NyyXxDnzp/PyniN0dveWqDozs5FxwA/w6Po9NNXX8pamGSd8du78GfT0BS/tOVKCyszMRs4Bn4gIVm45QGvLTCSd8Pmps6dQW13Ji7sOl6A6M7ORc8AntrcfY8/hTi5cOPQNTRUSpzVMZfO+jjGuzMxsdBzwiZVb2gHeNOABWmZPYX9HF7sPHR+rsszMRs0Bn1i15QA11RUsmTftTddpaZgKwFOb9o9VWWZmo5ZawEtqlvSYpHWS1kr6TFptFcPKLe2cv6Ce6so3/5XMm1HLpKoKB7yZjQtp9uB7gP8cEWcDlwE3SzonxfZG7Xh3Ly/sOMiyhfUnXa+yQpw6a4oD3szGhdQCPiJ2RsTK5PVhYB3QlFZ7hXh++0G6e+Ok4+/9Whqm8uLuw7Qf7RqDyszMRm9MxuAltQDLgOVDfHaTpBWSVrS1tY1FOSdYlccJ1n4LZ00BYM22g6nWZGZWqNQDXlIdcC/w2Yg44V7/iLgtIlojorWxsTHtcoa0cssBmmfV0jht8rDrNtXXIsHq5EvBzKxcpRrwkqrJhfvdEXFfmm2NVv8NTvn03gFqqis5c04dq7ceSLkyM7PCpHkVjYBvA+si4ta02inUjoPH2X2ok2XNJz/BOtAFC+pZs+0gEZFiZWZmhUmzB38F8FHgSkmrk5+rU2xvVFa+muuJXziCR/ItXVjP/o4utu4/llZZZmYFS+2ZrBHxa+DESV3KzKot7dRUV3D2vOl5b7M06e2v2nqAhbOnpFWamVlBJvydrCu3HOD8ppPf4DTY4lOmUVNdweqtPtFqZuVrQgf88e5e1uZxg9NgVZUVvKVphgPezMrahA74tTtyNzgty/MKmoGWNtezdschunr6UqjMzKxwEzrgX7vB6dSR9eABljbPpKunz89pNbOyNaEDfuWWAyyYWcucaTUj3nZpMqzjYRozK1cTO+Bfbc/7BqfB5s+ooaFusu9oNbOyNWEDfkf7MXYdOj7iE6z9JLG0uZ7V2xzwZlaeJmzAr9yS3OA0yh48wLKF9Wxs6+Dg0e5ilWVmVjQTNuBXbWlnctXIbnAarP+GpzXuxZtZGZqwAb9yywHOXzCDSVWj/xW8ZcGM3MySPtFqZmVoQgZ8Z08va7cfKmh4BmB6TTWLGusc8GZWliZkwD+//RBdvX2jPsE60NLmetZsbffMkmZWdiZkwK8qwgnWfkub69nX0cW2A55Z0szKywQN+Haa6muZM33kNzgN9vrMkh6mMbPyMuECPiJYvmk/rS2F994BFs+dxuSqCt/wZGZlZ8IF/Ma9Hew90smlp80uyv6qKyu4oLmepzfvL8r+zMyKZcIF/PKNuSC+9PRZRdvnZafPZu2Ogxw85huezKx8TLyA37SPhrrJnN4wtWj7vPz02fQFPL3JvXgzKx8TKuAjguUb93Pp6bPIPRO8OJYtrGdSVQVPbtxXtH2amRVqQgX8lv1H2XXoOJedVrzhGYCa6kouWjiTJ19xwJtZ+ZhQAf+rDXsBuHxRQ9H3ffmi2azbdYgDHV1F37eZ2WhMqIB/dN1uTp09hUWNxRt/73fFGQ1EwK82tBV932ZmozFhAv5YVy+/eWUfVy6ZU9Tx935Lm+tpqJvEwy/sLvq+zcxGY8IE/G9e2UtnTx9XLTkllf1XVoirlpzC4y+2+UHcZlYWJkzAP7J+D1MnVXJJkU+wDvSOc07hcGcPv/XVNGZWBiZEwHf39vHQ2l387uLGguZ/H85bz2ygtrrSwzRmVhYmRMA/tn4Pe4908f4LF6TaTk11JVcumcMDz+6gs6c31bbMzIYzIQL+nme20VA3md89qzH1tm64pJkDR7t5cK178WZWWpkP+H1HOnl0/R7+7YVNVFWmf7hXLGqgeVYt31++JfW2zMxOJvMBf+eTr9LTF3zgonSHZ/pVVIgbLl7Ikxv38fKeI2PSppnZUKpKXUCaDnR0ccevN/Hu8+Zy5inTxqzdD17czNcfe5lbH36Rb3zkojFr18yK53tD/Cv8w5cuLEElo5fpHvxtT2yko6uHz73jrDFtt6FuMp/8ndP5+XO7Xns8oJnZWMtswD+//SDffmIT110wn7PGsPfe75NvO52Guknccv8LvvHJzEoikwHf0dnDp3+wiplTq/nye88tSQ11k6u45dpzWbO1nf/+83UlqcHMJrbMjcEfPNrNH935NJv3dnD3Jy5j1tRJJavlmvPns/LVdu74f5toqJvEzW8/I5V5cMysePr6gl2HjrO9/RjHuno53t1LdaWoqa5kw+7DNM2sZcqk8RGdqVYp6V3A14BK4PaI+Ms023tiQxtf+elath44yt996EIuX1Sc564W4otXL2F/Ryd/9dBLrNt1mC++ewkLZk4pdVlmBuw90smLuw6//rP7MBt2H6aja+gbFb/5q40ANNXXsmhOHUvmTuPsedM4e950FjXWUT0Gl2KPRGoBL6kS+DrwDmAb8LSkn0XEC8Vs53h3L3c9uZmfrt7B2h2HOHX2FO76w0vLItwh91Duv/7gUs48ZRpfe2QDD6/dzduXNPKvF8/hnHnTmTujhvop1Uyuqix1qWaZERF09vRx+HgPRzp7aD/axa6Dx9lx8DjbDhzlpd25QN975PXnN8yaOonFp0zjA63NnDGnjvU7D1EzqZKaqkp6+4Jj3b0sW1jPln1H2bDnCC/vOcI/vrKPrt7cObbqSnHGnFzgL2qsY359DfNm1NJQN5lpNVVMnVzFlOpKKirG7l/xafbgLwFejoiNAJJ+AFwHFDXgqysr+NYTm5g/o4avXnsuN1zSXHZhKYmb334Gv7+siW89sZH/+9yuE+50nTqpkklVFVRWiArptT8rKqDCwzqZFjHoPXHyzwe9P3F/g7YvtL1hth+8xvDbj7C+YdYfvCCAY9299PYN/Yuqra7krFPquHLJHBbPnc6SudN4fvtBptVUv2G9c+bPOGHb65Y2veF9d28fm/Z2sG7nIV7YeYh1Ow/zxIa93Ldy+5BtSzB1UhXVlbm/45UVolKiYdpkfvaptw65TSE0+JdXtB1L1wPviohPJO8/ClwaEZ8atN5NwE3J28XAi6kUNDYagL2lLqJIsnQs4OMpZ1k6Fhj74zk1IoachyXNHvxQ3c4Tv3wjbgNuS7GOMSNpRUS0lrqOYsjSsYCPp5xl6VigvI4nzTMC24DmAe8XADtSbM/MzAZIM+CfBs6UdJqkScANwM9SbM/MzAZIbYgmInokfQp4kNxlkndExNq02isTmRhqSmTpWMDHU86ydCxQRseT2klWMzMrrfK6Kt/MzIrGAW9mllEO+CKQ9C5JL0p6WdIXSl1PISTdIWmPpOdLXUsxSGqW9JikdZLWSvpMqWsaLUk1kp6StCY5lq+WuqZikFQpaZWkB0pdS6EkbZb0nKTVklaUvB6PwRcmmZLhJQZMyQB8qNhTMowVSW8DjgB3RcR5pa6nUJLmAfMiYqWkacAzwPvG438f5WaqmxoRRyRVA78GPhMRvy1xaQWR9HmgFZgeEdeUup5CSNoMtEZEWdy45R584V6bkiEiuoD+KRnGpYj4FbC/1HUUS0TsjIiVyevDwDqg6eRblafI6X8OZHXyM657aJIWAO8Bbi91LVnkgC9cE7B1wPttjNMAyTpJLcAyYHlpKxm9ZDhjNbAHeDgixu2xJP4G+BMgK0/FCeAhSc8k07CUlAO+cHlNyWClJakOuBf4bEQcKnU9oxURvRGxlNyd4ZdIGrfDaJKuAfZExDOlrqWIroiIC4F3AzcnQ54l44AvnKdkKHPJePW9wN0RcV+p6ymGiGgHfgm8q8SlFOIK4Npk3PoHwJWSvlvakgoTETuSP/cAPyY3hFsyDvjCeUqGMpacmPw2sC4ibi11PYWQ1CipPnldC/wesL60VY1eRHwxIhZERAu5vzePRsSNJS5r1CRNTU7kI2kq8E6gpFejOeALFBE9QP+UDOuAH43nKRkkfR94ElgsaZukPyp1TQW6Avgoud7h6uTn6lIXNUrzgMckPUuuY/FwRIz7Swsz5BTg15LWAE8B/xwR/1LKgnyZpJlZRrkHb2aWUQ54M7OMcsCbmWWUA97MLKMc8GZmGeWANzPLKAe82RiQ9DFJ80e4TUtWpm220nDA27iXTNk83DqSVMr/3z8GjCjgzQrlgLeyJ+m/DXxQh6S/kPTp5EEe3wOee5PtWpIHfXwDWAk0S/pjSU9Lerb/gRkD1vtW8iCNh5KpAJC0VNJvk/V/LGmmpLMlPTWonWeT119O9v+8pNuSL5bryc13fndyJ22tpIskPZ7MOvhgMm89yfI1kp4Ebk7nN2oThQPexoNvA38AkPTCbwC2k5vI6b9GxDkn2XYxuYeXLEten5lstxS4aMBsf2cCX4+Ic4F24P3J8ruA/xIR55P7IvlKRKwDJkk6PVnng8CPktf/OyIuTh6WUgtcExH3ACuAjyQzQfYAfwdcHxEXAXcAf5Fs/w/ApyPi8hH/lswGccBb2YuIzcA+ScvITeC0CtgHPBURm4bZ/NUBTzx654DtVwJLyAU7wKaIWJ28fgZokTQDqI+Ix5PldwL9Xwg/Av5d8vqDwA+T12+XtFzSc8CVwLlD1LQYOA94OJnb/U+BBUO0951hjs3spKpKXYBZnm4nN449l1yPF6Ajj+0GriPgf0TENweukDwIpHPAol5yve+T+SHwT5LuI/ewpQ2SaoBvkHtk21ZJtwA1Q2wrYO3gXnoyU6Qnh7KicQ/exosfk5v7/GJyM3eOxoPAHyYP/0BSk6Q5b7ZyRBwEDkj6nWTRR4HHk89eIfdF8Ge83nvvD/O9SRvXD9jdYWBa8vpFoFHS5Ukd1ZLOTeZ4Pyjprcl6HxnlcZoB7sHbOBERXZIeA9ojojc3zfuI9/GQpLOBJ5PtjwA3kgvqN/MHwP+RNAXYCHx8wGc/BP4XcFqy/3ZJ3yI3Vr+Z3JS+/f4x2c8x4HJy4f+3ybBMFblH161N9n+HpKOM/ovMDPB0wTZOJCdXVwIfiIgNpa7HbDzwEI2VPUnnAC8DjzjczfLnHryNe5JmA48M8dFVEbFvrOsxKxcOeDOzjPIQjZlZRjngzcwyygFvZpZRDngzs4z6/0HNuREHA8cEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3Rc93nm8e87Awx6B0iQAKvYRHUZoqodNzlylR3LtqTIlh07ipy4JruJUjZls9lN3/jEWsuKW+ISWXGVE9myJVuyOklJlCg2kWIFSBAA0TEApv32j5kBQRAdc+cOOM/nHBwCdy7uvADJeeZXrznnEBGR/BXwuwAREfGXgkBEJM8pCERE8pyCQEQkzykIRETyXIHfBcxVfX29W716td9liIgsKs8991yXc65hsscWXRCsXr2a7du3+12GiMiiYmZHpnpMXUMiInlOQSAikucUBCIieU5BICKS5xQEIiJ5TkEgIpLnFAQiInlOQSAikucUBCIieW7RrSyW+fvWs0cnPX7rlSuzXImI5BK1CERE8pyCQEQkzykIRETynIJARCTPKQhERPKcgkBEJM8pCERE8pzWESwSWgMgIl5Ri0BEJM8pCERE8pynQWBmN5jZPjM7YGZ3TfL4682sz8x2pD7+1Mt6RETkbJ6NEZhZELgbuB5oBbaZ2QPOud0TTn3cOfcOr+oQEZHpedki2AIccM4ddM5FgPuAGz18PhERmQcvg6AJODbu69bUsYmuNrMXzezHZnaBh/WIiMgkvJw+apMccxO+fh5Y5ZwbNLO3AT8A1p91IbM7gDsAVq7UdEkRkUzyskXQCqwY93UzcHz8Cc65fufcYOrzB4FCM6ufeCHn3L3OuRbnXEtDQ4OHJYuI5B8vg2AbsN7M1phZCLgZeGD8CWbWaGaW+nxLqp5THtYkIiITeNY15JyLmdkngIeAIPAV59wuM7sz9fg9wE3Ax80sBgwDNzvnJnYfiYiIhzzdYiLV3fPghGP3jPv888DnvaxBRESmp5XFIiJ5TpvOLVLxhKNjYATnHKlhlimFIzES6nATkSkoCBahnnCE+7cd40h3mJ1tffzv91zE8uqSs8470TfMZ7+9g+2He1haWczHXruGooKgDxWLSC5T19Ai45zjq08eor1/hKvPq2ProW5u+sJTHO8dPuO83nCED315Ky+39XPLlpW09Q7z2L5On6oWkVymIFhkjnWH6RqM8M6Ll/POi5fzH3dezcBIjNu/snUsDE4NjvLBL2/lSHeYf/lQC3/57gv5tcuaePxAF6cGR33+CUQk16hraJF5sbWPgoCxeXklABcsr+Jfbm/ho1/bxls/9zjvvGQZD+5sZ2Akyq1bVnKoa4hDXUOsX1pBIuF4sbWXN25a6vNPISK5RC2CRSSecOxs62NjYwXFhaf7+q9aW8d/feq1rFtSzvefb6O4MMBHrlnDxsbKsXOqSgppqCjiaHfYj9JFJIepRbCIHOoaYnA0xiXN1Wc9trq+jO9+/Bpg6ttarqwtZdfxfhLOEZhhppGI5A+1CBaRI6eGMGD90vJ5ff/K2lKGo3G6NE4gIuMoCBaR9v4RastC854CuqK2FEgOOIuIpCkIFpH2vhEaq4rn/f0NFUUUFwY0TiAiZ1AQLBKjsTjdQ5EFBUHAjBU1pQoCETmDgmCR6OgfxQHLKucfBADLq0voHBglrj0nRCRFQbBItPeNANBYdfZWEnNRVxYi4aBvOJqJskTkHKAgWCTa+0cIFQSoLi1c0HVqykIAdA9FMlGWiJwDFASLRHv/CI2VxQue/1+nIBCRCRQEi8TJ/hGWVhYt+DqVJYUEzRQEIjJGQbAIDI3GCEfi1JaGFnytgBnVpYV0hxUEIpKkLSYWgRN9yV1FqyYJgqm2k5hObVmIHrUIRCRFLYJFoK03OWOoumRhA8VptWUhdQ2JyBgFwSKQvs/AQmcMpdWWhRiOxhmOxDNyPRFZ3NQ1lIMmdvf8bHc7AYOK4swEQU2qi6k7HKEptLB1CSKy+KlFsAj0hqNUFhcSDGRm6+i6ck0hFZHTFASLQO9wlKoMdQvB6RaBBoxFBBQEi0JvOJKxgWKA4sIgoYIA/SPaZkJEFAQ5L+Ec/cMxqjOwhmC8yuIC+kdiGb2miCxOCoIcNzgSI+4cVRlsEUBy4HlALQIRQUGQ83pTu4RmaupoWkVxAQNqEYgICoKc15vaCqK6JNNdQ4X0D0dxTvclEMl3CoIcl75vQOa7hgqIJRwj0URGrysii4+nQWBmN5jZPjM7YGZ3TXPeFWYWN7ObvKxnMRociVEQMIoLM/tXVZlanKaZQyLiWRCYWRC4G3grsBm4xcw2T3He3wAPeVXLYjY4GqO8qABb4H0IJqooTi4q1ziBiHjZItgCHHDOHXTORYD7gBsnOe+TwHeBDg9rWbQGR2OUF2d+J5B0i0Azh0TEyyBoAo6N+7o1dWyMmTUB7wHume5CZnaHmW03s+2dnZ0ZLzSXpVsEmZZuEWgtgYh4GQST9WVMnKLyT8AfOOem3QbTOXevc67FOdfS0NCQsQIXg6HRGGUeBEFRanWxWgQi4uXuo63AinFfNwPHJ5zTAtyX6v+uB95mZjHn3A88rGvRSDjnWYsAtLpYRJK8DIJtwHozWwO0ATcDt44/wTm3Jv25mX0N+E+FwGkjkTgJh2dB4PXq4snunnbrlSs9ez4RmR/PuoacczHgEyRnA+0B7nfO7TKzO83sTq+e91wyOJp8t+5dEGh1sYh4fGMa59yDwIMTjk06MOyc+7CXtSxGY0HgwawhSK8u7sc5t6DpqfO5b7KI5A6tLM5h6SDwYrAYki2NWMKNPY+I5CcFQQ7zumso3dLoGtQNakTymYIghw2OxjCgNBT05PrpgOkaHPXk+iKyOOjm9TlscCS5hiCQ4e0l0saCYCBzQRBPOB7Zc5LjfcNUlRRy46VNntUvIpmhFkEOG/JwDQGM7xrKXBBsO9zNo6900j0UYdvhHl442pOxa4uINxQEOcyrfYbSykIFGNCZoTGCkWicR/acZE19GZ958wZW1Zby45fbCWswWiSnKQhymJerigGCAaMkFMxYi+CJA10MReK89cJGAmbceGkTw5E4Tx08lZHri4g3FAQ5bHA0RplHA8Vp5UUFGRsjeKm1j3VLymmuKQWgsaqYVXVl7D7en5Hri4g3FAQ5KhpPEI07z9YQpJUXF3BqaOFdQ12Do3QNjnJ+Y8UZxzcvr6S9f4RTmpkkkrMUBDkqHEluyFoa8jgIigoy0jW090TyXf+mZZVnHN+c+nr3CbUKRHKVgiBHhSPJAVav1hCkZapraE/7AI2VxdSUhs44XlsWYllVsbqHRHKYgiBHnW4ReB8EQ5E4w5FpbwkxreFInCOnhtg4oVso7fxllRztDi/oOUTEOwqCHJXNriFY2FqCo91hEg7WLSmf9PE19WW41HkiknsUBDkq3TVU4nWLILVOoXMBQXCsJ4wBzTUlkz6+oqaUgMHhU0Pzfg4R8Y6CIEcNZ7FrCBa2zcSx7jBLK4spKpi81lBBgOXVJRxREIjkJAVBjgpH4hQGjcKgt39Fp7uG5jeFNJFwHOsJs6K2dNrzVteV0dozzGhM4wQiuUZBkKPCkZjn4wOw8DGCg11DjEQTrJiiWyhtdV0psYRjZ2vfvJ5HRLyjIMhR4Ujc824hgIJggMrignkv+NpxrBdgxhbByroyALYd1iZ0IrlGQZCjshUEAPUVRfPuGnrhaA9FBQEaKoqmPa+8qIC6spB2IxXJQQqCHJWtriGA+vKiec8a2nW8n+XVJbO658CK2lJeONaLc25ezyUi3lAQ5KistgjKQ/MaI0gkHPvaB2isKp7V+StqSugcGOV438icn0tEvKMgyEEJ5xjOahAUzWv66NHuMMPROMsqZxkEqXGEHUd75/xcIuIdBUEOGonGcXi/qjitvryI/pHYnKd27m0fAJh1i6CxqphQQUDjBCI5ZlZBYGbfNbO3m5mCIwvSi8m8XlWcVl+eHOg9NccB473t/ZjBkorZBUFBIMCFyyvHZhqJSG6Y7Qv7F4Bbgf1m9tdmtsnDmvJetjacS6svT+4YOtdxgr0nBlhTV0aoYPbvDy5bWcPOtj6i8cScnktEvDOr/8HOuYedc78OXA4cBn5mZk+Z2UfMrNDLAvPR6S2os9Q1VDH7FsG3nj069rHtcPecWy2XrqhmNJZg74mBedUqIpk367dyZlYHfBj4GPAC8DmSwfAzTyrLY9luETSkuobmMoU0EkvQPRSZ9fhA2qUrqgHYcUzjBCK5YrZjBN8DHgdKgXc6597lnPu2c+6TwOR7D8u8Zb9rKBkEc+ka6hgYwQGNs5wxlNZcU0J9eREvaJxAJGfMtu/hS865B8cfMLMi59yoc67Fg7ryWjgSx4DiwuwEQUkoSFkoSNfA7AeLO/qTobF0lgPFaWbGpSuqNYVUJIfMtmvof01y7OmZvsnMbjCzfWZ2wMzumuTxG83sJTPbYWbbzey6WdZzThuOxiguDM5qtW6mJLeZmEuLYJRgwKgpC8188gSXrazmYNcQfeHonL9XRDJv2haBmTUCTUCJmV0GpF+ZKkl2E033vUHgbuB6oBXYZmYPOOd2jzvtEeAB55wzs4uB+4G8n5E0HIlnbepoWl3Z3FYXdwyMUF8eIhiYe1hdlh4naO3lVzY0zPn7RSSzZuoa+lWSA8TNwD+OOz4A/NEM37sFOOCcOwhgZvcBNwJjQeCcGxx3fhmgTWiA4Wickix1C6XVlxfN6Q5inQOjLKuefuvpqVzUXIVZcsM6BYGI/6YNAufcvwL/ambvdc59d47XbgKOjfu6Fbhy4klm9h7g/wBLgLdPdiEzuwO4A2DlypVzLGPxyeb2Emn1FUVsPzK7mTzReHLG0CWpd/ZzVVFcyMalFWzXltQiOWHaMQIzuy316Woz+92JHzNce7I+g7Pe8Tvnvu+c2wS8G/jLyS7knLvXOdfinGtpaDj330GGI/GsDRSn1ZcX0ROOEJvFQq+uwVEcsGSGraenc9XaOrYf6SYS08IyEb/NNFhclvqzHKiY5GM6rcCKcV83A8enOtk590vgPDOrn+G657zhaPbHCJZUFOHc7G5Z2ZnaoG6mexBM56q1tYxEE+xs0+whEb/N1DX0xdSffzGPa28D1pvZGqANuJnkNhVjzGwd8GpqsPhyIAScmsdznTOcc4xE45RmuUWwLLUwrL1/ZMZFYh0Doxin1x/Mx5Y1dQA8c7Cb16yqnfd1RGThZrug7G/NrNLMCs3sETPrGtdtNCnnXAz4BPAQsAe43zm3y8zuNLM7U6e9F3jZzHaQnGH0AZfndy0ZHI2RcNnbcC4t/eLf3jc847kdA6PUloUoDM5/D8LashAbl1bwzMG8zn2RnDDbBWVvcc79fmpgtxV4H/AL4BvTfVNqEdqDE47dM+7zvwH+Zk4Vn+P6hpNz67M9ayi9QvjELG4a0zkwsqBuobSr1tZy//ZWovEE/7G99azHb73y3J8YIJILZvuWLr2x3NuAf3fOdXtUT94bC4Istwhqy0KEggHa+6cPgnjC0TUYWdBAcdrV59UxHI3z/CxnK4mIN2YbBD8ys71AC/CImTUAut+gB9KrbbPdIjAzGquKaZ+hRdAzFCGecLO+B8F0rl1XT0HAePSVzgVfS0Tmb7bbUN8FXA20OOeiwBDJxWGSYX61CCA5TjBT11BHBmYMpVUUF9KyuoZf7O1Y8LVEZP7mMtp3PvABM/sQcBPwFm9Kym+9Po0RQHKcYKYWQedA8vFMBAHA6zcuYW/7wFgAikj2zXbW0NeBvweuA65IfWjXUQ/42SJYVlVMe/8I003c6hgYpaqkMGML3t6wcQkAr5zUjWpE/DLbWUMtwOZ8n9qZDX3DUYJmhBYwNXO+GquKicQS9ISj1E6xq2jHwGjGWgMAG5aWs7yqmL3tA1yxWusJRPww21ebl4FGLwuRpN5wlOJQEMviFtRp6UVlJ6ZYS+CcozPDQWBmvOn8pRzoGNB9jEV8MtsgqAd2m9lDZvZA+sPLwvJV/3A066uK05am1hKcnGIK6fG+ESLxREamjo53/ealROOOAx2DM58sIhk3266hP/eyCDmtbzjqy/gAwLKq5LbSU80c2p/qx1/I1NFvPXv0rGOxRIKiggB7TvRz/rLKeV9bROZnVkHgnHvMzFYB651zD5tZKeDPq9U5rnc44suMIUjOBAoGbMqZQ3tOJINgrvcpnklBIMCGpRXsaR8g4VxG78w2WfBoxbLImWY7a+g3ge8AX0wdagJ+4FVR+czPFkEwYDRWFnO0Ozzp43tO9FNdWuhJfZuXVTI0GuPYFM8tIt6Z7RjB7wDXAv0Azrn9JG8kIxnWG4761iIAWLekfMq++t0n+lmW4dZA2oalFQQsGTYikl2zDYJR59zYRvVmVoBuK5lx8YRjYCTmW4sAkkHwaucgicSZf70j0TgHOwdprJrf7SlnUhIKsra+nN0ntJ5AJNtmGwSPmdkfkbyJ/fXAfwA/8q6s/DQw4t+q4rT1S8oZiSZo6z1zCum+9gES7vQUUy+cv7ySrsHRsRvfiEh2zDYI7gI6gZ3Ab5HcWvpPvCoqX/WG/VtVnLZuSTnAWd1D6S4bT4OgseKM5xKR7JjtpnMJkoPDv+2cu8k59y9aZZx56e0l/FpHAKeDYH/HmV00u0/0UxYKUjPFiuNMqC4NpVYZKwhEsmna6aOWXN76ZyTvNGapQ3Hgn51z/zML9eUVv/YZmjjFsryo4KwWwa7j/WxaVpnRqZ2TWbekgicOdDIajXv6PCJy2kwtgs+QnC10hXOuzjlXC1wJXGtmn/W8ujyT3nk0Uxu6zVdDRRH7xwXBwEiUF4/1ZmUvoHVLykk4OHRqyPPnEpGkmYLgQ8AtzrlD6QPOuYPAbanHJIPGuoZ8HCMAWFJRxIGOwbFdSJ9+9RSxhON1G+o9f+5VdaUUBEzbTYhk0UxBUOic65p40DnXyenbV0qG9IWTM3T9nDUEsKSymIGR2NjCsl/u76Q0FKRllfctgsJggDX1ZQoCkSyaKQgi83xM5qFvOEpxYYACH7agHm9TYwVm8L3n2wD45StdXL22jlBBduo6r6GcjoHRKTe/E5HMmul/9iVm1j/JxwBwUTYKzCd9w1GqS7yblTNbNaUhrltXz3eea+XVzkGOdod53YaGrD1/eubSE/vPaoyKiAemDQLnXNA5VznJR4VzTl1DGdYbjlJVkhu/1ve3rKCtd5gPfPEZCoPGGzdlb0eRxqpiykJBnjigIBDJBn/7IOQMfcNRqkpzIwiu37yUmtJCYokEX//olayoLc3acwfMOG9JOU8c6Jr2tpkikhmzvR+BZEHfcDSrL7jTKS4M8p2PX0N5UcHYDWuyaV1DOS+19vHKyUE2plYci4g3FAQ5pG84yoU50jUEyUFbv6THCR7f3znvIOgLR/nKk4cYGo1xXkM5b7lgKQUBNYJFJtL/ihySHCzOnSDwU3VpiLX1ZfMeJxiJxvnNr2/nUOcQJanxhqdfPZXhKkXODQqCHBGJJQhH4jkzWJwLrl1Xz9ZD3URic7+p/d2/OMDWQ93c1NLMx65by6bGCh7Z20F/atGeiJymIMgR6VXF1TkyWJwLrl1XTzgS58XW3jl9X99wlK89eZi3XdTIJc3VALz9omUkEo6f7+3wolSRRU1BkCPSQVCpFsGYq9fWEbC5ryf4+tOHGRiN8TtvWDd2rK68iAubqtjZ1kc0PvcWhsi5zNMgMLMbzGyfmR0ws7smefzXzeyl1MdTZnaJl/Xksr7h5EJtdQ2dVlVayEVNVTw5h3GCkWicrzx5mDduWsIFy6vOeOzC5ZUMR+M8e7A706WKLGqeBYGZBYG7gbcCm4FbzGzzhNMOAb/inLsY+EvgXq/qyXWnu4b8X1mcS65ZV8+OY70MjsZmdf5PXm6neyjCR69bc9Zj65dWEAoG+PHLJzJdpsii5mWLYAtwwDl3MHW/4/uAG8ef4Jx7yjnXk/ryGaDZw3pyWjoI1CI403Xr6oklHM/MMOPnW88e5VvPHuWfHt5PXVmIQ11DZ91noTAYYENjBQ/taiee0EI1kTQvg6AJODbu69bUsal8FPjxZA+Y2R1mtt3Mtnd2dmawxNyRvk2lguBMLatrKC8q4JG9J2c8t2NghMOnhrhide2UN9C5YHklXYOROQ9Ai5zLvAyCyf4nTvo2zMzeQDII/mCyx51z9zrnWpxzLQ0N2dv8LJvGBouLtcZvvKKCIL+yoYGH93SQmOFd/PbDPQTNuHxVzZTnrK0vA2DbIY0TiKR5GQStwIpxXzcDxyeeZGYXA18CbnTO5e2Kn95wlIqiAt+3oM5Fb968hM6BUV5q65vynGg8wfNHezh/eSXlRVOHaUVxIWvry9iqIBAZ4+WrzjZgvZmtMbMQcDPwwPgTzGwl8D3gg865VzysJef1hiNUl6lbaDJv2LiEYMB4ZM/U3UO7j/cTjsTZMovbaW5ZU8u2w90ztjBE8oVnQeCci5G86f1DwB7gfufcLjO708zuTJ32p0Ad8P/MbIeZbfeqnlzXE45SoxlDk6ouDdGyqob/2nliyt1Itx7uprYsxNqGshmvt2VNLf0jMfadHMh0qSKLkqf9EM65B51zG5xz5znn/ip17B7n3D2pzz/mnKtxzl2a+mjxsp5c1huOaOroNN7XsoKDnUOT7he0s7WPQ13TDxKPt2VNstWg7iGRJHVI54hki0BdQ1N5x8XLqC0L8bWnDp/12P99+BVKCoNcuWZ291RurilleVUxWw8rCERAQZAzeoYi6hqaRnFhkJuvWMHDe05yrDs8dvyFoz38fG8Hr11fT3FhcNbXu2xlDS9pCqkIoCDICdF4goHRmIJgBrddtYqigiCfvu8FRmNxugZH+ey3d1BXFuLqtXVzutZFzVUc6x6mZyjiUbUii4eCIAekF5PVaNbQtJZXl/AP77+E54/28v57nuZ99zxNe/8I937oNRTNoTUAcFFTch+indNMSRXJFwqCHNAbTr4r1WDxzN520TL+4l0XMBpLUFYU5Au3vYbXrJrd2MB4Fy5XEIikaRlrDuhJtwg0WDwrt1+zmtuvWb2ga1SVFrK6rpSdrQoCEbUIckBPqkWgMYLsuqi5Wi0CEdQiyAnpAcuaMgVBNqR3JY3FE7T1DnPvLw9SXlTArVeu9LkyEX8oCHKAuoYmN3Eb6bRMvWA3VZcAcLx3mA1LKzJyTZHFSF1DOaA3HCFUEKBkjjNfZGGWp4KgtWfY50pE/KUWQQ7oCUeoKS3EZrE9QrZM9W78XFJcGKS+vIjjvQoCyW9qEeQAbTjnn6bqYtoUBJLnFAQ5oGcoQrXGB3zRVFNK33CUgZGo36WI+EZBkAN6whFqNWPIF+kBY7UKJJ9pjCAH9IajWlU8B5kcv1heXYyhIJD8phaBz5xz9A5rC2q/FBUEqa8ook0zhySPKQh81j8SI55wGiz2UXN1iVoEktcUBD47NTgKoDECHzXVlDAwEqO9b8TvUkR8oSDwWddgcnuJ+vIinyvJX82pAeMXdaMayVMKAp91pVoECgL/LKsuIWBoJ1LJWwoCn40FQYW6hvxSGAywtLJYLQLJWwoCn3UNjGIGtRos9lVTdQk72/pwzvldikjWKQh81jkYobY0REFQfxV+aqopoTcc5Vi3Zg9J/tGrj8+6Bkc1PpADmmtKAQ0YS35SEPisa3CUhgoFgd+WVhYRCgZ0xzLJSwoCnyVbBBof8FtBIMD5yyt58ZhaBJJ/FAQ+6xqIqGsoR1zcVMXLbX0kEhowlvyiIPDR0GiM4WicenUN5YSLm6sYisQ52DXodykiWaUg8FHngBaT5ZKLm6sBeEkLyyTPeBoEZnaDme0zswNmdtckj28ys6fNbNTM/puXteSi06uKNUaQC9YtKaekMKggkLzj2f0IzCwI3A1cD7QC28zsAefc7nGndQOfAt7tVR25TNtL5JZgwLiwqVJTSCXveNki2AIccM4ddM5FgPuAG8ef4JzrcM5tA/LyPoGdqQ3nNH00d1y2soZdbf2MRON+lyKSNV4GQRNwbNzXraljc2Zmd5jZdjPb3tnZmZHickHXgLagzjWvWVVDJJ7QegLJK14GgU1ybF7z8pxz9zrnWpxzLQ0NDQssK3d0DY5SWxaiUNtL5IyWVTUAbD/c43MlItnj5StQK7Bi3NfNwHEPn2/Rae8bYYm6hXJKXXkRa+vLeO5It9+liGSNl0GwDVhvZmvMLATcDDzg4fMtOm29wzTXlPhdhkzwmlU1PHekRzuRSt7wLAicczHgE8BDwB7gfufcLjO708zuBDCzRjNrBX4X+BMzazWzSq9qyjVtvcMsr1YQ5JqW1TX0hKO82jnkdykiWeHZ9FEA59yDwIMTjt0z7vN2kl1Gead/JMrASIwmBUHO+NazR4HTC/3u/vkBrlhTy61XrvSzLBHPaZTSJ8d7k/veN6lrKOfUl4eoKC7gVW01IXnC0xaBTK2tJxkEL7f10z8c87kaGc/MOK+hnAMdgxonkLygFoFP0i2C6tJCnyuRyaytL2NwNEZHqptI5FymIPBJa+8wwYBRXqRGWS46r6EcgIOd6h6Sc5+CwCdtPcNUlRQSsMnW3YnfaspC1JQWauaQ5AUFgU+O9w6rWyjHnddQzsGuQaLxhN+liHhKQeCTtt5hakq0x1Au27C0gpFogueOaLsJObcpCHwQiSXoGBilSi2CnLZ+STlBM36xt8PvUkQ8pSDwwYm+YZyDGgVBTisqDLKmvoyfKwjkHKcg8MH+k8mZKA26IU3O29hYwf6OQY51h/0uRcQzCgIf7Ds5AMDSymKfK5GZbGqsAOCnu0/6XImIdzSJ3Qd72wdorimhqDDodykyg7ryIjYvq+RHLx7no9etmdc10nsYTaQ9jCRXqEXgg33t/WxcWuF3GTJL77p0OTuO9XL0lLqH5NykIMiySCzBwc4hNjYqCBaLd16yHIAHXmzzuRIRbygIsuzVzkFiCacgWESaqktoWVXDD3Yc1yZ0ck5SEGTZvvbkQPGmxry5/8454X0tzRzoGOTpV0/5XYpIxikIsmxv+wCFQWNtQ5nfpcgc3HhpE3VlIb78xKF5ff9oLM7BzkEOdJk4F3EAAArTSURBVAwyHIlnuDqRhdGsoSx7ua2PdUsqKAwqgxeT4sIgt121is89sp9XOwfHdiedSdfgKD/c0cb2wz3EU91KAYNr19Xza5c3UayZY5ID9GqUReFIjK2Hurn2vDq/S5F5uO2qVRQVBPi7n+yb1fkP7Wrnzf/4GNsOd3P5qhpuv3o1H7lmNZetrOHx/V289wtP0ReOely1yMwUBFn0zMFTROIJXr9xid+lyDw0VBTxqTet5ye72nl4mgVmw5E4f/z9nfzW15+juaaET71xPe+5rImNjRWsX1rBey9v5oNXrWL/yUFu/+pWhkZ1hzrxl4Igix7d10lJYZAr1tT4XYrMwbeePTr2UVFcwJKKIv7kBy9Puq5g1/E+3vn5J/jms0e543Vr+d7Hr2XJJCvIz19WyT/fehk72/r4vftf1Gwk8ZWCIEucczy6r5NrzqujqED9wotVQSDA+1tWMBKLc9M9T/HInpMMR+Ic6BjgL360i3d9/klO9o/wG9euYXVdGd95rnXKa/3qBY384Vs38ZNd7dz9iwNZ/ClEzqTB4izZ3zHI0e4wv/na+W1TILljeXUJ377jaj781a189F+3jx0vCBiXNFfz1gsbKZvlLUg/et0aXm7r4x9+9gqbl1fyxk1LvSpbZEoKgiz5wqOvUlwY4IYLl/ldimTAxsYKHvvvb+Dx/Z3sbOtjeVUJ166v57F9nbO+RnoPoktX1PDsoW5++5vP8+NPv4419ZpaLNmlrqEsONAxyA93tPGhq1fTUKGtp88VoYIAbzp/KZ958wbef8UKmqpL5n2d265cRcCMO/5tO4PzHDx2znGoa4gXj/Vy5NSQxh1k1tQi8FgsnuCv/ms3xYVBfut1a/0uR3JUTVmIW7as5GtPHeYz9+3gC7ddPuu1Jt1DEb742Kv8cMdx2vtHxo6XhoJc3FzFlWvq+Oz1G7wqXc4BCgIPxeIJ/uC7O/nFvk7+/J2bqdONaGQa5zWU86fv2MyfPbCLj3/jOT5/6+XTLjj76pOHePJAF4/v7yISS3D+skquXltHeXEBgyMxXu0aZPvhHp452M3uE/184g3ruGRFdRZ/IlksFAQeiMUT/HxvB3//0328cnKQz755Ax++VoPE55Kp7jGwULdfs5qAwf/44S7e9fkn+NubLuHSCS/ew5E4928/xt//9BWGRmNsXlbJ9ZuXnnWjoyvW1BIejfH0wVNsPdTNjbuf5LXr6/nkG9ezZU2tJ/XL4qQgyJBoPMHWQ93888/3s+t4PwMjMWrLQty6ZSX15SG/y5NF5INXr6a5tpS7vvsS7777SS5bWU3LqhpKQgUc7Bzk0X2dDI7GWFNfxq9etYqVtaVTXqu0qIA3nb+Uz91yGd945ghfevwg7//i05y/rJL3tzRz46VN1Jbp32e+s8U2oNTS0uK2b98+84kec87xaucQTx88xdOvdvHkgVP0DUcpDBrrl1Rw+cpqNjZWEgwYMPndqLx6VynnhuFInOeOdPPCsV66hyJE4gmWV5Vw9Xl1vO81yd1QzWxO14zEEjx/tIfnjvTQ1jtMYdC4bl09129u5M2bl7CkQrdPPVeZ2XPOuZZJH/MyCMzsBuBzQBD4knPuryc8bqnH3waEgQ87556f7pp+BEE4EmNf+wB7Tgywt72fPSf62XtigIHU7I5lVcVcu66e6zcv5UTvCKECTcaSzLplywpiCXfGAPJC30ic6BvmhaO97DreR09qz6MVNSXcvGUlW9bUcnFzlRY/nkN8CQIzCwKvANcDrcA24Bbn3O5x57wN+CTJILgS+Jxz7srprpupIHDOEU84YglHNJ6gfyRGbzhCz1CUYz1hjpwKc+TUEPvaBzjUNUT6t1RUEKCxspjGqmKWV5WwtqGM2rLQnN+ZieQK5xzt/SPsOdHPnhMDtPUOA8lprRcur+S8hnJW15exqq6UmtIQlcWFVJYUUBgMUBAwCoIBggGjMGgEA0ZBIDDWEvaqXucg4RyO1J+OM46lX9fS/y8NSP8XNQzHmddwCXA4EqljicTpz+OJ5Llx51LPlXo+SP15+vnh9NdMfHxCXQUBS/4Og0ZhIPln+vPg+GMBy8jry3RB4OUYwRbggHPuYKqI+4Abgd3jzrkR+DeX/O08Y2bVZrbMOXci08X85OUTfPbbLxJPOOKpv9zpFAaN5ppSNiwtZ019GY1VxSyrKqG6tJCAXvTlHGJmLKsqYVlVCW/ctJTB0RhHTw1x+FSY1p4w+0+2j7V+537t9IvwmS/Ixtir8lnHzMa9qLvTL9DOJf/MR8nANe547Vp+9y0bM3/9jF/xtCbg2LivW0m+65/pnCbgjCAwszuAO1JfDprZ7PYBXqADwKPTn1IPdGWhlFyV7z8/6HeQ7z8/ZPF38Hupj3laNdUDXgbBZG+bJ+b5bM7BOXcvcG8misokM9s+VVMrH+T7zw/6HeT7zw/nxu/Ay1HNVmDFuK+bgePzOEdERDzkZRBsA9ab2RozCwE3Aw9MOOcB4EOWdBXQ58X4gIiITM2zriHnXMzMPgE8RHL66Fecc7vM7M7U4/cAD5KcMXSA5PTRj3hVj0dyrrsqy/L95wf9DvL954dz4Hew6BaUiYhIZmnlk4hInlMQiIjkOQXBApjZ35nZXjN7ycy+b2Z5s8evmd1gZvvM7ICZ3eV3PdlkZivM7BdmtsfMdpnZp/2uyS9mFjSzF8zsP/2uxQ+pRbDfSb0O7DGzq/2uaT4UBAvzM+BC59zFJLfT+EOf68mK1PYhdwNvBTYDt5jZZn+ryqoY8HvOufOBq4DfybOff7xPA3v8LsJHnwN+4pzbBFzCIv1dKAgWwDn3U+dceu39MyTXQeSDse1DnHMRIL19SF5wzp1Ib47onBsg+Z+/yd+qss/MmoG3A1/yuxY/mFkl8DrgywDOuYhzrtffquZHQZA5vwH82O8ismSqrUHyjpmtBi4DnvW3El/8E/D7QMLvQnyyFugEvprqHvuSmZX5XdR8KAhmYGYPm9nLk3zcOO6cPybZXfBN/yrNqlltDXKuM7Ny4LvAZ5xz/X7Xk01m9g6gwzn3nN+1+KgAuBz4gnPuMmAIWJTjZbpD2Qycc2+e7nEzux14B/Amlz+LMvJ+axAzKyQZAt90zn3P73p8cC3wrtRW8sVApZl9wzl3m891ZVMr0OqcS7cGv8MiDQK1CBYgdeOdPwDe5ZwL+11PFs1m+5BzVuqGSl8G9jjn/tHvevzgnPtD51yzc241yb//n+dZCOCcaweOmVl6X+g3ceY2+4uGWgQL83mgCPhZar/1Z5xzd/pbkvem2j7E57Ky6Vrgg8BOM9uROvZHzrkHfaxJ/PFJ4JupN0QHWXzb5ADaYkJEJO+pa0hEJM8pCERE8pyCQEQkzykIRETynIJARCTPKQhE5sjMBv2uQSSTFAQiInlOQSAyT5b0d6m9p3aa2QdSx19vZo+O26f+m6nVyCI5SSuLRebv14BLSe5DXw9sM7Nfph67DLiA5B5MT5JcjfyEH0WKzEQtApH5uw74d+dc3Dl3EngMuCL12FbnXKtzLgHsAFb7VKPIjBQEIvM3XXfP6LjP46j1LTlMQSAyf78EPpC6b28DybtVbfW5JpE507sUkfn7PnA18CLJG/P8vnOu3cw2+VuWyNxo91ERkTynriERkTynIBARyXMKAhGRPKcgEBHJcwoCEZE8pyAQEclzCgIRkTz3/wEL0fiW6Jb3pAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Xsc = stdscaler_transform(X_num.drop(['price', 'zip_code'], axis=1))\n",
    "ysc = X_num['price']\n",
    "\n",
    "for column in Xsc.columns:\n",
    "    sns.distplot(Xsc[column])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.889795120954649,\n",
       " 71230.98320683662,\n",
       " 135220.2198053535,\n",
       "           true        pred      resid\n",
       " 16861   500000   456322.10   43677.90\n",
       " 9750    290000   270898.75   19101.25\n",
       " 19      450500   412581.04   37918.96\n",
       " 11346  1170000  1039626.00  130374.00\n",
       " 10347   337000   352497.50   15497.50\n",
       " 9002    410000   442011.80   32011.80\n",
       " 4973    450000   541846.16   91846.16\n",
       " 17232   394250   370728.00   23522.00\n",
       " 11014   315000   322084.58    7084.58\n",
       " 59      420000   338236.45   81763.55)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "def model_function_random_forest(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "    reg = RandomForestRegressor()\n",
    "    reg.fit(X_train, y_train)\n",
    "    predictions = reg.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    plot_feature_importance(reg.feature_importances_,X_train.columns,'RANDOM FOREST')\n",
    "    return reg.score(X_test, y_test), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "    \n",
    "scores_random_forest = model_function_random_forest(Xsc, ysc)\n",
    "scores_random_forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.6929909843985417,\n",
       " 0.691291342412403,\n",
       " 130858.69196335497,\n",
       " 225692.3252574602,\n",
       "           true          pred          resid\n",
       " 16861   500000  2.721513e+05  227848.685774\n",
       " 9750    290000  2.616026e+05   28397.374585\n",
       " 19      450500  4.792590e+05   28758.952601\n",
       " 11346  1170000  1.117064e+06   52935.578852\n",
       " 10347   337000  4.278890e+05   90888.976838\n",
       " 9002    410000  6.506542e+05  240654.178606\n",
       " 4973    450000  4.351727e+05   14827.277940\n",
       " 17232   394250  5.936145e+05  199364.511652\n",
       " 11014   315000  4.506646e+05  135664.571125\n",
       " 59      420000  4.692881e+05   49288.083661)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "def model_function_linear(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "    model = LinearRegression()\n",
    "    model.fit(X_train, y_train)\n",
    "    predictions = model.predict(X_test)\n",
    "    results = pd.DataFrame()\n",
    "    results['true'] = y_test\n",
    "    results['pred'] = predictions\n",
    "    results['resid'] = results.apply(lambda x: abs(x['true'] - x['pred']), axis=1)\n",
    "    return r2_score(y_test, predictions), 1 - (1-r2_score(y_test, predictions))*(len(y_test)-1)/(len(y_test)-X_test.shape[1]-1), mean_absolute_error(y_test, predictions), mean_squared_error(y_test, predictions, squared=False), results.head(10)\n",
    "\n",
    "scores_linear = model_function_linear(Xsc, ysc)\n",
    "scores_linear"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}