KathyZhang_ML_mp3_covid_death_forecast.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <font color='blue'> PPHA 30546 Machine Learning | Mini Project 3 - CDC COVID Death Forecast </font> \n",
    "\n",
    "Kathy Zhang & Yimeng Wu"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Overview\n",
    "\n",
    "It is early in July of 2020, and you have just been hired to work for the Centers for Disease Control and Prevention (CDC).2 As you probably know, the CDC is a national health protection agency tasked with protecting public health and safety by preventing and mitigating disease, injury, and disability in both the United States (US) and abroad. According to their website the CDC “conducts critical science and provides health information that protects our nation against expensive and dangerous health threats.” \n",
    "\n",
    "The number of COVID-19 deaths in the US has just surpassed 100,000. Things are looking grim around the world, but hope is on the way in the form of several potential vaccines. Your team at the CDC has been tasked with optimizing the distribution of a future vaccine to the places in the US that are most in need. To aid in doing so, you need a good prediction of which communities will be the most hard-hit when the vaccine is ready.\n",
    "\n",
    "You colleagues have asked you to develop a model that predicts COVID-19 deaths per capita using socio-economic, health, and weather/pollution data from any and all available sources. Predictions based on your analysis will help shape your office’s recommendations about how to best deploy limited vaccines. The project has two parts: understanding the available data and performing data analysis and answering questions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-01-16T02:10:13.296809Z",
     "start_time": "2024-01-16T02:10:10.114201Z"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# Import the libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import openpyxl\n",
    "import seaborn as sns\n",
    "sns.set_style(\"darkgrid\")\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import statsmodels.formula.api as smf\n",
    "from statsmodels.formula.api import ols\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import GridSearchCV, KFold\n",
    "from sklearn.model_selection import cross_validate\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis\n",
    "from sklearn.metrics import mean_absolute_error, r2_score, mean_squared_error\n",
    "from sklearn.metrics import accuracy_score, precision_score\n",
    "from sklearn.metrics import confusion_matrix, roc_curve\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import LassoLarsIC \n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.model_selection import GridSearchCV, KFold\n",
    "\n",
    "from sympy import symbols, Eq, solve, exp\n",
    "from IPython.display import Image\n",
    "from IPython.core.display import HTML \n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Understanding the Data\n",
    "\n",
    "### The full dataset is available for download from Canvas, but you should familiarize yourself with how it was constructed. The data was compiled from the following sources:\n",
    "\n",
    "### 1. NY Times: Daily data on COVID-19 cases and deaths at the county level comes from the New York Times GitHub Repository. Cumulative cases and deaths were calculated for coun- ties in the continental United States as of July 1, 2020.3\n",
    "### 2. Opportunity Insights: County characteristics, including baseline health measures, come from Bergeron, Chetty, Cutler, Scuderi, Stepner, and Turner (2016). The data are available for download on the Opportunity Insights Data Library webpage.4\n",
    "### 3. PM COVID: Annual, county-level data on air pollution (PM2.5) and weather (average winter/summer temperature and relative humidity), and baseline mortality come from Wu, Nethery, Sabath, Braun, and Dominici (2020). The pollution and weather measures used in this project are calculated by averaging across annual averages from 2000-2016"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3 Data Analysis\n",
    "\n",
    "### 1. The “Variable Description.xlsx” spreadsheet contains a list of variables that we’ll use for our analyses. Note that this is not a full list of all the variables in the dataset, although it’s close (we are ignoring a few perfectly co-linear predictors). Filter the full set of variables in the dataset down to the the Opportunity Insights and PM COVID variables listed in the spreadsheet along with county, state, and deathspc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(62, 4)\n"
     ]
    },
    {
     "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>Variable</th>\n",
       "      <th>Description</th>\n",
       "      <th>Source</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>casespc</td>\n",
       "      <td>COVID-19 Cases per 100,000 Population</td>\n",
       "      <td>NY Times</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>deathspc</td>\n",
       "      <td>COVID-19 Deaths per 100,000 Population</td>\n",
       "      <td>NY Times</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>intersects_msa</td>\n",
       "      <td>Urban Area</td>\n",
       "      <td>Opportunity Insights</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>cur_smoke_q1</td>\n",
       "      <td>BRFSS: Fraction Current Smokers in Q1</td>\n",
       "      <td>Opportunity Insights</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>cur_smoke_q2</td>\n",
       "      <td>BRFSS: Fraction Current Smokers in Q2</td>\n",
       "      <td>Opportunity Insights</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Count        Variable                             Description  \\\n",
       "0      1         casespc   COVID-19 Cases per 100,000 Population   \n",
       "1      2        deathspc  COVID-19 Deaths per 100,000 Population   \n",
       "2      3  intersects_msa                              Urban Area   \n",
       "3      4    cur_smoke_q1   BRFSS: Fraction Current Smokers in Q1   \n",
       "4      5    cur_smoke_q2   BRFSS: Fraction Current Smokers in Q2   \n",
       "\n",
       "                 Source  \n",
       "0              NY Times  \n",
       "1              NY Times  \n",
       "2  Opportunity Insights  \n",
       "3  Opportunity Insights  \n",
       "4  Opportunity Insights  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Read in the variable description data\n",
    "var_des = pd.read_excel('PPHA_30546_MP03-Variable_Description.xlsx')\n",
    "print(var_des.shape)\n",
    "var_des.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3107, 93)\n"
     ]
    },
    {
     "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>county</th>\n",
       "      <th>state</th>\n",
       "      <th>fips</th>\n",
       "      <th>cases</th>\n",
       "      <th>deaths</th>\n",
       "      <th>population</th>\n",
       "      <th>casespc</th>\n",
       "      <th>deathspc</th>\n",
       "      <th>cty</th>\n",
       "      <th>county_name</th>\n",
       "      <th>...</th>\n",
       "      <th>taxrate</th>\n",
       "      <th>tax_st_diff_top20</th>\n",
       "      <th>summer_tmmx</th>\n",
       "      <th>summer_rmax</th>\n",
       "      <th>winter_tmmx</th>\n",
       "      <th>winter_rmax</th>\n",
       "      <th>pm25</th>\n",
       "      <th>bmdeaths</th>\n",
       "      <th>bmcruderate</th>\n",
       "      <th>pm25_mia</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Autauga</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>1001</td>\n",
       "      <td>4434</td>\n",
       "      <td>152</td>\n",
       "      <td>918492</td>\n",
       "      <td>482.74780</td>\n",
       "      <td>16.548864</td>\n",
       "      <td>1001</td>\n",
       "      <td>Autauga</td>\n",
       "      <td>...</td>\n",
       "      <td>0.011183</td>\n",
       "      <td>0.0</td>\n",
       "      <td>306.02344</td>\n",
       "      <td>96.055420</td>\n",
       "      <td>288.08508</td>\n",
       "      <td>85.651848</td>\n",
       "      <td>11.712587</td>\n",
       "      <td>7893</td>\n",
       "      <td>859.29999</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Baldwin</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>1003</td>\n",
       "      <td>10465</td>\n",
       "      <td>278</td>\n",
       "      <td>3102984</td>\n",
       "      <td>337.25601</td>\n",
       "      <td>8.959118</td>\n",
       "      <td>1003</td>\n",
       "      <td>Baldwin</td>\n",
       "      <td>...</td>\n",
       "      <td>0.011756</td>\n",
       "      <td>0.0</td>\n",
       "      <td>305.51663</td>\n",
       "      <td>97.971542</td>\n",
       "      <td>290.20886</td>\n",
       "      <td>89.730972</td>\n",
       "      <td>10.077723</td>\n",
       "      <td>30292</td>\n",
       "      <td>976.20001</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Barbour</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>1005</td>\n",
       "      <td>3157</td>\n",
       "      <td>33</td>\n",
       "      <td>499262</td>\n",
       "      <td>632.33331</td>\n",
       "      <td>6.609756</td>\n",
       "      <td>1005</td>\n",
       "      <td>Barbour</td>\n",
       "      <td>...</td>\n",
       "      <td>0.012059</td>\n",
       "      <td>0.0</td>\n",
       "      <td>306.06226</td>\n",
       "      <td>97.371674</td>\n",
       "      <td>289.24210</td>\n",
       "      <td>88.633575</td>\n",
       "      <td>10.981967</td>\n",
       "      <td>5197</td>\n",
       "      <td>1040.90000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Bibb</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>1007</td>\n",
       "      <td>2291</td>\n",
       "      <td>24</td>\n",
       "      <td>397470</td>\n",
       "      <td>576.39569</td>\n",
       "      <td>6.038192</td>\n",
       "      <td>1007</td>\n",
       "      <td>Bibb</td>\n",
       "      <td>...</td>\n",
       "      <td>0.008007</td>\n",
       "      <td>0.0</td>\n",
       "      <td>305.98218</td>\n",
       "      <td>96.293076</td>\n",
       "      <td>287.36282</td>\n",
       "      <td>86.485870</td>\n",
       "      <td>11.998714</td>\n",
       "      <td>4089</td>\n",
       "      <td>1028.80000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Blount</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>1009</td>\n",
       "      <td>2082</td>\n",
       "      <td>15</td>\n",
       "      <td>997531</td>\n",
       "      <td>208.71532</td>\n",
       "      <td>1.503713</td>\n",
       "      <td>1009</td>\n",
       "      <td>Blount</td>\n",
       "      <td>...</td>\n",
       "      <td>0.007012</td>\n",
       "      <td>0.0</td>\n",
       "      <td>305.17886</td>\n",
       "      <td>94.630951</td>\n",
       "      <td>285.56567</td>\n",
       "      <td>85.449142</td>\n",
       "      <td>11.793022</td>\n",
       "      <td>9912</td>\n",
       "      <td>993.70001</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 93 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    county    state  fips  cases  deaths  population    casespc   deathspc  \\\n",
       "0  Autauga  Alabama  1001   4434     152      918492  482.74780  16.548864   \n",
       "1  Baldwin  Alabama  1003  10465     278     3102984  337.25601   8.959118   \n",
       "2  Barbour  Alabama  1005   3157      33      499262  632.33331   6.609756   \n",
       "3     Bibb  Alabama  1007   2291      24      397470  576.39569   6.038192   \n",
       "4   Blount  Alabama  1009   2082      15      997531  208.71532   1.503713   \n",
       "\n",
       "    cty county_name  ...   taxrate  tax_st_diff_top20 summer_tmmx  \\\n",
       "0  1001     Autauga  ...  0.011183                0.0   306.02344   \n",
       "1  1003     Baldwin  ...  0.011756                0.0   305.51663   \n",
       "2  1005     Barbour  ...  0.012059                0.0   306.06226   \n",
       "3  1007        Bibb  ...  0.008007                0.0   305.98218   \n",
       "4  1009      Blount  ...  0.007012                0.0   305.17886   \n",
       "\n",
       "   summer_rmax winter_tmmx  winter_rmax       pm25  bmdeaths bmcruderate  \\\n",
       "0    96.055420   288.08508    85.651848  11.712587      7893   859.29999   \n",
       "1    97.971542   290.20886    89.730972  10.077723     30292   976.20001   \n",
       "2    97.371674   289.24210    88.633575  10.981967      5197  1040.90000   \n",
       "3    96.293076   287.36282    86.485870  11.998714      4089  1028.80000   \n",
       "4    94.630951   285.56567    85.449142  11.793022      9912   993.70001   \n",
       "\n",
       "   pm25_mia  \n",
       "0         0  \n",
       "1         0  \n",
       "2         0  \n",
       "3         0  \n",
       "4         0  \n",
       "\n",
       "[5 rows x 93 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Read in the covid data\n",
    "covid = pd.read_csv('Data-Covid002.csv', encoding_errors= 'replace')\n",
    "print(covid.shape)\n",
    "covid.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filter columns\n",
    "selected_variables = var_des[var_des['Source'].isin([\n",
    "    'Opportunity Insights', 'PM_COVID'])].loc[:,'Variable'].tolist()\n",
    "filtered_variables = ['county', 'state', 'deathspc'] + selected_variables\n",
    "\n",
    "# Apply filter\n",
    "covid = covid.filter(items=filtered_variables)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3107, 63)\n"
     ]
    },
    {
     "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>county</th>\n",
       "      <th>state</th>\n",
       "      <th>deathspc</th>\n",
       "      <th>intersects_msa</th>\n",
       "      <th>cur_smoke_q1</th>\n",
       "      <th>cur_smoke_q2</th>\n",
       "      <th>cur_smoke_q3</th>\n",
       "      <th>cur_smoke_q4</th>\n",
       "      <th>bmi_obese_q1</th>\n",
       "      <th>bmi_obese_q2</th>\n",
       "      <th>...</th>\n",
       "      <th>subcty_exp_pc</th>\n",
       "      <th>taxrate</th>\n",
       "      <th>tax_st_diff_top20</th>\n",
       "      <th>pm25</th>\n",
       "      <th>pm25_mia</th>\n",
       "      <th>summer_tmmx</th>\n",
       "      <th>summer_rmax</th>\n",
       "      <th>winter_tmmx</th>\n",
       "      <th>winter_rmax</th>\n",
       "      <th>bmcruderate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Autauga</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>16.548864</td>\n",
       "      <td>1</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.238095</td>\n",
       "      <td>0.208333</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>0.238095</td>\n",
       "      <td>...</td>\n",
       "      <td>1059.66930</td>\n",
       "      <td>0.011183</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.712587</td>\n",
       "      <td>0</td>\n",
       "      <td>306.02344</td>\n",
       "      <td>96.055420</td>\n",
       "      <td>288.08508</td>\n",
       "      <td>85.651848</td>\n",
       "      <td>859.29999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Baldwin</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>8.959118</td>\n",
       "      <td>1</td>\n",
       "      <td>0.268097</td>\n",
       "      <td>0.233503</td>\n",
       "      <td>0.167464</td>\n",
       "      <td>0.176991</td>\n",
       "      <td>0.298050</td>\n",
       "      <td>0.262467</td>\n",
       "      <td>...</td>\n",
       "      <td>2209.91040</td>\n",
       "      <td>0.011756</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.077723</td>\n",
       "      <td>0</td>\n",
       "      <td>305.51663</td>\n",
       "      <td>97.971542</td>\n",
       "      <td>290.20886</td>\n",
       "      <td>89.730972</td>\n",
       "      <td>976.20001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Barbour</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>6.609756</td>\n",
       "      <td>0</td>\n",
       "      <td>0.228571</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.294118</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>...</td>\n",
       "      <td>1570.24830</td>\n",
       "      <td>0.012059</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.981967</td>\n",
       "      <td>0</td>\n",
       "      <td>306.06226</td>\n",
       "      <td>97.371674</td>\n",
       "      <td>289.24210</td>\n",
       "      <td>88.633575</td>\n",
       "      <td>1040.90000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Bibb</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>6.038192</td>\n",
       "      <td>1</td>\n",
       "      <td>0.244444</td>\n",
       "      <td>0.280000</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.150000</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>...</td>\n",
       "      <td>1338.44120</td>\n",
       "      <td>0.008007</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.998714</td>\n",
       "      <td>0</td>\n",
       "      <td>305.98218</td>\n",
       "      <td>96.293076</td>\n",
       "      <td>287.36282</td>\n",
       "      <td>86.485870</td>\n",
       "      <td>1028.80000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Blount</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>1.503713</td>\n",
       "      <td>1</td>\n",
       "      <td>0.304348</td>\n",
       "      <td>0.260870</td>\n",
       "      <td>0.352941</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.347826</td>\n",
       "      <td>0.318182</td>\n",
       "      <td>...</td>\n",
       "      <td>987.43884</td>\n",
       "      <td>0.007012</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.793022</td>\n",
       "      <td>0</td>\n",
       "      <td>305.17886</td>\n",
       "      <td>94.630951</td>\n",
       "      <td>285.56567</td>\n",
       "      <td>85.449142</td>\n",
       "      <td>993.70001</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 63 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    county    state   deathspc  intersects_msa  cur_smoke_q1  cur_smoke_q2  \\\n",
       "0  Autauga  Alabama  16.548864               1      0.333333      0.238095   \n",
       "1  Baldwin  Alabama   8.959118               1      0.268097      0.233503   \n",
       "2  Barbour  Alabama   6.609756               0      0.228571      0.250000   \n",
       "3     Bibb  Alabama   6.038192               1      0.244444      0.280000   \n",
       "4   Blount  Alabama   1.503713               1      0.304348      0.260870   \n",
       "\n",
       "   cur_smoke_q3  cur_smoke_q4  bmi_obese_q1  bmi_obese_q2  ...  subcty_exp_pc  \\\n",
       "0      0.208333      0.133333      0.375000      0.238095  ...     1059.66930   \n",
       "1      0.167464      0.176991      0.298050      0.262467  ...     2209.91040   \n",
       "2      0.181818      0.111111      0.294118      0.571429  ...     1570.24830   \n",
       "3      0.181818      0.150000      0.466667      0.375000  ...     1338.44120   \n",
       "4      0.352941      0.166667      0.347826      0.318182  ...      987.43884   \n",
       "\n",
       "    taxrate  tax_st_diff_top20       pm25  pm25_mia  summer_tmmx  summer_rmax  \\\n",
       "0  0.011183                0.0  11.712587         0    306.02344    96.055420   \n",
       "1  0.011756                0.0  10.077723         0    305.51663    97.971542   \n",
       "2  0.012059                0.0  10.981967         0    306.06226    97.371674   \n",
       "3  0.008007                0.0  11.998714         0    305.98218    96.293076   \n",
       "4  0.007012                0.0  11.793022         0    305.17886    94.630951   \n",
       "\n",
       "   winter_tmmx  winter_rmax  bmcruderate  \n",
       "0    288.08508    85.651848    859.29999  \n",
       "1    290.20886    89.730972    976.20001  \n",
       "2    289.24210    88.633575   1040.90000  \n",
       "3    287.36282    86.485870   1028.80000  \n",
       "4    285.56567    85.449142    993.70001  \n",
       "\n",
       "[5 rows x 63 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(covid.shape)  # effectively removes 30 columns\n",
    "covid.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "county             object\n",
       "state              object\n",
       "deathspc          float64\n",
       "intersects_msa      int64\n",
       "cur_smoke_q1      float64\n",
       "                   ...   \n",
       "summer_tmmx       float64\n",
       "summer_rmax       float64\n",
       "winter_tmmx       float64\n",
       "winter_rmax       float64\n",
       "bmcruderate       float64\n",
       "Length: 63, dtype: object"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check types\n",
    "covid.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Compute and report descriptive (summary) statistics for the subset of Opportunity Insights and PM COVID variables you filtered in previous question."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>deathspc</th>\n",
       "      <th>intersects_msa</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>23.790131</td>\n",
       "      <td>0.596717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>67.852145</td>\n",
       "      <td>0.490636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.802303</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>21.461759</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2279.610600</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          deathspc  intersects_msa\n",
       "count  3107.000000     3107.000000\n",
       "mean     23.790131        0.596717\n",
       "std      67.852145        0.490636\n",
       "min       0.000000        0.000000\n",
       "25%       0.000000        0.000000\n",
       "50%       3.802303        1.000000\n",
       "75%      21.461759        1.000000\n",
       "max    2279.610600        1.000000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>cur_smoke_q1</th>\n",
       "      <th>cur_smoke_q2</th>\n",
       "      <th>cur_smoke_q3</th>\n",
       "      <th>cur_smoke_q4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.212659</td>\n",
       "      <td>0.171048</td>\n",
       "      <td>0.134467</td>\n",
       "      <td>0.098316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.149348</td>\n",
       "      <td>0.128130</td>\n",
       "      <td>0.132181</td>\n",
       "      <td>0.110110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.198718</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.096535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.310931</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.148719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cur_smoke_q1  cur_smoke_q2  cur_smoke_q3  cur_smoke_q4\n",
       "count   3107.000000   3107.000000   3107.000000   3107.000000\n",
       "mean       0.212659      0.171048      0.134467      0.098316\n",
       "std        0.149348      0.128130      0.132181      0.110110\n",
       "min        0.000000      0.000000      0.000000      0.000000\n",
       "25%        0.000000      0.000000      0.000000      0.000000\n",
       "50%        0.250000      0.198718      0.142857      0.096535\n",
       "75%        0.310931      0.250000      0.200000      0.148719\n",
       "max        1.000000      1.000000      1.000000      1.000000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>bmi_obese_q1</th>\n",
       "      <th>bmi_obese_q2</th>\n",
       "      <th>bmi_obese_q3</th>\n",
       "      <th>bmi_obese_q4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.239166</td>\n",
       "      <td>0.214580</td>\n",
       "      <td>0.209621</td>\n",
       "      <td>0.186739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.165928</td>\n",
       "      <td>0.153237</td>\n",
       "      <td>0.175849</td>\n",
       "      <td>0.167227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.080128</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.272076</td>\n",
       "      <td>0.241590</td>\n",
       "      <td>0.223124</td>\n",
       "      <td>0.194118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.335532</td>\n",
       "      <td>0.304348</td>\n",
       "      <td>0.297220</td>\n",
       "      <td>0.266667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       bmi_obese_q1  bmi_obese_q2  bmi_obese_q3  bmi_obese_q4\n",
       "count   3107.000000   3107.000000   3107.000000   3107.000000\n",
       "mean       0.239166      0.214580      0.209621      0.186739\n",
       "std        0.165928      0.153237      0.175849      0.167227\n",
       "min        0.000000      0.000000      0.000000      0.000000\n",
       "25%        0.080128      0.000000      0.000000      0.000000\n",
       "50%        0.272076      0.241590      0.223124      0.194118\n",
       "75%        0.335532      0.304348      0.297220      0.266667\n",
       "max        1.000000      1.000000      1.000000      1.000000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>exercise_any_q1</th>\n",
       "      <th>exercise_any_q2</th>\n",
       "      <th>exercise_any_q3</th>\n",
       "      <th>exercise_any_q4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.455995</td>\n",
       "      <td>0.555671</td>\n",
       "      <td>0.603792</td>\n",
       "      <td>0.638727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.273874</td>\n",
       "      <td>0.322336</td>\n",
       "      <td>0.357861</td>\n",
       "      <td>0.376922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.312500</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.354167</td>\n",
       "      <td>0.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.566563</td>\n",
       "      <td>0.707143</td>\n",
       "      <td>0.778364</td>\n",
       "      <td>0.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.641509</td>\n",
       "      <td>0.769231</td>\n",
       "      <td>0.841804</td>\n",
       "      <td>0.890497</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       exercise_any_q1  exercise_any_q2  exercise_any_q3  exercise_any_q4\n",
       "count      3107.000000      3107.000000      3107.000000      3107.000000\n",
       "mean          0.455995         0.555671         0.603792         0.638727\n",
       "std           0.273874         0.322336         0.357861         0.376922\n",
       "min           0.000000         0.000000         0.000000         0.000000\n",
       "25%           0.312500         0.444444         0.354167         0.400000\n",
       "50%           0.566563         0.707143         0.778364         0.833333\n",
       "75%           0.641509         0.769231         0.841804         0.890497\n",
       "max           1.000000         1.000000         1.000000         1.000000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>brfss_mia</th>\n",
       "      <th>puninsured2010</th>\n",
       "      <th>reimb_penroll_adj10</th>\n",
       "      <th>mort_30day_hosp_z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3103.000000</td>\n",
       "      <td>3106.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.249437</td>\n",
       "      <td>18.469460</td>\n",
       "      <td>9302.737743</td>\n",
       "      <td>0.457806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.432757</td>\n",
       "      <td>5.536651</td>\n",
       "      <td>1590.926253</td>\n",
       "      <td>1.206493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.625483</td>\n",
       "      <td>3663.530000</td>\n",
       "      <td>-7.778000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.410248</td>\n",
       "      <td>8159.340000</td>\n",
       "      <td>-0.255867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>18.147072</td>\n",
       "      <td>9193.770000</td>\n",
       "      <td>0.400088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>21.961417</td>\n",
       "      <td>10285.430000</td>\n",
       "      <td>1.147822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>41.366287</td>\n",
       "      <td>18443.220000</td>\n",
       "      <td>8.472745</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         brfss_mia  puninsured2010  reimb_penroll_adj10  mort_30day_hosp_z\n",
       "count  3107.000000     3107.000000          3103.000000        3106.000000\n",
       "mean      0.249437       18.469460          9302.737743           0.457806\n",
       "std       0.432757        5.536651          1590.926253           1.206493\n",
       "min       0.000000        3.625483          3663.530000          -7.778000\n",
       "25%       0.000000       14.410248          8159.340000          -0.255867\n",
       "50%       0.000000       18.147072          9193.770000           0.400088\n",
       "75%       0.000000       21.961417         10285.430000           1.147822\n",
       "max       1.000000       41.366287         18443.220000           8.472745"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>adjmortmeas_amiall30day</th>\n",
       "      <th>adjmortmeas_chfall30day</th>\n",
       "      <th>med_prev_qual_z</th>\n",
       "      <th>primcarevis_10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3106.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3012.000000</td>\n",
       "      <td>3098.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.165483</td>\n",
       "      <td>0.108969</td>\n",
       "      <td>-0.148547</td>\n",
       "      <td>80.865348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.039408</td>\n",
       "      <td>0.023565</td>\n",
       "      <td>0.863881</td>\n",
       "      <td>7.401457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-4.853847</td>\n",
       "      <td>18.331749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.145312</td>\n",
       "      <td>0.096301</td>\n",
       "      <td>-0.615591</td>\n",
       "      <td>78.803466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.162727</td>\n",
       "      <td>0.107242</td>\n",
       "      <td>-0.090228</td>\n",
       "      <td>82.202779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.183402</td>\n",
       "      <td>0.120155</td>\n",
       "      <td>0.444429</td>\n",
       "      <td>84.957865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.444663</td>\n",
       "      <td>0.344451</td>\n",
       "      <td>3.478521</td>\n",
       "      <td>95.665079</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       adjmortmeas_amiall30day  adjmortmeas_chfall30day  med_prev_qual_z  \\\n",
       "count              3106.000000              3107.000000      3012.000000   \n",
       "mean                  0.165483                 0.108969        -0.148547   \n",
       "std                   0.039408                 0.023565         0.863881   \n",
       "min                   0.000000                 0.000000        -4.853847   \n",
       "25%                   0.145312                 0.096301        -0.615591   \n",
       "50%                   0.162727                 0.107242        -0.090228   \n",
       "75%                   0.183402                 0.120155         0.444429   \n",
       "max                   0.444663                 0.344451         3.478521   \n",
       "\n",
       "       primcarevis_10  \n",
       "count     3098.000000  \n",
       "mean        80.865348  \n",
       "std          7.401457  \n",
       "min         18.331749  \n",
       "25%         78.803466  \n",
       "50%         82.202779  \n",
       "75%         84.957865  \n",
       "max         95.665079  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>diab_hemotest_10</th>\n",
       "      <th>diab_eyeexam_10</th>\n",
       "      <th>diab_lipids_10</th>\n",
       "      <th>mammogram_10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3069.000000</td>\n",
       "      <td>3054.000000</td>\n",
       "      <td>3057.000000</td>\n",
       "      <td>3029.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>83.706025</td>\n",
       "      <td>66.080221</td>\n",
       "      <td>78.307420</td>\n",
       "      <td>63.110073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.594153</td>\n",
       "      <td>7.598549</td>\n",
       "      <td>7.854145</td>\n",
       "      <td>8.397699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>16.911765</td>\n",
       "      <td>31.372549</td>\n",
       "      <td>19.661336</td>\n",
       "      <td>30.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>81.108462</td>\n",
       "      <td>61.258165</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>57.943925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>84.782609</td>\n",
       "      <td>65.977073</td>\n",
       "      <td>79.759036</td>\n",
       "      <td>63.618290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>87.679083</td>\n",
       "      <td>70.906526</td>\n",
       "      <td>83.342526</td>\n",
       "      <td>68.907563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>94.482759</td>\n",
       "      <td>95.238095</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       diab_hemotest_10  diab_eyeexam_10  diab_lipids_10  mammogram_10\n",
       "count       3069.000000      3054.000000     3057.000000   3029.000000\n",
       "mean          83.706025        66.080221       78.307420     63.110073\n",
       "std            6.594153         7.598549        7.854145      8.397699\n",
       "min           16.911765        31.372549       19.661336     30.000000\n",
       "25%           81.108462        61.258165       75.000000     57.943925\n",
       "50%           84.782609        65.977073       79.759036     63.618290\n",
       "75%           87.679083        70.906526       83.342526     68.907563\n",
       "max          100.000000        90.000000       94.482759     95.238095"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>cs00_seg_inc</th>\n",
       "      <th>cs00_seg_inc_pov25</th>\n",
       "      <th>cs00_seg_inc_aff75</th>\n",
       "      <th>cs_race_theil_2000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.025892</td>\n",
       "      <td>0.024278</td>\n",
       "      <td>0.026463</td>\n",
       "      <td>0.075402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.030576</td>\n",
       "      <td>0.030757</td>\n",
       "      <td>0.032920</td>\n",
       "      <td>0.084131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-0.013363</td>\n",
       "      <td>-0.019502</td>\n",
       "      <td>-0.001993</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.005047</td>\n",
       "      <td>0.004164</td>\n",
       "      <td>0.003455</td>\n",
       "      <td>0.015591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.013647</td>\n",
       "      <td>0.013136</td>\n",
       "      <td>0.012577</td>\n",
       "      <td>0.047192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.036453</td>\n",
       "      <td>0.034737</td>\n",
       "      <td>0.037337</td>\n",
       "      <td>0.104508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.438241</td>\n",
       "      <td>0.749106</td>\n",
       "      <td>0.196959</td>\n",
       "      <td>0.712014</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cs00_seg_inc  cs00_seg_inc_pov25  cs00_seg_inc_aff75  \\\n",
       "count   3107.000000         3107.000000         3107.000000   \n",
       "mean       0.025892            0.024278            0.026463   \n",
       "std        0.030576            0.030757            0.032920   \n",
       "min       -0.013363           -0.019502           -0.001993   \n",
       "25%        0.005047            0.004164            0.003455   \n",
       "50%        0.013647            0.013136            0.012577   \n",
       "75%        0.036453            0.034737            0.037337   \n",
       "max        0.438241            0.749106            0.196959   \n",
       "\n",
       "       cs_race_theil_2000  \n",
       "count         3107.000000  \n",
       "mean             0.075402  \n",
       "std              0.084131  \n",
       "min              0.000000  \n",
       "25%              0.015591  \n",
       "50%              0.047192  \n",
       "75%              0.104508  \n",
       "max              0.712014  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>gini99</th>\n",
       "      <th>poor_share</th>\n",
       "      <th>inc_share_1perc</th>\n",
       "      <th>frac_middleclass</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3008.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3008.000000</td>\n",
       "      <td>3106.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.379021</td>\n",
       "      <td>0.141739</td>\n",
       "      <td>0.094808</td>\n",
       "      <td>0.554244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.086677</td>\n",
       "      <td>0.065460</td>\n",
       "      <td>0.050631</td>\n",
       "      <td>0.093099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.160954</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.018570</td>\n",
       "      <td>0.215630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.317518</td>\n",
       "      <td>0.095383</td>\n",
       "      <td>0.062577</td>\n",
       "      <td>0.491883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.369998</td>\n",
       "      <td>0.129621</td>\n",
       "      <td>0.083600</td>\n",
       "      <td>0.559830</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.429472</td>\n",
       "      <td>0.175282</td>\n",
       "      <td>0.113570</td>\n",
       "      <td>0.622758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.091437</td>\n",
       "      <td>0.569170</td>\n",
       "      <td>0.734770</td>\n",
       "      <td>0.875000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            gini99   poor_share  inc_share_1perc  frac_middleclass\n",
       "count  3008.000000  3107.000000      3008.000000       3106.000000\n",
       "mean      0.379021     0.141739         0.094808          0.554244\n",
       "std       0.086677     0.065460         0.050631          0.093099\n",
       "min       0.160954     0.000000         0.018570          0.215630\n",
       "25%       0.317518     0.095383         0.062577          0.491883\n",
       "50%       0.369998     0.129621         0.083600          0.559830\n",
       "75%       0.429472     0.175282         0.113570          0.622758\n",
       "max       1.091437     0.569170         0.734770          0.875000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>scap_ski90pcm</th>\n",
       "      <th>rel_tot</th>\n",
       "      <th>cs_frac_black</th>\n",
       "      <th>cs_frac_hisp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3106.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.000182</td>\n",
       "      <td>53.224564</td>\n",
       "      <td>8.744503</td>\n",
       "      <td>6.209190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.347960</td>\n",
       "      <td>18.502524</td>\n",
       "      <td>14.483719</td>\n",
       "      <td>12.050404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-4.258739</td>\n",
       "      <td>1.816347</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.082034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.964225</td>\n",
       "      <td>39.669796</td>\n",
       "      <td>0.264501</td>\n",
       "      <td>0.917235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-0.091105</td>\n",
       "      <td>51.328668</td>\n",
       "      <td>1.691121</td>\n",
       "      <td>1.783438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.818039</td>\n",
       "      <td>64.786780</td>\n",
       "      <td>10.031043</td>\n",
       "      <td>5.107685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>9.911112</td>\n",
       "      <td>164.527310</td>\n",
       "      <td>85.965088</td>\n",
       "      <td>97.539047</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       scap_ski90pcm      rel_tot  cs_frac_black  cs_frac_hisp\n",
       "count    3107.000000  3106.000000    3107.000000   3107.000000\n",
       "mean        0.000182    53.224564       8.744503      6.209190\n",
       "std         1.347960    18.502524      14.483719     12.050404\n",
       "min        -4.258739     1.816347       0.000000      0.082034\n",
       "25%        -0.964225    39.669796       0.264501      0.917235\n",
       "50%        -0.091105    51.328668       1.691121      1.783438\n",
       "75%         0.818039    64.786780      10.031043      5.107685\n",
       "max         9.911112   164.527310      85.965088     97.539047"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>unemp_rate</th>\n",
       "      <th>cs_labforce</th>\n",
       "      <th>cs_elf_ind_man</th>\n",
       "      <th>cs_born_foreign</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.049871</td>\n",
       "      <td>0.609344</td>\n",
       "      <td>0.159118</td>\n",
       "      <td>3.441958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.017738</td>\n",
       "      <td>0.070393</td>\n",
       "      <td>0.090862</td>\n",
       "      <td>4.836270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.016092</td>\n",
       "      <td>0.319209</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.037422</td>\n",
       "      <td>0.567037</td>\n",
       "      <td>0.088637</td>\n",
       "      <td>0.898505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.046908</td>\n",
       "      <td>0.616551</td>\n",
       "      <td>0.149391</td>\n",
       "      <td>1.727323</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.058742</td>\n",
       "      <td>0.657982</td>\n",
       "      <td>0.219933</td>\n",
       "      <td>3.922074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.176995</td>\n",
       "      <td>0.860937</td>\n",
       "      <td>0.485540</td>\n",
       "      <td>50.935669</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        unemp_rate  cs_labforce  cs_elf_ind_man  cs_born_foreign\n",
       "count  3107.000000  3107.000000     3107.000000      3107.000000\n",
       "mean      0.049871     0.609344        0.159118         3.441958\n",
       "std       0.017738     0.070393        0.090862         4.836270\n",
       "min       0.016092     0.319209        0.000000         0.000000\n",
       "25%       0.037422     0.567037        0.088637         0.898505\n",
       "50%       0.046908     0.616551        0.149391         1.727323\n",
       "75%       0.058742     0.657982        0.219933         3.922074\n",
       "max       0.176995     0.860937        0.485540        50.935669"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>mig_inflow</th>\n",
       "      <th>mig_outflow</th>\n",
       "      <th>pop_density</th>\n",
       "      <th>frac_traveltime_lt15</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3017.000000</td>\n",
       "      <td>3017.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.028677</td>\n",
       "      <td>0.027522</td>\n",
       "      <td>244.325026</td>\n",
       "      <td>0.403803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.019034</td>\n",
       "      <td>0.013780</td>\n",
       "      <td>1676.096088</td>\n",
       "      <td>0.137215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.099542</td>\n",
       "      <td>0.099878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.016502</td>\n",
       "      <td>0.018767</td>\n",
       "      <td>17.479568</td>\n",
       "      <td>0.299927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.024430</td>\n",
       "      <td>0.025111</td>\n",
       "      <td>43.130142</td>\n",
       "      <td>0.385816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.036320</td>\n",
       "      <td>0.033038</td>\n",
       "      <td>104.991115</td>\n",
       "      <td>0.499088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.168671</td>\n",
       "      <td>0.153256</td>\n",
       "      <td>66940.078000</td>\n",
       "      <td>0.817636</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mig_inflow  mig_outflow   pop_density  frac_traveltime_lt15\n",
       "count  3017.000000  3017.000000   3107.000000           3107.000000\n",
       "mean      0.028677     0.027522    244.325026              0.403803\n",
       "std       0.019034     0.013780   1676.096088              0.137215\n",
       "min       0.000000     0.000000      0.099542              0.099878\n",
       "25%       0.016502     0.018767     17.479568              0.299927\n",
       "50%       0.024430     0.025111     43.130142              0.385816\n",
       "75%       0.036320     0.033038    104.991115              0.499088\n",
       "max       0.168671     0.153256  66940.078000              0.817636"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>hhinc00</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>ccd_exp_tot</th>\n",
       "      <th>score_r</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3.107000e+03</td>\n",
       "      <td>3080.000000</td>\n",
       "      <td>3069.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>32853.502978</td>\n",
       "      <td>1.121801e+05</td>\n",
       "      <td>6.092697</td>\n",
       "      <td>0.077348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6975.837500</td>\n",
       "      <td>6.318905e+04</td>\n",
       "      <td>2.103573</td>\n",
       "      <td>9.007980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>10511.805000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>3.032457</td>\n",
       "      <td>-38.687138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>28733.524500</td>\n",
       "      <td>7.704740e+04</td>\n",
       "      <td>5.027049</td>\n",
       "      <td>-4.969633</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>32234.641000</td>\n",
       "      <td>1.007748e+05</td>\n",
       "      <td>5.785282</td>\n",
       "      <td>0.834938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>36039.471000</td>\n",
       "      <td>1.285012e+05</td>\n",
       "      <td>6.735288</td>\n",
       "      <td>5.990181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>77942.648000</td>\n",
       "      <td>1.333001e+06</td>\n",
       "      <td>53.258174</td>\n",
       "      <td>32.985218</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            hhinc00  median_house_value  ccd_exp_tot      score_r\n",
       "count   3107.000000        3.107000e+03  3080.000000  3069.000000\n",
       "mean   32853.502978        1.121801e+05     6.092697     0.077348\n",
       "std     6975.837500        6.318905e+04     2.103573     9.007980\n",
       "min    10511.805000        0.000000e+00     3.032457   -38.687138\n",
       "25%    28733.524500        7.704740e+04     5.027049    -4.969633\n",
       "50%    32234.641000        1.007748e+05     5.785282     0.834938\n",
       "75%    36039.471000        1.285012e+05     6.735288     5.990181\n",
       "max    77942.648000        1.333001e+06    53.258174    32.985218"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>cs_fam_wkidsinglemom</th>\n",
       "      <th>subcty_exp_pc</th>\n",
       "      <th>taxrate</th>\n",
       "      <th>tax_st_diff_top20</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3106.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.194598</td>\n",
       "      <td>2119.407531</td>\n",
       "      <td>0.023089</td>\n",
       "      <td>0.775634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.067828</td>\n",
       "      <td>999.833466</td>\n",
       "      <td>0.013848</td>\n",
       "      <td>1.470989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.024793</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.152436</td>\n",
       "      <td>1510.192750</td>\n",
       "      <td>0.014993</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.182469</td>\n",
       "      <td>1935.919400</td>\n",
       "      <td>0.020339</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.221578</td>\n",
       "      <td>2505.411100</td>\n",
       "      <td>0.027164</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.543878</td>\n",
       "      <td>20541.918000</td>\n",
       "      <td>0.209907</td>\n",
       "      <td>7.220000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cs_fam_wkidsinglemom  subcty_exp_pc      taxrate  tax_st_diff_top20\n",
       "count           3107.000000    3107.000000  3107.000000        3106.000000\n",
       "mean               0.194598    2119.407531     0.023089           0.775634\n",
       "std                0.067828     999.833466     0.013848           1.470989\n",
       "min                0.024793       0.000000     0.000000           0.000000\n",
       "25%                0.152436    1510.192750     0.014993           0.000000\n",
       "50%                0.182469    1935.919400     0.020339           0.000000\n",
       "75%                0.221578    2505.411100     0.027164           1.000000\n",
       "max                0.543878   20541.918000     0.209907           7.220000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>pm25</th>\n",
       "      <th>pm25_mia</th>\n",
       "      <th>summer_tmmx</th>\n",
       "      <th>summer_rmax</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>8.371871</td>\n",
       "      <td>0.003540</td>\n",
       "      <td>303.126997</td>\n",
       "      <td>88.970517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.565927</td>\n",
       "      <td>0.059405</td>\n",
       "      <td>3.173950</td>\n",
       "      <td>9.689271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>290.455540</td>\n",
       "      <td>31.643282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.309710</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>300.848035</td>\n",
       "      <td>88.052494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>8.784647</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>303.290440</td>\n",
       "      <td>91.320313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>10.483764</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>305.817430</td>\n",
       "      <td>94.812389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>15.786018</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>313.872680</td>\n",
       "      <td>99.778748</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              pm25     pm25_mia  summer_tmmx  summer_rmax\n",
       "count  3107.000000  3107.000000  3107.000000  3107.000000\n",
       "mean      8.371871     0.003540   303.126997    88.970517\n",
       "std       2.565927     0.059405     3.173950     9.689271\n",
       "min       0.000000     0.000000   290.455540    31.643282\n",
       "25%       6.309710     0.000000   300.848035    88.052494\n",
       "50%       8.784647     0.000000   303.290440    91.320313\n",
       "75%      10.483764     0.000000   305.817430    94.812389\n",
       "max      15.786018     1.000000   313.872680    99.778748"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>winter_tmmx</th>\n",
       "      <th>winter_rmax</th>\n",
       "      <th>bmcruderate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.000000</td>\n",
       "      <td>3107.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>280.404875</td>\n",
       "      <td>87.469432</td>\n",
       "      <td>1029.15597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.597855</td>\n",
       "      <td>4.811207</td>\n",
       "      <td>248.38181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>264.693820</td>\n",
       "      <td>58.159798</td>\n",
       "      <td>189.30000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>275.113020</td>\n",
       "      <td>85.093342</td>\n",
       "      <td>864.29999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>280.154690</td>\n",
       "      <td>88.028793</td>\n",
       "      <td>1036.30000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>285.543750</td>\n",
       "      <td>90.747704</td>\n",
       "      <td>1194.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>298.340360</td>\n",
       "      <td>97.672874</td>\n",
       "      <td>1978.60000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       winter_tmmx  winter_rmax  bmcruderate\n",
       "count  3107.000000  3107.000000   3107.00000\n",
       "mean    280.404875    87.469432   1029.15597\n",
       "std       6.597855     4.811207    248.38181\n",
       "min     264.693820    58.159798    189.30000\n",
       "25%     275.113020    85.093342    864.29999\n",
       "50%     280.154690    88.028793   1036.30000\n",
       "75%     285.543750    90.747704   1194.10000\n",
       "max     298.340360    97.672874   1978.60000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Display all column summry statistics\n",
    "total_columns = len(covid.columns)\n",
    "\n",
    "# Iterate through the DataFrame in chunks of 4 columns\n",
    "for start in range(0, total_columns, 4):\n",
    "    end = start + 4\n",
    "    # Use .iloc for integer-location based indexing\n",
    "    display(covid.iloc[:, start:end].describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Note that some variables have missing values. This causes problems when estimating the models. Normally we’d impute missing values, but to keep things simple, given the size of our data, you should drop all observations (rows) with missing values.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Arizona', 'North Dakota', 'Illinois', 'Maine', 'Minnesota', 'South Carolina', 'Louisiana', 'Nevada', 'Pennsylvania', 'Kansas', 'New Mexico', 'Iowa', 'Indiana', 'Georgia', 'Florida', 'West Virginia', 'Montana', 'New York', 'Kentucky', 'Washington', 'Michigan', 'Delaware', 'Colorado', 'Vermont', 'Connecticut', 'New Jersey', 'District of Columbia', 'Mississippi', 'Nebraska', 'California', 'New Hampshire', 'Idaho', 'Alabama', 'Oregon', 'South Dakota', 'Maryland', 'Tennessee', 'Wyoming', 'Arkansas', 'Oklahoma', 'Missouri', 'Rhode Island', 'Virginia', 'Texas', 'Utah', 'Wisconsin', 'Ohio', 'Massachusetts', 'North Carolina'}\n",
      "49\n"
     ]
    }
   ],
   "source": [
    "# Print the states and number of states before droping missing values\n",
    "print(set(covid['state']))\n",
    "print(len(set(covid['state'])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                Count NAs\n",
      "county                  0\n",
      "state                   0\n",
      "deathspc                0\n",
      "intersects_msa          0\n",
      "cur_smoke_q1            0\n",
      "...                   ...\n",
      "summer_tmmx             0\n",
      "summer_rmax             0\n",
      "winter_tmmx             0\n",
      "winter_rmax             0\n",
      "bmcruderate             0\n",
      "\n",
      "[63 rows x 1 columns]\n"
     ]
    }
   ],
   "source": [
    "# Check for NA values\n",
    "na_count = pd.DataFrame(np.sum(covid.isna(), axis = 0),\n",
    "                        columns = [\"Count NAs\"])\n",
    "print(na_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2915, 63)\n"
     ]
    }
   ],
   "source": [
    "# Drop NAs\n",
    "covid = covid.dropna()\n",
    "print(covid.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Create a separate dummy variable for each of the 48 states and the District of Columbia in the dataset (so you’ll create 49 dummy variables in total, but dropping observations with missing values may reduce this number).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Arizona', 'North Dakota', 'Illinois', 'Maine', 'Minnesota', 'South Carolina', 'Louisiana', 'Nevada', 'Pennsylvania', 'Kansas', 'New Mexico', 'Iowa', 'Indiana', 'Georgia', 'Florida', 'West Virginia', 'Montana', 'New York', 'Kentucky', 'Washington', 'Michigan', 'Delaware', 'Colorado', 'Vermont', 'Connecticut', 'Mississippi', 'Nebraska', 'California', 'New Hampshire', 'Idaho', 'Alabama', 'Oregon', 'South Dakota', 'Maryland', 'Tennessee', 'Wyoming', 'Arkansas', 'Oklahoma', 'Missouri', 'Rhode Island', 'Virginia', 'Texas', 'Utah', 'Wisconsin', 'Ohio', 'Massachusetts', 'North Carolina'}\n",
      "47\n"
     ]
    }
   ],
   "source": [
    "# Print the states and number of states after droping missing values\n",
    "print(set(covid['state']))\n",
    "print(len(set(covid['state'])))  # we lost 2 states after droping missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>county</th>\n",
       "      <th>deathspc</th>\n",
       "      <th>intersects_msa</th>\n",
       "      <th>cur_smoke_q1</th>\n",
       "      <th>cur_smoke_q2</th>\n",
       "      <th>cur_smoke_q3</th>\n",
       "      <th>cur_smoke_q4</th>\n",
       "      <th>bmi_obese_q1</th>\n",
       "      <th>bmi_obese_q2</th>\n",
       "      <th>bmi_obese_q3</th>\n",
       "      <th>...</th>\n",
       "      <th>state_South Dakota</th>\n",
       "      <th>state_Tennessee</th>\n",
       "      <th>state_Texas</th>\n",
       "      <th>state_Utah</th>\n",
       "      <th>state_Vermont</th>\n",
       "      <th>state_Virginia</th>\n",
       "      <th>state_Washington</th>\n",
       "      <th>state_West Virginia</th>\n",
       "      <th>state_Wisconsin</th>\n",
       "      <th>state_Wyoming</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Autauga</td>\n",
       "      <td>16.548864</td>\n",
       "      <td>1</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.238095</td>\n",
       "      <td>0.208333</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>0.238095</td>\n",
       "      <td>0.260870</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Baldwin</td>\n",
       "      <td>8.959118</td>\n",
       "      <td>1</td>\n",
       "      <td>0.268097</td>\n",
       "      <td>0.233503</td>\n",
       "      <td>0.167464</td>\n",
       "      <td>0.176991</td>\n",
       "      <td>0.298050</td>\n",
       "      <td>0.262467</td>\n",
       "      <td>0.193237</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Barbour</td>\n",
       "      <td>6.609756</td>\n",
       "      <td>0</td>\n",
       "      <td>0.228571</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.294118</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.545455</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Bibb</td>\n",
       "      <td>6.038192</td>\n",
       "      <td>1</td>\n",
       "      <td>0.244444</td>\n",
       "      <td>0.280000</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.150000</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>0.190476</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Blount</td>\n",
       "      <td>1.503713</td>\n",
       "      <td>1</td>\n",
       "      <td>0.304348</td>\n",
       "      <td>0.260870</td>\n",
       "      <td>0.352941</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.347826</td>\n",
       "      <td>0.318182</td>\n",
       "      <td>0.529412</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 109 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    county   deathspc  intersects_msa  cur_smoke_q1  cur_smoke_q2  \\\n",
       "0  Autauga  16.548864               1      0.333333      0.238095   \n",
       "1  Baldwin   8.959118               1      0.268097      0.233503   \n",
       "2  Barbour   6.609756               0      0.228571      0.250000   \n",
       "3     Bibb   6.038192               1      0.244444      0.280000   \n",
       "4   Blount   1.503713               1      0.304348      0.260870   \n",
       "\n",
       "   cur_smoke_q3  cur_smoke_q4  bmi_obese_q1  bmi_obese_q2  bmi_obese_q3  ...  \\\n",
       "0      0.208333      0.133333      0.375000      0.238095      0.260870  ...   \n",
       "1      0.167464      0.176991      0.298050      0.262467      0.193237  ...   \n",
       "2      0.181818      0.111111      0.294118      0.571429      0.545455  ...   \n",
       "3      0.181818      0.150000      0.466667      0.375000      0.190476  ...   \n",
       "4      0.352941      0.166667      0.347826      0.318182      0.529412  ...   \n",
       "\n",
       "   state_South Dakota  state_Tennessee  state_Texas  state_Utah  \\\n",
       "0               False            False        False       False   \n",
       "1               False            False        False       False   \n",
       "2               False            False        False       False   \n",
       "3               False            False        False       False   \n",
       "4               False            False        False       False   \n",
       "\n",
       "   state_Vermont  state_Virginia  state_Washington  state_West Virginia  \\\n",
       "0          False           False             False                False   \n",
       "1          False           False             False                False   \n",
       "2          False           False             False                False   \n",
       "3          False           False             False                False   \n",
       "4          False           False             False                False   \n",
       "\n",
       "   state_Wisconsin  state_Wyoming  \n",
       "0            False          False  \n",
       "1            False          False  \n",
       "2            False          False  \n",
       "3            False          False  \n",
       "4            False          False  \n",
       "\n",
       "[5 rows x 109 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "covid_new = pd.get_dummies(covid, columns = ['state'])\n",
    "covid_new.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. Split the sample into a training set (80%) and a test set (20%). Be sure to set a random state so you can replicate your work. There are odd patterns in the results with some splits of the data, so to avoid confusion and facilitate grading, please set random_state=11."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define X, y and test, training sets\n",
    "X = covid_new.copy()\n",
    "X = X.drop(['county'], axis=1)\n",
    "y = X.pop('deathspc') \n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X, y, test_size=0.2, random_state=11)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6. Using the training set, fit a model of COVID-19 deaths per capita (y =deathspc) as a function of the Opportunity Insights and PM COVID predictors listed in the spreadsheet, as well as state-level fixed effects (the state dummy variables) using OLS.\n",
    "### (a) Using the model you fit, calculate and report the MSE in both the training and test sets.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='blue'> Training set MSE is 1440.7, and test set MSE is 2408.75. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "ols = LinearRegression().fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficients: [ 2.37702564e+00 -1.03471715e+01 -1.61148248e+01  4.67713593e+00\n",
      "  7.46622702e-02 -1.58893213e+00  1.34545068e+01 -1.72708563e+01\n",
      "  6.36324857e+00 -9.38672528e+00  6.85699486e+00  3.37188762e+00\n",
      " -2.51138257e+00 -7.27164148e+00  4.14098624e-02 -9.30073703e-04\n",
      "  5.94541681e-01  2.11075403e+00 -1.71365314e+01  9.11057620e+00\n",
      " -2.80134636e-01 -8.49686889e-01 -9.26122723e-02 -3.73950393e-01\n",
      "  2.25634841e-02  7.54010877e+02 -5.15841419e+02 -2.24027029e+02\n",
      "  1.06511716e-01 -4.22160985e+01  3.20079680e+01 -1.59458593e+01\n",
      " -5.95117337e+01 -5.40874454e+00  1.34324468e-01  6.76379157e-01\n",
      " -1.45762130e-01 -2.41393005e+02 -5.31091177e+01  4.53444740e+01\n",
      "  1.03909357e+00 -9.85661503e+00 -2.24869965e+02  1.01949088e-02\n",
      " -1.80329786e+01  7.52128985e-04 -1.00579775e-05  1.34950098e+00\n",
      "  2.46993959e-01  2.92115799e+01 -6.42863930e-04 -9.42957664e+01\n",
      " -7.93339858e-01 -5.01525805e-01 -1.29624203e+01 -5.13678437e-01\n",
      " -7.68260713e-01  1.38200161e+00  5.53849066e-02 -2.72799877e-03\n",
      " -2.29050833e+01 -2.94350341e+01 -2.87963504e+01 -5.02418245e+01\n",
      "  8.23716594e+00  1.57093555e+02  3.61464687e+00 -3.86502651e+01\n",
      " -2.28491096e+00 -1.55605213e+01  9.02981850e+00  3.27859322e+01\n",
      "  1.34106231e+01 -7.12934740e+00  2.48139144e+00  6.01724605e+01\n",
      " -1.48303724e+00  2.19336755e-01  8.43106787e+01  2.02564414e+01\n",
      "  1.41898216e+01 -3.39428343e+00 -5.51585449e+00 -6.49674045e-01\n",
      "  6.60318091e+00 -3.71877823e+01 -3.43334114e+00 -2.43527533e+01\n",
      "  3.64378431e+01 -2.31403020e+01  1.54377476e+01  1.11800578e+01\n",
      " -1.12879104e+01 -1.84423241e+01  1.11731656e+01 -1.21882057e-01\n",
      " -4.29058492e+01  7.70761356e+00 -1.22569984e+01 -3.00848132e+01\n",
      " -3.13453279e+01  5.02861262e+00 -2.55946004e+01 -1.15976623e+01\n",
      " -6.90816830e+00  1.53026970e+00 -1.61944621e+01]\n",
      "Intercept: 51.8659986333785\n"
     ]
    }
   ],
   "source": [
    "# Examine coefficient estimates\n",
    "print('Coefficients:', ols.coef_)\n",
    "print('Intercept:', ols.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "OLS-Train MSE:  1440.7\n"
     ]
    }
   ],
   "source": [
    "y_pred_train = ols.predict(X_train)\n",
    "mse_train_ols = mean_squared_error(y_train, y_pred_train)\n",
    "print(\"OLS-Train MSE: \", round(mse_train_ols,2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "OLS-Test MSE:  2408.75\n"
     ]
    }
   ],
   "source": [
    "y_pred_test = ols.predict(X_test)\n",
    "mse_test_ols = mean_squared_error(y_test, y_pred_test)\n",
    "print(\"OLS-Test MSE: \", round(mse_test_ols,2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### (b) Why might you be concerned about overfitting in this context? Is there any evidence of overfitting? Briefly explain."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='blue'> \n",
    "\n",
    "#### Concerns\n",
    "\n",
    "- Complexity and dimentionality: The COVID-19 pandemic is influenced by a multitude of factors, and with the high dimensionality in this context, the complex interplay between these variables can lead to models that, while fitting the training data well, fail to generalize to unseen data or future scenarios as it captures noise in the training data,  which leads to unreasonable policy decisions. \n",
    "\n",
    "- Linear relationship: linear relationships are the easiest to interpret, but, with overfitting, some non-significant predictors can be falsely interpreted and used for policy.\n",
    "\n",
    "- Time constraint: This is crucial because the progression of the pandemic is inherently time-dependent, with dynamics that can change rapidly due to new virus strains, changes in public behavior, and policy interventions. A model trained on data from a specific time period might overfit to the conditions of that period and not perform well as the situation evolves.\n",
    "\n",
    "\n",
    "#### Evidence for overfitting\n",
    "\n",
    "As observed from our plots of flexibility versus error from PSet1, an increase in flexibility consistently leads to a reduction in training error. This trend occurs because a model tailored to the training set becomes increasingly adept at capturing the specific nuances of that data. However, this trend does not necessarily extend to the testing set. Beyond a certain threshold, where the testing error reaches its minimum, further increases in flexibility lead to a rise in testing error. While bias tends to decrease with greater model flexibility, variance, and consequently, the testing error, tend to escalate.\n",
    "\n",
    "A significant discrepancy between the model's performance on the training set and the test set might indicate overfitting given the huge gap between testing and training error--2408.75-1440.7 = 968.05. It suggests that the model performs well on the data it was trained on but poorly on unseen data, indicating that testing error may have past its minimum point for best bias and starts to surge as variance increases due to fitting these intricate patterns in the training data too closely. Intuitively, using 100+ variables, it is no wonder the model is prone to overfitted."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7. Use the training set to estimate ridge regression and the lasso analogs to the OLS model in the previous question. For each, you should report a plot of the cross-validation estimates of the test error as a function of the value of the hyperparameter (lambda) that indicates the tuned value of lambda. Hint: to do so you should be sure to scale your predictors. Then you should tune the hyperparameter by:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several ways to scale or normalize variables. Scaling here refers to transforming each predictor so that\n",
    "the mean is 0 and the variance is 1. Similar to the discussion in a previous assignment, there is a trade-off between\n",
    "scaling before splitting into training and test sets or after. The former uses more precise estimates of the population\n",
    "mean and variance. The latter does not bring information from the test set into the training set (which could result in\n",
    "inadvertent overfitting). Given the sample size in this question, you should scale after splitting your data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Scale the data\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(X_train)\n",
    "X_train = scaler.transform(X_train)\n",
    "X_test = scaler.transform(X_test)\n",
    "\n",
    "ridge = Ridge()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### (a) Calculating each model for a grid or range of values of lambda. You will want to adjust the values you search over based on what you learn in the subsequent steps, but start by using 100 values of lambda from 0.01 to 100."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hint: to do so, start by creating a range of 100 numbers from −2 to 2. You can do that by starting at −2 and incriminating each subsequent number by 1/25 . Let a denote those values and b denote the values of the grid for tuning λ.You can calculate b = 10^a."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "# initial attempt -- do not reveal a U-shaped curve\n",
    "# a = np.linspace(-2, 2, 100)  \n",
    "# lambda_range = np.power(10, a)\n",
    "\n",
    "a = np.linspace(-2, 3, 100)\n",
    "lambda_range = np.power(10, a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "def vector_values(grid_search, trials):\n",
    "    '''\n",
    "    Extracts mean and standard deviation scores from a GridSearchCV object.\n",
    "    ​\n",
    "    Parameters\n",
    "    ----------\n",
    "    grid_search : GridSearchCV object\n",
    "    The result of a grid search performed using scikit-learn's GridSearchCV method.\n",
    "    trials : int\n",
    "    The number of trials performed during the grid search.\n",
    "    ​\n",
    "    Returns\n",
    "    -------\n",
    "    mean_vec : ndarray of shape (trials,)\n",
    "    Array of mean scores, with the negative of each score stored in the array.\n",
    "    std_vec : ndarray of shape (trials,)\n",
    "    Array of standard deviation scores.\n",
    "    '''\n",
    "    mean_vec = np.zeros(trials)\n",
    "    std_vec = np.zeros(trials)\n",
    "    i = 0\n",
    "    final = grid_search.cv_results_\n",
    "    \n",
    "    for mean_score, std_score in zip(final[\"mean_test_score\"],\n",
    "                                     final[\"std_test_score\"]):\n",
    "        mean_vec[i] = -mean_score\n",
    "        std_vec[i] = std_score\n",
    "        i = i+1\n",
    "    return mean_vec, std_vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "def highlight_min(data, color='yellow'):\n",
    "    attr = f'background-color: {color}'\n",
    "    if data.ndim == 1:  # Series\n",
    "        is_min = data == data.min()\n",
    "        return [attr if v else '' for v in is_min]\n",
    "    return ''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### (b) Using 10-fold cross-validation (10FCV) (on the training data) to estimate the test error for each model at the given value of lambda. Use the KFold function to shuffle the data randomly for 10FCV. Please set random_state=25 when doing so."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a parameters grid\n",
    "param_grid = [{'alpha': lambda_range}]\n",
    "\n",
    "# Running Grid Search over the alpha (regularization) parameter\n",
    "kf = KFold(n_splits=10, random_state = 25, shuffle=True)\n",
    "grid_search_ridge = GridSearchCV(ridge, param_grid, cv=kf,\n",
    "                                 scoring='neg_mean_squared_error')\n",
    "grid_search_ridge.fit(X_train, y_train)\n",
    "\n",
    "# Extract results for all tested alphas\n",
    "tested_alphas_ridge = [params['alpha'] for params in grid_search_ridge.cv_results_['params']]\n",
    "\n",
    "# Assuming vector_values is properly defined to extract mean and standard deviation values\n",
    "mean_vec_ridge, std_vec_ridge = vector_values(grid_search_ridge, len(tested_alphas_ridge))\n",
    "results_cv_ridge = pd.DataFrame({'alpha': tested_alphas_ridge,\n",
    "                                 'MSE': mean_vec_ridge}) # 'standard deviation': std_vec_ridge\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a parameters grid\n",
    "param_grid = {'lasso__alpha': lambda_range}\n",
    "\n",
    "# Creating a pipeline and GridSearchCV object\n",
    "pipeline = make_pipeline(StandardScaler(), Lasso(random_state=25))\n",
    "grid_search = GridSearchCV(pipeline, param_grid, cv=10,\n",
    "                           scoring='neg_mean_squared_error')\n",
    "\n",
    "# Fitting the GridSearchCV object\n",
    "grid_search.fit(X_train, y_train)\n",
    "\n",
    "# Extract results and convert 'mean_test_score' to positive values\n",
    "# Note: the term mean_test_score refers to the average mean squared error (MSE) \n",
    "# across the cross-validation folds for each alpha value when using GridSearchCV.\n",
    "results = pd.DataFrame(grid_search.cv_results_)\n",
    "results['mean_test_score'] = -results['mean_test_score']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### (c) Plotting the 10FCV estimates of the test error as a function of the value of lambda."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mean_vec, std_vec = vector_values(grid_search_ridge, len(tested_alphas_ridge))\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8, 6))\n",
    "ax.set_title('Ridge Regression, with log(Alpha)', fontsize=12)\n",
    "log_alpha_param = np.log10(lambda_range) # log transform to aid with interpretability of axis\n",
    "ax.plot(log_alpha_param, mean_vec, label='Mean MSE')\n",
    "ax.errorbar(log_alpha_param, mean_vec, yerr=std_vec, fmt='o', ecolor='r',\n",
    "            capsize=5, alpha=0.5, label='MSE Std Dev')\n",
    "ax.set_ylabel(\"MSE\", fontsize=12)\n",
    "ax.set_xlabel(\"log(Alpha)\", fontsize=12)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAr0AAAIiCAYAAADFFHBOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/SrBM8AAAACXBIWXMAAA9hAAAPYQGoP6dpAACDTklEQVR4nOzdd3hUVeLG8e+dmkkjlUDoSJNiQJAiIGAXXQvqrnXtZYXFXrAgFtQVu4BtEbGsooBY14I/FQuKIAT7gii9pBDSM+3+/phkyJBCElImw/t5njyZuefce8+dO8m8OTn3XMM0TRMRERERkQhmaekGiIiIiIg0NYVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIi0Srq3kojUh0KviOy3888/n969e4d89enTh0MPPZQJEybw1ltvhdTv3bs3Tz755D63ef755zdls4NuueWWKu3v3bs3gwYN4i9/+Qtz585tlnY0t7qch3Dx5JNP0rt37+DzlStXcvnllwefb968md69e7No0aIGbd/v9zN27Fh69+7Njz/+WKW8Idvf3zaJSOOytXQDRCQy9O3blzvvvDP43OfzsX37dl544QVuuukmEhISGDNmDADz58+nXbt2LdXUaqWmpjJz5szgc9M0yc7O5rXXXuOBBx7A6XRyzjnntGALG184noeanHnmmYwePTr4/I033uD3339vtO1/9dVXZGdn0717d1577TXuvffeRtu2iIQHhV4RaRSxsbEMHDiwyvIjjjiCESNGsGjRomDora5eS3M4HNW2a+zYsRx99NEsWrQo4kJvOJ6HmrRr165JA/qiRYsYNGgQo0eP5qmnnuKWW24hNja2yfYnIs1PwxtEpEk5nU4cDgeGYQSX7f1v9a1btzJp0iQGDx7MyJEjqx1O4PF4eOihhzjiiCM45JBDuOSSS1i8eDG9e/dm8+bNwXorVqzgvPPOIyMjg6FDh3LzzTeTm5vb4Pbb7XZcLldI+yHQ03jiiSfSv39/xo4dy5NPPonP5wup8+abbzJ+/HgGDBjAySefzLJly+jbt2/w392LFi2ib9++vPHGG4wcOZKhQ4eybt06AJYsWcKECRMYMGAAI0eO5N5776W4uDi47dLSUqZNm8YRRxxB//79Of7445kzZ07I/ufNm8fxxx/PgAEDGD16NNOmTaOwsDBYvvd52LlzJ1OmTGHMmDEccsghnHHGGXzyySch2+zduzevvPIKt912G0OHDmXQoEFcffXVZGdn1/k1ffHFF+nTpw+7du0KLps1axa9e/dm2bJlwWVLliyhT58+7NixI2R4wy233MKbb77Jli1bqgwfyMrKYvLkyQwaNIihQ4dyxx13UFRUVGt7du/ezZIlSxg3bhwnnXQSJSUlVYbk7G3RokX07t2bzMxMTjvtNA455BD+8pe/8MEHH1Spu682lZaW8vDDD3PsscfSv39/Dj30UC666CJ++eWX2l9IEakXhV4RaRSmaeL1eoNfZWVlrF+/nilTplBUVMQpp5xS7XrFxcWcd955/O9//+Oee+7hjjvu4I033mDVqlUh9aZOncq8efM477zzmDVrFikpKdxxxx0hdb777jsuvPBCoqKieOyxx7j11ltZvnw5f//73yktLd3nMVRuv9vtZvPmzdx///388ccfnHrqqcF6zzzzDHfccQcjRozg6aef5txzz+W5554Lac/ixYu55ZZbOPTQQ5k9ezbHHXccV111VZVg7PP5eP7555k+fTpTpkzhoIMO4p133mHixIl0796dWbNmMWnSJN5++22uuuqq4MVb9913H0uXLuXmm29mzpw5HHXUUTz44IMsXLgQgHfffZcZM2Zw7rnnMmfOHCZOnMhbb73FPffcU+2xZ2dnc8YZZ7BixQquvfZannzySTp06MDEiRN5++23Q+o++uij+P1+HnnkEW666SY+/fRT7rvvvn2+vhXGjh2LaZp88803wWUVj7/77rvgsqVLl9K3b1/S0tJC1r/qqqsYM2YMqampzJ8/n7FjxwbLHn/8cdq3b8/s2bO54IILeP3110OGrVTnnXfewefz8Ze//IX09HSGDx/O/Pnz63QsV1xxBUcddRQzZ86kW7duXHPNNXz++echdfbVpptuuomFCxdy+eWX8/zzzzNlyhTWrl3L9ddfr4v1RBqRhjeISKP47rvv6NevX8gywzDo1asXjz/+OOPGjat2vTfffJOtW7fy7rvv0qNHDwAyMjI45phjgnU2btzIm2++yc0338xFF10EwOjRo8nOzubLL78M1nv44Yfp1q0bzzzzDFarNbitE088kYULF3LuuefW2P4tW7ZUaT9A165dufPOOzn77LMBKCgoYPbs2fztb3/j9ttvB2DUqFEkJCRw++23c9FFF9GzZ8/gMVeMDR09ejR2u52HH364yj6uvPLKYHAzTZOHHnqI0aNH89BDD4W048ILL+Tzzz9n7NixLF++nJEjR3LiiScCMGzYMKKjo0lOTgZg+fLldOzYkXPPPReLxcLQoUOJjo5m9+7d1R7/3Llzyc3N5cMPP6RDhw4AjBkzhgsvvJAHH3yQk046CYsl0E/Sq1cv7r///uC6a9asqbaHsyadO3emW7duLFu2jBNOOIGSkhJWrVpFv379QkLvF198wYQJE6pdPykpKWRISkUv+HHHHceUKVMAGDFiBF999VVIuK7OokWLOOKII0hNTQVgwoQJ3HjjjXz//fcceuihta57/vnnM3HiRCBwjk877TRmzZoVHMqzrza53W6Kioq4/fbbGT9+PABDhw6lsLCQBx54gOzs7GC7RGT/qKdXRBpFv379WLBgAQsWLGD27Nn06tWLrl278thjj3H88cfXuN6KFSvo3LlzMPACtG/fPmS86bfffotpmlW2c9JJJwUfl5SUkJmZyZgxY0J6nTt16sRBBx3EV199VWv7U1NTg+1//vnnGTJkCG3btuWBBx7gnHPOCQ5vWLVqFaWlpRx55JEhPcNHHnkkELggasOGDWzdurVKeysC6t4OPvjg4OP169ezffv2Kts/7LDDiI2NDR7HsGHDeP3117nssst4+eWX2bRpExMnTgyG5+HDh/PHH38wYcIEZs6cyQ8//MBf/vKXGmfEWL58OYMGDQoG3gonn3wyWVlZrF+/Prhs77HA7dq1o6SkpNbXd29jx47l66+/BgIzMdjtdv7+97+TmZmJ2+1m3bp1bN26NaQXty6GDBkS8rxjx47k5+fXWP/XX3/lp59+4thjjyU/P5/8/HyGDx9OdHR0nXp7TzvttOBjwzA45phjWLNmTch/Fmprk8PhYM6cOYwfP54dO3bwzTff8Nprr/Hpp58CgVAsIo1DPb0i0ihiYmIYMGBA8HlGRgYnn3wyF198MYsWLSIpKana9Xbv3k1iYmKV5ampqcFxohVjcit6MStUfp6fn4/f7+e5557jueeeq7I9p9NZa/sdDkdI+w899FBOP/10LrvsMt544w26desGQF5eHkDIdFmV7dy5s8b2pqSkVLtOdHR08HHF9u+66y7uuuuuarcPcNttt9GuXTvefvtt7rnnHu655x4GDRrEtGnT6NOnD+PHj8fv9/Of//yH2bNnB4cr3HDDDcEexcp2795Np06dqiyvaHPl4OhyuULqWCyWev8bfsyYMcydO5fNmzezbNkyDj30UEaMGEFZWRmZmZn8+OOPpKam0r9//3ptt75tW7BgAQBTpkwJ9sZW+O9//8utt95KmzZtaly/bdu2Ic+Tk5MxTbNer9cXX3zBfffdx/r164mJiaFPnz7B94SGN4g0HoVeEWkSKSkpTJ06lauvvprp06dX+299gMTERDZs2FBleUX4A4JjOrOzs0lPTw8ur3yBWkxMDIZhcOGFF1bbo7p38NgXl8vFAw88wN/+9jemTJnCq6++imEYxMfHA/DQQw/RtWvXKutVDrY5OTkhZXs/r07F9m+66SaGDh1apbwigDkcDv7xj3/wj3/8g61bt/Lpp58ye/Zsrr/+et577z0g0BN+0kknUVBQwJdffslzzz3HjTfeyODBg6uMk23Tpg1ZWVlV9lexrLo/TPbHkCFDiI2NZdmyZXzzzTccd9xxpKWl0bVrV7799ltWrlzJ2LFjq1xA2JjcbjfvvPMOxx57LOedd15I2ebNm7n11lt58803ufDCC2vcRl5eXsg5z87Oxmq1kpCQEPwDpTYbN25k4sSJHH300TzzzDN06tQJwzB45ZVX+OKLLxp8bCJSlYY3iEiTOf744xk9ejTvvvsuy5cvr7bO8OHD2bx5Mz/88ENwWW5uLqtXrw4+Hzx4MFarlY8//jhk3Y8++ij4ODY2lr59+7J+/XoGDBgQ/OrZsydPPvkk3377bb3bf8ghh/DXv/6VVatWsXjxYiDQg22329mxY0fIfmw2G4888gibN2+mXbt2dO7cudb21qR79+4kJyezefPmkO2npaXx8MMP8/PPP1NaWspxxx3H888/D0B6ejrnnnsuJ554Ilu3bgXgmmuuCY41jYuL44QTTuCqq67C6/VWG8YOO+wwVq1axZYtW0KWv/3226SmptKlS5d6v361sdvtjBw5kk8++YRffvklGPCHDx/OZ599xooVK2ocBw4Exxfvj//7v/8jLy+Ps846i2HDhoV8nX766XTt2nWfQxyWLFkSfGyaJh999BGDBw/G4XDUqQ0//vgjZWVlXH755XTu3DkY8isCr3p6RRqPenpFpEndeuutnHzyydx77728+eabwQvMKpxyyim8+OKLTJo0iWuvvZbY2Fieeuop/H5/sE6nTp04/fTTeeSRR/B4PPTp04ePP/44OO6xIgBdd911XH755Vx//fWcfPLJwZkRMjMzueqqqxrU/muuuYb//ve/PPzwwxxzzDEkJiZy6aWX8vjjj1NYWMiwYcPYsWMHjz/+OIZh0KdPHwzDYPLkydxwww3ceeedHHPMMfz666/MmjUrpL3VsVqtXHvttUydOhWr1cq4cePIz89n9uzZ7Nixg379+hEVFUW/fv2YOXMmdrud3r1788cff/Dmm29y3HHHAYHweOedd/Kvf/2LI444gvz8fGbOnEnXrl3p06dPlf1edNFFvP3221x44YVMmjSJhIQEFi9ezDfffMN9991Xr5CZm5vLxo0b6dGjR61z3Y4ZM4Zbb72V6Ojo4DCGYcOG8dprr+F0Ojn88MNrXDc+Pp7s7Gw+//zzkDHR9bFw4UKSk5MZPnx4teUnn3wyTzzxBN9++22Vsc4VHnzwQcrKyujWrVvwhhnz5s2rcxv69euHzWZjxowZXHzxxbjdbhYtWsRnn30GEDJNnYjsH/X0ikiT6t69O+effz6//fYbr776apVyh8PBvHnzOOyww5g+fTq33norI0aMqNLLd8cdd3DWWWfx/PPPc9VVV7F9+3b+8Y9/AHvGxI4aNYo5c+awfft2Jk+ezE033YTVamXu3LkNvhFDYmIiV199NVlZWcHQes0113DLLbfw8ccfc9lllzFjxgwGDx7Myy+/TFxcHAB/+ctfuPvuu1m2bBlXXnkl7733HrfddltIe2ty5pln8vDDD/P9999z5ZVXMm3aNDp27MhLL70UHHd79913M2HCBJ5//nkuvvhiZs+ezRlnnMG0adMAOOuss7j99ttZunQpV155JVOnTuWggw7i+eefx263V9lnamoqr776Kv369ePee+/l6quvZtu2bcyePZvTTz+9Xq/ZZ599xt/+9jd++umnWuuNGTMGwzA49NBDsdkCfTDDhg3DMAyGDRtW65CUCRMmBKdUq+iFr48dO3bw1Vdfcfzxx1f5Q6zCKaecgmEYvPbaazVuZ9q0abz++utMmjSJrKys4EWQddWlSxcefvhhduzYwT/+8Q+mTp0KwEsvvYRhGKxYsaJ+ByYiNTJM/e9ERMJcXl4eS5cuZfTo0SFjS//1r3+xaNGiBg1daGrvvvsuffv2pXv37sFln332GVdccQVvvfVWtb2tkeSmm27inHPOaVV3fauPRYsWMWXKFD755BM6duzY0s0RkTrQ8AYRCXsul4vp06dz8MEHc8EFFxAdHc3q1at5+eWXueKKK1q6edV6++23efTRR7nmmmto3749GzZs4IknnmDo0KERH3jXrVtHZmZmsNdZRCQcqKdXRFqFX375hccee4zVq1dTUlJC586dOeusszj33HOb9Ar/htq1axcPP/wwS5cuJTc3l5SUFI477jgmT55MTExMSzevSeXm5lJWVkb79u1builNRj29Iq2PQq+IiIiIRDxdyCYiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8TRP7z7k5BTQHPNbGAYkJ8c12/6k8ekctn46h62fzmHrpvPX+jX3OazYX10o9O6DadKsP3jNvT9pfDqHrZ/OYeunc9i66fy1fuF4DjW8QUREREQinkKviIiIiEQ8hV4RERERiXga07sfTNPE7/fh9/v3e1uGAaWlpXg87rAbAyP7ZrFYsFqtLd0MERERqYFCbwN5vR52787F4ylttG3m5loaJUBLy3A4omjTJqqlmyEiIiLVUOhtANM0ycnZjsVioU2bFKxWG4Zh7Pd2rVYDn0/dvK2NaZr4fF4KC/NYv349ycnpwP6/H0RERKTxKPQ2gNfrwTT9tGmTisPReD17NpsFr1c9va2TE6vVSl5eFl6vB5vN0dINEhERkUp0Idt+MAy9fLKH3g8iIiLhS5/SIiIiIhLxFHpFREREJOIp9B5gRo0awqhRQ9i+fXuVssWLFzBq1BDmzHmm2dv1/vvvMGrUECZPvrLa8ssvv5BRo4awbdvW4LI33niN8847k3HjRnDyycdx//13k5OTHSyfM+eZ4PHu/TV9+rSmPiQREREJI7qQ7QBks9n46qvPOf30v4UsX7r0s0aZhaKhbDYbmZmrKCgoIC4uLrg8OzuL3377JaTuG2+8xquvvsR1193MQQf1IDs7i6eeepLrr5/M88+/jMUS+Huuf/9DmD79wSr7cjo1tZiIiMiBRD29B6CMjEP58sulIcuKigr58ccf6Nmzdwu1ClJSUmnXrj3Lln0VsvyLLz7n4IP7hSz773/f4ayzzmXUqCNo3z6dAQMymDZtOuvW/Y+ff/4pWM9ms5GcnFLlKzY2tlmOSURERMKDQm8jMk2TEo+v4V/u+q9jNuD2baNHH8Hq1d9TVFQYXPb111+SkTGQ6OjokLqLFy/kzDNP5phjRjNp0uX8/vu6YFlW1k5uv/0mjj9+HOPGjeDii89lzZrVAGzbtpVRo4bw+ef/x1//egpHHnk4N910Dfn5u2tt26hRY/jqq89Dln3xxWccccTYkGWGYWH16lV4PJ7gsrZt03j55Tfo2bNn3V8MEREROSBoeEMjMU2TS1/LZM3W/Gbdb0Z6PM+dlVGvYQndu/cgJaUt33yzjKOOOgYIDG0YPXosH33032C9L79cyty5z3LTTbfTuXMXPvjgPSZPvoJXX32T+Ph47r77DmJj43jmmbn4/X6efvpJHn74AebNey24jRdfnMu0adMxTbjllut49dWXueKKiTW2bfToMdxyy3V4vV5sNhuFhYEe6EmTrmH27CeC9c44429Mnz6N004bz+GHj2Lw4MMYMWIkXbt2q8/LJyIiIgcI9fQ2otZ0D67Ro4/gq68CQxzcbjffffcNo0ePCanzn/+8yPnnX8TIkaPp1Kkzl132D9LS2vPRR+9jmiajR4/l2mtvpEuXrnTr1p0JE/7KH3+sD9nGJZdcQd++/enXrz/HHns8v/76c63tGjAgA6vVyqpVK4FAD/TAgYNwuUJ7oE844SRmzHicnj178dFH/+Wee6Zy6qknMG/enJB6a9as5phjRlf5ysxc1aDXTURERFon9fQ2EsMweO6sDEr3445qNqsFr69+60fZLA26+GzUqDHcfvvNeL1eVq5cTvfuPUhMTAqps2HDH8ye/STPPDMruMztdrNp00YMw+C0085gyZIP+fHHNWzY8Ce//fYrfn9o+zt27BR8HB0dg9frrbVdVquVww8fzVdfLeWww4ZVO7ShwogRIxkxYiTFxUWsWPEdb721iOeee4quXbszZsw4AHr3Ppg777y3yrqpqam1tkNEREQii0JvIzIMA5fd2uD1A7chbp7+4kMOGQgEekKXLv282mDp8/mYPPk6hgwZGrI8JiYGv9/PtddOpKCggKOOOoaRI4/A4/Fw2203htS12+0hz+syBnnUqDE8+eQjXHXV1Xz33Tdcd93NlJaWBMt37NjOSy/NZfLk63E4HERHx3DEEWMZPXoMV155MStWLA+GXqfTGRK8RUREZP8ZhQUYRUXVF3oKMXIKq/0PuBkTgxkbV01J01PoPUDZbDZGjBjJV18t5euvl3L++c9XqdOpUxeysnaGhMb77ruLI44YS3p6B1av/p533vmYxMREABYtegOoW7CtzdChw8nNzWXBgtfo0aMXiYmJbNu2J/Q6HA7eeWcxQ4eOCAnrhmEQExNLQkLCfu1fREREamfLXI39qy9CF/p8OL79Guw2XIOHYlqqdgR6Ro7GM3J0M7UylELvAWz06DHcd9/dpKd3ID29Q5Xys846lwceuJdOnTozYEAGb721iP/7v485//yLcDqdWCwWPvnkQ0aNGsMvv/zE888Hbmrhdrv3q10ul4shQ4bywgtzuPTSK6qUJyYmccopE7j//rvJzc3msMOGU1hYyBdffMYvv/zETTfdFqzr9XpDblhRwWq1KRyLiIg0kDdjIL4ee82W5HZjlJZgdzkoOe8CsDuqrGfGxDRTC6tS6D2ADR06Aq/XW+UCtgpHHXUsubm5/PvfT5Obm0u3bt35178epVOnzgBcf/0tvPDCv3nmmVl06tSFq6++gXvvvZO1a38jOTllv9o2evQYvv76C0aPHltt+eTJ19OuXTqLFr3Bk08+is1mIyPjUGbNepZ27doF6/344xpOOeX4Kut36NCR+fMX71cbRUREDlRmbFzVYQpuN8TGQrQTM60dZjWhtyUZ5v7+LzrCZWcXsPcr5PG4ycnZRnJye+yNeEIDY3obfiGctCyPx82uXTtISkrDZguvH3SpG8OAlJS4an/upXXQOWzddP5aObebmMcfIjraSfbl/2yW0FvxnqkLTVkmIiIiIhFPoVdEREREIp7G9IqIiIg0UK1Td9WiJafuOlAp9IqIiIg0UE1Td9m/+RoAz/DDwRpeU3cdqBR6RURERBqoxqm7SgLzy5eedwE4wmvqrgOVQq+IiIhIA9U0dZcZGwuAP61dtaFXmp9CbzPa17gfw2rB4qs6ZZnG/YiIiIjsH4XeZrSvcT++w0fiN6pOqKFxPyIiIiL7R6G3Ge1r3I/77xfis1Q9JRr3IyIirVm9ZjjwFGLkFGKg/3RK41LobUbVjvspK4U4P+DBcHnwt0mHanp7G8uoUUMAWLDg3ZDb9QIsXryAhx56gIsuuoxLLrkCgF9//YVnn53FDz+sAUx69erDhRdewmGHDQdg27atnHnmyTXu78svV1S7vLi4iOeee5rPPvuEvLxdpKW147jjxnPeeRdit9sB2LJlMxs3bmDEiJHVbmPOnGdYtWolM2c+W6Xs++9XMHnylcHnVquV+Pg2DB06nIkTryYpKbnGNouISOOq6wwHhgFEO3EVl2Ga+k+nNC6F3hZkzV2LY+172O3LMQwf1u8exZrYE3f34/El9dz3BhrIZrPx1Vefc/rpfwtZvnTpZxiGEXy+c+cOrr76Ss466zyuvvoGDAM+/vhDbrjhambPnkO/fv2DdZ97bh5t26bVuQ333juN/Pzd3H33/aSkpLJu3f945JEHycvbxbXX3gTAAw/cw8CBh9YYeuvirbc+AMDr9bJ16xZmz36cyZP/wXPPzcPlcjV4uyIiUnf1meEgOjmWkpxCQP/plMal0NtCrLlriVrzPJbiLEwzCtO04Y9KwJb9A5bCrZQecnGTBd+MjEP58sulIaG3qKiQH3/8gZ49eweXff75p7Rv34GLLrosuOySS65gzZpM3n//7ZDQm5CQSHJySp32X1hYyBdffMbzz78c3F/79ukUFxfz4IPTufrqG7BYLJiNcOP1ym1KS2vHgw8+zjnnnM7ixQs5++zz9nv7IiKyb3Wd4cAwgJQ4THsBjfARIBJCtyFuCaYfx/oPsJTm4EvoCTgACzji8CX2wlKag2P9B2BWncmhMYwefQSrV39PUVFhcNnXX39JRsZAoqOjg8ssFoPt27eyefOmkPVvv31acPhDQxgGGIbBd98tD1k+ZsyRzJ37CoZhMH36NFav/p65c59j0qTLAfjjj/X84x+XcNRRI5k8+Up2786r974TExM54oixLF36aXBZZuYqLrnkfI48ciR///vf+OyzTwDYsOFPRo0awtatW4J1N23ayOjRh7Fjx/YGHLmIiIi0FIXeFmAp2IJt1zr8semBBFiZYeCPSce2ax2Wgi3Vb2A/de/eg5SUtnzzzbLgsqVLP2P06LEh9Y488hicTifnnnsG1147kf/850XWr19Hamrb/RoTGxMTywknnMTs2YFe1yeffIRly77EYrHQuXNXDMPg6qtvoH//QzjrrPO4774ZuN1ubrrpGtLTO/D8868wduxRvPXWogbtv2vX7vz55x8A5ORkc9NN1zB+/Em8+OJrnHvuBUyffheZmavo0qUrPXr04vPP9wTkzz77hP79DyEtrV1NmxcRaVRGYQGWHdvr/WUUFrR000XCioY3tADDUwi+UkxbNFTz7xvT7oKi7YF6TWT06CP46qulHHXUMbjdbr777huuu+4mPvrov8E6iYlJPPfci8ybN4fPP/+U7777ltmzn2Dw4MOYNm06iYlJwbrnn//XkPHAAMceewI33nhrtfu/+ebb6dmzF++99zbz5/+H+fP/Q1JSMrfccgeHHz6K2NhYbDYbLpeL+Pg2fP31l+zevZsbbpiCy+WiS5eurFq1kl27cut97LGxsRQXB64iXrToDYYMGRoc6tGxYyf+97/feP31/5CRMYijjjqWzz//v+BQiE8//YTx40+q9z5FRBpKt7kVaRwKvS3AtMeCNQrDW4xpja5SbnhKwOoM1Gsio0aN4fbbb8br9bJy5XK6d+8REmIrtG2bxo033sr119/C//73K59++gkLFrzGv/51Lw888Eiw3owZj5Oa2jZk3ZhaLkCwWCycccZZnHHGWWRnZ7Fs2VfMn/8Kt99+M6+/vpiUlNSQ+n/+uZ6OHTuFXHx28MF9+frrL+t97EVFhcTEBF7bDRv+4KuvvuCYY/Z8MHi9Xjp16gzA0Ucfy3PPzSY7OwuPx8Pvv69l3Lij671PEZGG0m1uRRqHQm8L8Md1wJvYA1v2D/jiDwotNE0sRVvxpgzAH9ehydpwyCEDAVizZjVLl37OEUeMrVLnpZde4OCD+zJkyFAsFgt9+vSlT5++tG/fnpkzHwup265de9q3T6/Tvr//fgU//JDJBRdcAkBKSip/+cupjBkzjtNOG8+aNZkceWR1wTK0W9xms9dpf3v7/fd1dO8eeN19Ph/HHnsCf//7xXttO/Cj0b59On369OXzzz/F7XaTkTGozhfsiYg0Bt3mVqRxaExvSzAsuLsfjz8qGWveWsAN+DHcBVh3/Q9/VDLu7sc36Xy9NpuNESNG8tVXS/n666UcccS4KnV+/DGThQvnV1keGxtHQkJig/ddUJDPvHlzyMnJDlnuckVjtdpISEgACBku0a3bQWzatJHCwj1DPtau/a3e+87P380XX3zGuHFHAdCpUxc2b95Ex46dgl9ffPF5yDCPo48+lmXLvuSLLz7jqKOOrfc+RUREpOUp9LYQX1JPSg+5GG9yfwyjFItlN0ZpHt6UAU06XVllo0eP4Z133iIxMZn09Kq9yueddyHLln3FAw/cw6+//sLmzZv45JOPmT37Cc4669yQunl5u8jJya7y5fV6q2x35Mgj6NKlK9dccxVffrmUbdu2kpm5imnTbqVz5y4MHHgoAC6Xi82bN7FrVy6HHTaMtLR2PPDA3fz55x+8//47fPLJx/s8xop27NixnRUrlnPttZNo2zaNk046FYAJE84svwHHbDZt2shHH33As8/Ool279sFtHHnkMWRmruLXX39h7Ngj6/MSi0gY0oVhIgcmDW9oQb6knpQMvBLL0u2AB89hk3C36dKkPbyVDR06Aq/Xy+jRY6otHzAgg8cff5p58/7NtddOpKyslE6dOnPRRZfxl7+cGlL3sssuqHYbs2b9m4yMgSHLbDYbjz/+FP/+99M8+uiD5ORkEx/fhiOOGMvNN9+OxRI4/pNOOpX777+bDRv+4PnnX+HBBx/jX/+6l4svPo+DDupRHlh/rvUYTznleADsdjupqW0ZOfIILrjgEpxOJxAYlvGvfz3CU089yauvvkRKSlsmTbqGY489IbiNlJRUevc+GKczijZtEmrdn4iEP10YJnJgMszGuANABMvOrjpBtsfjJidnG8nJ7bHb6z6Oqtp7j7vduJ57OvDwH1fhs1T9O0T3Hm8dPB43u3btICkpDZtN4+taI8OAlJS4an/upXWoyznc1+/iksuurPHCsLD5Xex2E/3YQwAUX3ND6xzTW80xRMzPYCScn4Zwu4l5/CGio51kX/5PzHpkpIaqeM/UhXp6m1G1vQsQvBjB+cpL+P1Vf8rVuyAi0nh0YZjIgUmhtxlVO+1MJVarBZ+v6l3YNO2MiLRG1fao1kFY9aiKSMRQ6G1G1fYuVGKxWfB7m+bWwyIizU1jZ0UknCj0iohIk9BNFUQknCj07gddAyiV6f0g4SCchhRo7KyIhBOF3gawlv87zu0uw+FwtnBrJFy43WUYBlit+rGSlqMhBSIi1dOncwNYLFZcrlgKC3cB4HA4Q+4e1lB+v4HPp97C1sY0TdzuMgoLd5GcnITFYmndU+1Iq6YhBSIi1VPobaD4+CSAYPBtDBaLBb9fF7K1Vi5XLO3btycnp3DflSWiteQQAw0pEBGpnkJvAxmGQZs2ycTFJeLzVb3Vbv23B4mJMezaVaRewlbIarVhtVoapcdfWj8NMRARCT8KvfvJYrFgsex/r4lhQFRUFHa7R6FXpAnUuffVU4iRU0jFny8N6X3VEAMRkfCj0CsiB4S69L4aNitEO3EVlwX/+GxI76uGGIiIhB+FXhFpcc0xBrauva/RybGUVBqXrd5XEZHIoNDbzCyXX4Rr8cIay6NrWdfT9SC8/56L/5CBjd4ukZbUHGNg69L7ajgdkBKHaS/QMCMRkQij0NvMEhYvxFJLuau2sj9/x33OXyk762z8/frj79ELf7fuuke9NKum6JXVGFgREWlqCr3NLO/U03EtXogNsEKtAbg6jp3bcTzxKKbDiTe9A55xR+IZNAgzIREzKRkzJRWzbZqCsAQ1dkhtil5ZjYEVEZGmptDbzPzPzqXo2blYtm/D+PorbL/8iOWrr4hZ8W29tmO4y7D/uR77C3/AfxyYTif+hES83brjGXsU3v79MV0uzJRUiIltkluMSuvQ2CFVvbIiItIaKfS2EH+79jDhDHycAS/Pqzb0mjYbYGDarICB4feB34/h92OWhxTD48X0+6GkFEvZduz5u7GvWR1YL9qFt1MX/Ekp+PscjLdPb8yoKHBFY7pcEB2DGR0NMYHeNAXjprE/Pa3AfvfSNnZIVa+siIi0Rgq94czhAMMCsbH4UlIDy3w+wMSfnII1OwvLtq1gt4PHg+HxYJSVgc+H4fdDQT7WHTsAMD9bguF2Y1ptmPHxmHFx+F0u/Cmp+FPaBjbdfwDe/v0D9ctDMRASjOHADcd1Cq97zfEKYPshE9v3K0Pr1bGnFdjvXlqFVBEREYXe1sFiAZcr8L38uW9ABn6PG9vKFXgP7hsIw15PoNzjAW/gLnGWnBysubn4evTAtmYNRlkJ/qRkLLmB5dasLIzi7wAwv1oKgFFQgOl04OvaHfx+/G3TAsGonLdPX3y99vQcVg7IFBdjlBSHLitfDmboskpM06z2bmYhgbuoEKO4uObyutappm6NdSqxffct9u++BbsDsyIkVgqg3hGHQ1x0yByvAJ5DB1N6wcWhG6tHT6uGEoiIiOw/hd7WylEevFwuzNS21fb2Afjcbvw+HyUXXoLr2afA46H0b2cTNf9V8HgoG38izrcWB+r2ORjbLz9h++VnsFrwHdwPy47t2NesAsOCv306WAwcy7/BsnULQKDXOCEBf1w81s2boLQUnM7Asvg2gUaYJtZ1awP76NEzcPu5iuWbNwWWd+8BHndo400TIz8fs00bPMMPx7JlM9ZNG0Oq+Dt0xLJlM0Cd61S8VpaNG4J1/Z274OvUufZeVHcZmODJGIR36LDyZXsCaMl5FxDdLilkjleooWe8Hj2t6qUVERHZfy0eenfs2MH06dP55ptvcDqdjB8/nuuuuw6n08mmTZu44447WL16Nenp6dx6662MGjUquO7XX3/Nfffdx6ZNm8jIyGD69Ol06tQpWP7CCy8wZ84cCgsLOeGEE7jjjjtwuWqbFCwCORyYViv+tmmYSUkA+Pr2x0xLA8A7agyONZkAlJ1/Ae6iIlzPPQ1AyWVXQqXnpef+HRx2yNtN1CsvBtaZcCZmmziwOQLLPB7K/nJqYFlFr67bQ9QLc8AwKb3g0sA2KpaXb6fk3PMx9g5vbg9Rb/wH7M5Aj6bHXbUX12bH9Z+XAu2ra52K/VTX01vHXtRgEK0UQM20dtA+WXO8ioiIhKEWDb2maTJ58mTi4+N55ZVX2L17N7feeisWi4WbbrqJiRMn0qtXLxYuXMiSJUuYNGkS77//Punp6WzdupWJEyfyz3/+k9GjRzNr1iyuuuoq3n77bQzD4MMPP2TmzJnMmDGD5ORkpkyZwowZM5g6dWpLHnJYqxj7GdKLCMGA7Os/IBAC3W7MrwPjTL1jxu5ZVhGkK5ZVcLsxO3YM3UbF8vJ1/P0PqRow3W7MJcl72lJdj2Zdej3r0zOqXlQREZGIVN9pYhvV+vXrWb16Nffffz89e/ZkyJAhTJ48mXfffZdvvvmGTZs2cffdd3PQQQdxxRVXMHDgQBYuDNzN7I033qB///5cfPHF9OzZk/vvv58tW7awfPlyAF588UUuuOACxo0bxyGHHMJdd93FwoULKSnvxRMRERGRA0eLht7U1FT+/e9/k5KSErK8sLCQzMxM+vbtS3T0nhvzDh48mNWrVwOQmZnJkCFDgmUul4t+/fqxevVqfD4fP/zwQ0j5wIED8Xg8/Prrr017UCIiIiISdlp0eEN8fDyjR++Z/N7v9/Pyyy8zfPhwsrKyaNu2bUj95ORktm/fDlBreX5+PmVlZSHlNpuNhISE4Pp1Vc2EAs3LCHzt3Y6K50ZFWW3tNCrVo/bH+/y+V3tqXbbX/quU1bZOXcobs05D6lZXv9L3eq1Xn/fZ/qzbmNtoyu011Tb3sY96ncP93FeTHE847ruZ29LgcxhOr1FdtLb2VqeaY2jSn8HmFAnnpyEqHWdzHXd93istfiFbZTNmzODnn39mwYIFvPDCCzj2Gk/pcDhwuwNX+JeUlNRYXlpaGnxe0/p1lZzcDPPRxkUFH4ZkxYpPYYsFq92258xaLNijnYHHdlvgcQ2zNwBgtRKdEgfl69T6GGp/Xj5+t07LKtRUVts6dSlvzDoNqbt3/fL3Sp3eM/XdT2Ot25jbaMrtNdU267iPJvm5b47jCcd9t1Bb6n0Ow+k1qovW1t7q1HIMzfLZ25Qi4fw0RKXjTk4Ov+MOm9A7Y8YM5s2bx6OPPkqvXr1wOp3k5eWF1HG73URFBQKi0+msEmDdbjfx8fE4nc7g873L6zt7Q05OM1yJX1BKMoHAW3lXpmmCaWL6/fg9XjAq5uk18RSXAWD3eAOP9xF6i7MLiC5fp7bHQO3PywNunZZVqKmstnXqUt6YdRpSd6/6JTkFJLdPrtt7pr77aax1G3MbTbm9ptrmPvZhOB0kJ8c1zc99cxxPOO67mdtiGDTsHIbTa1QXra291anmGBp8/sJNJJyfhig/7uhoZ+Ac2pv+uCveM3URFqH3nnvu4dVXX2XGjBkcd9xxAKSlpbFu3bqQetnZ2cEhC2lpaWRnZ1cpP/jgg0lISMDpdJKdnc1BBx0EgNfrJS8vj9TU1Hq1rTx3thwz8GWahHQDV7Qp2L7a2miGHkdtj/f5vXJ79rVsr/1XKattnbqUN2adhtStrj51fM/Udz+NtW5jbqMpt9dU26zjPprk5745jicc991Cban3OQyn16guWlt7q1PLMbT4Z+/+ioTz0xBhfg5b9EI2gJkzZ/Laa6/xyCOPcOKJJwaXZ2Rk8NNPPwWHKgCsXLmSjIyMYPnKlXtu7VpSUsLPP/9MRkYGFouFAQMGhJSvXr0am81Gnz59muGoRERERCSctGjo/f3335k9ezaXXXYZgwcPJisrK/g1dOhQ2rdvz5QpU1i7di3PPvssa9as4YwzzgDg9NNP5/vvv+fZZ59l7dq1TJkyhY4dOzJsWOBOWeeccw5z5sxhyZIlrFmzhmnTpvHXv/71wLs5hYiIiIi07PCGTz75BJ/Px1NPPcVTTz0VUvbbb78xe/ZsbrvtNiZMmECXLl2YNWsW6enpAHTs2JEnn3yS++67j1mzZjFo0CBmzZoVuPgLOPHEE9myZQtTp07F7XZz7LHHcuONNzb7MYqIiIhIy2vR0Hv55Zdz+eWX11jepUsXXn755RrLx4wZw5gxYxq8fRERERE5MLT4mF4RERERkaam0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXhhE3rdbjcnnXQS3377bXDZihUrmDBhAgMHDuSUU07h66+/Dlnn3Xff5eijjyYjI4OJEyeSm5sbLDNNk4ceeojhw4czdOhQHnzwQfx+f7Mdj4iIiIiEj7AIvWVlZVx33XWsXbs2uCwnJ4crr7yS8ePH884773DCCSdw1VVXsX37dgDWrFnDbbfdxqRJk5g/fz75+flMmTIluP7cuXN59913mTlzJk888QTvvPMOc+fObfZjExEREZGW1+Khd926dfz1r39l48aNIcu///57rFYrl156KZ06deLKK6/E6XSyevVqAF5++WVOOOEETj31VPr06cODDz7I559/zqZNmwB48cUXmTx5MkOGDGH48OHccMMNvPLKK819eCIiIiISBlo89C5fvpxhw4Yxf/78kOUJCQnk5eXx0UcfYZomS5YsoaioiF69egGQmZnJkCFDgvXbt29Peno6mZmZ7Nixg23btnHYYYcFywcPHsyWLVvYuXNn8xyYiIiIiIQNW0s34Jxzzql2+ZAhQzj33HOZPHkyFosFn8/H/fffT/fu3QHYuXMnbdu2DVknOTmZ7du3k5WVBRBSnpKSAsD27durrFcbw6jX4TQ+I/C1dzsqnhsVZbW106hUj9of7/P7Xu2pddle+69SVts6dSlvzDoNqVtd/Urf67Vefd5n+7NuY26jKbfXVNvcxz7qdQ73c19NcjzhuO9mbkuDz2E4vUZ10draW51qjqFJfwabUyScn4aodJzNddz1ea+0eOitSVFREZs2bWLSpEmMGzeOjz76iHvvvZeMjAwOOuggSktLcTgcIes4HA7cbjelpaXB55XLIHDBXH0kJ8ft55HUQVxU8GFIVqz4FLZYsNpte86sxYI92hl4bLcFHlutNW/faiU6JQ7K16n1MdT+3OEAt7tuyyrUVFbbOnUpb8w6Dam7d/3y90qd3jP13U9jrduY22jK7TXVNuu4jyb5uW+O4wnHfbdQW+p9DsPpNaqL1tbe6tRyDM3y2duUIuH8NESl405ODr/jDtvQ++9//xvTNJk0aRIA/fr1Y82aNbz44ovcddddOJ3OKgHW7XbjcrlCAq7T6Qw+BnC5XPVqR05OAaa5v0ezDwWlJBMIvJV3ZZommCam34/f4wWjfDSKxcRTXAaA3eMNPN5H6C3OLiC6fJ3aHgO1Py8PuHVaVqGmstrWqUt5Y9ZpSN296pfkFJDcPrlu75n67qex1m3MbTTl9ppqm/vYh+F0kJwc1zQ/981xPOG472Zui2HQsHMYTq9RXbS29lanmmNo8PkLN5Fwfhqi/Lijo52Bc2hv+uOueM/URdiG3p9++ok+ffqELDv44IODMzykpaWRnZ0dUp6dnU1qaippaWkAZGVl0bFjx+BjgNTU1Hq1ozx3thwz8GWahHQDV7Qp2L7a2miGHkdtj/f5vXJ79rVsr/1XKattnbqUN2adhtStrj51fM/Udz+NtW5jbqMpt9dU26zjPprk5745jicc991Cban3OQyn16guWlt7q1PLMbT4Z+/+ioTz0xBhfg5b/EK2mrRt25Z169aFLFu/fn0wxGZkZLBy5cpg2bZt29i2bRsZGRmkpaWRnp4eUr5y5UrS09PrNZ5XRERERCJD2Pb0nnnmmZxzzjm88MILHHXUUXzyySd8+eWXvPnmmwCcffbZnH/++QwcOJABAwYwffp0xo4dS6dOnYLlDz30EO3atQPg4Ycf5uKLL26x4xERERGRlhO2oXfgwIE8+eSTPPHEEzz++ON069aNZ599lp49ewIwaNAg7r77bp544gl2797NyJEjueeee4LrX3LJJeTk5DBp0iSsVitnnHEGF154YQsdjYiIiIi0pLAKvb/99lvI86OOOoqjjjqqxvoTJkxgwoQJ1ZZZrVamTJkScpc2ERERETkwhe2YXhERERGRxqLQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiIyP4z/RhGPpCNpWAzmP6WblEIW0s3QERERERaN2vuWhxr38NuXw6YRK8oxJvUC3f34/El9Wzp5gFN0NNbWlrKr7/+2tibFREREZEwZM1dS9Sa57Hl/IhpRgGJmM4EbNk/ELXmeay5a1u6iUA9Qu+oUaP45ZdfQpbNnTuX3NzckGW//fYbp512WuO0TkRERETCl+nHsf4DLKU5+BJ6Ag7AgumIw5fYC0tpDo71H4TFUIc6D2/Izs7G4/EEn/t8Ph588EGGDh1KUlJSkzRORERERMJHqcdHdpGb3GIPOUVu3Dl/csj61eSacezenk33HYW4HDbaAxgG/ph0bLvWYSnYgj++U4u2fb/G9Jqm2VjtEBEREZEW4PH5ySlyk1PsIbfIXf7YTW6Rh5zi8uflQbfI7QtZt7exkX/a8lhvusDvJc3to8TjJ9VnYrOCaXdB0XYMT2ELHd0eupBNREREpDEFZzHwYCnYjD+pKxjNO2GWz2+yq8RTHlYrgqun0uM9z3eXeuu1bafNQnKMg+RoO72d7WlXFE9qlBObI5rOJdHExzjwWQ1MwPCUgNWJaY9tmgOtB4VeERERkUZSeRYDw/BhWfEo3uTGmcXANE12l3orBdmqobaiZzavxIO/Hv+Qt1oMkqPtgTAb4yA52kFyjJ2kaEdwWVJ5eYzDimEY5Y3y41r5M7bsH/DFt8Xxpx27zUpxoMFYirbiTRmAP67Dfh17Y1DoFREREWkEFbMYWIqzMM0oTNOGPyowi4GlcCulh1xcJfiapkmR2xccPpCzj+EF3nokWQNIrAiy5SE2EF4rguye5/FRNiwVQbY+DAvu7sdjKdyKNW8t4AYsGO4CLCU78Ecl4+5+fLP3dFdnv0Ov0ZAXSERERCSS7D2LgZmN1+8nq8xBsb8jUdvXsSH3Vd6P+xvZxd7gGNqcIjdl3vrNbNAmykZS+fCCPb2we0JscrSDpBgHCS47NkvT5zRfUk9KD7kYx9r3sBnrgRKMsjy8KQPCap7eeoXeiRMn4nA4QpZdeeWV2O324HO32904LRMREREJI9Vd8FXRO0v+Rk7IXs5OXyy73JsZuGU3AMutO/BbLMTiINH4mW+8P7LZTK2y7RiHNWQIQSC42st7aPcML0iKduCwtXyv6d58ST0pGXgl1i+2Y7MZFA+5DF9i17Do4a1Q59CruXeb0O6dcJ0TnJbA/yJ2+eGpspZulYiISNMJg4u9ALx+k7zi0PGwlcNsbqWy/Fou+OptbOUIWzE7zDZUDKY1gDiHFWeUgxibky6UcH6HBPwpPaqMk3XZrc10xE3IsGCa8YATf1zHsAq8UI/Qe//99zdlOw5YbWZ3w4YHYiv1oKcCd9ihyA/PtFjTREREmkRNF3t5DjoeUg7d7+37TZPdJZ7ghV4VYbbymNmKx3klHuozAeveF3xVhNauFhd9t6XQLzoOpzOWRHMTVsOg/6EdwGrFKCvAUppCp6G98Men7/cxSv012oVsubm5uklFPbWZ3Q2b6am+0DAgxgL/BOZpPmQREYkMtV3sZS3cCokxYFQNhaZpUlDmDZmxIHuvabgqguyuYje+enx0WgxIjK40tKDyeNmKYQb7uuDL7IJr5TeBWQyiErBWrhNmsxgcqOoVejdt2sQLL7zA4YcfzlFHHQXAkiVLmDZtGjk5OaSkpDBlyhTGjx/fJI2NKDn/CwbemoaYG4YRuJvfeUWwpgB2xddSW0REJMztdbGXhWx8pslun4tiaxfsWf8j68OXeS/mdHKKfSG9srnFbjz1SbJAgsseMu1WUrSdlL16aJNjHLSJsmPd3wu+qp3FwBaWsxgcqOocejdt2sSZZ55JWVkZffv2BeCPP/7gmmuuISkpiVtuuYX169dzww030LZtW4YMGdJkjY4E8QtPrRJfa/pRNlxgH7oF365s/OvaQl58UzdPRETCWZiMh61OxRRce4+JzS12Y+Zt5MgdK8jyx5JXtpW+W3ZjmrDc2Fp+sZedxG3f85H3oGov9gKIc9pqnD+2cg9tosuOzdrMN4SoNIuB01gfGLpRGn6zGByo6hx6n376aZKSkpg3bx6pqYE34ty5c/H5fDz00EMMHToUCMze8Nxzzyn07oPFU1S/FQywJpVhPXQTvt3RWKwd8fs6Nk3jREQkbDXlzQ9qsvdcssGbIZTPYlA54OYWe2qcgqu3sZFDbYVsMuPA78Ms7+2xWw2cDhux9ni6Wks4LTUOd1KXqkE22oEzDGcuqKxiFgPL0u2Ah5IhE8Pqj5IDWZ1D79dff82kSZOCgRdg6dKltG3bNhh4AY499limTJnSuK2MQH57DHjy67+iDaxJxVhYgteThpFqxbAUYNIGDX0QEYlsDbn5QU0qB9mKGyHkFu+ZOza3UsCtLcjWJNpuDU65lVTeG9vd6qBXVhI9XTE4nLGkeDdit1oYcFgnsFqxlBXgwkK3gQfji+vUkJcoPARnMSAsZzE4UNU59GZnZ9O5c+fg802bNrF9+3ZOPvnkkHpxcXEUFdWzF/MAlH/6YpJfOzIkphrUPMQhhAEGPuyOrdAfDD7G52+L39cZ05eCacaiACwiEmGqGQ8LYDri8EW1wbrrfzjWf8BOVxdySryBsFr5NrUVc8sGe2fduOs5RjbGYQ3OFVtxc4Tg94pxs+VBN6q6KbjMg3Ct/C5wsVdsEvbyXlsfgGliFG2FrkN0sZc0iTqH3piYGPLz9/RMLl++HMMwGD58eEi9TZs2kZCQ0GgNjFjJvfASOAH7FU+tYCUXqy0P0/wdvy8Rvz8Vn7crfl+H/d26iMiBIYzHyPpNk/wSL4VZ62m7+WfyLQkUbC+k7e5SPH6TH3/NotgHFreFqLVf8/DS9BrHw1an8k0RkspnMEiqNKwgKXofQbY+9nGxl+lKhj4nBV57TVwkjazOoXfgwIG8//77wVkb3nrrLaxWK2PGjAnWMU2T119/nUMOOaTxWxqBdk/cTJtZHfc/+BoAJgZerLZsLOZubLYN+Hzt8Hp646fzvrYgInLAaokxsm6vn9xiN7tKPMFhBZWHEuwKPveQVz79Vm9jI/+07WS9aQV/MUMLAjcx2pRXgt9iwYqVboabGEqq3N0rJMzutWy/g2w91Xaxl+eg43Gl9obsgmZtkxwY6hx6L7vsMi644AK2b9+O3+9n1apV/O1vfyM5ORmAZcuWMW/ePFavXs3cuXObrMGRJhB8e2I3iwJz8zaYiYkFAz/gwzB8WG1/YrVux+dvj2fLYAxjN+DFUriZwJ/Q6gUWkQNbY42RrTxjQW6Rm9ySwPddxYE7ee0qDg22hWW+erfV4ozFbnXR1Q6GLZrUPAc2i4VR3ZKIinIQSxGxfoNDh4/CntS1Aa9G86npYi/DEh696xKZ6hx6Bw8ezHPPPcfTTz9NTk4Ol156KVdffXWw/IYbbqC4uJhp06ZVGfIgtdvd5nZS7rsaLo4qvxVx/cJoxdWvpmlilI/3NbEBfgyjBKv1D+KWXg9OA9OMw/J9CYZtI35/OwVgETlw7WOMrCX3f3h/e591B11Ebokv0DNbHLjL165qembrOz7WajGCQwoSo+0kR9uDN0ioGFKQ5Ap8T3DZsVvAtXJTYDxsfCfsuS4AUtJiwWLBumsj3rYD8CW2kv/u6WIvaWb1ujnFiBEjGDFiRLVlTz31FF27diU+XnPINki2Hx4shq5WGO+AJCvlCbZOTAJ9vX6zPPzixzD85eHXh+ErAyNQy5K3Fpt9E/A7lm/KwLYrEIALNuOP1a0RRaQJteDYWdM0yS/1sqvEQ16xh9LcP+m9fg27iKdwVx4dc4rx+k2+z9xGkQ9sXj+JxjIeXdaxzmNko+1WEkPGxlZ6HAy3ge/xUbbATYjqQTc/EGm4Oofe+kxDZhgG9913X4MadMD70wezS6CrFXN0FEY7CzihyjQPlZ+WT/tgGOCv6PXFj89vAn5sFhOfJQorxRhGMdbiHZhmdODxrl+x2AuA37F8fTee1Awslo2YZoxCsIg0qsYeO2uaJoVlgR7YvJLy8a8lHsrYzpacwvIhBZ6QMp9/T29sYIxsLutNJ/jLiCkJ3CVzV4mnfIysk3bsop3TgyMmOjS4lt/8IGmvC8Caenysbn4g0nB1Dr1vvvkmhmGQlpaGZR9jbur7l6tU408f7DQx25iYvR3Q34oRH+itrbioNSQHlz+xGoHhDoEeXwsWi4nXD7vK/CT6vRhY8Hn92EwPhuHF4inANB0Yhg+jaCvO/A1YnTvx++OxrHDjbdM9NASH+TgxEalGGMxMUJexs97EHhS5fcHxr5XDasUQgl3FnkBPbYmHXcUevP76X+If4wj0xnZ0pJLgiWWAw4rFHkPHwihsFgtJfdriirITbRbj8tkYMWwk/vjwmTNWNz8QaZg6h94TTjiBzz77DLfbzfHHH8+JJ57I4MGDm7JtApAP/ObEV9gBUouxts+H+BKINTENH4bfj1l+v3DDCERhwwCrYcHAhmG48ZoWLJgYhonbY6HU4sHh9WK3+PFZ7Vh9BjaLFWvBVkxnGwKR2ocJOH9/OzQEJ/fC3XFc1Q9QEQlLLTEzQQWf3yS/NDD7QELmYly528iydSKlYAten5tfNrop8SbQtmwdP/z8HI+7T8Ttq3+nSYzDSoLLTlK0nYRoO+mJMbgskOCykxgd+EpyOUiItpPosuOouKOX6ce18n/lY2TTsW91AhCXEFU+RnYH3pQB4TlnrMbDitRbnUPvo48+SklJCZ9++invv/8+F110ESkpKYwfP54TTzyRgw8+uCnbKRiQE4cvJxaiffgO7YTVvhlb0SbMNgaGpQzTDNwtxzAMTNOGxQiMebBZ7CQ5ALcFq8Uo//Lj9RuU+sBT4iHa4cbuNsjDQaLHgd1WhC3nJwzTT0UI9jsTsG9dFrjww5EH2PZ8gFYXhPVLWGSPFuhtbcy7d/lNk8IyL3klXvLKe1rzSjzsrvS4ctnuEg/5pV5MoKORxbW2lewyYyn272Lo7lIAfs8uwm+xkEcb2hkbaevPZjOpRNutJESXh9iKMOtylI+VrbzMXuW2tIYBKSlxZGcXBC/yrdE+5ozVGFmRyFKvC9lcLhfjx49n/PjxFBYW8vHHH/P+++/zwgsv0LFjR0466SROPPFEunXr1lTtFQwodeD3dsPv7Qo/fYk3oy8WWx4W+yas1m0EfnH7MHGAaQF8YHFhYMVpN7G7osHvwW/6KbW5sBhl2K0+SvwOdpb48Bd7SYwuJm9XHiWOtrRz23FYS6A0F0vpLoyinYCB35+MP6qWIFzRkxQG/1oVaUkt0tta28wEzjaYub9R+su7/K/7xeSV+vYKr1UD7O4SD/WcnCAozekh0eLHY48nxWolOdeBzWowtHMCUU4HLptJUukmBgzqgSu9f7POG6sxsiIHjnqF3spiY2M57bTTOO2008jLy+Pjjz/mv//9L08//TS9evVi0aJFjdlOqVYgAJv+ZHy+tvh8PfFZt2C1/4bVuj0we4NpAyyYtigMijFNE9MWg9XYhYGVNoltsXi2YxgmXmccMUY0zuIsrBaTPF8UhSUeokr8uOxlbP7fDyTYPJi2BNr6duD1lmF43NirCcIVPUnuLkdh2/lj9R/2MZ1qD8MKy9LU9nqPmY6ujb6LxuxtrU6Z1x/sVd1dGvieX+rBzNvEYRtXs4tYCnbkcFBWEV6/yXffb6HYB9GmSaLxLY9+27ned+9q4wr0sia4bOXfA1/VLY+PsuMo3EzMt8vwu+IxrdHY8wJTbbVNjwerFaOsAIs1BktiEv5mvlECaIysyIGiwaG3srKyMkpKSigtLcXn87Fly5Z6b8PtdjNhwgTuuOMOhg0bBsDWrVu58847Wb58OW3btuXaa69l/PjxwXXeffddHnvsMbKyshg1ahT33HMPSUlJQOCq3ocffpgFCxbg9/s544wzuOGGG/Z5EV7rZuD3dcTv6xAIv7Y/sViKwPBiOmMxC4oBL/hK8fsdGIaFwBjgMsCKKzaBLs5obKVgYqNtQiIxHpOoksCQiRijmBxfDF6vSZzHw+Zd+RT99D3tbQX4rPG08+6k1F1CqSeK+DY9cWatInrFY/ij21X5sLfm/ILpTK6x56vOPWMKxpGric9tde8xX0ovGHw6GI00a8k+5oG17vofjvUfUJzQnTIf7C4PrPml3kDvaqk3GGIrHlcE3IplZV5/tbvubWykvS2P9aYL/MW0LfMCUOz24bdYKMFJupFH5xgfMdGxIUG1TaUgG7I8qtJ42Hrwx3XAm9ijfOzsQXu9RiaWoq0tP3ZWY2RFIl6DQ++OHTv44IMP+OCDD8jMzCQ6Opqjjz6aK664gpEjR9ZrW2VlZVx//fWsXbs2uMzr9XLFFVfQsWNH3nzzTZYvX85NN91Ejx496NWrF2vWrOG2227jrrvuok+fPkyfPp0pU6bwzDPPADB37lzeffddZs6cidfr5cYbbyQ5OZlLLrmkoYfciuwJv4ZRCFYfJeMuwfXyPAwjF3+39jhWf4LFsgujNAe/34VhRGH4SjBK3JimHcOw0MZp0Cbaia3IxO+PwdUmmjhLPJ6yUqxlVhw2E4dRyC6/E6/PINHrZcvuIj7/YRt+i8E42w7SLbn8L64rXQshyubH43cR50jGueVzTEtUtT1f7i5H4djwyT57xpokGCtEN0wjv25NPRygpt5Xa9YP8E0W1t7n402s335M06TE46egzEtBmZfCUi++XRvo8eePFBhtKC7Ip31eKT7T5OffsijxG1g9VqJ/X8ZDX3bgD19Kw4/HgPiowLyv8VF22rhsdLP6Sd/dhlSnE4s9ms7F0VgtBukD2uF02nH5i3G4bQwbNqLpZybQ2FkRCQP1Cr2Vg+7q1atxuVyMGzeOSy+9lNGjR+NwOOrdgHXr1nH99ddj7nXFweeff862bdt49dVXiY2NpXv37ixdupRVq1bRq1cvXn75ZU444QROPfVUAB588EHGjRvHpk2b6NSpEy+++CKTJ09myJAhQOCOcY8//vgBEnorBO7AhmnFH9cZ02yDabahePAkjO/cGEYO/q7tcaz+EigJ1DdsmP5kDKMQS9kuMKyYph3TbIPNZiXRbmIx3OCIo0dKPP7EMgqtbfCUlWJ323DaAj1BDn8RDn8RBX4rf2bvxppXAsDy1VsYbF1LJ6sb0wBzt4nFEsWu3RZiojqTvHs9UWvmYDjiau0Zc5t+on54ofGCMQ0MWnuFPfyJdTs1+xMSGyNgNmJIbeyA2tTDAWrsfbXHUWaPx8z/nYIf3+F/3S6iwF0eYkt9FJaH2YIyb+Bx6Z7nBaWBZXuPeQ3MA5vNetMO/hKGFpYBsGFXSfk8sBa6GWU4/MWBY7cYtImy0SYYYG3Eu+y0KX9csbxNlJ14155lMQ5r1akiTT+ulT+U9662xf6nHQBXjKN8ZoJtzdq7qrGzItLS6hx6zz77bDIzM3E6nYwZM4bHH3+cMWPG4HQ696sBy5cvZ9iwYVx77bUMHDgwZPmIESOIjY0NLps9e3bwcWZmJpdddlnwefv27UlPTyczMxOHw8G2bds47LDDguWDBw9my5Yt7Ny5k7Zt2+5Xm1u98n/jmWZ8MACDB/fIE7FlZeLMeofAW6MiBCdimjHgMLAUbsWMSgEzKjAPsNVOnN2PxSwDeyzdkxPpNLgj7uJsHJsslJoxdHcm0iZrO2VeP4mWEhKNAnL80cRSSnEplHhg+e85+C0W2lHKCMuvrLT0p9S+g0HZxTisBts27yYqykG8kUT8lp+xFORASe3/Mq5rMIaGBa3qwh47+mNtd2StvYT7ExIbI2A2Zkht9IBax+EAJYkHBUO61+enyO2j2OOjqMxHkdsbeO7e87jyc1fRFk7IXUGuGUfhhh303V6Az2/yzfKNeA0LsXhJNJbz6Hdd6jXWNfiaWAzinDbinFba2ZOJc8fQ226ALYa03U6shsHIbok4HA5izGJi/PDY4OG4UroSba8mvDZUGPauauysiLSkOofeVatWYbVa6dGjB7m5ubz88su8/PLL1dY1DIN58+bVabvnnHNOtcs3bdpEhw4deOihh3jrrbdITExk8uTJHH300QDVhtfk5GS2b99OVlYWQEh5SkrgX4fbt2+vV+ht8ftsGIGvvdtR8dyoKKutnUaleuz12GIBAuPYvJ1G4e14ONYvdgAeykadiH1nJs7sdzCMfHxRXTFNP6YjDoPdGIYV02EPCcIV245yOHFYvNhj25KR3gF71kYAeg1IwbLVSSlR+D0GrhgrxW4b6fFOCrxguq3Y8VLihRy3m92lgTskfb95d3nPmI++xib8bGaDmYbXsplh2wuwWQx+/zULh9NOvBFDQu5qvJu2kOTdRllsd9p6dmC1gNsSgyOhJ7a8tTj/+ICSpMD4QucfgaDlT+iJ1QgELRxx+KPaYNn1vz11yz+cK8KeUZwFBMKeGZUA29YQtfPPGsNeTevZsn/AWriV0oyaQ+L+rNuY2wgy/fV63bx+k1KPj1KPjxKPn1Jv+fdKz60Fmxj4+xoKrfEU5+fTMa8Uv2ny29psykwDm8eOa8NynlnVjfXeZIrcvhrHtNakt7GFw22FbDDjwO8Oru8vv+NLqRFFrHU3B7n8xLtiiY2ylYfYwNee59a9nge+ouyWPcHV9ONauQ5r1g/44ztiz4oCICUtDiwWLLs24Ws3AGt69yYJfv7knpRlBHpXbWvXAz4sZXn4UgO9q/6knnW903njqfQ7x4zvWP47qPFU/j1XvxX3Wrelf/fvS2trb3WqOYYGn79wEwnnpyEqHWdzHXd93it1Dr2Ve033Hoqwt32V10VxcTFvvvkm48eP5+mnn+bbb79l8uTJzJ8/nwEDBlBaWlplOIXD4cDtdlNaWhp8XrkMAhfM1Udyctx+HkkdxEUFH4beZa38p8ZiwWq37TmzFgv26PIedrst8NhayxXPVivRKXFQvk6tjwFcgd4t18ATwDwOVuwCPNhOvAZMD/z8Nmx+EyiBuIPAZoAjDnJ2Axbs1jIozoH49lhd8ThcTrAH3mr2mFhwOonyFENqB5KK3YBBryGdwWrF3G3Dv349J7RrS56RRHTJn3h8fvq1j6fQa0JpPg63Fa/PR7EZBV6TMq+fMmBjXgl+S1l5MN6Bn52sMNMo9mcxdEchAMtXbMZvsdDGYpL6+zJeyuxOtMPGhWUrKLW3gdw8euaVYrUYbN1WgNVuw2UmEr3xJ36N/gV/m85E2Qx6/O9d2hRm443vQayZjcUw8Djb4ItOwpb7G1GbP8HXuT82qxVb+dzIhmnCz/8H/nxI7w9/5AVek4QUSGoLWb/i2vEp9BhYHgwq8fvrta5pmvj8Jr7y716/id/nw77xYwz3LjxJvYEcTBOKbHH44uKx7/of+T+/xx+9L6PMB2VeH26vH7fPj9vrp8wb+B547CO6aAtjtq4m34inJGcX3XNK8JsmP/6yE7dpYPdaiV3/DU9804nfPcl46jDf1d63hXWUDwf4Pae4/I8e6GYUU+DNI7f8oqMKTpuF2PIQGuMIfI912ohxBr7HOq3EOu2k+dvQb8NX9I1qg8URQ5J/EzaLQb+RXXE6Hdg9BRglURw37ihI7LLPNu/T4NPhmywo/BPsPsAe+Bkp3A6J7WDw6cSkttn//dQk5VDo0he+ywU82I65BlK6E9NSF/W63aG/cxowLK4u6v27u5na1WhaW3urU8sxNMtnb1OKhPPTEJWOOzk5/I67zqH3pZdeasp2VGG1WklISGDatGlYLBb69evHihUreP311xkwYABOp7NKgHW73bhcrpCAWzH8oqKuy+WqVztycuowwfn+KiglmUDgrbwr0zTBNDH9fvwe756eIIuJpzgQBuweb+DxPkJvcXYB0eXr1PYYqOZ5FBBFsTcx8AbudTExH2wEPJQcNhEDD451/8W59m3AhzdvJ77kg/EeNAHHn59gbP0Rq6cYsOEtcWPzeDB8XjzRnbF5fgMIHIPFgiU/F6NNF6J9BUTFt8UeHXiLJndJKO8Zy8EfdxiW4p0MsidQ7HcSVfYnXr+Jo3sSpX4wywqIL47C7/exy4jH6TGxWQx8lW5XWuh3kGrmkpe3izygwFbIejMO/AW0KQgc/3cbdgV7l7sZWTz5xyp+M3PKJ9r/tnyi/c0M3bwbgOXmn/gt5f8e//UzHv04NeTf450t2Vxr+4zdRhxl5noGb8wFYJW5Hr/FQozpJeGXz3jm03ZsM0L/G5HOTq7wfU4ecZTwB4M25IIJK/2/47NYiDY9tPnpUx77IIVN/pRq51MNtHtpebs3MHRzXqDdbCxvt0mi8Q2PftupTv/W721spGdwdgAPicWBn7Ftu0uD41WjjVJMbyEeMym4nsUAl91KlM1ClN0aeGy34LJb6YibpMI4ou1W/NZo0nY7sRgGw7skYHPYiTGLifalMrXfoVgTuxDjsBLtsBLjsGKz1nVMdFtc/r6B3teoeOzW8j8mfX68ZW4chVspSuhDiTcByn8G9ouRjrX3+YGxrN7f2PMz0icwpMRIb5z91MbtDv05zi1q2v3tsy2Vfsc08oeiYQQ+bOv9u7uJ29XoWlt7q1PNMTT4/IWbSDg/DVF+3NHRzsA5tDf9cVe8Z+qiUaYsawpt27bFMIyQKca6devGb78FQlJaWhrZ2dkh62RnZ5OamkpaWhoAWVlZdOzYMfgYIDW1fmP0ynNnyzEDX2b5v16Di809383yOrVto/Jx1PZ4n99NAAt+f6CXzRfbERwOvBldMD4vH6c3eM84PV98Fxxr38NK4MIVo2w37k6jsRbvxCjNwTQD4wypGGfoSsHd52+B2Rvy1lYtj0qmtM+ZOP74iKjsH7DHH4TdGXgbx7SNLb9AZzu+TgOxFu+kZ8W8oCV/ANB3aCfcWPCV7IZSuKPXoRR7fPT49Qs6W1yU+J20y3fiN6FvWhweDGzeQlzeWHrFtCOGNrQryyG52E+BGUOMPzCG0zRNrBYwDYMS00k7dhFDSchpcJnF2E03+X4H+P14y0N4iceH32LixkaCUYa7LJ9sMyFk3WQjH9NWRo6ZBH5fsNe01OvHbwEvdpINN05/cY03EIihBCceSojCagmETzBw2S1gtWEjhmQjnwFtDJIc8YELEq0GDqsFu9VS5Xmqz06XHYm0t0fht7joUOTCYhjE90rBarfh9BXh8po8mDEUS0JnXOUB1241ah63avpxrfxlz21ht5cPB2gfX35ut+BtP4CoHn2qDAeo+8+phbJuxxNVsLXa9xiJ7XB3Ox4TS+0/V/XgTeyJN+PKan9GGmsftdr757yFf6c1R1vq/bs7nF6jumht7a1OLcfQ4p+9+ysSzk9DhPk5DNvQm5GRwVNPPYXP58Na3ov5+++/06FDh2D5ypUrmTBhAgDbtm1j27ZtZGRkkJaWRnp6OitXrgyG3pUrV5Kenq6L2JpKDXNc1nThinXX77Vexe1v06XWcrdhqfUCnbLyYLz3vKAWi0GUxcDq3Ym33QB69+wDgKukP+nlde0bA0GrbbfE8qCVjTdlCL0HHwmGBUt+EjHffsZQV2IgULs3ANB/aGei46Ip3pUNJVYOHTKKspiOweEFlvwOpK76hsMcifiMKGI8GwGTDgPagdWKxVOAvQym9x9CWUzHkN+RzqK2dP5xGYMcbfBZXbTxBPbZ8ZD2GFYrVk8hDo9Bj4HD8cd3Ctxq2jD23HbaMLAXbibuu+8wK9rt3wxAv8Ed99wgoBQGDDusblNYmV1xrVxR/hqnYt8Q+Is+Lim6/HXbiLf9ANI79aj7eNVmuviqppkEfKkDYPDpgd7Xxv5lrXlgRURaVNiG3pNOOolZs2Zx1113cckll/Dll1/yxRdf8PrrrwOB2STOP/98Bg4cyIABA5g+fTpjx46lU6dOwfKHHnqIdu3aAfDwww9z8cUXt9jxHNCq+bDf11XcdSnf1/RH+wrGlcNTfYJWXSfatyd2xl452ER3w962N66KcG0PlNmCU0hl420/gC5de1UNRGYvXDl9AvuMPgh7+V2rbNH2wLqlO/G2HUBy+1ouiGrTEV9j3iCgiQJqc01tVd17zEzuGhhf29TDDUREpNmFbeiNjY1l7ty5TJs2jZNOOon09HQeffRR+vXrB8CgQYO4++67eeKJJ9i9ezcjR47knnvuCa5/ySWXkJOTw6RJk7BarZxxxhlceOGFLXQ0Uq199Xzto7wxgnHlbdU5aNUS9sjagOmqIeztT0hsjIDZBCG1qQJqs01ttdd7zFDvq4hIxAqr0FsxXrdCjx49apwWDWDChAnB4Q17s1qtTJkyhSlTpjRqGyXM7Gcw3p+6e4c9f1kedM6gNG0cvhrm6d2fkNgYAbMpQmqTBVQNBxARkUYUVqFXpEnUJzzVo+7eYa90yESieg7Al1tU63jQ/QmJjREwmySkKqCKiEiYU+gV2R97h726zn26PyGxMQKmQqqIiBxg9EknIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEvLAJvW63m5NOOolvv/22SllBQQGjR49m0aJFIcvfffddjj76aDIyMpg4cSK5ubnBMtM0eeihhxg+fDhDhw7lwQcfxO/3N/lxiIiIiEj4CYvQW1ZWxnXXXcfatWurLZ8xYwY7d+4MWbZmzRpuu+02Jk2axPz588nPz2fKlCnB8rlz5/Luu+8yc+ZMnnjiCd555x3mzp3bpMchIiIiIuGpxUPvunXr+Otf/8rGjRurLV+xYgXffPMNqampIctffvllTjjhBE499VT69OnDgw8+yOeff86mTZsAePHFF5k8eTJDhgxh+PDh3HDDDbzyyitNfjwiIiIiEn5aPPQuX76cYcOGMX/+/CplbrebO+64g6lTp+JwOELKMjMzGTJkSPB5+/btSU9PJzMzkx07drBt2zYOO+ywYPngwYPZsmVLlR5jEREREYl8tpZuwDnnnFNj2dNPP03fvn0ZNWpUlbKdO3fStm3bkGXJycls376drKwsgJDylJQUALZv315lvdoYRp2rNg0j8LV3OyqeGxVltbXTqFSP2h/v8/te7al12V77r1JW2zp1KW/MOg2pW139St/rtV593mf7s25jbqMpt9dU29zHPup1DvdzX01yPOG472ZuS4PPYTi9RnXR2tpbnWqOoUl/BptTJJyfhqh0nM113PV5r7R46K3JunXreO2113j77berLS8tLa3S++twOHC73ZSWlgafVy6DQO9xfSQnx9WrfoPERQUfhmTFik9hiwWr3bbnzFos2KOdgcd2W+Cx1Vrz9q1WolPioHydWh9D7c8dDnC767asQk1lta1Tl/LGrNOQunvXL3+v1Ok9U9/9NNa6jbmNptxeU22zjvtokp/75jiecNx3C7Wl3ucwnF6jumht7a1OLcfQLJ+9TSkSzk9DVDru5OTwO+6wDL2maXL77bczefLkYA/t3pxOZ5UA63a7cblcIQHX6XQGHwO4XK56tSUnpwDTrO8R1FNBKckEAm/lXZmmCaaJ6ffj93jBKB+NYjHxFJcBYPd4A4/3EXqLswuILl+ntsdA7c/LA26dllWoqay2depS3ph1GlJ3r/olOQUkt0+u23umvvtprHUbcxtNub2m2uY+9mE4HSQnxzXNz31zHE847ruZ22IYNOwchtNrVBetrb3VqeYYGnz+wk0knJ+GKD/u6Ghn4Bzam/64K94zdRGWoXfr1q2sWrWK3377jX/9618AlJSUcOedd/L+++/z73//m7S0NLKzs0PWy87OJjU1lbS0NACysrLo2LFj8DFQ5YK4fSnPnS3HDHyZJiHdwBVtCravtjaaocdR2+N9fq/cnn0t22v/VcpqW6cu5Y1ZpyF1q6tPHd8z9d1PY63bmNtoyu011TbruI8m+blvjuMJx323UFvqfQ7D6TWqi9bW3urUcgwt/tm7vyLh/DREmJ/DsAy9aWlpfPTRRyHLzj//fM4//3xOPvlkADIyMli5ciUTJkwAYNu2bWzbto2MjAzS0tJIT09n5cqVwdC7cuVK0tPT6zWeV0REREQiQ1iGXpvNRpcuXaosS05ODvbinn322Zx//vkMHDiQAQMGMH36dMaOHUunTp2C5Q899BDt2rUD4OGHH+biiy9u3gMRERERkbAQlqG3LgYNGsTdd9/NE088we7duxk5ciT33HNPsPySSy4hJyeHSZMmYbVaOeOMM7jwwgtbrsEiIiIi0mLCKvT+9ttvNZb93//9X5VlEyZMCA5v2JvVamXKlCkhd2kTERERkQNTi9+cQkRERESkqSn0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0CsiIiIiEU+hV0REREQinkKviIiIiEQ8hV4RERERiXgKvSIiIiIS8Wwt3YADjeXyi3AtXli3ysXFge9FhbBtK0RHQ/9Dmq5xIiIiIhFKobeZJSxeWOfudaPSdwvgLy7Gt3s3JCY2TeNEREREIpSGNzSzvFNPxwP4G7CuBbD+uR5KShq5VSIiIiKRTaG3mfmfnUvBmt8omng1vpRUsFpDvyzWWte3lJVh3fAHFBY2U4tFREREWj8Nb2gB/nbtcV9/E2aHDkS/NA/rti3g82EYBqa/vA/YNDGKiqpd3ygqwrZ8Gb7+GeDzNmPLRURERFonhd4WYsbG4TnrXPKHjcD14lwc/7cES1HhntALmG0SMEpLMDweMPesa/h82NetxZqdhRkTC243uFwtcBQiIiIirYNCbwsyY+MwB2RQcv3NeHsfTPz/fYh/02bw7zXi1zSx5ORglBQHwq8BpsuF4fZglOVi2fAn/u4HgcPRIschIiIiEu4UesOAv117PGefA2ecQsm7H2D/bnlgujJvpaELfj+2tb9iyc4BA/xJyQBYduUGLm5zuzE7dcaMiVH4FREREdmLQm+YMGPjICUdz2kxmAmJOBcvxLp1M/h8eypFx+DvFI1lxw4s+fmAiVFaijU7C2tuLv4Nf+Lt3Qd/124KviIiIiKVKPSGGTM2Ds/YI/H27Yf96y+xr14FHk/gq0LvgwM9vz//iMVvBnp9DQMsVqybNwW2k94B0+FQ+BURERFBoTcsmbFxmLFxuNum4TnmeGzffYv9qy9Cgy/gGXIY9jWZYO65ys3w+wLBd9tWfO3TA72+ushNREREDnAKvWGsIvx6xh4JgP27bwMzNVSEX4sVz8BDwe8LCb+G3wcV4RcwO3XG0Ly+IiIicgBT6G0FgkMeDhsW6PWtR/i1bt4E27dh/+JzjMJCTIe9BY9EREREpGUo9LYSlXt96xN+Db8PPH7s33yN7ecf8aV3wLJzB7jLwOFs2YMSERERaSYKva1Mg8IvgNeD4XFj3bQR13NPYezcidmho4Y9iIiIyAFBobeVqjH8wp4AXBF+AfyBqc8Mvw+Ki7Fu2ghbtwSHPQAYhQWBqdNEREREIoxCbyu3d/gFQgNwJZ6RowMPfD4MqxUA+w+Z2DJXBdZbk4nn8FHN03ARERFptYzCAoyiotCFbjcUFoLPjbFjO4a96rSpZkxMi3WwKfRGiIrwC4QEYIoKMYqLQyu7PVBSCkDpSScTVf6m9bVrj2XHdl3wJiIiIrWyZa4OTKe6t9hYiHbievXlyjOqBnlGjt7TCdfMFHojUOUAbP/qiz1vSncZhrt83G9MDABRb72J7ecfAXC99AIAtu9XBC5427E9uJ4uehMREZEK3oyB+Hr0rLYsOjmWkpzqrxkyy/NHS1DojXCV35RVhj2Uj/319usfeF5WCn7fngvenp4FgLF9G2aHjoEQ7HBg2tULLCIiciCr3MFWmWEAKXGY9oJqe3pbkkJvhKtx2AN1GPsL4PNh3bEdsrOIenkeWK14hg5v8naLiIi0BjWNba24SLyiw2hvLTm29UAVNqHX7XYzYcIE7rjjDoYNCwSz1atX88ADD/Dbb7/Rtm1bLr30Us4888zgOl9//TX33XcfmzZtIiMjg+nTp9OpU6dg+QsvvMCcOXMoLCzkhBNO4I477sB1AN+Sd++/yvYOwTWN/40qH/9bdtx4cNgxbXYclX6YzcRETA1/EBGRA1BNY1vN2FgAol59udr1WnJs64EqLEJvWVkZ119/PWvXrg0uy8rK4rLLLuPss8/mgQce4KeffmLKlCmkpqYyduxYtm7dysSJE/nnP//J6NGjmTVrFldddRVvv/02hmHw4YcfMnPmTGbMmEFycjJTpkxhxowZTJ06tQWPNLzsHYJDxv9WrpeWBoBzyYeBBT5fcMaHqJfn4TliLJ7K4VlEROQAUdvY1tq05NjWA1WLh95169Zx/fXXY+418GPJkiWkpKRw3XXXAdC1a1e+/fZb3nnnHcaOHcsbb7xB//79ufjiiwG4//77GTlyJMuXL2fYsGG8+OKLXHDBBYwbNw6Au+66i0suuYQbb7zxgO7trU21P7g19P4a2dlAoPfXTGgTnPUBKvX+6t82IiIS4Woa2yrhp8VDb0VIvfbaaxk4cGBw+ejRozn44IOr1C8sD1aZmZkMGTIkuNzlctGvXz9Wr17NkCFD+OGHH5g0aVKwfODAgXg8Hn799VcGDRrUdAfUilX3g1tt76+7bM/sD4sXBpb5fNi+XwGAy+fDM/xwvEPLe3/LxzZpGjQRkQNTvca9egoxcgox0LhXaVwtHnrPOeecapd37NiRjh07Bp/n5OTw3nvv8c9//hMIDH9o27ZtyDrJycls376d/Px8ysrKQsptNhsJCQls3769Xu0zjHpVb7CK/TTX/urKN7Bq72+1F8DZrHtCLmBfswr7rz+Vb8SHfc0qfJ06B46v4hiNvY5772PfV3lj1mlI3erqU8dzWN/9NNa6jbmNptxeU21zH/to0p/D5jiecNx3M7elwecwnF6jumhl7bWvqWFO17jAuFfXa5XGvUY7cRWXARr32ho1d56pz35aPPTWRWlpKf/85z9JSUnhb3/7GwAlJSU49roa0uFw4Ha7KS0tDT6vrrw+kpOb9y/M5t7fPqVU054u7eD4o6qvX1gIe/817/FA4W7sDgdRnkIwKv6ad4MvcD6incDex+52lxdAdEpctVe/NlqdhtTdu355++t0Duu7n8ZatzG30ZTba6pt1nEfTfJz2BzHE477bqG21PschtNrVBetrb1Hjoahdf8va3TFg9hYiAuzz0Wpk7DLM7SC0FtUVMRVV13Fn3/+yX/+85/geFyn01klwLrdbuLj43E6ncHne5fXdzxvTk7zzDNnGIE3SHPtb7/ZY6tf/POq6v+ab5MU+D7r6T3LfD7s3wR6jIs//RLP2CND13G7iS7/a784u6DGQNsodRpSd6/6JTkFJLdPrts5rO9+GmvdxtxGU26vqba5j30YTkfT/Rw2x/GE476buS0N/l0aTq9RXbS29kKNnxuVVTl/ZUBZQZM3TRpPc+eZiv3VRViH3sLCQi699FI2btzIvHnz6Nq1a7AsLS2N7PKLqSpkZ2dz8MEHk5CQgNPpJDs7m4MOOggAr9dLXl4eqamp9WqDadKsIbS599fYPIcMxHtQNVex1nQ75KzAOfSmtcfYe+iJ241ZFrgbnGkC1b0ulV6v/arTkLrV1aeO57C++2msdRtzG025vabaZh330SQ/h81xPOG47xZqS73PYTi9RnupaTwsBYHxsMb2yJsHtrV/Fkp4nsOwDb1+v59JkyaxefNmXnrppWB4rZCRkcHKlSuDz0tKSvj555+ZNGkSFouFAQMGsHLlypA5f202G3369GnW4zjQ1HQV676mQ4ta8NqeWyRX8Pmq3A2uyvq6O5yIRDjNAyvSOMI29C5YsIBvv/2Wp556ivj4eLKysgCw2+0kJCRw+umnM2fOHJ599lnGjRvHrFmz6NixYzDknnPOOUydOpVevXrRtm1bpk2bxl//+ldNV9ZC9jWPYU0Xx5ndumM6HDX/Utfd4UQkwmkeWJHGEbah98MPP8Tv93PFFVeELB86dCgvvfQSHTt25Mknn+S+++5j1qxZDBo0iFmzZmGUX8Z34oknsmXLFqZOnYrb7ebYY4/lxhtvbIlDEfY9j2GVu8NVp5ohEnvfHU69wSKyL63ttrGaB1akcYRV6P3tt9+Cj+fMmbPP+mPGjGHMmDE1ll9++eVcfvnljdI2aVp1+aVe4xCJff2LT73BIlKJhguIHJjCKvSK1KZO/+JraG9wK77gQ0TqR8MFRA5MCr3SajR6b7C7LHjxnOewYYGba4TxvzhFWptwHUag4QIiByaFXoko9enBqXzxnD1zFbZfAneQ0784RRqHhhGISDhR6JWIUp8enDpdPFdZ+dAJMzo60EMFIb1Wxo7tYLiD94wPtkk9w9KMwql3VcMIRCScKPTKAau+/+Lc19AJ16svB+8ZX3lCbs+hg/ENyAhdKQz+xStNpyWDZzj1rmoYgYiEE4VekTqqS69VdHIsJTmFIctsP2QSNe/5KnXrEkK8GQP3OzyFU89fS6rz6+ApDOmtb8jr0JLBU72rIiLVU+gVqaN99VoZBpASh2kPvd+4JyYG7949vXXZX0xMo4Snxg5gTRGimyOY1+V1MAyq9NY3JIi2ZPBU76qISPUM0wy3OyOHl+zsgma5d7RhQEpKXLPtTxpfU5zDasNgHeyzp7ee26ispmEe+1JbeGyKbe6trq9DcnIsOZV66yOtxzvS6Xdp66bz1/o19zms2F9dqKdXJIw1Rq9dY/f8NUUvZnP0jNbldaipt15ERFo/hV4RqZem+Pe5/iUvIiJNzdLSDRARERERaWoKvSIiIiIS8RR6RURERCTiKfSKiIiISMRT6BURERGRiKfQKyIiIiIRT6FXRERERCKeQq+IiIiIRDyFXhERERGJeAq9IiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4tpZuQLgzjObdT3PtTxqfzmHrp3PY+ukctm46f61fc5/D+uzHME3TbLqmiIiIiIi0PA1vEBEREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIiIiEvEUekVEREQk4in0ioiIiEjEU+gVERERkYin0BsGysrKuPXWWxkyZAijRo3i+eefb+kmyV527NjB5MmTGTp0KKNHj+b++++nrKwMgE2bNnHhhRcycOBAxo8fz5dffhmy7tdff81JJ51ERkYGf//739m0aVNLHIKUu/zyy7nllluCz3/++WfOPPNMMjIyOP300/nxxx9D6r/77rscffTRZGRkMHHiRHJzc5u7yVLO7XZz1113cdhhh3H44YfzyCOPUHFTUZ3H8Ldt2zauuOIKDj30UI488kheeOGFYJnOX3hzu92cdNJJfPvtt8Fl+/vZ98ILLzB69GgGDRrErbfeSklJSZMfh0JvGHjwwQf58ccfmTdvHnfeeSczZ87kgw8+aOlmSTnTNJk8eTIlJSW88sorPProo3z66ac89thjmKbJxIkTSUlJYeHChZxyyilMmjSJrVu3ArB161YmTpzIhAkTWLBgAUlJSVx11VXo7t8t47333uPzzz8PPi8uLubyyy9nyJAhLFq0iEGDBnHFFVdQXFwMwJo1a7jtttuYNGkS8+fPJz8/nylTprRU8w949957L19//TVz5szh4Ycf5vXXX2f+/Pk6j63ENddcQ3R0NIsWLeLWW2/lscce4+OPP9b5C3NlZWVcd911rF27Nrhsfz/7PvzwQ2bOnMndd9/NvHnzyMzMZMaMGU1/MKa0qKKiInPAgAHmN998E1w2a9Ys87zzzmvBVkll69atM3v16mVmZWUFl73zzjvmqFGjzK+//tocOHCgWVRUFCy74IILzCeeeMI0TdN87LHHQs5lcXGxOWjQoJDzLc1j165d5hFHHGGefvrp5s0332yapmm+8cYb5pFHHmn6/X7TNE3T7/ebxxxzjLlw4ULTNE3zxhtvDNY1TdPcunWr2bt3b3Pjxo3NfwAHuF27dpl9+/Y1v/322+CyZ555xrzlllt0HluBvLw8s1evXuZvv/0WXDZp0iTzrrvu0vkLY2vXrjVPPvlk8y9/+YvZq1ev4GfX/n72nXPOOcG6pmma3333nXnIIYeYxcXFTXo86ultYb/++iter5dBgwYFlw0ePJjMzEz8fn8LtkwqpKam8u9//5uUlJSQ5YWFhWRmZtK3b1+io6ODywcPHszq1asByMzMZMiQIcEyl8tFv379guXSfP71r39xyimn0KNHj+CyzMxMBg8ejGEYABiGwaGHHlrj+Wvfvj3p6elkZmY2a9sFVq5cSWxsLEOHDg0uu/zyy7n//vt1HluBqKgoXC4XixYtwuPxsH79er7//nsOPvhgnb8wtnz5coYNG8b8+fNDlu/PZ5/P5+OHH34IKR84cCAej4dff/21SY9HobeFZWVlkZiYiMPhCC5LSUmhrKyMvLy8lmuYBMXHxzN69Ojgc7/fz8svv8zw4cPJysqibdu2IfWTk5PZvn07wD7LpXksW7aMFStWcNVVV4Us39f52blzp85fmNi0aRMdOnRg8eLFHH/88Rx11FHMmjULv9+v89gKOJ1Opk6dyvz588nIyOCEE07giCOO4Mwzz9T5C2PnnHMOt956Ky6XK2T5/nz25efnU1ZWFlJus9lISEho8nNqa9Ktyz6VlJSEBF4g+NztdrdEk2QfZsyYwc8//8yCBQt44YUXqj1/FeeupvOrc9t8ysrKuPPOO5k6dSpRUVEhZfs6P6WlpTp/YaK4uJgNGzbw2muvcf/995OVlcXUqVNxuVw6j63E77//zrhx47joootYu3Yt99xzDyNGjND5a4X2dc5qKy8tLQ0+r2n9pqLQ28KcTmeVk1zxfO8PaGl5M2bMYN68eTz66KP06tULp9NZpUfe7XYHz11N5zc+Pr65mnzAmzlzJv379w/pra9Q0/nZ1/nbu9dDmp7NZqOwsJCHH36YDh06AIGLZV599VW6dOmi8xjmli1bxoIFC/j888+JiopiwIAB7Nixg6eeeopOnTrp/LUy+/PZ53Q6g8/3Lm/qc6rhDS0sLS2NXbt24fV6g8uysrKIiopSMAoz99xzD3PnzmXGjBkcd9xxQOD8ZWdnh9TLzs4O/tumpvLU1NTmabTw3nvvsWTJEgYNGsSgQYN45513eOeddxg0aJDOXyuSmpqK0+kMBl6Abt26sW3bNp3HVuDHH3+kS5cuIZ05ffv2ZevWrTp/rdD+nLOEhAScTmdIudfrJS8vr8nPqUJvCzv44IOx2WwhFzatXLmSAQMGYLHo9ISLmTNn8tprr/HII49w4oknBpdnZGTw008/Bf9dA4Hzl5GRESxfuXJlsKykpISff/45WC5N76WXXuKdd95h8eLFLF68mCOPPJIjjzySxYsXk5GRwapVq4LT6Jimyffff1/j+du2bRvbtm3T+WsBGRkZlJWV8ccffwSXrV+/ng4dOug8tgJt27Zlw4YNIb1769evp2PHjjp/rdD+fPZZLBYGDBgQUr569WpsNht9+vRp0nYrVbUwl8vFqaeeyrRp01izZg1Llizh+eef5+9//3tLN03K/f7778yePZvLLruMwYMHk5WVFfwaOnQo7du3Z8qUKaxdu5Znn32WNWvWcMYZZwBw+umn8/333/Pss8+ydu1apkyZQseOHRk2bFgLH9WBo0OHDnTp0iX4FRMTQ0xMDF26dOH4448nPz+f6dOns27dOqZPn05JSQknnHACAGeffTZvvfUWb7zxBr/++is33XQTY8eOpVOnTi18VAee7t27M3bsWKZMmcKvv/7KF198wbPPPsvZZ5+t89gKHHnkkdjtdm6//Xb++OMP/u///o+nn36a888/X+evFdrfz75zzjmHOXPmsGTJEtasWcO0adP461//2vRDVpp0QjSpk+LiYvOmm24yBw4caI4aNcqcO3duSzdJKnnmmWfMXr16Vftlmqb5559/mueee67Zv39/88QTTzS/+uqrkPU/++wz89hjjzUPOeQQ84ILLtDcki3s5ptvDpnzMzMz0zz11FPNAQMGmGeccYb5008/hdRfuHChOWbMGHPgwIHmxIkTzdzc3OZuspTLz883b7zxRnPgwIHmiBEjzCeffDI4t6vOY/hbu3ateeGFF5qHHnqoefTRR5tz587V+WtFKs/Ta5r7/9n3zDPPmCNGjDAHDx5sTpkyxSwtLW3yYzBMU7eGEhEREZHIpuENIiIiIhLxFHpFREREJOIp9IqIiIhIxFPoFREREZGIp9ArIiIiIhFPoVdEREREIp5Cr4iIiIhEPIVeEREREYl4Cr0iIq3I9ddfT+/evXn++eerlB155JHccsst9dpeQ9YREWmNFHpFRFqJgoIClixZQq9evZg/fz66oaaISN0p9IqItBLvvvsuALfddht//vkn33zzTQu3SESk9VDoFRFpJRYuXMiIESMYPnw4Xbp04bXXXqux7ubNm+nduzfvvfceV155JRkZGYwdO5ZZs2bh9/tD6no8Hh588EFGjhzJwIEDufjii9mwYUNInTfeeIMJEyYwcOBADjnkEE455RT++9//Nslxiog0BYVeEZFWYO3atfzwww+ceuqpAJx66ql88sknZGdn17retGnTiI2N5cknn+SUU05h5syZPPzwwyF13n//fdauXcsDDzzAnXfeyY8//si1114bLH/llVeYOnUqRx99NM888wwPPfQQDoeDG264ge3btzf6sYqINAVbSzdARET2beHChSQkJHDkkUcCcNppp/Hkk0+yYMECrrzyyhrX69evHw899BAARxxxBMXFxcybN49//OMfxMbGApCWlsbs2bOx2+0AbNiwgaeeeorCwkJiY2PZtGkTl1xyCVdddVVwux06dGDChAmsXLmSE088sakOW0Sk0ainV0QkzHk8Ht5++22OPvpoSktLyc/PJyYmhsGDB/P6669XGa5QWUXPcIXjjjsOj8fDqlWrgssOOeSQYOAF6NixIwD5+fkA3HLLLdxwww3k5+ezevVq3nrrLV555RUA3G53Yx2miEiTUk+viEiY++yzz8jJyWHBggUsWLCgSvkXX3zBmDFjql03LS0t5HlSUhIAu3fvDi6Ljo4OqWOxBPpDKsL0xo0bmTp1KsuWLcNut9O9e3f69OkDoBkkRKTVUOgVEQlzCxcupFOnTkyfPj1kuWmaTJo0iddee63G0Ltr166Q5zk5OQAkJyfXad9+v5/LL78cu93OggULOPjgg7HZbKxbt4633nqrAUcjItIyFHpFRMJYVlYWX3zxBZdeeinDhg2rUn788cezaNEiduzYUe36S5YsYfz48cHnH374IS6Xi4yMjDrtf9euXfzxxx/ceuutDBgwILh86dKlALUOrRARCScKvSIiYWzx4sV4vd4aLxY79dRTeeONN3j99derLf/vf/9LcnIyY8aMYfny5bzyyitce+21VYY01CQ5OZkOHTrwyiuv0K5dO+Lj4/niiy948cUXASgpKWnYgYmINDNdyCYiEsYWLVpEz5496dWrV7XlgwcPpmPHjrzxxhv4fL4q5VdffTW///47V111FR9++CFTp07l8ssvr1cbZs+eTVpaGrfccgvXXHMNmZmZPPXUU3Tv3p0VK1Y06LhERJqbYeoqBBGRiLN582aOOuoo7r//fiZMmNDSzRERaXHq6RURERGRiKfQKyIiIiIRT8MbRERERCTiqadXRP6/3TqQAQAAABjkb32PrygCgD3pBQBgT3oBANiTXgAA9qQXAIA96QUAYE96AQDYk14AAPYCIusJxYlAvZQAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8, 6))\n",
    "ax.set_title('Ridge Regression, with Alpha', fontsize=12)\n",
    "ax.plot(lambda_range, mean_vec, label='Mean MSE')\n",
    "ax.errorbar(lambda_range, mean_vec, yerr=std_vec, fmt='o', ecolor='r',\n",
    "            capsize=5, alpha=0.5, label='MSE Std Dev')\n",
    "ax.set_ylabel(\"MSE\", fontsize=12)\n",
    "ax.set_xlabel(\"Alpha\", fontsize=12)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArwAAAIoCAYAAACLTTgWAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/SrBM8AAAACXBIWXMAAA9hAAAPYQGoP6dpAABxJ0lEQVR4nO3de3zO9f/H8ee1zWZsDm02jMSYc5swE4olIql0pJRvSn3R+qUi9U0OSc7lWFRKfCPkEKrvl0lnRY1KNCQzx2tszGGz6/r8/tCur8tO1zXbddge99vN7eb6fN7X5/O69t7Hnt57f94fk2EYhgAAAIAyysfdBQAAAAClicALAACAMo3ACwAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDKNwAu42DPPPKPGjRvr3XffdXcpLrVlyxY1btxYjRs31tdff51vm71799raHDx40MUVFi4xMVEPP/yw2rRpo5YtW+rmm2/W+PHjlZaW5u7SSlT//v3Vv3//Qtvk9lFBf6ZMmeKiagv33nvvqUOHDrr22ms1Z84cd5dTbDNnzlTjxo2L/f5vvvlGjRs31m233VZix7/SmgBX83N3AUB5cvr0aW3YsEFRUVFaunSp/vGPf8hkMrm7LJfy8fHRZ599po4dO+bZt379ejdUVLSVK1dq5MiRuv/++zVgwAAFBgZqz549mjdvnjZt2qQVK1aoatWq7i7Tpe6++27dc889+e4LDw93cTV5ZWZmauLEiercubMeeeQR1alTx90luc2KFSsUFRWlP/74Q9u2bVPr1q3dXRLgcozwAi60du1aSdKLL76o/fv36/vvv3dzRa533XXX6b///a9ycnLy7Fu/fr2aNm3qhqoKN3v2bN16660aPXq0unTpori4OD344IOaN2+eUlJStGzZMneX6HI1a9ZUTExMvn9q1arl7vKUkZEhq9Wqrl27qm3bth5RkzucOnVKGzZs0COPPKL69etryZIl7i4JcAsCL+BCK1asUPv27RUXF6d69erZ/fB55JFH1KdPnzzvGTx4sHr37m17vXXrVj344IOKjo5WbGysRowYoRMnTtj2f/zxx2rWrJmWLVumDh06KDY2Vnv27JHFYtG8efPUq1cvXXvttYqJidH999+fJ3R/8cUX6tOnj6699lp1795da9eu1c0336yZM2fa2qSnp2vUqFG6/vrr1bJlS91777367rvvHPoa9OzZU+np6XnOu2vXLu3fv189evTI854//vhDjz/+uK677jpdd911GjJkiFJSUvK8f+jQoYqLi1Pz5s3VqVMnvfLKKzp//rytTePGjbV48WK9+OKLio2NVatWrfTUU0/JbDYXWrPZbJZhGHm2N2nSRCNHjlSLFi1s27KysjR+/Hhdf/31atWqlYYPH65///vfdr/+zW/aQO6Ujy1btti2/fjjjxo4cKDatm2rFi1aKD4+XjNnzpTVapUkHTx4UI0bN9aCBQt0yy23KDo6WitWrHD4a3bo0CENHTpUrVu3VocOHbRgwYJCvw7OKqi+mTNn6uabb9asWbMUGxurjh07KiMjQxaLRYsXL9Ztt92ma6+9Vp07d9aUKVOUlZVlO+bzzz+vhx9+WC+//LKuu+469ezZUxaLxe68H3/8seLj4yVJL7zwgt3Xfv369erTp49atWqlDh06aNSoUcrIyLDtL6i2/DhyHZw4cUJjxoxRly5d1KJFC8XGxmrIkCF5puysWrVKd955p6Kjo9W5c2dNnTpV2dnZdm2++OIL9e7dWy1btlT37t21atWqIvvgk08+UU5Ojjp16qTevXvr888/V3p6eqHv6d+/v55//nm9+eabuv7669W6dWsNHjxYqampedoWVZMj1yXgCkxpAFwkOTlZv/zyi9544w1J0h133KE5c+bIbDYrNDRUvXv31ogRI/TXX3+pXr16ki6Oznz55Zd6+umnJV0MQP/4xz8UFxen119/XRkZGXrjjTf00EMPafny5apYsaIkyWKx6N1339X48eN18uRJRUZGatKkSfrwww9tc4iPHj2q2bNn66mnntIXX3yhwMBAff/99xo8eLC6dOmip556Sn/99Zdefvllu8CRlZWlhx9+WGazWU8//bTCwsK0YsUKPfroo3r77bfVvn37Qr8ODRs2VKNGjfJMa1i3bp1iY2NVo0YNu/Z//vmn7r//fjVo0EATJ05UTk6O5s6dq759+2r16tUKCQnRsWPH9MADDygmJkavvfaa/P399eWXX2rBggUKCwvToEGDbMebPn26br75Zk2bNk0pKSmaMGGCfH19NW3atAJr7ty5s9atW6esrCz16NFDbdu2tf3afsCAAXZtn332WX311Vf6v//7P9WvX1+LFy/W66+/XujXJD+7du3SgAEDdMstt2j69OkyDEOffPKJZs2apQYNGujWW2+1tZ05c6ZefPFFBQUFKTo62qGv2dmzZ/Xggw/Kz89P48aNk4+Pj2bMmKEDBw6oVatWRdZntVrzHaWXJD8/+x8tl9e3bNkyHTp0SJs3b9b06dOVnp6uqlWr6sUXX9Tq1av12GOPqU2bNtq5c6dmz56t33//XW+//bZt+s/WrVsVEBCg2bNn6+zZs/L19bU7X+fOnTVr1iwNHTpU//znP9W5c2dJ0pw5czRjxgz169dPTz/9tFJSUvTGG28oKSlJH330ke36ya+2yzlyHRiGoccff1wZGRl69tlnFRoaqt27d+v111/Xyy+/rHfeeUeStHjxYo0dO1b33HOPhg0bppSUFE2aNEkZGRkaO3as7ZyjRo3S//3f/yksLEzz58/X888/ryZNmqhJkyYF9tOKFSvUqVMnhYaG6o477tDMmTO1cuVK/eMf/yi0fzdu3Kjq1avrX//6l6xWq6ZOnar+/ftr3bp1CgwMdKgmZ65LoNQZAFxiwoQJRmxsrJGVlWUYhmEcOnTIaNKkiTF37lzDMAzjzJkzRkxMjDFr1izbe5YtW2Y0adLEOHLkiGEYhnHfffcZvXr1MnJycmxt9u3bZzRt2tRYtGiRYRiGsWLFCiMqKspYtWqV3fmHDRtmvPfee3bbPv/8cyMqKsr4+eefDcMwjH79+hm9e/c2rFarrc3atWuNqKgoY8aMGYZhGMbSpUuNqKgoIykpydbGarUaDzzwgNGnT58CP//3339vREVFGd9//70xa9YsIzY21rhw4YJtf3x8vPHRRx/Z6k9JSbHVff311xunT5+2tT158qTRunVr47XXXjMMwzC++uor44EHHrBrYxiG0atXL+ORRx6xvY6KijL69u1r1+b55583YmJiCqzbMAzj1KlTxpNPPmk0btzYiIqKMqKiooyuXbsaEyZMsPWNYRjG7t27jaioKFtfGIZh5OTkGN27dzeioqJs2x588EHjwQcfLPDrYxiGsXLlSuPRRx81LBaLrY3FYjFat25tvPTSS4ZhGEZKSooRFRVlvPDCC3bHcuRrtmjRIqNx48ZGcnKyrc2hQ4eM5s2b56ntcrlfg4L+pKWlFVrfjBkzjKioKOPHH3+0bUtOTjaioqKMt956y67tqlWrjKioKOOLL74wDMMwRowYYURFRRmHDx8utMbcc69YscIwDMNIT083WrRoYfva5frxxx/t+iy/2vLjyHVw5MgRo3///nmONW7cOKNFixaGYVzs0/bt2xuDBw+2a/P2228bd955p5GdnW2rafPmzbb9f/31lxEVFWW8//77Bda4a9cuIyoqyvjss89s2wYMGGB0797drl3u8XM9+OCDRvPmzY0DBw7Ytv32229GVFSU8e9//9vuPYXV5Oh1CbgCUxoAF7hw4YLWrFmjrl276vz58zp16pQqV66s1q1b66OPPpLValWlSpXUtWtXuxu31q1bp/bt2ys8PFznzp3T9u3bdeONN8owDOXk5CgnJ0d169ZVZGSkvvnmG7tzXj4XdurUqXr44Yd14sQJbd26VStWrNCaNWskSdnZ2crOztbPP/+sbt262d1Id8stt9iN2H333XeqUaOGmjdvbqvBYrGoS5cu+vXXXwv89e+lLp/WsH37dh09elTdunXL0/b7779XbGysKlasaDtfUFCQ2rRpo2+//VaS1LFjRy1atEgBAQHas2ePNm7cqLlz5+rEiRN5fi0cExNj97pmzZo6d+5cofUGBwdrxowZ2rBhg0aNGqXu3bvr1KlTtl/V//zzz5IujjxK0k033WR7r6+vr3r16lXk1+Ryd9xxh+bPn68LFy5o165d+vzzzzVjxgxZLBZduHDBru3lfe3I12zr1q26+uqr1bBhQ9v7atWqlefrU5B7771Xy5cvz/dPlSpVCq0vv+0//PCDJNmNXOe+9vX1tZvqUa1aNdWsWdOhOnMlJSUpOzs7T1+0adNGERERtvMXVXMuR66D8PBwLVy4UK1bt9bBgwf1zTff6IMPPtBPP/1k+778888/lZaWpptvvtnu+AMHDtTHH3+sChUq2NWaK/cmvFOnThVY44oVK1SlShW1adNGp06d0qlTp9S9e3f9+eefRd4/cN1116lu3bq2182aNVPdunX1448/2rUrrCZnrkugtDGlAXCBL774QmlpabZAcLmvvvpKN954o26//XatWbNGu3btUmhoqLZs2aJXX31V0sUfIlarVfPnz9f8+fPzHCMgIMDudaVKlexe//LLLxozZox++eUXBQYGqmHDhqpdu7YkyTAMpaeny2KxKCQkxO59vr6+qlatmu11enq6jh8/rubNm+f7WY8fP17kigX169dX06ZNbdMa1q9fr44dO+b7vvT0dK1fvz7fFRyuuuoqSRd/vT5t2jQtXrxYZ8+eVa1atXTttdfm+ZpIsvt1rHRx1Qgjn/m5+alTp44eeOABPfDAA7JardqwYYOef/55jRs3Th9//LEt7OfWlas4qxacP39e48aN0+rVq5WTk6M6deqoVatW8vPzy1Pv5X3tyNcsIyND1atXz7O/Ro0aRc5plqSwsDC1bNnSoc9yeX25KleubPt77tfu8iktfn5+ql69uk6fPp3v+xyVe/zQ0NA8+0JDQ+2O78g5HL0O1qxZo2nTpunw4cOqVq2amjZtaps6kXscSXmuu/xc+nX08bk4XlXQ927uf7JPnTql66+/Ps/+JUuWKC4ursBz5fc9GxISkuc/tIXV5Mx1CZQ2Ai/gAitWrFDdunU1fvx4u+2GYWjo0KFasmSJbrzxRrVv3141atTQp59+qho1aiggIMA26lm5cmWZTCYNGDAgzyiYlDfIXSozM1OPPvqoGjdurHXr1qlBgwby8fHR5s2b9fnnn0u6+MOsQoUKecKO1Wq1u8klODhY11xzTYFrrTq6/FPPnj31zjvv6OWXX9Znn32mZ599Nt92wcHBuv766/Odc5g78jxv3jy99957GjNmjLp166bg4GBJF5fOulKff/65Xn75ZX344YeqX7++bbuPj4+6deumH3/8UR999JEk2QKk2Wy2/WdCkk6ePJnnuJffaHX27Fm71+PHj9fnn3+u119/Xddff70tWBQ1R1py7GtWvXp1/fXXX3n2F3VDU2nJ/c/O8ePHFRERYdt+4cIFnTx5Mt9wXpzjm81mNWjQwG7f8ePH7UYzHeHIdbB161aNGDFC/fv318CBA20hctKkSdq2bZsk2UbDL73xVLr4PbNz506H5lPnZ9OmTTp58qTGjRtnuycg14cffqgNGzYoLS2twKCd3/es2WzW1Vdf7XANpXldAs5iSgNQyo4fP66vvvpKt956q9q1a2f3Jy4uTrfccos2b96so0ePytfXV7fddps2bdqkzz77TF27drUFnaCgIDVr1kz79u1Ty5YtbX8aNWqkmTNn2v3K93L79u1Tenq6HnroITVs2NA2EvPll19KuhhqfX19dd1112njxo12701MTLS7OSk2NlaHDx9WSEiIXR3ffPON3n777Tw3EBWkR48eSk9P15tvvqmMjAy7aQCXyl1lomnTprZztWjRQu+9957++9//SpK2bdumhg0b6q677rL9UD169Kj++OMP24oGxdWoUSOlp6fr/fffz3f//v37FRUVJUmKi4uTyWTSZ599Ztdm06ZNdq+DgoJ05MgRu225AejS1+3atbP7Hvj111914sSJIj+TI1+zuLg4HTx4UL/88ovtfSdOnFBSUlKhxy4tsbGxki5O47nUunXrZLFYrnjt2OjoaPn7+9uWBsy1detWHTp0SNddd53T9RZ1Hfz888+yWq168sknbWHXYrHYppVYrVY1aNBA1atXz/M9snr1ag0aNCjP9BVHrVixQjVr1tQ999yT59+d/v3768KFC7YVPfKzbds2u9D766+/6uDBgw79h+vSY5TWdQk4ixFeoJStWrVKOTk5+Y7KShfnai5btkwfffSRnnzySd1+++1699135ePjk2fqwrBhwzRo0CA988wz6t27t201hu3bt2vw4MEF1lC/fn0FBQXpzTfflJ+fn/z8/PT555/bplfkzmFNSEhQ//79lZCQoLvvvluHDh2yrSqRO6+3T58+WrRokf7xj3/oiSeeUK1atfTtt99q/vz5evDBB+3mHBambt26atmypd566y3dfPPNBf7ae/Dgwbr//vv1+OOPq2/fvgoICNDSpUu1YcMGzZgxQ5JsT9KaN2+eYmJi9Ndff+mtt95SdnZ2kfNzi9KgQQMNGjRIb731lg4dOqTevXurZs2aSktL0+rVq/Xdd9/ZlvO65pprdP/992v69OnKyclRs2bNtGbNGtsc31xdunRRYmKiJkyYoPj4eG3dujXPck7XXnutPv30U3344YeKjIzUrl27NHfuXJlMpiI/kyNfs9tvv10LFy7U0KFD9fTTTysoKEhz5851OIgcOXKkwHAcGBjo9FO4GjZsqDvvvFMzZszQuXPn1LZtW/3++++aNWuW2rVrp06dOjl1vMtVq1ZNgwYN0uzZs1WhQgV16dJFBw8e1BtvvGE7tzMcuQ6uvfZaSdLYsWN11113KSMjQ4sXL9auXbskXRzVDwoK0pNPPqmxY8cqJCRE8fHx+vPPPzVjxgw98MADxXqgybFjx/TVV1/p4YcfzvfBNq1bt9bVV1+tpUuX6rHHHsv3GOfOndOjjz6qf/7znzpz5oymT5+uqKgop+ajl+Z1CTiLwAuUso8//liNGjWyjQJernXr1qpTp46WLVumwYMHq0mTJoqKitLJkyfzjKZ07NhR77zzjmbNmqWEhARVqFBBzZs314IFCwq92Sg4OFhz5szRpEmT9NRTT6ly5cpq2rSpFi1apMcee0xbt25VfHy82rRpo5kzZ+qNN97Q4MGDFRERoZdeeklPP/20bU5jpUqVtHjxYk2dOlWTJ0/W6dOnFRERoWeeeUaPPPKIU1+bnj176pdffinwPwPSxbVuFy9erOnTp2v48OEyDENRUVGaPXu2bVT48ccf18mTJ7Vw4ULNnj1btWrV0u233y6TyaS33npLp06dynMjlTOGDRumpk2batmyZXrllVeUmZlpuxlo+fLldstCjRo1SqGhofrggw906tQp3XjjjerXr58WL15sa3PXXXfpwIEDWrlypZYsWaK2bdtqxowZ6tu3r63N888/rwsXLuj1119Xdna26tSpo3/+85/as2ePEhMT80yJcPZr5u/vr/fff1+vvvqqxo8fL5PJpHvvvVd169Z16HHJBc1Hzz3/6tWrizzG5caPH6969eppxYoVmj9/vsLCwvTQQw9p8ODBtt9KXIknn3xSoaGhWrRokZYuXapq1arplltu0f/93/8V+B+ugjhyHbRr106jRo3SggUL9Nlnnyk0NFTt2rXTrFmzNGTIEG3btk033nijHnjgAVWqVEnvvPOOli5dqpo1a+qxxx4rMIwWZdWqVbJYLOrZs2eBbW6//XbNnDlTX331Vb7727Rpo7i4OL344ouSpPj4eA0fPlz+/v4O11Ha1yXgDJPh6N0aAMq8jRs3qmbNmnY34iQnJ6tXr16aM2dOgdMOULiZM2dq1qxZ2r17t7tLAYqU+1CUDz74wM2VACWHEV4ANl9//bXWr1+vZ599VvXr19fRo0c1d+5cNWjQwO4hEQAAeBMCLwCbESNGqGLFipo7d66OHTumatWqqVOnTnrmmWdYSggA4LWY0gAAAIAyjWXJAAAAUKYReAEAAFCmEXgBAABQphF4AQAAUKaxSkMh0tJOyxW39JlMUkhIsMvOh5JHH3o/+tD70Yfejf7zfq7uw9zzOYLAWwjDkEsvOlefDyWPPvR+9KH3ow+9G/3n/TyxD5nSAAAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDLNz90FAABQnlishpJSM2TOzFZokL9iIqrK18eU73ZJDrf19GMUeewz2Yqsna0GwRXkY/LOz+iSr5OHHyPr4CkFWK227Z6CwAsAgIskJps1NXGPjmVm27aFBfmre5Mwfb7rmN32KhX9ZJKUcT6nyLaefgxPr89TjuHp9Tl7jGfiGyq+Uag8gckwDMPdRXgqs/m0XPHVMZmk0NBgl50PJY8+9H70offz9D5MTDZrxJqd7i4DcKmJvZuVWujNveYdwRxeAABKmcVqaGriHneXAbjctE17ZbG6/3+gBF4AAEpZUmqG3a97gfLi6OksJaVmuLsM5vACAFAaLr3h588TZ9xdDuA2Zg/4zx6BFwCAEpbfzWlAeRUa5O/uEtw7peHo0aNKSEhQbGysOnXqpAkTJigrK0uSdOjQIT322GOKjo7WzTffrPXr19u9d+3ateratauio6M1ZMgQnThxwrbPMAxNmTJFcXFxio2N1aRJk2S1Wl362QAA5VPuzWmEXUAKDw6wLWfmTm4LvIZhKCEhQefOndPixYs1ffp0bdq0Sa+//rpycnL0+OOPy8/PTytXrtTAgQM1fPhw/fHHH5KkHTt26MUXX9TQoUO1dOlSnTp1SiNHjrQde8GCBVq7dq1mzZqlGTNm6JNPPtGCBQvc9VEBAOUEN6cB9oZ1ifSI9XjdFnj37dunpKQkTZgwQY0aNVKbNm2UkJCgtWvXavPmzTp8+LAmT56sBg0a6P7779cNN9ygn3/+WZK0aNEi9ejRQ3fccYeaNGmiSZMmafPmzUpJSZEkLVy4UAkJCWrTpo3i4uL07LPPavHixe76qACAcqI4N6eFBweof5s6Crvs175VK/qpakU/h9p6+jE8vT5POYan1+fsMUpzSTJnuW0d3lOnTmn79u3q1KmTbdvatWv10ksv6d5779WhQ4c0c+bMfN/bvXt3PfbYY7r77rtt2+Lj4zVs2DC1bdtWN9xwgzZs2KC6detKkg4ePKibbrpJX331lcLCwhyuMS3NdevwhoQEu+x8KHn0ofejD72fJ/ThZ78f07/W7Sqy3SNxddUgpLJCK/urVZ3/Panq54MXnziWu11Snm0FtfX0YxR17LSz2WpQq5oiq/zvSWve9hld8XXy5GMkpWbovMlHFQ3XPGkt95p3qK2nPHjCarWqX79+ql69ukwmkyIiIhQQEKDVq1erevXqSkhIUNeuXSVJrVq10htvvKEbbrjB9v577rlH3bt3V1xcnO666y7t2LFDAQEBkqTz588rOjpay5Yt07XXXuuWzwcAKPu+25umvvO/L7Ldh4/FqX1kiAsqAiB50CoNkydP1s6dO7V8+XK9+uqrWrlypXr27Kk333xTW7ZsUUJCgpYuXaqWLVvq/Pnz8ve3Hzr39/dXdna2zp8/b3t96T5Jys527tdMjPDCUfSh96MPvZ8n9GGD4AoKC/IvdFpDeHCAGgRXkNl82oWVeT5P6D9cGVf3oTMjvB4ReCdPnqz3339f06dPV1RUlHx9fVWtWjWNHj1aPj4+at68ubZu3aqPPvpILVu2VEBAQJ7wmp2drcDAQLtwmzvCm9s2MDDQqboMQy696Fx9PpQ8+tD70Yfez5196GMy6Zn4hoU+QnhYl0j5mEx8nxWAa9D7eWIfuv1Ja+PGjdOCBQs0efJkde/eXZIUFhama665Rj4+/yuvfv36Onz4sCQpPDxcZrPZ7jhms1k1atRQeHi4JOn48eO2fbl/r1GjRql+FgAArr+muir5++bZ7mk38QDliVtHeGfNmqUlS5Zo2rRpuuWWW2zbo6OjNXfuXFksFvn6XvxHY+/evYqIiLDt37Ztm/r06SNJOnz4sA4fPqzo6GiFh4erdu3a2rZtm+rUqSNJ2rZtm2rXru3UDWsAABTH6l+O6Gy2RbWqBOhf3aJ08uwFhQb5u+QmHgD5c1vg3bt3r+bMmaNBgwapdevWdiOyvXr10uzZszVmzBgNHDhQX3/9tb766it99NFHkqS+ffuqf//+iomJUcuWLTV+/Hh17tzZtipD3759NWXKFNWsWVOSNHXqVD3yyCOu/5AAgHIlx2LVoq0HJUkPx9ZVbL3qbq4IgOTGwLtx40ZZLBbNnTtXc+fOtdu3e/duLViwQKNHj1avXr1Uu3ZtTZ8+Xc2bN5d0cZWGsWPHasaMGcrIyFCHDh00btw42/sHDhyotLQ0DR06VL6+vrr77rs1YMAAV348AEA5krsk0392HdOR01m6qlIF9Wpe091lAfibxyxL5onMZtfdZRgaGuyy86Hk0Yfejz70fu7qw8Rks6Ym7rFbmSHI31cv3dKY+bpO4Br0fq7uw9zzOcLtN60BAOCtEpPNGrFmZ55lyDKzLRqxZqcSk80FvBOAKxF4AQAoBovV0NTEPYW2mbZpryxWhisBdyPwAgBQDEmpGYU+YEKSjp7OUlJqhosqAlAQAi8AAMVgLiLsOtsOQOkh8AIAUAyhQf5FN3KiHYDSQ+AFAKAYYiKqKqyIMBseHKCYiKouqghAQQi8AAAUg6+PSc/ENyy0zbAukTxdDfAABF4AAIopvlGohnS6Js/28OAATezdjHV4AQ/htietAQBQFoRUujitISqssh5qU1ehQf6KiajKyC7gQQi8AABcgf0nzkqSomtXVfemYW6uBkB+mNIAAMAV+DPtYuC95qpKbq4EQEEIvAAAXIHcEd4GIQRewFMReAEAKKasHKtSM85Lkq4h8AIei8ALAEAxpZw8J6shBQf4KaRSBXeXA6AABF4AAIrpzxP/m79rMrEqA+CpCLwAABTTn2lnJEn1QwLdXAmAwhB4AQAopj/TzklihQbA0xF4AQAoptwVGupzwxrg0Qi8AAAUg8Vq6MBJ1uAFvAGBFwCAYjiUcV7ZFkMBfj6qVaWiu8sBUAgCLwAAxZC7QkO96oHy9WGFBsCTEXgBACiG/WnM3wW8BYEXAIBiuHQNXgCejcALAEAx/MkIL+A1CLwAADjJMAzbkmSM8AKej8ALAICTjmdm60y2Rb4m6erqPGUN8HQEXgAAnJQ7fzeiWqAq+PKjFPB0XKUAADjJtkID0xkAr0DgBQDASX/ySGHAqxB4AQBw0n4CL+BVCLwAADgpd0kyVmgAvAOBFwAAJ2Scu6ATZy9IIvAC3oLACwCAE3KnM4QHB6iSv6+bqwHgCAIvAABO+JMVGgCvQ+AFAMBBFquh7/86KUkK9PeVxWq4uSIAjiDwAgDggMRks3rP36KNf5glSZv+fp2YbHZzZQCKQuAFAKAIiclmjVizU8cys+22H8vM1og1Owm9gIcj8AIAUAiL1dDUxD2Ftpm2aS/TGwAPRuAFAKAQSakZeUZ2L3f0dJaSUjNcVBEAZxF4AQAohLmIsOtsOwCuR+AFAKAQoUH+JdoOgOsReAEAKERMRFWFFRFmw4MDFBNR1UUVAXAWgRcAgEL4+pj0THzDQtsM6xIpXx+TiyoC4CwCLwAARYhvFKqJvZvJdFmmDQ8O0MTezRTfKNQ9hQFwiJ+7CwAAwBvEXl1Nxt8rj73UvZEiqgYqJqIqI7uAFyDwAgDggJT0c5KkqypVUO8WtdxcDQBnMKUBAAAHpJy8GHjrVgt0cyUAnEXgBQDAAbkjvHWrE3gBb0PgBQDAAbkjvFcTeAGvQ+AFAMABB06elyTVYUoD4HUIvAAAOODg31MaribwAl6HwAsAQBEys3J08twFSVKd6hXdXA0AZxF4AQAowoG/5++GVPZXZX9W9AS8DYEXAIAi5E5nqFuN0V3AGxF4AQAowgHW4AW8GoEXAIAisAYv4N0IvAAAFIE1eAHvRuAFAKAIKemswQt4MwIvAACFOH0+R+l/L0nGHF7AOxF4AQAoxIG/5++GVvZXJX9fN1cDoDgIvAAAFOLgSZYkA7wdgRcAgEIcYIUGwOsReAEAKEQKa/ACXo/ACwBAIXLX4GVJMsB7EXgBAChE7ggvS5IB3ovACwBAAU6dv6CM8zmSmMMLeDMCLwAABcgd3a0R5K/ACixJBngrAi8AAAXgCWtA2UDgBQCgALkjvFcTeAGvRuAFAKAArMELlA0EXgAACmBbg5fAC3g1Ai8AAAU4mM5jhYGygMALAEA+Ms5dsiQZc3gBr+bWwHv06FElJCQoNjZWnTp10oQJE5SVlWXX5vTp0+rUqZM+/vhju+1r165V165dFR0drSFDhujEiRO2fYZhaMqUKYqLi1NsbKwmTZokq9Xqks8EAPB+Fquh/+4+JkmqVtFPFXwZHwK8mduuYMMwlJCQoHPnzmnx4sWaPn26Nm3apNdff92u3eTJk3Xs2DG7bTt27NCLL76ooUOHaunSpTp16pRGjhxp279gwQKtXbtWs2bN0owZM/TJJ59owYIFrvhYAAAvl5hsVu/5WzRx415JUvr5HPWev0WJyWY3VwaguNwWePft26ekpCRNmDBBjRo1Ups2bZSQkKC1a9fa2mzdulXff/+9atSoYffeRYsWqUePHrrjjjvUpEkTTZo0SZs3b1ZKSookaeHChUpISFCbNm0UFxenZ599VosXL3bp5wMAeJ/EZLNGrNmpY5nZdtuPZWZrxJqdhF7AS7kt8NaoUUNvv/22QkND7bZnZmZKkrKzs/XSSy9p1KhR8vf3t2uzfft2tWnTxva6Vq1aql27trZv366jR4/q8OHDatu2rW1/69atlZqammekGACAXBaroamJewptM23TXlmshosqAlBS/Nx14ipVqqhTp06211arVYsWLVJcXJwk6c0331SzZs3UsWPHPO89duyYwsLC7LaFhIToyJEjOn78uCTZ7c8N1UeOHMnzvsKYTI5/niuRex5XnQ8ljz70fvSh97vSPkxKzcgzsnu5o6ezlJSaoTZXVyveSVAgrkHv5+o+dOY8bgu8l5s8ebJ27typ5cuXa8+ePVqyZInWrFmTb9vz58/nGfX19/dXdna2zp8/b3t96T7p4qixM0JCgp1qf6VcfT6UPPrQ+9GH3q+4fZh18JRj7Xx8FBrK90lp4Rr0fp7Yhx4ReCdPnqz3339f06dPV6NGjdS3b18lJCTkme6QKyAgIE94zc7OVmBgoF24DQgIsP1dkgIDnVtWJi3ttAwX/ObKZLr4zeGq86Hk0Yfejz70flfahwEOruYTYLXKbD7t/AlQKK5B7+fqPsw9nyPcHnjHjRunDz/8UJMnT1b37t2Vmpqqn3/+Wbt379bEiRMlSefOndPLL7+s9evX6+2331Z4eLjMZvsbB8xms2rUqKHw8HBJ0vHjx1WnTh3b3yXlufmtKIYhl150rj4fSh596P3oQ+9X3D6MiaiqsCD/Qqc1hAcHKCaiKt8jpYhr0Pt5Yh+6NfDOmjVLS5Ys0bRp03TLLbdIksLDw/Wf//zHrl3//v3Vv39/9e7dW5IUHR2tbdu2qU+fPpKkw4cP6/Dhw4qOjlZ4eLhq166tbdu22QLvtm3bVLt2bafm7wIAyhdfH5OeiW+oEWt2FthmWJdI+fowyRTwNm4LvHv37tWcOXM0aNAgtW7d2jYKK0n16tWza+vn56eQkBDb6G3fvn3Vv39/xcTEqGXLlho/frw6d+6sunXr2vZPmTJFNWvWlCRNnTpVjzzyiIs+GQDAW8U3CtXE3s00av0uZeX8b4pDeHCAhnWJVHyj/KfaAfBsbgu8GzdulMVi0dy5czV37ly7fbt37y70va1atdLYsWM1Y8YMZWRkqEOHDho3bpxt/8CBA5WWlqahQ4fK19dXd999twYMGFAaHwMAUMbENwrVrCB/paSf14DYuoq7prpiIqoysgt4MZNheNosC89hNrtu0nVoaLDLzoeSRx96P/rQ+5VUH567YNGNM76RIek//4xT9Ur+Rb4HV45r0Pu5ug9zz+cIHg4OAMAl9prPyJAUUtmfsAuUEQReAAAu8cfxM5KkRjUqu7kSACWFwAsAwCWSj118xH2jUAIvUFYQeAEAuMQe898jvGEEXqCsIPACAPA3wzCUbJvSEOTmagCUFAIvAAB/O3TqvM5kW1TB16Rrqjv3OHoAnovACwDA3/b8Pbpb/6pK8vPlRyRQVnA1AwDwN9sKDWFMZwDKEgIvAAB/s83fZYUGoEwh8AIA8Lc9x/9ekow1eIEyhcALAICks9kWHUw/L4nAC5Q1BF4AAHRx/V1DUiiPFAbKHAIvAABiOgNQlhF4AQDQJSs0EHiBMofACwCAxBPWgDKMwAsAKPeshqG9ZkZ4gbKKwAsAKPcOZfzvkcL1eKQwUOYQeAEA5V7udIYGIZV5pDBQBnFVAwDKvT3csAaUaQReAEC59wdLkgFlmp+7CwAAwF0sVkNJqRnaceiUJCkytJKbKwJQGgi8AIByKTHZrKmJe3QsM9u2bfSnf+i5mxoqvlGoGysDUNKcntLw9ttv68iRI6VRCwAALpGYbNaINTvtwq4kmc9ka8SanUpMNrupMgClwenA++abb+rChQulUQsAAKXOYjU0NXFPoW2mbdori9VwUUUASpvTgbdXr16aO3eu9u/fr+zs7KLfAACAB0lKzcgzsnu5o6ezlJSa4aKKAJQ2p+fwfvnllzp06JBWrlyZ7/7ff//9iosCAKC0mIsIu862A+D5nA68r732WmnUAQCAS4QG+ZdoOwCez+nAGxsbK0nav3+/9u7dK6vVqvr166thw4YlXhwAACUtJqKqwoL8C53WEB4coJiIqi6sCkBpcjrwnjp1SiNHjtTGjRtVtWpVWSwWnTlzRm3bttXs2bMVHBxcGnUCAFAifH1Meia+oUas2Vlgm2FdIuXrY3JhVQBKk9M3rb3yyis6cuSI1q9fry1btmjr1q365JNPdPbsWU2YMKE0agQAoETFNwrVxN7NFBTga7c9PDhAE3s3Yx1eoIxxeoQ3MTFRCxYsUIMGDWzbGjZsqFGjRumxxx4r0eIAACgt8Y1C9e2+NK3+9ahubBiivtdFKCaiKiO7QBnkdOANCAiQj0/egWGTySSLxVIiRQEA4Ap/njgnSerWuIZa163m3mIAlBqnpzTEx8drzJgxOnDggG3b/v379corr+jGG28s0eIAACgthmFoX9oZSVKDkMpurgZAaXJ6hPe5557TkCFD1L17d1WpUkXSxRvZOnXqpJdeeqnECwQAoDQcy8xWZpZFvibp6uqB7i4HQClyOvCmp6frgw8+0K5du7Rv3z4FBASofv36dnN6AQDwdLmju3WrB8rfz+lfeALwIk4H3r59++qtt95SixYt1KRJk9KoCQCAUrfPfFaSFBnKdAagrHP6v7ShoaFKS0srjVoAAHCZ/83freTmSgCUNqdHeJs1a6bBgwerZcuWioiIkL+//aMXWYsXAOAN9qVdHOHlhjWg7HM68EpS7969S7oOAABcxjAM25SGBqGM8AJlndOBNzIyUr169VLNmjVLox4AAErdkdNZOnvBIj8fk66uxgoNQFnn9BzeN998UxcuXCiNWgAAcInc0d2rqwfKz5cVGoCyzumrvFevXpo7d67279+v7Ozs0qgJAIBSlXvDGis0AOWD01MavvzySx06dEgrV67Md//vv/9+xUUBAFCa9tpuWGP+LlAeOB14X3vttdKoAwAAl9ln/ntJMkZ4gXLB6cAbGxub73bDMHTs2LErLggAgNJkNQz9yQgvUK44NIf3pptu0smTJ+22zZs3T6dOnbK9TktLU+fOnUu0OAAAStqhjPM6n2NVBV+T6rBCA1AuOBR4U1NTZbVa7ba9+eabysjIsNtmGEbJVQYAQCnIfeDENVdVkp+Pyc3VAHCFYq/Fkl+4NZn4hwMA4Nls83eZzgCUGyw+CAAoV3JHeFmSDCg/CLwAgHJlHzesAeWOQ4HXZDLlma7A9AUAgLexWA3tP5EbeBnhBcoLh5YlMwxDHTp0yLOtW7dudq8JwQAAT5aacV5ZOVYF+PmodtWK7i4HgIs4FHgXLlxY2nUAAFDqcm9Yu+aqSvJlhQag3HAo8Bb0sAkAALyFxWroq31pkqRqgX6yWA1CL1BOcNMaAKDMS0w2q/f8LVrz61FJ0pa/0tV7/hYlJpvdXBkAVyDwAgDKtMRks0as2aljmdl2249lZmvEmp2EXqAcIPACAMosi9XQ1MQ9hbaZtmmvLFaeFAqUZQReAECZlZSakWdk93JHT2cpKTXDRRUBcAeHblrr37+/w0uOsaIDAMBTmIsIu862A+CdHAq87dq1s/395MmTWrp0qbp27aqWLVuqQoUK+v3337V+/Xo98MADpVYoAADOCg3yL9F2ALyTQ4F36NChtr8PGDBAL7zwgvr162fXpm3btlq6dGnJVgcAwBWIiaiqsCD/Qqc1hAcHKCaiqgurAuBqTs/hTUpKUvv27fNsj46O1u7du0ukKAAASoKvj0nPxDcstM2wLpGsxwuUcU4H3mbNmmnevHnKysqybcvMzNSMGTMUExNTkrUBAHDF4huF6rXbmubZHh4coIm9mym+UagbqgLgSg5NabjUuHHjNGjQIHXo0EH16tWTYRjav3+/ateurbfeeqs0agQA4Io0DQ+WJPn6SKO6N7ZNY2BkFygfnA68kZGR+vTTT/Xtt99q7969kqRGjRrp+uuvl5+f04cDAKDU/Zl2VpJ0zVWV1LNZuJurAeBqxUqo/v7+ioiI0IULF3T99dfrxIkT8vX1LenaAAAoEX+euBh4619Vyc2VAHAHpwNvRkaGnnrqKf3www+SpM8//1zjx49XSkqK5s2bp4iIiBIvEgCAK7H/khFeAOWP0zetvfLKKwoMDNT333+vgIAASdL48eNVs2ZNvfLKKyVeIAAAV8o2whtC4AXKI6cD71dffaVhw4apSpUqtm0hISEaOXKkfvzxxxItDgCAK2UYhvYTeIFyzenAK8luSbJcJ06c4KY1AIDHSTt7QafO58jHJF1dncALlEdOB95evXpp/PjxSk5Olslk0tmzZ/X999/rpZdeUs+ePUujRgAAii13/m7tqhUV4FescR4AXs7pIdnhw4dr2rRp6tOnjy5cuKA77rhDvr6+uvvuuzV8+PDSqBEAgGLLnb/LDWtA+eV04N2xY4eefvpp/d///Z9SUlJksVhUt25dVa5cuTTqAwDgiuSO8LIkGVB+Of27nSFDhujPP/9UxYoV1ahRIzVp0qTYYffo0aNKSEhQbGysOnXqpAkTJtjmByclJen+++9Xq1at1L17dy1btszuvd9++6169eql6OhoPfTQQ0pJSbHb/95776lTp05q1aqVXnjhBZ07d65YNQIAvBsrNABwOvA2atRIO3bsuOITG4ahhIQEnTt3TosXL9b06dO1adMmvf766zp+/Lgee+wxxcbGauXKlUpISNC4ceP0xRdfSJIOHTqkIUOGqE+fPlq+fLmuuuoqDR48WIZhSLq4NvCsWbM0duxYvf/++9q+fbsmT558xTUDALwPKzQAcHpKQ9WqVfXyyy9rxowZqlOnjvz9/e32L1y40KHj7Nu3T0lJSfrmm28UGhoqSUpISNDEiRN19dVXKzQ0VMOGDZMkXXPNNdqyZYs++eQTde7cWcuWLVOLFi30yCOPSJImTJigDh066IcfflC7du20cOFCPfzww+rSpYskacyYMRo4cKCee+45BQYGOvuRAQBeKjMrR8czsyUxhxcoz5wOvE2bNlXTpk2v+MQ1atTQ22+/bQu7uTIzM9WpU6d8z5GZmSlJ2r59u9q0aWPbHhgYqObNmyspKUlt2rTRL7/8oqFDh9r2x8TE6MKFC9q1a5datWp1xbUDALzDn3/P360R5K+gAJbOBMorp6/+S4PklahSpYo6depke221WrVo0SLFxcWpTp06qlOnjm1fWlqa1q1bpyeffFKSdPz4cYWFhdkdLyQkREeOHNGpU6eUlZVlt9/Pz0/VqlXTkSNHnKrRZCrOJ3Ne7nlcdT6UPPrQ+9GH3i+/PrRNZ7iqEn3r4bgGvZ+r+9CZ8zgdeM+dO6elS5dqz549slgstu3Z2dnauXOnPv30U2cPKUmaPHmydu7cqeXLl9ttP3/+vJ588kmFhobqvvvus9Vw+VQKf39/ZWdn6/z587bX+e13RkhIsLMf44q4+nwoefSh96MPvd+lfXj03EFJUtOIqgoNpW+9Adeg9/PEPnQ68P7rX//St99+q+uvv16fffaZevToob/++ivPNAJnTJ48We+//76mT5+uqKgo2/YzZ85o8ODB2r9/v/7973/b5t8GBATkCa/Z2dmqUqWKAgICbK8v3+/s/N20tNP6+z64UmUyXfzmcNX5UPLoQ+9HH3q//Ppw58F0SVKtyhVkNp92X3EoEteg93N1H+aezxFOB94vv/xSb7zxhq6//nolJydrwIABatGihV577TUlJyc7Xey4ceP04YcfavLkyerevbtte2Zmph599FEdOHBA77//vq655hrbvvDwcJnNZrvjmM1mNW3aVNWqVVNAQIDMZrMiIyMlSTk5OUpPT1eNGjWcqs0w5NKLztXnQ8mjD70ffej9Lu3DfWn/e+gE/eoduAa9nyf2odPLkmVlZdnCZ6NGjfTrr79Kku677z5t3brVqWPNmjVLS5Ys0bRp03TrrbfatlutVg0dOlQHDx7UBx98oEaNGtm9Lzo6Wtu2bbO9PnfunHbu3Kno6Gj5+PioZcuWdvuTkpLk5+enJk2aOPtxAQBe6vwFiw5lXJzmxgoNQPnmdOCNjIzUt99+K+li4M0NlqdPn7Y9NMIRe/fu1Zw5c/TYY4+pdevWOn78uO3P8uXLtWXLFr3yyiuqUqWKbXt6erok6a677tJPP/2kefPmKTk5WSNHjlSdOnXUrl07SVK/fv30zjvvaMOGDdqxY4dGjx6te++9lyXJAKAcOXDynAxJVSr66apKFdxdDgA3KtYqDU899ZSsVqtuv/123XrrrXriiSe0e/duu1UXirJx40ZZLBbNnTtXc+fOtdvXsWNHWa1WPf7443bbY2Nj9cEHH6hOnTqaOXOmXn31Vc2ePVutWrXS7NmzZfr7dr1bb71VqampGjVqlLKzs9WtWzc999xzzn5UAIAXy12h4ZqrKtl+PgAon0yG4fwsi5SUFFmtVtWrV0+7du3S6tWrVb16dfXv379MjaKaza6bdB0aGuyy86Hk0Yfejz70fpf34Vvf7Nfb3x/Q7S1r6l/dooo+ANyKa9D7uboPc8/niGKtwl23bl3b35s0acLcWACAx7l0DV4A5ZvTgTc+Pr7QXw1t3LjxigoCAKAk2FZoCCHwAuWd04E392lnuXJycpSSkqKPP/5YTz31VIkVBgBAceVYDR04eU4SI7wAihF477zzzny3R0dH691339U999xzxUUBAHAlUtPPKcdqqKKfj2pWCXB3OQDczOllyQrSsGFD/fLLLyV1OAAAisViNbThj+OSpBpB/twABcD5Ed4ff/wxz7YzZ87k+4AIAABc6bNfD2vUql91LPPi4+VT0s+r9/wteia+oeIbhbq5OgDu4nTg7d+/f55tFSpUUMuWLfXKK6+USFEAADgr8Q+zRqzZqcsHdI9lZmvEmp2a2LsZoRcop5wOvLt27SqNOgAAKDaL1dCUxD15wu6lpm3aqxsjQ+Trw0MogPLG6cB76NAhh9vWrl3b2cMDAOC0pNQM2zSGghw9naWk1Ay1rlvNNUUB8BhXtA5v7kPaLl+X1zAMmUwm/f777yVQIgAAhTMXEXadbQegbHE68L7++uuaOXOmnnvuObVq1Ur+/v767bffNH78ePXq1Uu33HJLadQJAECBQoP8S7QdgLLF6cA7ceJETZ06Vdddd51tW5s2bTRu3Dg98cQTGjhwYIkWCABAUWIiqiosyF/HM7MLnMcbHhygmIiqLq0LgGdweh3ezMxM21SGS506dUoXLlwokaIAAHCGr49Jz8Y3LLTNsC6R3LAGlFNOB97bb79dw4cP1yeffKLk5GT98ccfWrFihZ5//nn169evNGoEAKBI8VGhmvvgdbo804YHB7AkGVDOOT2lYcSIEfL399f48eOVnp4uSapVq5YGDRqkBx54oKTrAwDAYdfVqy6rIZkkjeoepVpVKyomoioju0A553TgrVChgoYPH67nnntOJ0+eVMWKFVWpUqXSqA0AAKf8duiUJKneVYHq1aKmm6sB4CmcCrw7duxQkyZN5O/vL5PJpJ9++knfffedrrrqKt11112qWZN/XAAA7rPz78DbOCzIzZUA8CQOzeE1m8264447dN999+ngwYOSpDfffFNPPvmkfvvtN/3888+6/fbbtWfPnlItFgCAwvx2KEMSgReAPYcC7/Tp01W5cmV98cUXatCggTIyMjRnzhx16tRJS5Ys0dtvv61+/fpp6tSppV0vAAAFyp3SEEXgBXAJhwLv5s2b9dxzzyk8PNz2+sKFC7rvvvtsbW6++WZt3bq1dKoEAKAImVk5+ivtrCRGeAHYcyjwZmRkKCwszPb6u+++k5+fn9q3b2/bFhwcrJycnJKvEAAAByQfPyPp4jJk1QIruLkaAJ7EocB79dVXKzk5WZJ04cIFbd68WbGxsXarM3zzzTeqW7du6VQJAEARdh/LlCRFhVV2cyUAPI1DqzTcd999Gj16tP7xj39o69atOnHihAYMGCDpYgD+8ssvNX36dD3xxBOlWSsAAAXaffRi4G3CdAYAl3Eo8D700EOSpFWrVslkMum1115Tp06dJEnjx4/XsmXLdP/99+vhhx8uvUoBAChE7ggv83cBXM5kGIZxJQc4cuSI/P39ddVVV5VUTR7DbD6tK/vqOMZkkkJDg112PpQ8+tD70YfeLTvHqhtmfiOL1dAnj8WqZpWK7i4JTuIa9H6u7sPc8znC6SetXY6HTQAA3G1f2hlZrIaqBlZQzSoB7i4HgIdx6KY1AAA8We50hua1q8hkMrm5GgCehsALAPB6fxy7uCRZs1pV3FwJAE9E4AUAeD3bCG8EgRdAXsWaw5uZmak9e/YoJydHl9/z1rZt2xIpDAAAR1gNw/bQiea1q7q5GgCeyOnAu3r1ao0ePVrnzp3Ls89kMun3338vkcIAAHBEyslzOnvBogA/HzUIraz0k2fcXRIAD+N04J0+fbruueceJSQkKCiItQ4BAO6VO52hYWhl+fkyUw9AXk7/y5Cenq6HHnqIsAsA8Ai7/75hrXE4P5cA5M/pwNulSxf95z//KY1aAABw2h9/j/BG1ajs5koAeCqnpzSEh4dr+vTp+vTTT1WvXj1VqFDBbv+ECRNKrDgAAApisRr6+WC6fjl0ShKBF0DBnA68GRkZ6tWrV2nUAgCAQxKTzZqauEfHMrNt20Z88rvG+fqqTU2mNgCw53TgZQQXAOBOiclmjVizM8/245nZ+ueinzSxdzN1aRTqhsoAeCqnA69hGNq4caOSk5NlsVhs27Ozs7Vz5069/fbbJVogAAC5LFZDUxP3FNpm6qa9uiEyRL4+PGIYwEVOB95x48Zp+fLlatasmXbs2KFWrVrpwIEDMpvN6tu3b2nUCACAJCkpNcNuGsPlDElHT2cpKTVDretWc1ldADyb06s0rF+/XlOmTNGSJUt09dVXa/To0dq0aZNuvfVWXbhwoTRqBABAkmQuJOwWpx2A8sHpwJuZmakWLVpIkqKiorRjxw75+fnp8ccf1+bNm0u8QAAAcoUG+ZdoOwDlg9OBt27dutq58+LNAo0aNdKOHTskXZzbe/r06ZKtDgCAS8REVFVYIWHWJCk8OEAxEVVdVxQAj+f0HN5HHnlEzz33nMaPH6+ePXuqT58+8vPz088//6zWrVuXRo0AAEiSfH1Meia+Yb6rNOR6pkskN6wBsOP0CO8999yjefPmqV69eoqMjNSsWbN0/PhxtWjRgiXLAAClLr5RqCbe1lSXZ9rw4ADNffA6xUexJBkAe06P8EpS27ZtJV18CEWHDh3UsWNHmUz8bxoA4BqNw4NkNSQfk/RStyjVqlpRrepUVXhYFZnNTK8DYM/pEV7DMDR37ly1a9dO7du3V2pqqp577jmNGjVK2dncFQsAKH0/H8yQJDWvGaxeLWqqdd1qTGMAUCCnA+/s2bO1Zs0avfbaa/L3v3jjwJ133qlvvvlGkyZNKvECAQC4XG7gbVWHm9MAFM3pwLty5UqNHTtWXbp0sU1j6NChgyZOnKhPP/20xAsEAOBySamnJBF4ATjG6cCblpamsLCwPNurVKmis2fPlkhRAAAUxJyZpQMnz8kkKbo2gRdA0ZwOvHFxcXrnnXfstmVmZmratGlq165diRUGAEB+fv57dLdRjcoKrlise68BlDNOB97Ro0dr586d6tChg7KysjR48GDdeOONSk1N1b/+9a/SqBEAABvm7wJwltP/Na5Zs6aWL1+u7777Tvv27VNOTo7q16+vjh07ysfH6fwMAIBTcgPvdQReAA4q9u+C2rdvr/bt25dkLQAAFCrj3AXtMZ+RJMUQeAE4yKHA27RpU4cP+Pvvvxe7GAAACpO7OsM1VwXqqkr+bq4GgLdwKPCGhIQoLS1N0dHR6tatm5o3b86T1QAALpc7nSEmgtFdAI5zKPB+/fXXSkpK0oYNG/TRRx8pKytLN910k7p27arY2Fjm7gIAXCIplRvWADjP4Tm8MTExiomJ0bPPPqu9e/dqw4YNmjp1qg4ePKjOnTura9eu6tixowICAkqzXgBAOXU226JdR09L4oY1AM4p1k1rkZGRioyM1OOPP66jR49q1apVGj58uKxWq37++eeSrhEAUM5ZrIZWbD8kiyFdVamCagQxuALAccVepSElJUUbN25UYmKifvrpJ9WvX1833XRTSdYGAIASk82amrhHxzKzJUknzl5Q7/lb9Ex8Q8U3CnVzdQC8gVOBNykpSYmJidq4caP279+v6667TjfddJPGjx+vunXrllaNAIByKjHZrBFrdubZfiwzWyPW7NTE3s0IvQCK5FDgffHFF7V582adPXtWHTt21KBBg3TjjTeqWrVqpVweAKC8slgNTU3cU2ibaZv26sbIEPn6sHIQgII5FHhXrFghPz8/NW/eXCdPntSKFSu0YsWKfNsuXLiwRAsEAJRPSakZtmkMBTl6OktJqRlqXbeaa4oC4JUcCrxDhw4t7ToAALBjLiLsOtsOQPlF4AUAeKTQIMeepOZoOwDlF0+MAAB4pJiIqgorIsyGBwfw1DUARSLwAgA8kq+PSc/ENyy0zbAukdywBqBIBF4AgMeKbxSq2lUq5tkeHhzAkmQAHFbsB08AAFDaUjPO6dCp8zJJmtS7qbJyDIUG+SsmoiojuwAcVqzAu2bNGr333ns6cOCAVq5cqYULF6pGjRoaNGhQSdcHACjHNu9JkyRdV7eqOjeq4eZqAHgrp6c0/Pvf/9akSZPUp08fXbhwQZLUokULvfPOO5o1a1aJFwgAKL82JZslSV0aMnUBQPE5HXg/+OADvfLKK3rwwQfl43Px7bfffrsmTZqkZcuWlXiBAIDyyXwmW9tTT0mSbmwY4uZqAHgzpwPvoUOHFBkZmWd73bp1lZ6eXhI1AQCgL/eYZUhqXjNYNfO5cQ0AHOV04I2OjtaqVavsthmGoXfffVfXXnttSdUFACjnNiVfnL/bmdFdAFfI6ZvW/vWvf2nQoEH64osvlJ2drTFjxmj//v06f/685s+fXxo1AgDKmdPnc/RjSrokqQtLjwG4Qk6P8EZFRenzzz/X/fffr4ceekgNGjTQwIED9fnnn6tp06ZOHevo0aNKSEhQbGysOnXqpAkTJigrK0uSlJKSogEDBigmJkY9e/bU119/bffeb7/9Vr169VJ0dLQeeughpaSk2O1/77331KlTJ7Vq1UovvPCCzp075+xHBQC4yVf70mSxGmoQUkn1rqrk7nIAeLliLUsWEBCge+6554pObBiGEhISVKVKFS1evFgZGRl64YUX5OPjo+HDh2vIkCGKiorSihUrtGHDBg0dOlTr169X7dq1dejQIQ0ZMkRPPvmkOnXqpNmzZ2vw4MFas2aNTCaTPv/8c82aNUuTJ09WSEiIRo4cqcmTJ2vUqFFXVDMAoHRZrIaSUjO05KdUSdysBqBkOB14mzRpIpMp/8W+K1SooBo1aqhHjx566qmnVKFChQKPs2/fPiUlJembb75RaOjFX1clJCRo4sSJuuGGG5SSkqIlS5aoUqVKioyM1HfffacVK1boySef1LJly9SiRQs98sgjkqQJEyaoQ4cO+uGHH9SuXTstXLhQDz/8sLp06SJJGjNmjAYOHKjnnntOgYGBzn5kAIALJCabNTVxj45lZtu2rf7liJqEB/NENQBXxOnAO3r0aM2aNUtPPvmkYmJiZBiGfv31V82cOVN33XWXoqKiNHv2bBmGoeeee67A49SoUUNvv/22LezmyszM1Pbt29WsWTNVqvS/X2O1bt1aSUlJkqTt27erTZs2tn2BgYFq3ry5kpKS1KZNG/3yyy8aOnSobX9MTIwuXLigXbt2qVWrVg5/1gJyfYnLPY+rzoeSRx96P/rQvRL/MGvEmp15tp84e0Ej1uzUpN7NFB9VeOilD70b/ef9XN2HzpzH6cD7zjvv6NVXX9UNN9xg29akSRPVqlVLY8eOVUJCgsLDw/Xkk08WGnirVKmiTp062V5brVYtWrRIcXFxOn78uMLCwuzah4SE6MiRI5JU6P5Tp04pKyvLbr+fn5+qVatme7+jQkKCnWp/pVx9PpQ8+tD70YeuZ7Eamr75hwL3myRN37xPd8Vd49DjhOlD70b/eT9P7EOnA6/ZbFbNmjXzbA8NDdXRo0clXRy9PXPmjFPHnTx5snbu3Knly5frvffek7+/v91+f39/ZWdf/DXXuXPnCtx//vx52+uC3u+otLTTMgyn3lIsJtPFbw5XnQ8ljz70fvSh+2w9kK7DGecL3G9IOpxxXv9NOqg2V1crsB196N3oP+/n6j7MPZ8jnA68HTp00NixYzVx4kRFRERIklJTUzV+/HjFxcXJYrFoxYoVioqKcviYkydP1vvvv6/p06crKipKAQEBeR5ikZ2drYoVLy48HhAQkCe8Zmdnq0qVKgoICLC9vny/s/N3DUMuvehcfT6UPPrQ+9GHrmfOdGwwwpyZ7VDf0Ifejf7zfp7Yh04vS/bKK6+oQoUKuummmxQXF6d27dqpa9euCggI0Lhx47R582Z9+OGHGjFihEPHGzdunBYsWKDJkyere/fukqTw8HCZzWa7dmaz2TZNoaD9NWrUULVq1RQQEGC3PycnR+np6apRo4azHxcAUMpCg/yLbuREOwC4nNMjvNWqVdOCBQv0559/6o8//pCvr68aNmyoa665RpJ0/fXX67vvvitwJYdLzZo1S0uWLNG0adN0yy232LZHR0dr3rx5On/+vG1Ud9u2bWrdurVt/7Zt22ztz507p507d2ro0KHy8fFRy5YttW3bNrVr106SlJSUJD8/PzVp0sTZjwsAKGUxEVUVFuRvtzrD5cKDAxQTUdWFVQEoS5we4ZUujphWqlRJLVu2VLNmzVShQgX9+eefWr9+vSpWrOhQ2N27d6/mzJmjxx57TK1bt9bx48dtf2JjY1WrVi2NHDlSycnJmjdvnnbs2KG7775bknTXXXfpp59+0rx585ScnKyRI0eqTp06toDbr18/vfPOO9qwYYN27Nih0aNH695772VJMgDwQL4+Jj0T37DQNsO6RDp0wxoA5MfpEd4NGzbopZdeyjPHVrp4s1rPnj0dOs7GjRtlsVg0d+5czZ07127f7t27NWfOHL344ovq06eP6tWrp9mzZ6t27dqSpDp16mjmzJl69dVXNXv2bLVq1UqzZ8+2Be1bb71VqampGjVqlLKzs9WtW7dCV4wAALhXl4YhqlHZX8fP2I/yhgcHaFiXSNbhBXBFTIbh3LTiHj16qG3bthowYID69u2refPmKT09XePGjdPgwYPVp0+f0qrV5cxm191lGBoa7LLzoeTRh96PPnSvHYdOaeCHSfL3NWnibc10Jtui0CB/xURUdXhklz70bvSf93N1H+aezxFOj/CmpKTorbfe0tVXX60WLVro+PHj6tq1q3x8fDRp0qQyFXgBAK6x9reL66R3bVxDHSN5nDCAkuX0HN4qVaro3LlzkqT69etr165dkqQGDRro4MGDJVsdAKDMO3/Bov/sOi5Juq153nXeAeBKOR14b7zxRo0ZM0Z79uxRu3bttHr1av32229aunRpnqefAQBQlE17zDqTbVHtKgG6ri4rMQAoeU4H3hdffFH16tXTr7/+qq5duyo6Olp33323Fi9e7PDauwAA5Prk14tP6ezVvKZ8HFjlBwCc5fQc3i+++ELDhw9X9erVJUlTpkzR6NGjFRAQoAoVKpR4gQCAssdiNZSUmqHk42f044F0SdKtzcPdWxSAMsvpwDtmzBgtXbrUFnglKSgoqESLAgCUXYnJZk1N3GP3oIkKvibtOpap2lUrurEyAGWV01Ma2rVrp7Vr1yo727FnnwMAkCsx2awRa3bmearaBYuhEWt2KjHZXMA7AaD4nB7hTUtL05w5c/Tmm2/qqquuUkBAgN3+jRs3llhxAICyw2I1NDVxT6Ftpm3aqxsjQ3iqGoAS5XTgvffee3XvvfeWRi0AgDIsKTUjz8ju5Y6ezlJSaoZa163mmqIAlAtOB94777zT9veMjAwFBwfLZDLZHusLAEB+zEWEXWfbAYCjnJ7DaxiG5s6dq3bt2ql9+/ZKTU3Vc889p1GjRjGvFwBQoNAg/xJtBwCOcjrwzp49W2vWrNFrr70mf/+L/yjdeeed+uabbzRp0qQSLxAAUDbERFRVWBFhNjw4QDERPHwCQMlyOvCuXLlSY8eOVZcuXWzTGDp06KCJEyfq008/LfECAQBlg6+PSc/ENyy0zbAukdywBqDEOR1409LS8n2EcJUqVXT27NkSKQoAUDbFNwpVVI3KebaHBwdoYu9mim8U6oaqAJR1Tt+0FhcXp3feeUdjx461bcvMzNS0adPUrl27Ei0OAFC2HDh5Tn8cPyNJGtezsUwyKTTIXzERVRnZBVBqnA68o0eP1tChQ9WhQwdlZWVp8ODBOnTokGrXrq25c+eWRo0AgDJiedIhSVLHBlfplqY8ShiAazgdeGvWrKnly5fru+++0759+5STk6P69eurY8eO8vFxeoYEAKCcOHfBok9+OyJJuiemtpurAVCeOB14X3rpJd16662Ki4tT+/btS6MmAEAZ9Onvx5SZZVHdahUVd011d5cDoBxxOvCePXtWQ4YMUWBgoLp3766ePXuqdevWpVEbAKAMsFgN/XwwXQu+PyBJ6hNdSz48rAiACzkdeKdOnars7Gx9/fXX+u9//6vBgwcrMDBQPXr0UM+ePdWyZcvSqBMA4IUSk82amrjH7pHC/96aqtpVA1mRAYDLmAzDMK7kANnZ2Xrvvff05ptv6ty5c/r9999Lqja3M5tP68q+Oo4xmaTQ0GCXnQ8ljz70fvRhyUtMNmvEmp0F7i/pZcjoQ+9G/3k/V/dh7vkc4fQIryRZLBZt2bJF//nPf7RhwwZZLBbddtttuvXWW4tzOABAGWOxGpqauKfQNtM27dWNkSEsRwag1DkdeJ9//nlt2rRJVqtVXbt21YQJE3T99dfL19dXx44dK40aAQBeJik1w24aQ36Ons5SUmqGWtet5pqiAJRbTgfe7OxsjR8/XjfccIP8/f2VnZ2tzz77TCtXrtR3332n3377rTTqBAB4EXMRYdfZdgBwJZwOvNOmTZMkbdu2TatWrdJnn32mzMxMRUZG6oUXXijxAgEA3ic0yL9E2wHAlXAq8KampmrVqlVavXq1UlJSVKVKFdtjhXv06FFaNQIAvExMRFWFBfkXOq0hPDhAMRFVXVgVgPLKocC7YsUKrVq1Slu3blVYWJji4+PVrVs3tW3bVtHR0WrUqFFp1wkA8CK+Pibd2jxcC7akFNhmWJdIblgD4BIOBd4XX3xR9erV08SJE9W7d+/SrgkAUAb8dvi0JKlSBV+dvWCxbQ8PDtCwLpGswwvAZRwKvK+++qrWrVunkSNHasKECercubO6du2qjh07lnZ9AAAvlHw8Uz8cSJevSVr80HU6ejpL5sxshQb5KyaiKiO7AFzKocDbp08f9enTRydOnNCnn36q9evXa+jQoapYsaKsVqu2bNmievXqqUKFCqVdLwDAC/x7W6okKT6qhupUC1SdaoFurghAeebjTOOrrrpKDzzwgBYvXqxNmzZpyJAhatq0qcaNG6dOnTppwoQJpVUnAMDDWayGtqWka3lSqtbvPCpJeqB1hJurAoBiPmlNkmrWrKlHH31Ujz76qPbv36+1a9dq/fr1GjlyZEnWBwDwAonJZk1N3GO3KkMFH5OOZmaruRvrAgDJyRHeglxzzTUaOnSo1q9fXxKHAwB4kcRks0as2ZlnCbILVkMj1uxUYrLZTZUBwEUlEngBAOWTxWpoauKeQttM27RXFqvhoooAIC8CLwCg2JJSMwp9uIQkHT2dpaTUDBdVBAB5EXgBAMVmLiLsOtsOAEoDgRcAUGyhQf4l2g4ASgOBFwBQbDERVRVWRJgNDw5QTERVF1UEAHkReAEAxebrY9Iz8Q0LbTOsSyRPVgPgVgReAMAV6dIwROH5jPKGBwdoYu9mim8U6oaqAOB/iv3gCQAAJOnXw6d1NDNb/r4mTbytmc5kWxQa5K+YiKqM7ALwCAReAMAVWf3LEUnSzU3C1DEyxM3VAEBeTGkAABTbmewc/Wf3MUnSHS1qurkaAMgfgRcAUGz/3XVc5y5YVa96oKIjqri7HADIF4EXAFBsq3+9OJ3h9pY1ZTIxXxeAZ2IOLwDAKRaroaTUDP16+LR+PXxaPibp1ubh7i4LAApE4AUAOCwx2aypiXt07JJHBVfw9VFS6imWHwPgsZjSAABwSGKyWSPW7LQLu5KUlWPViDU7lZhsdlNlAFA4Ai8AoEgWq6GpiXsKbTNt015ZrIaLKgIAxxF4AQBFSkrNyDOye7mjp7OUlJrhoooAwHEEXgBAkcxFhF1n2wGAKxF4AQBFCg3yL9F2AOBKBF4AQJFiIqoqrIgwGx4coJiIqi6qCAAcR+AFABTJ18ekZ+IbFtpmWJdI+frw8AkAnofACwBwSHyjUMXk8/jg8OAATezdjHV4AXgsHjwBAHDIqfMXtPPIaUnSyK4NVdnfT6FB/oqJqMrILgCPRuAFADjks9+PK9tiqFGNyrrz2loymQi5ALwDUxoAAA755NcjkqTeLWoSdgF4FQIvAKBIu49latexTFXwNemWpmHuLgcAnELgBQAUKXd098bIEFULrODmagDAOczhBQAUyGI19OOBk1rzd+C9tXm4mysCAOcReAEA+UpMNmtq4h4du+RxwRP+m6xn4g2WIAPgVZjSAADIIzHZrBFrdtqFXUk6lpmtEWt2KjHZ7KbKAMB5BF4AgB2L1dDUxD2Ftpm2aa8sVsNFFQHAlSHwAgDsJKVm5BnZvdzR01lKSs1wUUUAcGUIvAAAO+Yiwq6z7QDA3Qi8AAA7oUH+JdoOANyNwAsAsBMTUVVhRYTZ8OAAxURUdVFFAHBlCLwAADu+PibdEBlSaJthXSLl68PjhQF4B9bhBQDYsVgN/XAgXZIU5O+rzGyLbV94cICGdYlkHV4AXoXACwCw88Uesw6cPKcqFf20cmBbJR8/I3NmtkKD/BUTUZWRXQBeh8ALALAxDEPvbUmRJN0TU1tVKlZQ67rV3FsUAFwhAi8AQBaroaTUDH23/4R2HctUgK9J97eKcHdZAFAiPOKmtezsbPXq1Utbtmyxbdu6dav69OmjmJgY3X777fr222/t3rN27Vp17dpV0dHRGjJkiE6cOGHbZxiGpkyZori4OMXGxmrSpEmyWq0u+zwA4E0Sk83qPX+Lnvhoh97/4aAkycfHpJ94sASAMsLtgTcrK0vDhg1TcnKybVtaWpqeeOIJ9ezZU5988ol69OihwYMH68iRI5KkHTt26MUXX9TQoUO1dOlSnTp1SiNHjrS9f8GCBVq7dq1mzZqlGTNm6JNPPtGCBQtc/tkAwNMlJps1Ys3OPE9WO3fBqhFrdiox2eymygCg5Lg18O7Zs0f33nuvDhw4YLf9p59+kq+vrx599FHVrVtXTzzxhAICApSUlCRJWrRokXr06KE77rhDTZo00aRJk7R582alpFycd7Zw4UIlJCSoTZs2iouL07PPPqvFixe7+uMBgEezWA1NTdxTaJtpm/bKYjVcVBEAlA63Bt4ffvhB7dq109KlS+22V6tWTenp6frPf/4jwzC0YcMGnTlzRlFRUZKk7du3q02bNrb2tWrVUu3atbV9+3YdPXpUhw8fVtu2bW37W7durdTUVB07dsw1HwwAvEBSakaekd3LHT2dpSSmNgDwcm69aa1fv375bm/Tpo0eeOABJSQkyMfHRxaLRRMmTFCDBg0kSceOHVNYWJjde0JCQnTkyBEdP35ckuz2h4ZeXC/yyJEjed5XGJOLVt7JPY+rzoeSRx96v/LYh+YzhYfdS9t5w9elPPZhWUL/eT9X96Ez5/HIVRrOnDmjlJQUDR06VF26dNF//vMfvfLKK4qOjlZkZKTOnz8vf3/7x176+/srOztb58+ft72+dJ908eY4Z4SEBF/hJ3GOq8+Hkkcfer/y1IeRtR37NzGydjWFhnrP16U89WFZRP95P0/sQ48MvG+//bYMw9DQoUMlSc2bN9eOHTu0cOFCjRkzRgEBAXnCa3Z2tgIDA+3CbUBAgO3vkhQYGOhUHWlpp2W4YOqayXTxm8NV50PJow+9X3nswwbBFVQjyF/HC5nWEB4coAbBFWQ2n3ZhZcVTHvuwLKH/vJ+r+zD3fI7wyMD722+/qUmTJnbbmjZtalvJITw8XGaz/Z3DZrNZNWrUUHh4uCTp+PHjqlOnju3vklSjRg2n6jAMufSic/X5UPLoQ+9XnvrQx2RSw5BKhQbeYV0i5WMyedXXpDz1YVlE/3k/T+xDty9Llp+wsDDt2WN/5/C+fftsATY6Olrbtm2z7Tt8+LAOHz6s6OhohYeHq3bt2nb7t23bptq1azs1fxcAyiKL1dC2lHR9/vsxfbz9kL77K12SVC2wgl278OAATezdTPGNQt1QJQCULI8c4b3nnnvUr18/vffee7rpppu0ceNGff3111q5cqUkqW/fvurfv79iYmLUsmVLjR8/Xp07d1bdunVt+6dMmaKaNWtKkqZOnapHHnnEbZ8HADxBYrJZUxP35FmZISaiit68N1pJqRkyZ2YrNMhfMRFV5evD3UMAygaPDLwxMTGaOXOmZsyYoTfeeEP169fXvHnz1KhRI0lSq1atNHbsWM2YMUMZGRnq0KGDxo0bZ3v/wIEDlZaWpqFDh8rX11d33323BgwY4KZPAwDul/uAifwkpZ7S5r1pjOYCKLNMhuFpsyw8h9nsuknXoaHBLjsfSh596P3Kch9arIZ6z99S6Jq74cEBWv1orFeP6pblPiwP6D/v5+o+zD2fIzxyDi8AoOTwgAkA5R2BFwDKOHMRYdfZdgDgbQi8AFDGhQb5F93IiXYA4G0IvABQxsVEVFWNyoWH2fDgAMVEVHVRRQDgWh65SgMA4MpYrIZtmbGrKldQSOUKOn6m8AdMePMNawBQGAIvAJQxBa2362O6+ICJE2cv2LaFBwdoWJdIliQDUKYReAGgDClsvV2rIT13U0NVD6zAAyYAlCsEXgAoIyxWQ1MT9xTa5vUv9nn9ersA4CxuWgOAMoL1dgEgf4zwAoCXy71BbeMfZofas94ugPKGwAsAXqygG9QKw3q7AMobAi8AeKnCblArCOvtAiiPmMMLAF7IkRvU8sN6uwDKI0Z4AcALOXKD2qVYbxdAeUbgBQAvcenT0/48ccah99wTU1s3RYWy3i6Aco3ACwBeoDg3p0nSTVGhal23WukUBQBegsALAB6uODenSdygBgC5uGkNADxYcW9Ok7hBDQByMcILAB7M2ZvTJG5QA4DLEXgBwMMU5+a0R9rVVYOQygoN8ucGNQC4DIEXADxIcW9Oi61XnZvTAKAABF4A8BDcnAYApYOb1gDAA3BzGgCUHkZ4AcADcHMaAJQeAi8AeACzg2GXm9MAwHkEXgBwo9wVGfalObYaAzenAYDzCLwA4CbOrsjAzWkAUDwEXgBwg+KsyMDNaQBQPAReAHAxZ1dk4OY0ALgyBF4AcDFHV2QYGFdXba+uzs1pAHCFCLwA4ALFeVxw/asqc4MaAJQAAi8AlLLiPi44NMi/lCoCgPKFwAsApYjHBQOA+/FoYQAoJTwuGAA8AyO8AFCCLp2rm3Y2m8cFA4AHIPACQAkp7lxdHhcMAKWLwAsAJaC4c3UlHhcMAKWNObwAcIWuZK4uN6cBQOljhBcAiil3vu4Pf510ehpDLm5OA4DSR+AFgGIo7nzdXNycBgCuQ+AFACcVd77u050bKKSSPzenAYCLEXgBoAiXLjV2VaUKxZqvGx4coPtaRRByAcANCLwAUIgrnbqQi7m6AOA+BF4A+JvFamjrgXSZM7MVGuSvk2cvaOTa36/omMzVBQD3I/ACgKTEP8yavvkHHc44b9t2JQOyA+Pqqu3V1ZmrCwAegMALoNwr6CY0q1G844UHB+ix9tcQdAHAQxB4AZQ7JXETWmGYrwsAnoXAC6BcKamb0PLDfF0A8EwEXgBl2qWjuSnp5/TWt3+V2LHDgvz18i2NdfLsBdbWBQAPRuAFUGaV5miuJD0T31Cx9aqXyrEBACWHwAugTCru09Dy42Oyv4GNqQsA4F0IvADKhNK8EW18r6aqHljBtj4vUxcAwLsQeAF4vdKausBILgCUDQReN8sdlco6eEoBVqtt5OjS0SpGlID/ufzaKImnoeWqWSVAo7pH6cQZbkIDgLKEwOtG+Y1KhQX5q3uTMH2+61ie7c/EN9SNkSH5BuH8ArIkh9s6ewzAHfK7Zkry23F07+ZqUzNIRjEfOAEA8EwEXjcp6IaaY5nZ+mDrwXy3j1izU1Ur+injfI5te0EBuUpFP5kkh9o6e4ySCt5AURxZUqy4T0O7VHhwgJ7pEqlbWtSS2Xz6yg8IAPAoJsNgLKMgZvPpUhnpsVgN9Z6/pdSWSnKFKw3eZS00m0xSaGhwqX3PlEelvaTYoOvr6epqgbbvJz9fE33o5bgOvRv95/1c3Ye553MEI7xukJSa4dVhV7IPsFLBI9OnLmuX27YkRqudDc0lNfUDJa805+VejhvRAKD8IfC6gdnLw25JcWVoLqmpHwWFbJNJ+m5vmvYeSldoZUKzM0p7Xi5PQwMAMKWhEKU1JL8tJV1PfLSj5A8Ml7g8ZJdUaJau/CZDTw9yrhzJzTWxdzOHRnP5dar3ow+9G/3n/ZjSADsxEVUVFuTv9dMayqvLR6YLGoF25ubDkrjJMCzIX093icz3AQnOhubSWPXj5NkLmv7F3lIbyeVpaACAgjDCW4jS/B9KST72FChIcUJzaa36Udom3HZlT0NjdMn70Yfejf7zfp48wkvgLURpd1h+cxfDgwPUrXGNPGHh8hFBABeV1EguP2y9H33o3eg/7+fJgZcpDW4U3yjUNpczy8fH7klrQzrVz/Nr4s1705wKyJL9iFxBbZ05BsEbnuDyJcU8fe4yAMC9GOEthKv/h+LI+Vw557KgY1xp8CY0w1GunJfL6JL3ow+9G/3n/Tx5hJfAWwhPDLyegtAMV7jSebnO8MbrEPboQ+9G/3k/Aq+XIvCWLleG5pKY+kHILj2esMJCeb0OyxL60LvRf96PwOulCLyez9XLbeUXsgnNV86VI7kF4Tr0fvShd6P/vB+B10sReJGf/IKwySTtO33BoSetXWloLmi7N/KktXK5Dr0ffejd6D/vR+D1UgReOMrZPnTVgx2cDc2luepHeHCA/q9zA7eP5BaE69D70Yfejf7zfgReL0XghaM8pQ9LYopHaa764SnhNj+e0ocoPvrQu9F/3s+TAy/r8AJliK+PSa3rViu17SVxDAAAXM3H3QUAAAAApckjAm92drZ69eqlLVu22LYdOnRIjz32mKKjo3XzzTdr/fr1du9Zu3atunbtqujoaA0ZMkQnTpyw7TMMQ1OmTFFcXJxiY2M1adIkWa1Wl30eAAAAeA63B96srCwNGzZMycnJtm05OTl6/PHH5efnp5UrV2rgwIEaPny4/vjjD0nSjh079OKLL2ro0KFaunSpTp06pZEjR9rev2DBAq1du1azZs3SjBkz9Mknn2jBggUu/2wAAABwP7fO4d2zZ4+eeeYZXX7f3ObNm3X48GF9+OGHCgoKUoMGDfTll1/q559/VlRUlBYtWqQePXrojjvukCRNmjRJXbp0UUpKiurWrauFCxcqISFBbdq0kSQ9++yzeuONNzRw4EBXf0QAAAC4mVtHeH/44Qe1a9dOS5cuzbO9ffv2CgoKsm2bM2eO7rvvPknS9u3bbWFWkmrVqqXatWtr+/btOnr0qA4fPqy2bdva9rdu3Vqpqak6duxYKX8iAAAAeBq3jvD269cv3+0pKSmKiIjQlClTtHr1alWvXl0JCQnq2rWrJOnYsWMKCwuze09ISIiOHDmi48ePS5Ld/tDQi4vaHzlyJM/7CmNy0QpKuedx1flQ8uhD70cfej/60LvRf97P1X3ozHk8clmys2fPauXKlerZs6fefPNNbdmyRQkJCVq6dKlatmyp8+fPy9/f3+49/v7+ys7O1vnz522vL90nXbw5zhkhIY6t7VZSXH0+lDz60PvRh96PPvRu9J/388Q+9MjA6+vrq2rVqmn06NHy8fFR8+bNtXXrVn300Udq2bKlAgIC8oTX7OxsBQYG2oXbgIAA298lKTAw0Kk60tJct3BySEiwy86Hkkcfej/60PvRh96N/vN+ru7D3PM5wiMDb1hYmEwmk3x8/jfFuH79+tq9e7ckKTw8XGaz2e49ZrNZNWrUUHh4uCTp+PHjqlOnju3vklSjRg2n6jAMufSic/X5UPLoQ+9HH3o/+tC70X/ezxP70O3LkuUnOjpaycnJslgstm179+5VRESEbf+2bdts+w4fPqzDhw8rOjpa4eHhql27tt3+bdu2qXbt2k7N3wUAAEDZ4JGBt1evXrJarRozZoz++usvLV68WF999ZXuvfdeSVLfvn21evVqLVu2TLt27dLw4cPVuXNn1a1b17Z/ypQp2rJli7Zs2aKpU6fqoYcecudHAgAAgJt45JSGoKAgLViwQKNHj1avXr1Uu3ZtTZ8+Xc2bN5cktWrVSmPHjtWMGTOUkZGhDh06aNy4cbb3Dxw4UGlpaRo6dKh8fX119913a8CAAW76NAAAAHAnk3H5Ux9gYza7btJ1aGiwy86Hkkcfej/60PvRh96N/vN+ru7D3PM5wiNHeD0F6/DCUfSh96MPvR996N3oP+/nyevwMsILAACAMs0jb1oDAAAASgqBFwAAAGUagRcAAABlGoEXAAAAZRqBFwAAAGUagRcAAABlGoEXAAAAZRqBFwAAAGUagRcAAABlGoEXAAAAZRqB10tkZWXpqaee0oMPPqh7771XSUlJ7i4JV2DDhg0aOXKku8uAA6xWq0aOHKn7779fAwcOVFpamrtLQjFx3Xknfv55vwsXLmjYsGHq16+fBg4cqBMnTri8BgKvl1i+fLkaNGigRYsW6bXXXtOECRPcXRKKaeLEiZoyZYoMw3B3KXDAf//7X1WsWFFLlizRXXfdpbfeesvdJaEYuO68Fz//vN/69esVHh6uf//737r11ls1f/58l9fg5/Izolhuv/12mUwmSZLFYlGFChXcXBGK69prr9WNN96oVatWubsUOOCnn35Shw4dJEmdOnXSvHnz3FwRioPrznvx88/73X777br11lslSUeOHFHVqlVdXgOB18MsXbpUH3zwgd22d955R+Hh4ZKkEydOaPjw4Ro+fLg7yoODCuvHHj16aMuWLW6qDM7KzMxUUFCQJKly5co6c+aMmytCcXDdea/c64+ff97Nz89PgwYN0i+//KIFCxa4/vwuPyMKdd999+m+++7Ld9+ff/6phIQEPf3002rfvr2LK4MzCutHeJegoCBbyD1z5oyCg4PdXBFQ/vDzr2yYN2+e/vrrLw0aNEiff/65S8/NHF4vcfjwYf3zn//U+PHjFR8f7+5ygHIjJiZG33zzjSTpyy+/VKtWrdxcEVC+8PPP+y1dulSLFi2SJFWqVEk+Pq6PnwTeUpadna1evXrZ/SotKytLL7zwgtq0aaOOHTvq3XffLfI4c+bM0dmzZzV58mT1799fCQkJpVk2LlNS/Qj3c7Yvu3XrpnPnzun+++/XkiVL9MQTT7ijbFyC69G7Odt//PzzPM72YY8ePfT111/rwQcf1FNPPaVx48a5vGamNJSirKwsPfPMM0pOTrbbPmnSJP366696//33dejQIY0YMUK1a9fWLbfcUuCx3PHNgYtKsh9ztWvXTu3atSutklGA4vSlr68vd4V7kCu5Hrnu3K84/cfPP89SnD6sUqWK3nzzTTdVfBGBt5Ts2bNHzzzzTJ4lcM6ePatly5Zp/vz5at68uZo3b67k5GQtXrzYoaAE16Ifyw760vvRh96N/vN+3tyHTGkoJT/88IPatWunpUuX2m3ftWuXcnJy7OYBtm7dWtu3b5fVanV1mSgC/Vh20Jfejz70bvSf9/PmPmSEt5T069cv3+3Hjx9X9erV5e/vb9sWGhqqrKwspaen66qrrnJViXAA/Vh20Jfejz70bvSf9/PmPmSE18XOnTtn9w0hyfY6OzvbHSWhGOjHsoO+9H70oXej/7yfN/QhgdfFAgIC8nR+7uuKFSu6oyQUA/1YdtCX3o8+9G70n/fzhj4k8LpYeHi4Tp48qZycHNu248ePq2LFiqpSpYobK4Mz6Meyg770fvShd6P/vJ839CGB18WaNm0qPz8/JSUl2bZt27ZNLVu2dMtCzCge+rHsoC+9H33o3eg/7+cNfegZVZQjgYGBuuOOOzR69Gjt2LFDGzZs0LvvvquHHnrI3aXBCfRj2UFfej/60LvRf97PG/qQVRrcYOTIkRo9erQefvhhBQUF6cknn1S3bt3cXRacRD+WHfSl96MPvRv95/08vQ9NxuWrBwMAAABlCFMaAAAAUKYReAEAAFCmEXgBAABQphF4AQAAUKYReAEAAFCmEXgBAABQphF4AQAAUKYReAEAAFCmEXgBAABQphF4AcALffzxx2rcuLGWLVtmtz0+Pl4ff/xxke8/ePCgGjdurIMHD5ZWiQDgMQi8AOCF1q1bp6uvvlqrV692dykA4PEIvADgZdLS0vTdd99pyJAh2rp1q1JSUtxdEgB4NAIvAHiZzz77TMHBwerdu7fCwsIKHOXt37+/Zs2apb59+yo6Olr9+vXT3r177dps2LBBXbt2VXR0tJ544gllZGTY9i1btky33HKLWrRooXbt2mnMmDGyWCyl+tkAoDQQeAHAy6xbt06dO3eWj4+P4uPjtWrVKhmGkW/bt956S927d9fHH3+s8PBwDRo0SNnZ2bb9K1eu1LRp07Rw4UL99ttvmj9/viTphx9+0CuvvKJhw4bps88+05gxY7R8+XJt3LjRJZ8RAEoSgRcAvMjhw4f1008/qWvXrpKkbt26KSUlRdu2bcu3/Q033KABAwYoMjJS48aN04kTJ/TNN9/Y9j/33HO69tprFR0drR49emjXrl2SpEqVKmn8+PHq1q2b6tSpo1tuuUXNmjVTcnJy6X9IAChhfu4uAADguHXr1ikgIEAdO3aUJMXGxqpq1apauXKl2rRpk6f9ddddZ/t7UFCQ6tevr71796pRo0aSpKuvvtq2Pzg4WFlZWZKkFi1aqGLFipoxY4b27Nmj3bt366+//rKdFwC8CSO8AOBF1q1bp/Pnz6t169Zq1qyZrr32WmVkZOizzz7T+fPn87T387Mf17BYLPLx+d8//Zf+/VJfffWV+vTpI7PZrE6dOmnGjBl24RkAvAkjvADgJf7880/t3LlT//rXv9SuXTvb9j179ujpp5/Wf//73zzvyZ2iIEmnT5/WgQMH1Lhx4yLPtWzZMt111116+eWXJUk5OTk6cOCA4uLiSuCTAIBrEXgBwEusW7dO1apV03333Sd/f3/b9qioKM2ePVurVq3K855PPvlE7dq1U8uWLfXGG2+odu3aateunY4cOVLouapVq6aff/5Zu3fvlo+Pj9566y0dP37c7oY3APAWTGkAAC+xbt063XbbbXZhN1ffvn317bff6ujRo3bbb7vtNi1ZskR9+vTRmTNnNH/+/DzTHPIzdOhQhYSE6L777tM//vEPBQQEqG/fvvr9999L7PMAgKuYjILWsgEAeLX+/fsrNjZWTz75pLtLAQC3YoQXAAAAZRqBFwAAAGUaUxoAAABQpjHCCwAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDKNwAsAAIAyjcALAACAMo3ACwAAgDKNwAsAAIAy7f8Bx5Rb5zQ60yAAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Lasso\n",
    "# Plotting mean test scores for each alpha value\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.semilogx(lambda_range, results['mean_test_score'], marker='o')\n",
    "plt.xlabel('Alpha')\n",
    "plt.ylabel('Average Mean Squared Error')\n",
    "plt.title('Average Mean Squared Error for each Alpha')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### (d) Choosing the optimal value of lambda."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='blue'> The optimal value of lambda for ridge is 247.71, which gives the best MSE possible of 1594.35. The optimal value of lambda for lasso is 0.327, which gives the best MSE possible of 1629.5. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_1122c_row0_col1 {\n",
       "  background-color: yellow;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_1122c\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_1122c_level0_col0\" class=\"col_heading level0 col0\" >alpha</th>\n",
       "      <th id=\"T_1122c_level0_col1\" class=\"col_heading level0 col1\" >MSE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_1122c_level0_row0\" class=\"row_heading level0 row0\" >87</th>\n",
       "      <td id=\"T_1122c_row0_col0\" class=\"data row0 col0\" >247.707636</td>\n",
       "      <td id=\"T_1122c_row0_col1\" class=\"data row0 col1\" >1594.345332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1122c_level0_row1\" class=\"row_heading level0 row1\" >86</th>\n",
       "      <td id=\"T_1122c_row1_col0\" class=\"data row1 col0\" >220.513074</td>\n",
       "      <td id=\"T_1122c_row1_col1\" class=\"data row1 col1\" >1594.375914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1122c_level0_row2\" class=\"row_heading level0 row2\" >88</th>\n",
       "      <td id=\"T_1122c_row2_col0\" class=\"data row2 col0\" >278.255940</td>\n",
       "      <td id=\"T_1122c_row2_col1\" class=\"data row2 col1\" >1594.903311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1122c_level0_row3\" class=\"row_heading level0 row3\" >85</th>\n",
       "      <td id=\"T_1122c_row3_col0\" class=\"data row3 col0\" >196.304065</td>\n",
       "      <td id=\"T_1122c_row3_col1\" class=\"data row3 col1\" >1594.907631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1122c_level0_row4\" class=\"row_heading level0 row4\" >84</th>\n",
       "      <td id=\"T_1122c_row4_col0\" class=\"data row4 col0\" >174.752840</td>\n",
       "      <td id=\"T_1122c_row4_col1\" class=\"data row4 col1\" >1595.858304</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x108e7335bd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "top_5_results_cv_ridge = results_cv_ridge.sort_values(by='MSE',\n",
    "                                                      ascending=True).head(5)\n",
    "\n",
    "styled_top_5 = top_5_results_cv_ridge.style.apply(highlight_min, subset=['MSE'])\n",
    "display(styled_top_5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best alpha: 247.70763559917089\n",
      "Best MSE: 1594.3453320838667\n"
     ]
    }
   ],
   "source": [
    "# Getting the best alpha value and corresponding MSE\n",
    "best_alpha = grid_search_ridge.best_params_['alpha']\n",
    "best_mse = -grid_search_ridge.best_score_\n",
    "print(f\"Best alpha: {best_alpha}\")\n",
    "print(f\"Best MSE: {best_mse}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_e139d_row30_col1 {\n",
       "  background-color: yellow;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_e139d\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_e139d_level0_col0\" class=\"col_heading level0 col0\" >params</th>\n",
       "      <th id=\"T_e139d_level0_col1\" class=\"col_heading level0 col1\" >mean_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_e139d_row0_col0\" class=\"data row0 col0\" >{'lasso__alpha': 0.01}</td>\n",
       "      <td id=\"T_e139d_row0_col1\" class=\"data row0 col1\" >1645.888398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_e139d_row1_col0\" class=\"data row1 col0\" >{'lasso__alpha': 0.011233240329780276}</td>\n",
       "      <td id=\"T_e139d_row1_col1\" class=\"data row1 col1\" >1645.653939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_e139d_row2_col0\" class=\"data row2 col0\" >{'lasso__alpha': 0.012618568830660204}</td>\n",
       "      <td id=\"T_e139d_row2_col1\" class=\"data row2 col1\" >1645.392368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_e139d_row3_col0\" class=\"data row3 col0\" >{'lasso__alpha': 0.014174741629268055}</td>\n",
       "      <td id=\"T_e139d_row3_col1\" class=\"data row3 col1\" >1645.111803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_e139d_row4_col0\" class=\"data row4 col0\" >{'lasso__alpha': 0.015922827933410922}</td>\n",
       "      <td id=\"T_e139d_row4_col1\" class=\"data row4 col1\" >1644.806451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_e139d_row5_col0\" class=\"data row5 col0\" >{'lasso__alpha': 0.01788649529057435}</td>\n",
       "      <td id=\"T_e139d_row5_col1\" class=\"data row5 col1\" >1644.467543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_e139d_row6_col0\" class=\"data row6 col0\" >{'lasso__alpha': 0.02009233002565047}</td>\n",
       "      <td id=\"T_e139d_row6_col1\" class=\"data row6 col1\" >1643.989950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_e139d_row7_col0\" class=\"data row7 col0\" >{'lasso__alpha': 0.022570197196339202}</td>\n",
       "      <td id=\"T_e139d_row7_col1\" class=\"data row7 col1\" >1643.437662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_e139d_row8_col0\" class=\"data row8 col0\" >{'lasso__alpha': 0.025353644939701114}</td>\n",
       "      <td id=\"T_e139d_row8_col1\" class=\"data row8 col1\" >1642.804946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_e139d_row9_col0\" class=\"data row9 col0\" >{'lasso__alpha': 0.02848035868435802}</td>\n",
       "      <td id=\"T_e139d_row9_col1\" class=\"data row9 col1\" >1642.089233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_e139d_row10_col0\" class=\"data row10 col0\" >{'lasso__alpha': 0.03199267137797385}</td>\n",
       "      <td id=\"T_e139d_row10_col1\" class=\"data row10 col1\" >1641.323644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_e139d_row11_col0\" class=\"data row11 col0\" >{'lasso__alpha': 0.03593813663804628}</td>\n",
       "      <td id=\"T_e139d_row11_col1\" class=\"data row11 col1\" >1640.569649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_e139d_row12_col0\" class=\"data row12 col0\" >{'lasso__alpha': 0.040370172585965536}</td>\n",
       "      <td id=\"T_e139d_row12_col1\" class=\"data row12 col1\" >1639.831433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_e139d_row13_col0\" class=\"data row13 col0\" >{'lasso__alpha': 0.04534878508128582}</td>\n",
       "      <td id=\"T_e139d_row13_col1\" class=\"data row13 col1\" >1639.066257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_e139d_row14_col0\" class=\"data row14 col0\" >{'lasso__alpha': 0.0509413801481638}</td>\n",
       "      <td id=\"T_e139d_row14_col1\" class=\"data row14 col1\" >1638.336928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_e139d_row15_col0\" class=\"data row15 col0\" >{'lasso__alpha': 0.05722367659350217}</td>\n",
       "      <td id=\"T_e139d_row15_col1\" class=\"data row15 col1\" >1637.694318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_e139d_row16_col0\" class=\"data row16 col0\" >{'lasso__alpha': 0.06428073117284319}</td>\n",
       "      <td id=\"T_e139d_row16_col1\" class=\"data row16 col1\" >1637.155946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_e139d_row17_col0\" class=\"data row17 col0\" >{'lasso__alpha': 0.07220809018385464}</td>\n",
       "      <td id=\"T_e139d_row17_col1\" class=\"data row17 col1\" >1636.692634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "      <td id=\"T_e139d_row18_col0\" class=\"data row18 col0\" >{'lasso__alpha': 0.08111308307896872}</td>\n",
       "      <td id=\"T_e139d_row18_col1\" class=\"data row18 col1\" >1636.293197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "      <td id=\"T_e139d_row19_col0\" class=\"data row19 col0\" >{'lasso__alpha': 0.09111627561154892}</td>\n",
       "      <td id=\"T_e139d_row19_col1\" class=\"data row19 col1\" >1635.633694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "      <td id=\"T_e139d_row20_col0\" class=\"data row20 col0\" >{'lasso__alpha': 0.10235310218990264}</td>\n",
       "      <td id=\"T_e139d_row20_col1\" class=\"data row20 col1\" >1634.748728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "      <td id=\"T_e139d_row21_col0\" class=\"data row21 col0\" >{'lasso__alpha': 0.11497569953977356}</td>\n",
       "      <td id=\"T_e139d_row21_col1\" class=\"data row21 col1\" >1633.842948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "      <td id=\"T_e139d_row22_col0\" class=\"data row22 col0\" >{'lasso__alpha': 0.1291549665014884}</td>\n",
       "      <td id=\"T_e139d_row22_col1\" class=\"data row22 col1\" >1633.421563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row23\" class=\"row_heading level0 row23\" >23</th>\n",
       "      <td id=\"T_e139d_row23_col0\" class=\"data row23 col0\" >{'lasso__alpha': 0.14508287784959395}</td>\n",
       "      <td id=\"T_e139d_row23_col1\" class=\"data row23 col1\" >1632.814226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row24\" class=\"row_heading level0 row24\" >24</th>\n",
       "      <td id=\"T_e139d_row24_col0\" class=\"data row24 col0\" >{'lasso__alpha': 0.16297508346206444}</td>\n",
       "      <td id=\"T_e139d_row24_col1\" class=\"data row24 col1\" >1632.263568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row25\" class=\"row_heading level0 row25\" >25</th>\n",
       "      <td id=\"T_e139d_row25_col0\" class=\"data row25 col0\" >{'lasso__alpha': 0.18307382802953678}</td>\n",
       "      <td id=\"T_e139d_row25_col1\" class=\"data row25 col1\" >1631.823397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row26\" class=\"row_heading level0 row26\" >26</th>\n",
       "      <td id=\"T_e139d_row26_col0\" class=\"data row26 col0\" >{'lasso__alpha': 0.20565123083486514}</td>\n",
       "      <td id=\"T_e139d_row26_col1\" class=\"data row26 col1\" >1631.388370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row27\" class=\"row_heading level0 row27\" >27</th>\n",
       "      <td id=\"T_e139d_row27_col0\" class=\"data row27 col0\" >{'lasso__alpha': 0.23101297000831592}</td>\n",
       "      <td id=\"T_e139d_row27_col1\" class=\"data row27 col1\" >1630.840238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row28\" class=\"row_heading level0 row28\" >28</th>\n",
       "      <td id=\"T_e139d_row28_col0\" class=\"data row28 col0\" >{'lasso__alpha': 0.25950242113997357}</td>\n",
       "      <td id=\"T_e139d_row28_col1\" class=\"data row28 col1\" >1630.418515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row29\" class=\"row_heading level0 row29\" >29</th>\n",
       "      <td id=\"T_e139d_row29_col0\" class=\"data row29 col0\" >{'lasso__alpha': 0.2915053062825176}</td>\n",
       "      <td id=\"T_e139d_row29_col1\" class=\"data row29 col1\" >1629.913758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row30\" class=\"row_heading level0 row30\" >30</th>\n",
       "      <td id=\"T_e139d_row30_col0\" class=\"data row30 col0\" >{'lasso__alpha': 0.32745491628777285}</td>\n",
       "      <td id=\"T_e139d_row30_col1\" class=\"data row30 col1\" >1629.499994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row31\" class=\"row_heading level0 row31\" >31</th>\n",
       "      <td id=\"T_e139d_row31_col0\" class=\"data row31 col0\" >{'lasso__alpha': 0.36783797718286326}</td>\n",
       "      <td id=\"T_e139d_row31_col1\" class=\"data row31 col1\" >1629.817696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row32\" class=\"row_heading level0 row32\" >32</th>\n",
       "      <td id=\"T_e139d_row32_col0\" class=\"data row32 col0\" >{'lasso__alpha': 0.41320124001153363}</td>\n",
       "      <td id=\"T_e139d_row32_col1\" class=\"data row32 col1\" >1631.029486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row33\" class=\"row_heading level0 row33\" >33</th>\n",
       "      <td id=\"T_e139d_row33_col0\" class=\"data row33 col0\" >{'lasso__alpha': 0.464158883361278}</td>\n",
       "      <td id=\"T_e139d_row33_col1\" class=\"data row33 col1\" >1633.406269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row34\" class=\"row_heading level0 row34\" >34</th>\n",
       "      <td id=\"T_e139d_row34_col0\" class=\"data row34 col0\" >{'lasso__alpha': 0.5214008287999684}</td>\n",
       "      <td id=\"T_e139d_row34_col1\" class=\"data row34 col1\" >1636.964136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row35\" class=\"row_heading level0 row35\" >35</th>\n",
       "      <td id=\"T_e139d_row35_col0\" class=\"data row35 col0\" >{'lasso__alpha': 0.5857020818056667}</td>\n",
       "      <td id=\"T_e139d_row35_col1\" class=\"data row35 col1\" >1641.189689</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row36\" class=\"row_heading level0 row36\" >36</th>\n",
       "      <td id=\"T_e139d_row36_col0\" class=\"data row36 col0\" >{'lasso__alpha': 0.6579332246575679}</td>\n",
       "      <td id=\"T_e139d_row36_col1\" class=\"data row36 col1\" >1646.127881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row37\" class=\"row_heading level0 row37\" >37</th>\n",
       "      <td id=\"T_e139d_row37_col0\" class=\"data row37 col0\" >{'lasso__alpha': 0.739072203352578}</td>\n",
       "      <td id=\"T_e139d_row37_col1\" class=\"data row37 col1\" >1652.298549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row38\" class=\"row_heading level0 row38\" >38</th>\n",
       "      <td id=\"T_e139d_row38_col0\" class=\"data row38 col0\" >{'lasso__alpha': 0.8302175681319743}</td>\n",
       "      <td id=\"T_e139d_row38_col1\" class=\"data row38 col1\" >1658.796294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row39\" class=\"row_heading level0 row39\" >39</th>\n",
       "      <td id=\"T_e139d_row39_col0\" class=\"data row39 col0\" >{'lasso__alpha': 0.9326033468832199}</td>\n",
       "      <td id=\"T_e139d_row39_col1\" class=\"data row39 col1\" >1664.207987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row40\" class=\"row_heading level0 row40\" >40</th>\n",
       "      <td id=\"T_e139d_row40_col0\" class=\"data row40 col0\" >{'lasso__alpha': 1.0476157527896652}</td>\n",
       "      <td id=\"T_e139d_row40_col1\" class=\"data row40 col1\" >1669.507154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row41\" class=\"row_heading level0 row41\" >41</th>\n",
       "      <td id=\"T_e139d_row41_col0\" class=\"data row41 col0\" >{'lasso__alpha': 1.1768119524349978}</td>\n",
       "      <td id=\"T_e139d_row41_col1\" class=\"data row41 col1\" >1675.574236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row42\" class=\"row_heading level0 row42\" >42</th>\n",
       "      <td id=\"T_e139d_row42_col0\" class=\"data row42 col0\" >{'lasso__alpha': 1.3219411484660286}</td>\n",
       "      <td id=\"T_e139d_row42_col1\" class=\"data row42 col1\" >1682.388246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row43\" class=\"row_heading level0 row43\" >43</th>\n",
       "      <td id=\"T_e139d_row43_col0\" class=\"data row43 col0\" >{'lasso__alpha': 1.484968262254465}</td>\n",
       "      <td id=\"T_e139d_row43_col1\" class=\"data row43 col1\" >1691.055879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row44\" class=\"row_heading level0 row44\" >44</th>\n",
       "      <td id=\"T_e139d_row44_col0\" class=\"data row44 col0\" >{'lasso__alpha': 1.6681005372000592}</td>\n",
       "      <td id=\"T_e139d_row44_col1\" class=\"data row44 col1\" >1701.654503</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row45\" class=\"row_heading level0 row45\" >45</th>\n",
       "      <td id=\"T_e139d_row45_col0\" class=\"data row45 col0\" >{'lasso__alpha': 1.873817422860383}</td>\n",
       "      <td id=\"T_e139d_row45_col1\" class=\"data row45 col1\" >1713.362782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row46\" class=\"row_heading level0 row46\" >46</th>\n",
       "      <td id=\"T_e139d_row46_col0\" class=\"data row46 col0\" >{'lasso__alpha': 2.1049041445120196}</td>\n",
       "      <td id=\"T_e139d_row46_col1\" class=\"data row46 col1\" >1725.863300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row47\" class=\"row_heading level0 row47\" >47</th>\n",
       "      <td id=\"T_e139d_row47_col0\" class=\"data row47 col0\" >{'lasso__alpha': 2.364489412645407}</td>\n",
       "      <td id=\"T_e139d_row47_col1\" class=\"data row47 col1\" >1737.577192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row48\" class=\"row_heading level0 row48\" >48</th>\n",
       "      <td id=\"T_e139d_row48_col0\" class=\"data row48 col0\" >{'lasso__alpha': 2.656087782946687}</td>\n",
       "      <td id=\"T_e139d_row48_col1\" class=\"data row48 col1\" >1748.752743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row49\" class=\"row_heading level0 row49\" >49</th>\n",
       "      <td id=\"T_e139d_row49_col0\" class=\"data row49 col0\" >{'lasso__alpha': 2.9836472402833403}</td>\n",
       "      <td id=\"T_e139d_row49_col1\" class=\"data row49 col1\" >1762.585946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row50\" class=\"row_heading level0 row50\" >50</th>\n",
       "      <td id=\"T_e139d_row50_col0\" class=\"data row50 col0\" >{'lasso__alpha': 3.351602650938841}</td>\n",
       "      <td id=\"T_e139d_row50_col1\" class=\"data row50 col1\" >1779.488218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row51\" class=\"row_heading level0 row51\" >51</th>\n",
       "      <td id=\"T_e139d_row51_col0\" class=\"data row51 col0\" >{'lasso__alpha': 3.7649358067924674}</td>\n",
       "      <td id=\"T_e139d_row51_col1\" class=\"data row51 col1\" >1801.670159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row52\" class=\"row_heading level0 row52\" >52</th>\n",
       "      <td id=\"T_e139d_row52_col0\" class=\"data row52 col0\" >{'lasso__alpha': 4.229242874389499}</td>\n",
       "      <td id=\"T_e139d_row52_col1\" class=\"data row52 col1\" >1829.894732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row53\" class=\"row_heading level0 row53\" >53</th>\n",
       "      <td id=\"T_e139d_row53_col0\" class=\"data row53 col0\" >{'lasso__alpha': 4.750810162102798}</td>\n",
       "      <td id=\"T_e139d_row53_col1\" class=\"data row53 col1\" >1864.721882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row54\" class=\"row_heading level0 row54\" >54</th>\n",
       "      <td id=\"T_e139d_row54_col0\" class=\"data row54 col0\" >{'lasso__alpha': 5.336699231206307}</td>\n",
       "      <td id=\"T_e139d_row54_col1\" class=\"data row54 col1\" >1898.908576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row55\" class=\"row_heading level0 row55\" >55</th>\n",
       "      <td id=\"T_e139d_row55_col0\" class=\"data row55 col0\" >{'lasso__alpha': 5.994842503189409}</td>\n",
       "      <td id=\"T_e139d_row55_col1\" class=\"data row55 col1\" >1928.833179</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row56\" class=\"row_heading level0 row56\" >56</th>\n",
       "      <td id=\"T_e139d_row56_col0\" class=\"data row56 col0\" >{'lasso__alpha': 6.7341506577508214}</td>\n",
       "      <td id=\"T_e139d_row56_col1\" class=\"data row56 col1\" >1960.618251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row57\" class=\"row_heading level0 row57\" >57</th>\n",
       "      <td id=\"T_e139d_row57_col0\" class=\"data row57 col0\" >{'lasso__alpha': 7.56463327554629}</td>\n",
       "      <td id=\"T_e139d_row57_col1\" class=\"data row57 col1\" >1995.857268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row58\" class=\"row_heading level0 row58\" >58</th>\n",
       "      <td id=\"T_e139d_row58_col0\" class=\"data row58 col0\" >{'lasso__alpha': 8.497534359086439}</td>\n",
       "      <td id=\"T_e139d_row58_col1\" class=\"data row58 col1\" >2037.626198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row59\" class=\"row_heading level0 row59\" >59</th>\n",
       "      <td id=\"T_e139d_row59_col0\" class=\"data row59 col0\" >{'lasso__alpha': 9.545484566618338}</td>\n",
       "      <td id=\"T_e139d_row59_col1\" class=\"data row59 col1\" >2092.796448</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row60\" class=\"row_heading level0 row60\" >60</th>\n",
       "      <td id=\"T_e139d_row60_col0\" class=\"data row60 col0\" >{'lasso__alpha': 10.722672220103231}</td>\n",
       "      <td id=\"T_e139d_row60_col1\" class=\"data row60 col1\" >2162.722194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row61\" class=\"row_heading level0 row61\" >61</th>\n",
       "      <td id=\"T_e139d_row61_col0\" class=\"data row61 col0\" >{'lasso__alpha': 12.045035402587823}</td>\n",
       "      <td id=\"T_e139d_row61_col1\" class=\"data row61 col1\" >2245.650997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row62\" class=\"row_heading level0 row62\" >62</th>\n",
       "      <td id=\"T_e139d_row62_col0\" class=\"data row62 col0\" >{'lasso__alpha': 13.530477745798061}</td>\n",
       "      <td id=\"T_e139d_row62_col1\" class=\"data row62 col1\" >2324.757280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row63\" class=\"row_heading level0 row63\" >63</th>\n",
       "      <td id=\"T_e139d_row63_col0\" class=\"data row63 col0\" >{'lasso__alpha': 15.199110829529332}</td>\n",
       "      <td id=\"T_e139d_row63_col1\" class=\"data row63 col1\" >2397.978294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row64\" class=\"row_heading level0 row64\" >64</th>\n",
       "      <td id=\"T_e139d_row64_col0\" class=\"data row64 col0\" >{'lasso__alpha': 17.073526474706906}</td>\n",
       "      <td id=\"T_e139d_row64_col1\" class=\"data row64 col1\" >2490.994697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row65\" class=\"row_heading level0 row65\" >65</th>\n",
       "      <td id=\"T_e139d_row65_col0\" class=\"data row65 col0\" >{'lasso__alpha': 19.179102616724887}</td>\n",
       "      <td id=\"T_e139d_row65_col1\" class=\"data row65 col1\" >2574.043380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row66\" class=\"row_heading level0 row66\" >66</th>\n",
       "      <td id=\"T_e139d_row66_col0\" class=\"data row66 col0\" >{'lasso__alpha': 21.544346900318846}</td>\n",
       "      <td id=\"T_e139d_row66_col1\" class=\"data row66 col1\" >2602.241122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row67\" class=\"row_heading level0 row67\" >67</th>\n",
       "      <td id=\"T_e139d_row67_col0\" class=\"data row67 col0\" >{'lasso__alpha': 24.20128264794381}</td>\n",
       "      <td id=\"T_e139d_row67_col1\" class=\"data row67 col1\" >2613.603294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row68\" class=\"row_heading level0 row68\" >68</th>\n",
       "      <td id=\"T_e139d_row68_col0\" class=\"data row68 col0\" >{'lasso__alpha': 27.1858824273294}</td>\n",
       "      <td id=\"T_e139d_row68_col1\" class=\"data row68 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row69\" class=\"row_heading level0 row69\" >69</th>\n",
       "      <td id=\"T_e139d_row69_col0\" class=\"data row69 col0\" >{'lasso__alpha': 30.538555088334157}</td>\n",
       "      <td id=\"T_e139d_row69_col1\" class=\"data row69 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row70\" class=\"row_heading level0 row70\" >70</th>\n",
       "      <td id=\"T_e139d_row70_col0\" class=\"data row70 col0\" >{'lasso__alpha': 34.30469286314919}</td>\n",
       "      <td id=\"T_e139d_row70_col1\" class=\"data row70 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row71\" class=\"row_heading level0 row71\" >71</th>\n",
       "      <td id=\"T_e139d_row71_col0\" class=\"data row71 col0\" >{'lasso__alpha': 38.53528593710527}</td>\n",
       "      <td id=\"T_e139d_row71_col1\" class=\"data row71 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row72\" class=\"row_heading level0 row72\" >72</th>\n",
       "      <td id=\"T_e139d_row72_col0\" class=\"data row72 col0\" >{'lasso__alpha': 43.28761281083057}</td>\n",
       "      <td id=\"T_e139d_row72_col1\" class=\"data row72 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row73\" class=\"row_heading level0 row73\" >73</th>\n",
       "      <td id=\"T_e139d_row73_col0\" class=\"data row73 col0\" >{'lasso__alpha': 48.62601580065353}</td>\n",
       "      <td id=\"T_e139d_row73_col1\" class=\"data row73 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row74\" class=\"row_heading level0 row74\" >74</th>\n",
       "      <td id=\"T_e139d_row74_col0\" class=\"data row74 col0\" >{'lasso__alpha': 54.62277217684343}</td>\n",
       "      <td id=\"T_e139d_row74_col1\" class=\"data row74 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row75\" class=\"row_heading level0 row75\" >75</th>\n",
       "      <td id=\"T_e139d_row75_col0\" class=\"data row75 col0\" >{'lasso__alpha': 61.35907273413169}</td>\n",
       "      <td id=\"T_e139d_row75_col1\" class=\"data row75 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row76\" class=\"row_heading level0 row76\" >76</th>\n",
       "      <td id=\"T_e139d_row76_col0\" class=\"data row76 col0\" >{'lasso__alpha': 68.92612104349695}</td>\n",
       "      <td id=\"T_e139d_row76_col1\" class=\"data row76 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row77\" class=\"row_heading level0 row77\" >77</th>\n",
       "      <td id=\"T_e139d_row77_col0\" class=\"data row77 col0\" >{'lasso__alpha': 77.4263682681127}</td>\n",
       "      <td id=\"T_e139d_row77_col1\" class=\"data row77 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row78\" class=\"row_heading level0 row78\" >78</th>\n",
       "      <td id=\"T_e139d_row78_col0\" class=\"data row78 col0\" >{'lasso__alpha': 86.97490026177834}</td>\n",
       "      <td id=\"T_e139d_row78_col1\" class=\"data row78 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row79\" class=\"row_heading level0 row79\" >79</th>\n",
       "      <td id=\"T_e139d_row79_col0\" class=\"data row79 col0\" >{'lasso__alpha': 97.70099572992247}</td>\n",
       "      <td id=\"T_e139d_row79_col1\" class=\"data row79 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row80\" class=\"row_heading level0 row80\" >80</th>\n",
       "      <td id=\"T_e139d_row80_col0\" class=\"data row80 col0\" >{'lasso__alpha': 109.74987654930568}</td>\n",
       "      <td id=\"T_e139d_row80_col1\" class=\"data row80 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row81\" class=\"row_heading level0 row81\" >81</th>\n",
       "      <td id=\"T_e139d_row81_col0\" class=\"data row81 col0\" >{'lasso__alpha': 123.28467394420659}</td>\n",
       "      <td id=\"T_e139d_row81_col1\" class=\"data row81 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row82\" class=\"row_heading level0 row82\" >82</th>\n",
       "      <td id=\"T_e139d_row82_col0\" class=\"data row82 col0\" >{'lasso__alpha': 138.48863713938718}</td>\n",
       "      <td id=\"T_e139d_row82_col1\" class=\"data row82 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row83\" class=\"row_heading level0 row83\" >83</th>\n",
       "      <td id=\"T_e139d_row83_col0\" class=\"data row83 col0\" >{'lasso__alpha': 155.56761439304722}</td>\n",
       "      <td id=\"T_e139d_row83_col1\" class=\"data row83 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row84\" class=\"row_heading level0 row84\" >84</th>\n",
       "      <td id=\"T_e139d_row84_col0\" class=\"data row84 col0\" >{'lasso__alpha': 174.7528400007683}</td>\n",
       "      <td id=\"T_e139d_row84_col1\" class=\"data row84 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row85\" class=\"row_heading level0 row85\" >85</th>\n",
       "      <td id=\"T_e139d_row85_col0\" class=\"data row85 col0\" >{'lasso__alpha': 196.30406500402725}</td>\n",
       "      <td id=\"T_e139d_row85_col1\" class=\"data row85 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row86\" class=\"row_heading level0 row86\" >86</th>\n",
       "      <td id=\"T_e139d_row86_col0\" class=\"data row86 col0\" >{'lasso__alpha': 220.51307399030458}</td>\n",
       "      <td id=\"T_e139d_row86_col1\" class=\"data row86 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row87\" class=\"row_heading level0 row87\" >87</th>\n",
       "      <td id=\"T_e139d_row87_col0\" class=\"data row87 col0\" >{'lasso__alpha': 247.70763559917089}</td>\n",
       "      <td id=\"T_e139d_row87_col1\" class=\"data row87 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row88\" class=\"row_heading level0 row88\" >88</th>\n",
       "      <td id=\"T_e139d_row88_col0\" class=\"data row88 col0\" >{'lasso__alpha': 278.2559402207126}</td>\n",
       "      <td id=\"T_e139d_row88_col1\" class=\"data row88 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row89\" class=\"row_heading level0 row89\" >89</th>\n",
       "      <td id=\"T_e139d_row89_col0\" class=\"data row89 col0\" >{'lasso__alpha': 312.5715849688235}</td>\n",
       "      <td id=\"T_e139d_row89_col1\" class=\"data row89 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row90\" class=\"row_heading level0 row90\" >90</th>\n",
       "      <td id=\"T_e139d_row90_col0\" class=\"data row90 col0\" >{'lasso__alpha': 351.11917342151276}</td>\n",
       "      <td id=\"T_e139d_row90_col1\" class=\"data row90 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row91\" class=\"row_heading level0 row91\" >91</th>\n",
       "      <td id=\"T_e139d_row91_col0\" class=\"data row91 col0\" >{'lasso__alpha': 394.4206059437656}</td>\n",
       "      <td id=\"T_e139d_row91_col1\" class=\"data row91 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row92\" class=\"row_heading level0 row92\" >92</th>\n",
       "      <td id=\"T_e139d_row92_col0\" class=\"data row92 col0\" >{'lasso__alpha': 443.06214575838777}</td>\n",
       "      <td id=\"T_e139d_row92_col1\" class=\"data row92 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row93\" class=\"row_heading level0 row93\" >93</th>\n",
       "      <td id=\"T_e139d_row93_col0\" class=\"data row93 col0\" >{'lasso__alpha': 497.7023564332114}</td>\n",
       "      <td id=\"T_e139d_row93_col1\" class=\"data row93 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row94\" class=\"row_heading level0 row94\" >94</th>\n",
       "      <td id=\"T_e139d_row94_col0\" class=\"data row94 col0\" >{'lasso__alpha': 559.0810182512223}</td>\n",
       "      <td id=\"T_e139d_row94_col1\" class=\"data row94 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row95\" class=\"row_heading level0 row95\" >95</th>\n",
       "      <td id=\"T_e139d_row95_col0\" class=\"data row95 col0\" >{'lasso__alpha': 628.0291441834247}</td>\n",
       "      <td id=\"T_e139d_row95_col1\" class=\"data row95 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row96\" class=\"row_heading level0 row96\" >96</th>\n",
       "      <td id=\"T_e139d_row96_col0\" class=\"data row96 col0\" >{'lasso__alpha': 705.4802310718645}</td>\n",
       "      <td id=\"T_e139d_row96_col1\" class=\"data row96 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row97\" class=\"row_heading level0 row97\" >97</th>\n",
       "      <td id=\"T_e139d_row97_col0\" class=\"data row97 col0\" >{'lasso__alpha': 792.482898353917}</td>\n",
       "      <td id=\"T_e139d_row97_col1\" class=\"data row97 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row98\" class=\"row_heading level0 row98\" >98</th>\n",
       "      <td id=\"T_e139d_row98_col0\" class=\"data row98 col0\" >{'lasso__alpha': 890.2150854450392}</td>\n",
       "      <td id=\"T_e139d_row98_col1\" class=\"data row98 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e139d_level0_row99\" class=\"row_heading level0 row99\" >99</th>\n",
       "      <td id=\"T_e139d_row99_col0\" class=\"data row99 col0\" >{'lasso__alpha': 1000.0}</td>\n",
       "      <td id=\"T_e139d_row99_col1\" class=\"data row99 col1\" >2613.724845</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x108e7260650>"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Lasso\n",
    "# Applying the highlighting function only to 'mean_test_score' column\n",
    "results_style = results[['params', 'mean_test_score']].style.apply(\n",
    "    highlight_min, subset=['mean_test_score'])\n",
    "\n",
    "results_style"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best alpha: 0.32745491628777285\n",
      "Best MSE: 1629.4999941754438\n"
     ]
    }
   ],
   "source": [
    "# Getting the best alpha value and corresponding MSE\n",
    "best_lasso_alpha = grid_search.best_params_['lasso__alpha']\n",
    "best_lasso_mse = grid_search.best_score_\n",
    "\n",
    "print(f\"Best alpha: {best_lasso_alpha}\")\n",
    "print(f\"Best MSE: {-best_lasso_mse}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### (e) Re-estimating your model on the full training set using that optimal value of lambda."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='blue'> \n",
    "Based on the calculations of training and testing MSE using optimal values of lambda:\n",
    "\n",
    "- Ridge Training MSE:  1465.1888754008635\r",
    "- Ridge Testing MSE:  2402.320060977949\n",
    "- Lasso Training MSE:  1470.1107209605289\r",
    "- Lasso Testing MSE:  2397.1356134923935*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Ridge(alpha=247.70763559917089)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Ridge</label><div class=\"sk-toggleable__content\"><pre>Ridge(alpha=247.70763559917089)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "Ridge(alpha=247.70763559917089)"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "optimal_ridge = Ridge(alpha=best_alpha)\n",
    "optimal_ridge.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ridge Training MSE:  1465.1888754008635\n",
      "Ridge Testing MSE:  2402.3200609779497\n"
     ]
    }
   ],
   "source": [
    "# Estimating using the optimal value of lambda\n",
    "y_pred_train_ridge = optimal_ridge.predict(X_train)\n",
    "y_pred_test_ridge = optimal_ridge.predict(X_test)\n",
    "mse_train_ridge = mean_squared_error(y_train, y_pred_train_ridge)\n",
    "mse_test_ridge = mean_squared_error(y_test, y_pred_test_ridge)\n",
    "\n",
    "print('Ridge Training MSE: ', mse_train_ridge)\n",
    "print('Ridge Testing MSE: ', mse_test_ridge)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-4 {color: black;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-4 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Lasso(alpha=0.32745491628777285)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" checked><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Lasso</label><div class=\"sk-toggleable__content\"><pre>Lasso(alpha=0.32745491628777285)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "Lasso(alpha=0.32745491628777285)"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "optimal_lasso = Lasso(alpha=best_lasso_alpha)\n",
    "optimal_lasso.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Lasso Training MSE:  1470.1107209605289\n",
      "Lasso Testing MSE:  2397.1356134923935\n"
     ]
    }
   ],
   "source": [
    "# Estimating using the optimal value of lambda\n",
    "y_pred_train_lasso = optimal_lasso.predict(X_train)\n",
    "y_pred_test_lasso = optimal_lasso.predict(X_test)\n",
    "mse_train_lasso = mean_squared_error(y_train, y_pred_train_lasso)\n",
    "mse_test_lasso = mean_squared_error(y_test, y_pred_test_lasso)\n",
    "\n",
    "print('Lasso Training MSE: ', mse_train_lasso)\n",
    "print('Lasso Testing MSE: ', mse_test_lasso)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8. Using the ridge regression and the lasso models you trained based on the optimal values of lambda you found in the previous question, calculate and report the training and test set prediction errors (MSE) for each model. Did ridge regression and/or the lasso improve your prediction over OLS? Which model performs the best? Briefly explain which model you would recommend to the CDC and why."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='blue'> \n",
    "Based on the calculations of training and testing MSE using optimal values of lambda:\n",
    "\n",
    "- Ridge Training MSE:  1465.1888754008635\n",
    "- Ridge Testing MSE:  2402.3200609779497\n",
    "- Lasso Training MSE:  1470.1107209605289\n",
    "- Lasso Testing MSE:  2397.13561349239351\n",
    "\n",
    "Ridge trainig MSE < Lasso Training MSE <Lassoe Testing MSE <Ridgeo Testing MSE, hence lasso gives the best prediction. Compared to OLS testing MSE of 2408.75, both ddige regression and lasso improve the prediction over OLS. It is interesting to notice that ridge has a lower training MSE but a higher testing MSE compared to lasso, suggesting that ridge might be slightly overfitting to the training data, capturing noise along with the underlying data pattern.\n",
    "\n",
    "\n",
    "When recommending a model to the CDC for the purpose of predicting COVID-19 deaths per capita using socio-economic, health, and environmental data, we need to consider the test MSE (generalization), model complexity, and interpretability. I would recommend lasso since it has superior generalization capability, coupled with its feature selection property, makes it a more suitable choice for developing a predictive model for the CDC. Though it holds a slightly higher training MSE, it does have the best predictability on the test set, which suggests a better generalization to new data compared to the other models. Lasso, with its inherent feature selection capability (due to L1 regularization), likely provides a simpler model that generalizes better. In terms of model simplicity and interpretability, lasso tends to produce sparser models by setting some coefficient estimates to 0, effectively selecting a subset of the most important features. This can lead to simpler, more interpretable models, which is valuable for understanding the factors influencing COVID-19 deaths per capita and for communicating these findings to policymakers and the public."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ridge Training MSE:  1465.1888754008635\n",
      "Ridge Testing MSE:  2402.3200609779497\n"
     ]
    }
   ],
   "source": [
    "optimal_ridge = Ridge(alpha=best_alpha)\n",
    "optimal_ridge.fit(X_train, y_train)\n",
    "\n",
    "# Estimating using the optimal value of lambda\n",
    "y_pred_train_ridge = optimal_ridge.predict(X_train)\n",
    "y_pred_test_ridge = optimal_ridge.predict(X_test)\n",
    "mse_train_ridge = mean_squared_error(y_train, y_pred_train_ridge)\n",
    "mse_test_ridge = mean_squared_error(y_test, y_pred_test_ridge)\n",
    "\n",
    "print('Ridge Training MSE: ', mse_train_ridge)\n",
    "print('Ridge Testing MSE: ', mse_test_ridge)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Lasso Training MSE:  1470.1107209605289\n",
      "Lasso Testing MSE:  2397.1356134923935\n"
     ]
    }
   ],
   "source": [
    "optimal_lasso = Lasso(alpha=best_lasso_alpha)\n",
    "optimal_lasso.fit(X_train, y_train)\n",
    "\n",
    "# Estimating using the optimal value of lambda\n",
    "y_pred_train_lasso = optimal_lasso.predict(X_train)\n",
    "y_pred_test_lasso = optimal_lasso.predict(X_test)\n",
    "mse_train_lasso = mean_squared_error(y_train, y_pred_train_lasso)\n",
    "mse_test_lasso = mean_squared_error(y_test, y_pred_test_lasso)\n",
    "\n",
    "print('Lasso Training MSE: ', mse_train_lasso)\n",
    "print('Lasso Testing MSE: ', mse_test_lasso)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.7"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "vscode": {
   "interpreter": {
    "hash": "26de051ba29f2982a8de78e945f0abaf191376122a1563185a90213a26c5da77"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}