{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unsupervised Learning: Clustering and Decomposition\n",
    "\n",
    "We'll look at COVID cases per day for each county in WI.  We'll use three unsupervised learning techniques to understand multi-county patterns:\n",
    "\n",
    "1. KMeans (clustering)\n",
    "2. AgglomerativeClustering (clustering)\n",
    "3. PCA (decomposition)\n",
    "\n",
    "First, let's import all that we need:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import geopandas as gpd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.cluster import KMeans, AgglomerativeClustering\n",
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Mapping"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll be creating maps of WI.  I already grabbed shapefile of the counties from here: https://data-wi-dnr.opendata.arcgis.com/datasets/8b8a0896378449538cf1138a969afbc6_3.  I renamed it \"wi.zip\".  We rename \"Saint Croix\" to \"St. Croix\" for consistency with our later dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>OBJECTID</th>\n",
       "      <th>DNR_REGION</th>\n",
       "      <th>DNR_CNTY_C</th>\n",
       "      <th>COUNTY_NAM</th>\n",
       "      <th>COUNTY_FIP</th>\n",
       "      <th>SHAPEAREA</th>\n",
       "      <th>SHAPELEN</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COUNTY_NAM</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Kenosha</th>\n",
       "      <td>321</td>\n",
       "      <td>Southeast Region</td>\n",
       "      <td>30</td>\n",
       "      <td>Kenosha</td>\n",
       "      <td>59</td>\n",
       "      <td>7.210454e+08</td>\n",
       "      <td>123267.303863</td>\n",
       "      <td>MULTIPOLYGON (((699813.437 246226.688, 699794....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lafayette</th>\n",
       "      <td>322</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>33</td>\n",
       "      <td>Lafayette</td>\n",
       "      <td>65</td>\n",
       "      <td>1.641795e+09</td>\n",
       "      <td>164707.650040</td>\n",
       "      <td>POLYGON ((503148.082 260278.466, 503292.672 26...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Rock</th>\n",
       "      <td>323</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>54</td>\n",
       "      <td>Rock</td>\n",
       "      <td>105</td>\n",
       "      <td>1.879382e+09</td>\n",
       "      <td>174114.587670</td>\n",
       "      <td>POLYGON ((600602.683 264347.425, 603850.419 26...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Walworth</th>\n",
       "      <td>324</td>\n",
       "      <td>Southeast Region</td>\n",
       "      <td>65</td>\n",
       "      <td>Walworth</td>\n",
       "      <td>127</td>\n",
       "      <td>1.492598e+09</td>\n",
       "      <td>154833.279357</td>\n",
       "      <td>POLYGON ((658404.520 263083.277, 658417.776 26...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Green</th>\n",
       "      <td>325</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>23</td>\n",
       "      <td>Green</td>\n",
       "      <td>45</td>\n",
       "      <td>1.512855e+09</td>\n",
       "      <td>155741.104373</td>\n",
       "      <td>POLYGON ((571551.903 263810.562, 571555.279 26...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            OBJECTID            DNR_REGION  DNR_CNTY_C COUNTY_NAM COUNTY_FIP  \\\n",
       "COUNTY_NAM                                                                     \n",
       "Kenosha          321      Southeast Region          30    Kenosha         59   \n",
       "Lafayette        322  South Central Region          33  Lafayette         65   \n",
       "Rock             323  South Central Region          54       Rock        105   \n",
       "Walworth         324      Southeast Region          65   Walworth        127   \n",
       "Green            325  South Central Region          23      Green         45   \n",
       "\n",
       "               SHAPEAREA       SHAPELEN  \\\n",
       "COUNTY_NAM                                \n",
       "Kenosha     7.210454e+08  123267.303863   \n",
       "Lafayette   1.641795e+09  164707.650040   \n",
       "Rock        1.879382e+09  174114.587670   \n",
       "Walworth    1.492598e+09  154833.279357   \n",
       "Green       1.512855e+09  155741.104373   \n",
       "\n",
       "                                                     geometry  \n",
       "COUNTY_NAM                                                     \n",
       "Kenosha     MULTIPOLYGON (((699813.437 246226.688, 699794....  \n",
       "Lafayette   POLYGON ((503148.082 260278.466, 503292.672 26...  \n",
       "Rock        POLYGON ((600602.683 264347.425, 603850.419 26...  \n",
       "Walworth    POLYGON ((658404.520 263083.277, 658417.776 26...  \n",
       "Green       POLYGON ((571551.903 263810.562, 571555.279 26...  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wi = gpd.read_file(\"zip://wi.zip\")\n",
    "wi = wi.set_index(wi[\"COUNTY_NAM\"].str.replace(\"Saint\", \"St.\")).copy()\n",
    "wi.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa9e603a160>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wi.boundary.plot(color=\"black\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## COVID Data\n",
    "\n",
    "Let's use wi-covid.csv that we used in previous lectures:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NAME</th>\n",
       "      <th>DATE</th>\n",
       "      <th>POS_7DAYAVG</th>\n",
       "      <th>DTH_NEW</th>\n",
       "      <th>DTH_NEW_2WK</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Adams</td>\n",
       "      <td>2020-03-16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ashland</td>\n",
       "      <td>2020-03-16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Barron</td>\n",
       "      <td>2020-03-16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Bayfield</td>\n",
       "      <td>2020-03-16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Brown</td>\n",
       "      <td>2020-03-16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       NAME        DATE  POS_7DAYAVG  DTH_NEW  DTH_NEW_2WK\n",
       "0     Adams  2020-03-16          0.0      0.0          0.0\n",
       "1   Ashland  2020-03-16          0.0      0.0          0.0\n",
       "2    Barron  2020-03-16          0.0      0.0          0.0\n",
       "3  Bayfield  2020-03-16          0.0      0.0          0.0\n",
       "4     Brown  2020-03-16          0.0      0.0          0.0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"wi-covid.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For clustering, we'll want each county on its own row.  Historical data should be spread across columns, so that we can group rows based on year-long trends.  Pandas's `pivot` function can automatially take three columns and put the values of one to the index, the values of another to the columns, and the values of the third to the cells of the final table:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>DATE</th>\n",
       "      <th>2020-03-16</th>\n",
       "      <th>2020-03-17</th>\n",
       "      <th>2020-03-18</th>\n",
       "      <th>2020-03-19</th>\n",
       "      <th>2020-03-20</th>\n",
       "      <th>2020-03-21</th>\n",
       "      <th>2020-03-22</th>\n",
       "      <th>2020-03-23</th>\n",
       "      <th>2020-03-24</th>\n",
       "      <th>2020-03-25</th>\n",
       "      <th>...</th>\n",
       "      <th>2021-02-21</th>\n",
       "      <th>2021-02-22</th>\n",
       "      <th>2021-02-23</th>\n",
       "      <th>2021-02-24</th>\n",
       "      <th>2021-02-25</th>\n",
       "      <th>2021-02-26</th>\n",
       "      <th>2021-02-27</th>\n",
       "      <th>2021-02-28</th>\n",
       "      <th>2021-03-01</th>\n",
       "      <th>2021-03-02</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NAME</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Adams</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>3.00</td>\n",
       "      <td>2.71</td>\n",
       "      <td>2.57</td>\n",
       "      <td>3.14</td>\n",
       "      <td>2.14</td>\n",
       "      <td>1.86</td>\n",
       "      <td>1.29</td>\n",
       "      <td>1.29</td>\n",
       "      <td>1.57</td>\n",
       "      <td>1.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ashland</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Barron</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>7.71</td>\n",
       "      <td>7.29</td>\n",
       "      <td>7.57</td>\n",
       "      <td>7.71</td>\n",
       "      <td>8.14</td>\n",
       "      <td>7.71</td>\n",
       "      <td>8.29</td>\n",
       "      <td>9.00</td>\n",
       "      <td>8.86</td>\n",
       "      <td>8.57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bayfield</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.14</td>\n",
       "      <td>...</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Brown</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.29</td>\n",
       "      <td>...</td>\n",
       "      <td>23.29</td>\n",
       "      <td>23.57</td>\n",
       "      <td>18.86</td>\n",
       "      <td>18.71</td>\n",
       "      <td>21.57</td>\n",
       "      <td>20.86</td>\n",
       "      <td>20.14</td>\n",
       "      <td>19.57</td>\n",
       "      <td>19.71</td>\n",
       "      <td>18.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 352 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "DATE      2020-03-16  2020-03-17  2020-03-18  2020-03-19  2020-03-20  \\\n",
       "NAME                                                                   \n",
       "Adams            0.0         0.0        0.00        0.00         0.0   \n",
       "Ashland          0.0         0.0        0.00        0.00         0.0   \n",
       "Barron           0.0         0.0        0.00        0.00         0.0   \n",
       "Bayfield         0.0         0.0        0.00        0.25         0.2   \n",
       "Brown            0.0         0.0        0.33        0.50         0.4   \n",
       "\n",
       "DATE      2020-03-21  2020-03-22  2020-03-23  2020-03-24  2020-03-25  ...  \\\n",
       "NAME                                                                  ...   \n",
       "Adams           0.00        0.00        0.00        0.00        0.00  ...   \n",
       "Ashland         0.00        0.00        0.00        0.00        0.00  ...   \n",
       "Barron          0.00        0.00        0.00        0.00        0.00  ...   \n",
       "Bayfield        0.17        0.14        0.14        0.14        0.14  ...   \n",
       "Brown           0.50        0.43        0.43        0.43        0.29  ...   \n",
       "\n",
       "DATE      2021-02-21  2021-02-22  2021-02-23  2021-02-24  2021-02-25  \\\n",
       "NAME                                                                   \n",
       "Adams           3.00        2.71        2.57        3.14        2.14   \n",
       "Ashland         0.43        0.43        0.71        0.57        0.71   \n",
       "Barron          7.71        7.29        7.57        7.71        8.14   \n",
       "Bayfield        0.14        0.14        0.29        0.29        0.14   \n",
       "Brown          23.29       23.57       18.86       18.71       21.57   \n",
       "\n",
       "DATE      2021-02-26  2021-02-27  2021-02-28  2021-03-01  2021-03-02  \n",
       "NAME                                                                  \n",
       "Adams           1.86        1.29        1.29        1.57        1.29  \n",
       "Ashland         0.71        0.57        0.57        0.43        0.29  \n",
       "Barron          7.71        8.29        9.00        8.86        8.57  \n",
       "Bayfield        0.29        0.29        0.29        0.43        0.43  \n",
       "Brown          20.86       20.14       19.57       19.71       18.00  \n",
       "\n",
       "[5 rows x 352 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.pivot(df, index=\"NAME\", columns=\"DATE\", values=\"POS_7DAYAVG\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Clustering: KMeans and Agglomerative"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's use KMeans clustering to group similar rows.  Note that for this example, we're just clustering based on total cases (for many use cases, it may make sense to scale by population or other otherwise standardize the data)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
       "       n_clusters=8, n_init=10, n_jobs=None, precompute_distances='auto',\n",
       "       random_state=None, tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "km = KMeans(n_clusters=8)\n",
    "km.fit(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 4, 1, 7, 1, 1, 4, 4, 1, 4, 1, 3, 5, 1, 1, 4, 5, 1, 5, 1, 4,\n",
       "       1, 1, 1, 1, 1, 4, 1, 5, 1, 5, 1, 1, 1, 4, 5, 1, 1, 1, 2, 4, 1, 1,\n",
       "       0, 4, 1, 1, 4, 4, 1, 5, 1, 5, 1, 4, 1, 4, 5, 4, 1, 1, 1, 1, 4, 1,\n",
       "       5, 6, 4, 1, 0, 4], dtype=int32)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clusters = km.predict(df)\n",
    "clusters"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's put those cluster IDs back in the GeoDataFrame so we can plot it.  The cluster IDs are in an order corresponding to the DataFrame we fit to (`df`).  This is probably the same as the order in the `wi` DataFrame, but we should explicitly assign the index to each cluster ID before adding it to a different DataFrame, just to be safe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>OBJECTID</th>\n",
       "      <th>DNR_REGION</th>\n",
       "      <th>DNR_CNTY_C</th>\n",
       "      <th>COUNTY_NAM</th>\n",
       "      <th>COUNTY_FIP</th>\n",
       "      <th>SHAPEAREA</th>\n",
       "      <th>SHAPELEN</th>\n",
       "      <th>geometry</th>\n",
       "      <th>cluster</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COUNTY_NAM</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Kenosha</th>\n",
       "      <td>321</td>\n",
       "      <td>Southeast Region</td>\n",
       "      <td>30</td>\n",
       "      <td>Kenosha</td>\n",
       "      <td>59</td>\n",
       "      <td>7.210454e+08</td>\n",
       "      <td>123267.303863</td>\n",
       "      <td>MULTIPOLYGON (((699813.437 246226.688, 699794....</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lafayette</th>\n",
       "      <td>322</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>33</td>\n",
       "      <td>Lafayette</td>\n",
       "      <td>65</td>\n",
       "      <td>1.641795e+09</td>\n",
       "      <td>164707.650040</td>\n",
       "      <td>POLYGON ((503148.082 260278.466, 503292.672 26...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Rock</th>\n",
       "      <td>323</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>54</td>\n",
       "      <td>Rock</td>\n",
       "      <td>105</td>\n",
       "      <td>1.879382e+09</td>\n",
       "      <td>174114.587670</td>\n",
       "      <td>POLYGON ((600602.683 264347.425, 603850.419 26...</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Walworth</th>\n",
       "      <td>324</td>\n",
       "      <td>Southeast Region</td>\n",
       "      <td>65</td>\n",
       "      <td>Walworth</td>\n",
       "      <td>127</td>\n",
       "      <td>1.492598e+09</td>\n",
       "      <td>154833.279357</td>\n",
       "      <td>POLYGON ((658404.520 263083.277, 658417.776 26...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Green</th>\n",
       "      <td>325</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>23</td>\n",
       "      <td>Green</td>\n",
       "      <td>45</td>\n",
       "      <td>1.512855e+09</td>\n",
       "      <td>155741.104373</td>\n",
       "      <td>POLYGON ((571551.903 263810.562, 571555.279 26...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            OBJECTID            DNR_REGION  DNR_CNTY_C COUNTY_NAM COUNTY_FIP  \\\n",
       "COUNTY_NAM                                                                     \n",
       "Kenosha          321      Southeast Region          30    Kenosha         59   \n",
       "Lafayette        322  South Central Region          33  Lafayette         65   \n",
       "Rock             323  South Central Region          54       Rock        105   \n",
       "Walworth         324      Southeast Region          65   Walworth        127   \n",
       "Green            325  South Central Region          23      Green         45   \n",
       "\n",
       "               SHAPEAREA       SHAPELEN  \\\n",
       "COUNTY_NAM                                \n",
       "Kenosha     7.210454e+08  123267.303863   \n",
       "Lafayette   1.641795e+09  164707.650040   \n",
       "Rock        1.879382e+09  174114.587670   \n",
       "Walworth    1.492598e+09  154833.279357   \n",
       "Green       1.512855e+09  155741.104373   \n",
       "\n",
       "                                                     geometry  cluster  \n",
       "COUNTY_NAM                                                              \n",
       "Kenosha     MULTIPOLYGON (((699813.437 246226.688, 699794....        5  \n",
       "Lafayette   POLYGON ((503148.082 260278.466, 503292.672 26...        1  \n",
       "Rock        POLYGON ((600602.683 264347.425, 603850.419 26...        5  \n",
       "Walworth    POLYGON ((658404.520 263083.277, 658417.776 26...        4  \n",
       "Green       POLYGON ((571551.903 263810.562, 571555.279 26...        1  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wi[\"cluster\"] = pd.Series(clusters, index=df.index)\n",
    "wi.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa9e626ea90>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wi.plot(column=wi[\"cluster\"], cmap=\"tab10\", figsize=(8,8))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compactly, we could have done all the fitting and plotting in just three lines (note that clusters will probably be similar, but the colors for each cluster might change each time we run this):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa9e62dcfd0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clusters = KMeans(n_clusters=8).fit_predict(df)\n",
    "wi[\"cluster\"] = pd.Series(clusters, index=df.index)\n",
    "wi.plot(column=wi[\"cluster\"], cmap=\"tab10\", figsize=(8,8))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can simply replace `KMeans` with `AgglomerativeClustering` to see how the latter algorithm groups states:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa9e698a390>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clusters = AgglomerativeClustering(n_clusters=8).fit_predict(df)\n",
    "wi[\"cluster\"] = pd.Series(clusters, index=df.index)\n",
    "wi.plot(column=wi[\"cluster\"], cmap=\"tab10\", figsize=(8,8))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PCA: Principal Component Analysis\n",
    "\n",
    "For this analysis, we'll be looking at a tranpose of the earlier data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shape: (352, 72)\n",
      "variance: 3230.8970796538392\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>NAME</th>\n",
       "      <th>Adams</th>\n",
       "      <th>Ashland</th>\n",
       "      <th>Barron</th>\n",
       "      <th>Bayfield</th>\n",
       "      <th>Brown</th>\n",
       "      <th>Buffalo</th>\n",
       "      <th>Burnett</th>\n",
       "      <th>Calumet</th>\n",
       "      <th>Chippewa</th>\n",
       "      <th>Clark</th>\n",
       "      <th>...</th>\n",
       "      <th>Vernon</th>\n",
       "      <th>Vilas</th>\n",
       "      <th>Walworth</th>\n",
       "      <th>Washburn</th>\n",
       "      <th>Washington</th>\n",
       "      <th>Waukesha</th>\n",
       "      <th>Waupaca</th>\n",
       "      <th>Waushara</th>\n",
       "      <th>Winnebago</th>\n",
       "      <th>Wood</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DATE</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-03-16</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.00</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-17</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-18</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-19</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.50</td>\n",
       "      <td>2.25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-20</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.60</td>\n",
       "      <td>2.40</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 72 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "NAME        Adams  Ashland  Barron  Bayfield  Brown  Buffalo  Burnett  \\\n",
       "DATE                                                                    \n",
       "2020-03-16    0.0      0.0     0.0      0.00   0.00      0.0      0.0   \n",
       "2020-03-17    0.0      0.0     0.0      0.00   0.00      0.0      0.0   \n",
       "2020-03-18    0.0      0.0     0.0      0.00   0.33      0.0      0.0   \n",
       "2020-03-19    0.0      0.0     0.0      0.25   0.50      0.0      0.0   \n",
       "2020-03-20    0.0      0.0     0.0      0.20   0.40      0.0      0.0   \n",
       "\n",
       "NAME        Calumet  Chippewa  Clark  ...  Vernon  Vilas  Walworth  Washburn  \\\n",
       "DATE                                  ...                                      \n",
       "2020-03-16     0.00       0.0    0.0  ...     0.0    0.0       0.0       0.0   \n",
       "2020-03-17     0.00       0.0    0.0  ...     0.0    0.0       0.0       0.0   \n",
       "2020-03-18     0.00       0.0    0.0  ...     0.0    0.0       0.0       0.0   \n",
       "2020-03-19     0.25       0.0    0.0  ...     0.0    0.0       0.5       0.0   \n",
       "2020-03-20     0.20       0.2    0.0  ...     0.0    0.0       0.6       0.0   \n",
       "\n",
       "NAME        Washington  Waukesha  Waupaca  Waushara  Winnebago  Wood  \n",
       "DATE                                                                  \n",
       "2020-03-16        0.00      0.00      0.0       0.0       2.00  1.00  \n",
       "2020-03-17        0.00      0.50      0.0       0.0       1.00  0.50  \n",
       "2020-03-18        0.67      0.67      0.0       0.0       0.67  0.33  \n",
       "2020-03-19        0.50      2.25      0.0       0.0       1.00  0.25  \n",
       "2020-03-20        0.60      2.40      0.0       0.0       0.80  0.20  \n",
       "\n",
       "[5 rows x 72 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dft = df.T\n",
    "print(\"shape:\", dft.shape)\n",
    "print(\"variance:\", dft.values.var())\n",
    "dft.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have 352 rows, but of course there will be similarities across them.  Can we approximate the table by finding a few key rows (called components), such that every row in approximately some weighted combination of those component rows?  In particular, how much of the 3230.9 variance can be explained by these few rows?\n",
    "\n",
    "A principal component analysis (PCA) can help us find such components and tell us how much of the variance those components can explain.\n",
    "\n",
    "Passing 0.95 tells the PCA that we want it to find however many component rows are necessary to explain 95% of the variance in our original table (alternatively, we could pass an integer to require an exact number of components):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(copy=True, iterated_power='auto', n_components=0.95, random_state=None,\n",
       "    svd_solver='auto', tol=0.0, whiten=False)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(0.95)\n",
    "pca.fit(dft)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`explained_variance_ratio_` tells us how many components are necessary, and how much variance they each explain.  In this case, one component row can explain 90.6% of the variance, and a second component row can explain an additional 4.5%."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.90638618, 0.04510856])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see what those two rows look like in a table.  `components_` gives us a numpy array, but we can use the same column names from our original DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>NAME</th>\n",
       "      <th>Adams</th>\n",
       "      <th>Ashland</th>\n",
       "      <th>Barron</th>\n",
       "      <th>Bayfield</th>\n",
       "      <th>Brown</th>\n",
       "      <th>Buffalo</th>\n",
       "      <th>Burnett</th>\n",
       "      <th>Calumet</th>\n",
       "      <th>Chippewa</th>\n",
       "      <th>Clark</th>\n",
       "      <th>...</th>\n",
       "      <th>Vernon</th>\n",
       "      <th>Vilas</th>\n",
       "      <th>Walworth</th>\n",
       "      <th>Washburn</th>\n",
       "      <th>Washington</th>\n",
       "      <th>Waukesha</th>\n",
       "      <th>Waupaca</th>\n",
       "      <th>Waushara</th>\n",
       "      <th>Winnebago</th>\n",
       "      <th>Wood</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.014671</td>\n",
       "      <td>0.012164</td>\n",
       "      <td>0.063659</td>\n",
       "      <td>0.012197</td>\n",
       "      <td>0.186703</td>\n",
       "      <td>0.013653</td>\n",
       "      <td>0.012156</td>\n",
       "      <td>0.039235</td>\n",
       "      <td>0.081403</td>\n",
       "      <td>0.034584</td>\n",
       "      <td>...</td>\n",
       "      <td>0.017245</td>\n",
       "      <td>0.018460</td>\n",
       "      <td>0.065934</td>\n",
       "      <td>0.013056</td>\n",
       "      <td>0.120046</td>\n",
       "      <td>0.379391</td>\n",
       "      <td>0.034502</td>\n",
       "      <td>0.018779</td>\n",
       "      <td>0.124864</td>\n",
       "      <td>0.068892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.010148</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>-0.026789</td>\n",
       "      <td>-0.001165</td>\n",
       "      <td>0.488611</td>\n",
       "      <td>-0.003380</td>\n",
       "      <td>0.000764</td>\n",
       "      <td>0.125926</td>\n",
       "      <td>0.006220</td>\n",
       "      <td>0.003825</td>\n",
       "      <td>...</td>\n",
       "      <td>0.003234</td>\n",
       "      <td>0.010982</td>\n",
       "      <td>-0.023864</td>\n",
       "      <td>-0.011774</td>\n",
       "      <td>0.010618</td>\n",
       "      <td>-0.225074</td>\n",
       "      <td>0.099319</td>\n",
       "      <td>0.058002</td>\n",
       "      <td>0.452854</td>\n",
       "      <td>-0.009747</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 72 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "NAME     Adams   Ashland    Barron  Bayfield     Brown   Buffalo   Burnett  \\\n",
       "0     0.014671  0.012164  0.063659  0.012197  0.186703  0.013653  0.012156   \n",
       "1     0.010148  0.000019 -0.026789 -0.001165  0.488611 -0.003380  0.000764   \n",
       "\n",
       "NAME   Calumet  Chippewa     Clark  ...    Vernon     Vilas  Walworth  \\\n",
       "0     0.039235  0.081403  0.034584  ...  0.017245  0.018460  0.065934   \n",
       "1     0.125926  0.006220  0.003825  ...  0.003234  0.010982 -0.023864   \n",
       "\n",
       "NAME  Washburn  Washington  Waukesha   Waupaca  Waushara  Winnebago      Wood  \n",
       "0     0.013056    0.120046  0.379391  0.034502  0.018779   0.124864  0.068892  \n",
       "1    -0.011774    0.010618 -0.225074  0.099319  0.058002   0.452854 -0.009747  \n",
       "\n",
       "[2 rows x 72 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "components = pd.DataFrame(pca.components_, columns=dft.columns)\n",
    "components"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's visualize those components on a map.  We'll use the `coolwarm` colormap (https://matplotlib.org/stable/gallery/color/colormap_reference.html).  We want blues to be negative, reds to be positive, and gray to be about zero.  This means we'll need to pass in a vmin and vmax that are equally distance from zero.  Let's first find a value for this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6830682912657138"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cap = components.values.max()\n",
    "cap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's add the rows of the components table as columns to the GeoDataFrame (`wi`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>OBJECTID</th>\n",
       "      <th>DNR_REGION</th>\n",
       "      <th>DNR_CNTY_C</th>\n",
       "      <th>COUNTY_NAM</th>\n",
       "      <th>COUNTY_FIP</th>\n",
       "      <th>SHAPEAREA</th>\n",
       "      <th>SHAPELEN</th>\n",
       "      <th>geometry</th>\n",
       "      <th>cluster</th>\n",
       "      <th>pc0</th>\n",
       "      <th>pc1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COUNTY_NAM</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Kenosha</th>\n",
       "      <td>321</td>\n",
       "      <td>Southeast Region</td>\n",
       "      <td>30</td>\n",
       "      <td>Kenosha</td>\n",
       "      <td>59</td>\n",
       "      <td>7.210454e+08</td>\n",
       "      <td>123267.303863</td>\n",
       "      <td>MULTIPOLYGON (((699813.437 246226.688, 699794....</td>\n",
       "      <td>1</td>\n",
       "      <td>0.112478</td>\n",
       "      <td>-0.083256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lafayette</th>\n",
       "      <td>322</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>33</td>\n",
       "      <td>Lafayette</td>\n",
       "      <td>65</td>\n",
       "      <td>1.641795e+09</td>\n",
       "      <td>164707.650040</td>\n",
       "      <td>POLYGON ((503148.082 260278.466, 503292.672 26...</td>\n",
       "      <td>2</td>\n",
       "      <td>0.013154</td>\n",
       "      <td>0.014847</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Rock</th>\n",
       "      <td>323</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>54</td>\n",
       "      <td>Rock</td>\n",
       "      <td>105</td>\n",
       "      <td>1.879382e+09</td>\n",
       "      <td>174114.587670</td>\n",
       "      <td>POLYGON ((600602.683 264347.425, 603850.419 26...</td>\n",
       "      <td>1</td>\n",
       "      <td>0.120181</td>\n",
       "      <td>0.062958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Walworth</th>\n",
       "      <td>324</td>\n",
       "      <td>Southeast Region</td>\n",
       "      <td>65</td>\n",
       "      <td>Walworth</td>\n",
       "      <td>127</td>\n",
       "      <td>1.492598e+09</td>\n",
       "      <td>154833.279357</td>\n",
       "      <td>POLYGON ((658404.520 263083.277, 658417.776 26...</td>\n",
       "      <td>0</td>\n",
       "      <td>0.065934</td>\n",
       "      <td>-0.023864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Green</th>\n",
       "      <td>325</td>\n",
       "      <td>South Central Region</td>\n",
       "      <td>23</td>\n",
       "      <td>Green</td>\n",
       "      <td>45</td>\n",
       "      <td>1.512855e+09</td>\n",
       "      <td>155741.104373</td>\n",
       "      <td>POLYGON ((571551.903 263810.562, 571555.279 26...</td>\n",
       "      <td>2</td>\n",
       "      <td>0.020049</td>\n",
       "      <td>0.014773</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            OBJECTID            DNR_REGION  DNR_CNTY_C COUNTY_NAM COUNTY_FIP  \\\n",
       "COUNTY_NAM                                                                     \n",
       "Kenosha          321      Southeast Region          30    Kenosha         59   \n",
       "Lafayette        322  South Central Region          33  Lafayette         65   \n",
       "Rock             323  South Central Region          54       Rock        105   \n",
       "Walworth         324      Southeast Region          65   Walworth        127   \n",
       "Green            325  South Central Region          23      Green         45   \n",
       "\n",
       "               SHAPEAREA       SHAPELEN  \\\n",
       "COUNTY_NAM                                \n",
       "Kenosha     7.210454e+08  123267.303863   \n",
       "Lafayette   1.641795e+09  164707.650040   \n",
       "Rock        1.879382e+09  174114.587670   \n",
       "Walworth    1.492598e+09  154833.279357   \n",
       "Green       1.512855e+09  155741.104373   \n",
       "\n",
       "                                                     geometry  cluster  \\\n",
       "COUNTY_NAM                                                               \n",
       "Kenosha     MULTIPOLYGON (((699813.437 246226.688, 699794....        1   \n",
       "Lafayette   POLYGON ((503148.082 260278.466, 503292.672 26...        2   \n",
       "Rock        POLYGON ((600602.683 264347.425, 603850.419 26...        1   \n",
       "Walworth    POLYGON ((658404.520 263083.277, 658417.776 26...        0   \n",
       "Green       POLYGON ((571551.903 263810.562, 571555.279 26...        2   \n",
       "\n",
       "                 pc0       pc1  \n",
       "COUNTY_NAM                      \n",
       "Kenosha     0.112478 -0.083256  \n",
       "Lafayette   0.013154  0.014847  \n",
       "Rock        0.120181  0.062958  \n",
       "Walworth    0.065934 -0.023864  \n",
       "Green       0.020049  0.014773  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wi[\"pc0\"] = components.iloc[0]\n",
    "wi[\"pc1\"] = components.iloc[1]\n",
    "wi.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Component 1')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmgAAAE/CAYAAADyjD+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9eZAsW1rY9/tyraW7er/7e/OGxyC20GDmGUYRkgKHwiMgJA0hhxDCYkYEIWxjK6TQHxK2ZYNAUmBHyCEISRPCYYlBBlkyNmIsg9AEDuxwWIN5AzIjYOANM+/dpe/tvfYlt+M/Mqu6qrq6u6o6syqr+/zi1u2qU5nnnMys/PI753yLKKXQaDQajUaj0eQHY9kd0Gg0Go1Go9GMohU0jUaj0Wg0mpyhFTSNRqPRaDSanKEVNI1Go9FoNJqcoRU0jUaj0Wg0mpyhFTSNRqPRaDSanKEVNI1Go9FoNJqcoRU0zUyIyHeKyNsi0hSRlyLyCyLyB5fdr0UhIkpEvvyabb5TRN4TkZaI/HMR2V5U/zQazeVo+XW1/BKRhyLyKRHZT7Z9Y3G904yjFTTN1IjIXwb+DvC3gPvA68DfBz66zH7lCRH5GuAfAN9FfI7axOdIo9EsES2/piIC/iXwHyy7IxpAKaVf+nXtC9gAmsCfumIbl1gA7ievvwO4yXffBDwH/gpwCLwEvg34VuB3gVPgvxiq6weBnwH+KdAAfg344ND3XwX8MlAFfhP4E0Pf/QTw94D/Pdn3V4A3h77/SuDTSZu/A3z7NPsC/xeggFZyLv70hHPwt4CfHvr8JuAB68u+hvqlX3f1peXXdPJrqB4r2faNZV+7u/zSM2iaafkDQAH42Su2+S+BDwNfB3wQ+Abgrw19/yCp4zHwXwP/PfBngQ8Bfwj4r0Tk/UPbfxT4n4Ft4KeBfy4itojYwP8G/CvgHvAXgJ8Skd83tO93AH8d2AK+APxNABEpEwu3n072/Q7g74vIV1+3r1LqDyfff1AptaaU+qcTzsHXAP9f/4NS6veIFbSvuOykaTSazNHyazr5pckRWkHTTMsOcKyUCq7Y5j8EfkgpdaiUOiIWEt819L0P/E2llA/8T8Au8KNKqYZS6jeB3yIWjH0+q5T6mWT7/45YOH44ea0BP6KU8pRS/wfwL4A/M7Tvzyql/t+kvz9FLHQB/hjwrlLqHymlAqXUrwP/C/Cnpth3GtaA2lhZDVifoQ6NRpMuWn5pVg5r2R3QrAwnwK6IWFcIuUfAe0Of30vKBnUopcLkfSf5ezD0fYdYcPV51n+jlIpE5PlQfc+UUtFYW4+HPr8aet8eqvd9wDeKSHXoewv4x1PsOw1NoDJWViFebtBoNMtByy/NyqFn0DTT8q+BHrHdxWXsEwuQPq8nZfPyWv+NiBjAE87tQ15LyobbejFFnc+A/1MptTn0WlNK/Sc36Ocwv8nQKFpEvozYtuV3U6pfo9HMjpZfmpVDK2iaqVBK1YjtLv6eiHybiJQSe4pvEZH/NtnsnwB/TUT2RGQ32f5/vEGzHxKRPykiFvCXiAXsZ4gNX9vAX0n68E3AHydedriOfwF8hYh8V98eRET+XRH5qin7dAB82RXf/xTwx0XkDyX2Ij8E/K9KKT2DptEsCS2/BlwnvxCRAvGgEsBNPmuWgFbQNFOjlPrbwF8mNpw9Ih7N/WfAP082+RvA28BvAJ8j9lz6Gzdo8ueAPw2cEduC/EmllK+U8ogF2rcAx8Su8h9TSn1+imNoAB8hNqTdJ14O+G84F0jX8YPAJ0WkKiLfPqH+3wT+Y2JF7ZDY9uz7pqxbo9FkhJZfwDXyK6FDbKoB8HnOl3M1C0ZU7FKr0eQKEflB4MuVUn922X3RaDSaWdDyS5MGegZNo9FoNBqNJmdoBU2j0Wg0Go0mZ+glTo1Go9FoNJqcoWfQNBqNRqPRaHKGVtA0Go1Go9FocsatyySwu7ur3njjjWV3Q6PRLIjPfvazx0qpvWX3Iw20/NJo7h6XybBbp6C98cYbvP3228vuhkajWRAi8t71W60GWn5pNHePy2SYXuLUaDQajUajyRlaQdNoNBqNRqPJGVpB02g0Go1Go8kZWkHTaDQajUajyRlaQdNoNBqNRqPJGVpB02g0Go1Go8kZWkHTaDQajUajyRlaQdNoNBqNRqPJGVpB02g0Go1Go8kZWkHTaDQajUajyRlaQdPcGK/XQyk1UhZF0YUyAN/zaLeaE7/TaDSaRaOUotfrXSiPwvBiWRTRbrfxfX8RXdPccbSCprkxhmkO3kdRLNRiBWxMaQtD9p+9x+HLfVQULbKLGo1GMxERwUxk2PDAMpowiDw6OuLVy5d02u2F9lFzN9EKmuZGRFGEZVmICGcnx3i9HlEUYZomQRCMbOt5HkopCsXSiFKn0Wg0y6CvkFmWRbPRoNvt4nseAIYI4dgsWq/bBaBULi+8r5q7h1bQNHOjlGL/6bt4iUCzHQczUdYAup0O3vDSQVyMZVuL7qpGo9Fc4PT0lFq1Gn8QoVAoIEb8WIyUolGvj+4gMjLjptFkiVbQNNeilBqMKocREUpr6xgidFotgiDAth067Ra9bgfLsvD98/163S7Fcpnt3XuL7L5Go7nj9CbIL4BSsYhpmkRRRKvVAsA0TapnZxdWAXzfR4BHjx4NBqEaTZboqQzNlVRPT6jXquzdf4jtOPS6HWzHxTAMgsBne3cPgE6nTbm8RhgE8UyaGc+kDQuy9coGG5tbyzoUjUZzx+j1ehwcHlIoFLi3t0cYhgRBgOu6BEFAsVRCRPA8j1LyPvB9ymtriAg7u7uDukzT5LXXX1/i0WjuGnoGTXMlpfIapVJ5sCypFKgoim03IoVSCs/rsbZeQQyDo4OXGIaBYRiAotNu0U5Gpoahf24ajWZx2LZNuVymWCgA8bKlIl4VCKNooJzZts3a2hrHR0d4vo9t2yilCIKAWq2GUkrLL83C0TNomolEUYhhmIgIW7t7GIZBvVbl9OgQMQxef/+b2I5D9fSE6ukJWzu7lMpr7N1/iNfzqFcP6LRjxay8XtFGtRqNZmFEUYRhGIgIa+UyruvS7XY5Oj7G8zwe3L9PuVzG932eP3uG4zg8evyY9UoF0zQ5PT2lXqsN6llfW0O03ZlmwWgFTTMRwzBRSnFydEC328V1C/S6HSCeQXv14lk8u1Zeo3p6ytnJMesbG3Q7HQ5f7g/qKJZKbO3sXtWURqPRpEp/tqvRaHB0fIzruvi+T5SE9zmrVjmrVnn08CGu69Lr9Wg2Gqytr/Ps6VPCMEREsG2bvb097XWuWQpaQdNMpN1qcXp8SJAEZOwrZ3163S5KKUrlNR48fgIoRASlFGvrFSqbW1i2jdfrYVnnPzOllDaw1Wg0mRIEAUdHR7Q7sdwaD0Tb/9zpdLh3/z6+7+M4DkopXNdlY3Nz4Dzguu5gPy2/NItEL6rfcYKxiNhBEBCG4Yhydhler8fJ0SGFYhHTsqmdnVFeW2f3/gMcN3YkKBSLQJxBoN53Z9doNJoU6NuJ9QnDkCiKqNZqA+XsKl4dHABQLBZpNBpEUcSDhw8pFos4jhOH3RAhCkMajcaF2I4aTZboGbQ7TBwNW2HZNp12m2KpxMtn712apmkSu/cf0G41KRRLrFU2gHj2zTRN3MQw1/c8zk6P2dm9p0efGo0mFXzfp9vtsra2RqfToVAoUK1WqdXr08uv3d1BMNq1chnLsgiCgE67zXqlAsRK4OnZGcViEdu2MzsejWYcraDdUVrNBkevXrK9u4fv+dSqp5imdSFy9lWsb2zg9bo0ajVK5TUMw8Dr9Th8+QLTsnjw6Am242A7DvcePMrwaDQazV0iDENe7O/jOA6WbXN0dIRpWfQm5AW+DNM0KZdK7L94waPHj7ESz82DV6/ieoBKpYKIsLur7Wg1i+faJU4R+X0i8m+GXnUR+Usisi0inxaRd5K/W8n2IiI/JiJfEJHfEJGvH6rr48n274jIx4fKPyQin0v2+TFJplkua0MzH2EYUq9VOTk65Pgwnto/PT7i9PiQMAjwet2Z6isUS4RBwPbePZRStFtNXu0/p7S2xsbmFmLo2TLNctHy6/aglKLVanFyesr+y5eEYUin02F/fx8/COgmdrHTUnBdoihia3s7njnzfV69fEkURWxvb+vZMs3SuVZBU0r9jlLq65RSXwd8CGgDPwt8P/BLSqkPAL+UfAb4FuADyet7gU9ALKyAHwC+EfgG4AeGBNYngD8/tN83J+WXtaGZkW6nzYv3vsTp0SGNWvXGycrdQpFSeY31jU1EhFcvnlOvVnn05HXuPXgUOwlYWsBplouWX7eDKIrYf/mSVwcHVKvVQXq5eRERtre3cRyHtbU1Tk9OeP78OZtbWzx57TU2t7YoJvazGs2ymNVJ4I8Av6eUeg/4KPDJpPyTwLcl7z8K/KSK+QywKSIPgT8KfFopdaqUOgM+DXxz8l1FKfUZFQ9/fnKsrkltaKZAKUWzXqNRq3L06uXAxfymGKbJ/YdxupNOu83J0QG9bocoCgmj6ZdINZoFo+XXitHpdqk3GhwcHtLtzjbDfxX39vZwHIcgCKhWq1SrVaIoIggCbSeryQ2z2qB9B/BPkvf3lVIvk/evgPvJ+8fAs6F9nidlV5U/n1B+VRuaK/B9H6/bpVGv0p3Ck2kWCsUi9x89QUQ4PT6iXj3DLRTZ2buPaZnYtpNqexpNimj5tQJEUUSv16PValFvNGZatpyG1197DSuxV3vx/HlsY7YXp6zrZxzQaPLA1AqaiDjAnwD+8/HvlFJKRNK9i2ZoQ0S+l3g5gtdXIFdap9WgVT2lXNmiuF65UV1RFNKqVfF7HcR2CfyAVrORUk9HEcOgsrk9GGGKCMVSmd37DzB1IEdNjtHyKz38IOTotIEhwv3dyo1mnJRStNo96q0urgUqCmi1WgQzOCvNQqVSwbLO8wQbhsGDhw8paMVMk0NmmUH7FuDXlFIHyecDEXmolHqZTPMfJuUvgNeG9nuSlL0Avmms/JeT8icTtr+qjRGUUj8O/DjAW2+9lamgvQlRFIGKqB7sEwY+3XaTXfN9FEprM9WjlCLwPTqNGs3qCVEizEKrcGPbsqsQoFgqDT5vbu/o5QDNqqDl1w1RShGGEWe1FtV6e1D2YG9jZjkQhhHNdpfTaotOL463uF4UvF66M/7jlIrFQV9t2+Z9b7yhZZgmt8xig/ZnOF8eAPgU0Pdk+jjwc0PlH0u8oT4M1JJp/l8EPiIiW4lx7UeAX0y+q4vIhxPvp4+N1TWpjZUjDAIO3v1d9n/v84RBEgBWKU72n9KuVwdK1lUEvkft+ICDd9/h4N13qJ8cju6X8lLAOFEUDRKfR1GE73mpLz9oNBmh5dcNeXlY43ffPeD4rDkoO6u3OTiu0/OuDmoNscyo1tu89+KE3/nSK14cVAfK2aJoNOLVBaXUSOonjSaPTDWDJiJl4N8H/qOh4h8B/pmIfA/wHvDtSfnPA98KfIHYY+q7AZRSpyLyw8CvJtv9kFLqNHn/fcBPAEXgF5LXVW2sHIZpsnnvESf7T0fKVRRx+uo5tltg5+FrWE7s+u1124RBgGlZRGFIt9WgXa8BlytEi1CW6tVTymtxzDPHdanX61QqN1um1WiyRMuvdLi3W6HV6eEHo4PJ01qLs3qbR/c2qazFS4U9L6DnBRiGIECn53NWaxGElytEcoVsS4tWu43neTiOg+M4NBoNyuXyIHenRpMn5LbNgLz11lvq7bffXnY3gDiNkpXE0gkDn4N3v0B0jZej5bgEvjfzbJgCQiN7A/3y2jp7Dx6et6sUQeBr5wDN0hCRzyql3lp2P9IgT/IrDCMQMBPl5fmrU+rNqz0pbcskihThHDNT60VmjsU4K6Zp8uTx45H8wEEQYBiGVtI0S+MyGaYzCWRIu35Gp9nALRaJInWtcgYQeL1rt1kGYhhsbG6xsbU9Uu71unTabSzbpry2ru05NJpbQhhFPHt5imNbOLZJq3197LHx2bU8US6X2UmC0vYJw5BmM16yXV9f185OmlyhFbQMUUrh9zr4GRu+ApkrRuW1dTa3dy6UK6Wonp4A0Ot02N7T+TY1mtuAUmqwVLmgBjOrWkS4f++ibBIRmo0GnufRqNd58PChziCgyQ16TjdDlFqcAapItpeyMBZVOw6CW+dg/8WgrFGvpRoQV6PRLI/bZP1SKBQuKGee5/Fyf3+QlcD3ffZfvMDr5XMVQ3P30DNoGaKixUk4EeMq/4Eb4bgupXIcCiSKIurVM+rVs4mKWLvV5GD/OfcePtbLBRrNChMtWENTGQkwEWFrK87KpZSi025zeno6MV1UPwn7/fv3KQ2FFNJoloFW0DJkoQ4YQmYKWnkt9tJs1GucHh9dG2ut1+3y8tlT9h48xC0UiKJIG+BqNCvGwh3IMmrOdRwc28bzPA4PDq7N46miiFcvX7K9s8PGxkZcppSWYZqFoxW0DFnkEicZ2n2dnRzRbNTwZ0hQHAQ+L58/pVBep91us16psLOzo4WcRrMiLHIFIGkxk1q7vR4HBwf0ut2ZlM7TkxN6vR7tTgfLNLl37x6u62bSR41mElpBy5IFjkBj+4rs2psmkO4keonbfKNep91q4RYKbG1t4bouSik9MtVocsrClzizbC6RNbMSBEEclzKKeP7iBaVSiUKhwNbmJoBeHdBkilbQMiTLtEsXG8tr9ecze2EY0m61iKKI7a0tWu023W6XR48eac9PjSZnLD5GZnbtzVvzuFRqt9t0Oh2KhQKe53FWrXL/3j2dy1OTCVpBy5DF2qAZQIYxiFI8lm6nw34nDj1imia9bhfDNHEcHexWo8kLCzdBy6Hb6KQeKaV4sb8/+Ox5HlEUaacCTepoBS1DFmqDRrZtxWE80m8jDEP29/cpFos4jkN5bU2PRjWaHLDoJc5VnUU/Oj7GMAw2N+Kk8RsbsyeP12gmoRfPM2SxI8JsBcK8Amfa3TqdDrVajaPDw1yOpDWau4a+D6cniiJOz844uSR8h0YzD1pBy5DbpKBlbuSW4Ps+jUZjIW1pNJrLWbQXZx4Vwnmk6vHJSer90NxNtIKWIQt1EshYP5tXdM4jdE9PTnIprDWau8SilzizZF7xOM8Z6Ha7tFqtOVvUaM7RClqGLDTVU9YNzC2rZ+9ZFEWcnZ3N26BGo0mB26SgLfpIjvUgU5MCWkHLkEXeoOqWLHH26XW7C21Po9GMsmgFI8v25pWO8+4XBAHhnLEjNZo+WkHLkEUKuLw6Dc17Bjqdjk5arNEsET0BdLNhaa1eT60fmruJVtAyQimlJRzc6BxUq9UUO6LRaGZBOwncjFqtRrRIO2TNrUMraFlxy4TNMo6m2Wzi+/4SWtZoNNoG7YZtKkWtVltCy5rbglbQMmLxo8Gs3TiXI6y73a4ehWo0S+A22aAtC8/z9CBTMzc6k0BG3DZhM/fR3PA8HB0ecgQUikVKxSLFUgnHcXSkbo0mYxYrwm6XvOzTbLVotlrYtk2pWBwkW9cJ1jXToBW0jFhsmiey9xKYU1qnpah2Ox26nQ6cnmKaJsVikXK5RKFYxDT1z1ijSZvbNshcJr7vU/P9geNAsVikXCpRLBSw9YBTcwn6yZYRt024zXs8SqnUlccwDGP7tF6HwOvhuC6FYplCKVbY4ryhGo3mJtwmG7S5yegcdDodVBRxdnSAaZoUSmWKpRLFYgnT0o9lTYz+JWTEwmfQckg8Ksx+ZOj1eni9HvXqKSJCoVhKBF4Zy7b16FSjmQMdJmgxC69hGNJq1Gk14tk1x3UplsqxHCsW9IDzDqMVtIxYtIt6Xim4TizkFCgUKlL4QZBZe0opOu0WnXaLM8C0rHNhVypjmmZmbWs0t4nFKmj51NBEBNdxBp8V8YAwS42yP+CsnZ0POIulEoVSCdvWy6F3Ca2gZcWilwdyeNMqpfC6nZEyxy2wSJ+mMAho1ms067WkfXcwu+YWilrYaTSXsEgRJguYaZ+HZWc0GR5wApimNVDWinrAeevRClpG6CXOS1iyHB4sh56NLofq0alGM8oibdBEVLbribfEni4MA5qNOs2h5dB4hq2Mq71Dbx1aQcuI2+YkkB75UYAmjU4L/dFpsayNdTV3moUuceZILlxLjgZx5/a3Z/FybLFIsRjLMMdx9YBzxdFPoIwIFYRGPCMTvwAEMQRRgIBhmERRRN+WXpQM9JeLt1WyU3+Y2feO7AtRMXAcFxUbew3+xnv23/f3SUrV+WtRSI7jHYVhMGKsaztOvBxaLOMWi3p0qrlTOHRBFGIY8YNeJFakkoe+IGCYRIrku2HGJJgkskvF38VSQCXvFYaAoRJbr7696pBcmiSj+mU6kHWMUopuu0233YYTMExzoKwVSyUsy152FzUzohW0DFFMyMkZnr+1HSO1KNNuwcbzbpBcPBG+sRw2YoVx8J1B4HtpdJMgCCi4sSCO61d4QUQYhlftthR8z8P3PBrVMxChUCgmoTzKOK4enWpuL7HcihUfFYWXD6tMhzC8uYJkWxa9YH5Z2PcYHx4Qn8uvdIeFBdeN3ySDZAX0ejeQvRkRhSGtZoNWswEkA86+w0GxiGFo+7W8oxW0jFBTjOrSnda/oQhSKhm1Aoz2PU1FJAwCwjEvTtNxyaF+NopSdDttup02cIxhmmxu77BW2dSKmubWMfWselqz7ze6hSTpxuWrAaaT3uzRuOOAWyikVneWDAactSoAa+sVNnd2sbQpR27RazYZMZWB7ao813OqgCzTbiUKQ06PDnn57N2BDZtGc1uYVkFLK5xQ9ndyfk0rlkWzUefFe1+ienqil4lzilbQMqDX68Z2ANeST8VHMz2+53G4/5zD/ef4N1li1mhyQhiGNBqNqbZVqSk+WhYuA6UU1dMTXjx9l2ajrp3bcoae20yZIPB5+ezpVNumOjGV4X015JqQCcacJyJPoqTTbtF52qKytU1lc1vHJ9KsLIcHB3Q6nes3JMV7cJX1s1ug1IRBwPHBKxq1Ktu7e7iF4rK7pEHPoKVCFEWDkcd0M2fpo7IUcDld4swj3U6H50/fo1Y906NRzUqglBoscSmlplbOYlZDNug7cTp63S6vnr3L8csXBCk5sGnmR8+g3ZDA99l/9h6mZVHZ3KJZry+nI1kqA1rRmJ7kYXd6ckKj0eD+g4eJE4HCNOPbTTsVaPLEwcEBnXabjc3NpalbWd8T+o6bjXazTqfVYPv+Q0prFbxuF6dQQCmlww0tEK2g3ZB69Ywoiog8j5PDgxn3XhGxkbVCcZsUlqFD8T2P50/fw7ZtTMsi8H1My6JQKLC2XsFxHKIoot1u43senu8jIty7d295/dfcKXq9Hu1W7ORSPTubaV8ZxFS8OVlLgEyTFGRY9zJRSnHyap+afYxhGigFvtejsrmNUyhRWlsDoNloICL0el3arRYPHz3WQb5TQp/FGYmiiKNX+ximSam8Rj1xWZ4HpaLRYLM3IUMpYYiQpY/P3Ie/Inqd7/uDeHdBENDrdgnDiJ3dXVqtFkeHhyPbO46D1+uxtb2Nbdv0ej0cR6eh0qRDrVql2WyyXqnQajZvVJdjmXjBzWPkZK7kZLgKcOvuyrFnUuB7DCdQrp+d4JZ6uMU4l/Hx0eGIOcfp6QlRGFLZ3KRYLOH7HqZp6Zm3OdAK2gwopWg1G3QSO7PWlJ5Ol+H1eliWRTAWF2wexMhOTERRlBiNTormnbRPbAcnQ2Wj3gX9TAgX34th4DjOxLZF5IIt1+BIo5uft5viuAUwjMFxTpu/0Ov1ePdLX5r43enJCQDtdhvXdel0OpRKJfbu3Rs4HyiltMKmmZkwCDg7i2f9e0dHN6pLKYVlGakoaJ7njUiFtDFME9c0x0TY8AeZUDa83QRXqSQxgjEkv67r//Ax5iGrimla2Pa5GqCUwuteb0etopCn706WX83kudhutykUi3i9HiLC3v0HFIvFQTtafl2PVtCmoNvt0Gm1aDUaBDeIdj2J5d+i1xNFEb3u9YbDYlpzxtO5/pwaIrlQyC4gQrc7e3iNaYRTFEUDg+12u83+ixeUymXCMEQpxdbWFp7nUSqV9OhUcylBENBpt2k0GnTHgqzmhazNXDvd6zOhFF0HfwrlZJxppZLjung5yzhgmAZeZzSO4zRqUzRlZPHukMPJ0cErHNfFdQu0Wk329u7FNm2meekA/a6jFbQrUCri6OAVnVY7Xo7MpI1VUNGmY5rsCbePxY0Cfd+nVj1fUu8vT9m2TWVjg0qlokelmhHq9Tpnp6e5TKU2TJr2bPN3IuMOLPv4lkwYhnTa7cEK1P6L50A8A1kql9ne3tG2a2NMNewWkU0R+RkR+byI/LaI/AER2RaRT4vIO8nfrWRbEZEfE5EviMhviMjXD9Xz8WT7d0Tk40PlHxKRzyX7/JgkT5nL2lgEntfj8OVL2s1mZsoZoD0kV545r1+KipTv+5wcH/Ps6dNUlstvG3dRfkVRxOnpKSfHx7lXzu4MeRw75eDxE0URzUaD58+eUq/Vlt2dXDHtusiPAv9SKfWVwAeB3wa+H/glpdQHgF9KPgN8C/CB5PW9wCcgFlbADwDfCHwD8ANDAusTwJ8f2u+bk/LL2siUMAx59eL5QlL4pHd/5ODuz3D2Jr8zjfMec/rHE0WRDpA7mTslvwBOjo+pnulYfDORaTDJvJKf30cURTov6BjXng0R2QD+MPDnAJRSHuCJyEeBb0o2+yTwy8BfBT4K/KSKJcNnktHrw2TbTyulTpN6Pw18s4j8MlBRSn0mKf9J4NuAX0jqmtRGpjTrtanX2G/KbRKgkwz600IBBcdhkFzm/L+L26pxh4ThLdXElYx4nzhZvFIRgsH4L8AtFAi9HjJs75UjmR5FEYcHB+zu7WlFLeEuyq8gCKZO1XRTUkv0lJIz+806kW31pggFu39fqrE/QwefyLb+eHfieRl4Z8UnTiVyTSXvVRRiFCsjNm+GYWBFPmKag/MtqAtybpmcnZ4CUCqXl9yTfDCNuvp+4Aj4RyLyQeCzwF8E7iulXibbvALuJ+8fA8+G9n+elF1V/nxCOVe0kRm+79FsLDDYbGpSadnSLXtl0+/NEuH8hlwirKMwYFiiWbabbgM3RES0w8Aod0p+KbURA60AACAASURBVKVmjmd2swbTqigHI52sTdCigLCXXaaZST6pw0RRhIoC1JCzlWGn0Vp6BIGP484rU28f00hyC/h64BNKqX8HaDE2Va/60w8ZclUbIvK9IvK2iLx9dEPX8Uathu9d7/GTP/Ig4DKMNbRg4/dIhdimgQlYArZlEU3y4J37kNM/V67rsru3px0FRrlT8svzPOrLymZyA3Lxk83aSWDReG0cQhwCHENRsM1JgZLmqjoLUS8i3H/wUC9zDjGNgvYceK6U+pXk888QC7yDZOqf5G8/2uYL4LWh/Z8kZVeVP5lQzhVtjKCU+nGl1FtKqbf29vamOKTL0Q+3fLLo6yJAGPgoFRJFIaHfI/TTVNzTPR7DMHjw8KGePbuIll+a6chaP1vwGq4Kg2TGLEQFHmGvfVHqLH1d+Zyd3T0KRZ2kfZhrpblS6hXwTER+X1L0R4DfAj4F9D2ZPg78XPL+U8DHEm+oDwO1ZJr/F4GPiMhWYlz7EeAXk+/qIvLhxPvpY2N1TWojEzyvR+3sNMsmNPOS2wfPvAIuXcEYRREnJycjy8xKKcLwbnt13iX5BVzISpE1t8rJKXPyowydM2efMlDs6rUqvd5onL7wjnulTzuX+BeAnxIRB/gi8N3Eyt0/E5HvAd4Dvj3Z9ueBbwW+ALSTbVFKnYrIDwO/mmz3Q32DW+D7gJ8AisTGtb+QlP/IJW1kQqeVvdfmMFka1d828iq+8zRj0Ww0UEpRLpXwWjU6jRqFtQo7D1+7fufbzZ2QX0EQ0MtZINRpydFtdLeYVz/LIO+D53kcHR5SqWwgkU/z7IgoDHn45lff2ZUBuW0KwltvvaXefvvtmfdr1uscH77KoEeXk6aC5haKU0X7zxKFZCZpTdNE/OUe3ySs4jrdOR6KbqEwVwaCabAsC6MXe/FZtkNl9z6mZSFiYDkOhnG7PDxF5LNKqbeW3Y80mFd++b7Pq5cvBzlfF4XtFOh6N5/lsG0rzvm4RIquPVcmgWkp2AbhIh2dpsAwLfw5wnwahok/dZSu2SmYEHkdEGFta5fiWoUojLALBQzDzNXAOA0uk2F33hrP8zzOjo9minlmmCYoNWdao3PSnUG7XYr2OFnm6bsJecind4GxRMenL4ecD0W4//qb2G5hCR3TpE0UhtRqNaq12tSZPESMJMyCurH8uZOhw24R81/+Bck9pWieHtE8PXeeWd++R2U3c4foXHDnFbTjg5cz50ezQw8V+rHSIEYcF6v/F+L3MjybFGcR90I1uqae6ijglkvKnM705rFXV/ZJKapHr9h5+Fo80NCsNI1mk7MZw2rYboFGt6/MKUwxMAwwBBBBYnFFkg4BI/lFCRHemI1QLm+A3JJHGT3f0Dfry35Vj1q1E9zyGm7x9sdKu9MKmlLRnCE1YuEmACpChfHn6360RmF9JCigpHrD3m5Jmdel+HmvYOD7OI49dNn6AXiHAusqhRkFoMZDScrgjyTLyiJG8lQVUNGVwSd77Sb7X/w8W/cfUa4sLPuQJgNunnxbCJXiPC735ffZmntxSSu1QLW5VF5SJofLcqNBvWfYL4ooOHEQtYHESgLsqqEg4pZtE/a6jP5SZOiPxNfeMBBASfw59McGAkNEYcjxsy9SKK+z8/iNmfu+StxpBc33/bke/CqaM3bMeEH+7lfNjMz7gArDcKociYby4Yqlq/lXKBS1o1dYtotbLM1bi2bJ9OYYYKp5Bc+k3dIaOOVBFuZzDJgx8x+0710/OLAsCxVebRc5bw+6rQb1kwMqO7d3ufNuukYkdNvzGYTOnTx9TJilaeiYC9mSwxFi1mR9xJkZw4pQWt/E1EEhV5YgCOabQZtTqcpSP8sFd0985eO5MSdOsXTrlznvtHSOui3ssIuIIIaZ2I4ZsaJhyGD5qG9DRvI5bF0+/XoVmS7TLflOE8mluXzm5HXp9ToMw6Sye//Ouq/fBjw/pEeB2I5MMIzEfkwkHnn3bclEABXbmAFekL/fbC50oxW9l29EnJAzM7K8rmtbe7iltQxbWD53WkEzTDOxI1OoJKBnlreoG3ZQUZjYDAnK93ESB4NzRXBIKST+o5J1+uG+jbxXIIbgOO550t2xAxlNxntuJ6CGy/qVDYpmOBsiS1cSbyfZiLgoDKge7rP94Mn1G2tyiWn2lWshVBCG/RswozyJmIRSGIgpQ4RIgWWbiR1R/9eaSKvB7K8asjEbli+DdxiGgeM4gy8mH4E6329sm76sGpVZ6k7qXHkiy9NfPXiOU/wKTPP2qjG398im4KZhMmZFhQFEsd3RrKJ0fDsZey/lTXopxCOa1IYafJbzycQxQ3UxDKJwsefzbpDdGLRdr2I7Bda3dzNrQ5Md0Zy2sPMiiUPBsGMLwJUeKVNScAyCBeVAHjYb6HuqxpOMWn6lTZbOH1EYcvL8XfZef/PWxUXrc6fXNxYesHNFlzjP5/NUMvumUCqKX1FEFIULUHbzeQOurA1aQq+z2OwZmvQwjMXeE5fNa6XCAg9FDWSYIkriWYZ3dHCZ9WnPegjh9zpEUzhbrSp3egZtffcexcoGvVYTRLAdl7OXz4gyyl+o1Nz+U9fXnVG905K9fF32EU4m815lpqDFdkml9Y2M6tdkjevYfPnr9+n0fLo9n7WSy2mtSWNOG9lryectuDLkc4i5wojgFsu32tHp9h7ZFIgY2G4R2y0Oyvbe+ADHT3+PMJO0I1lKuGV7CSy/C8sh24POaonAdly2Hz7BtO1M6tcsBtu2sG2Lylosw4oFh4PjGmd1PTM6O9ney3kVj9nmhM7uqHcevo51yzOi3GkFbRKmZWPZTuoKmohxq0dQfU+xLLHdwrWBr4d7IcMl/ZmoS/YPPI8oyuFUeUY/Gt/rUj16xd6TN7JpQLMURIRiwclEQctyifM2y8Y+hmEiYwrFsPOXDP4b3WIaVBQSTBGXbDJZyu7sruzJ/ns8+LKvzKz+PKAVtAm4axV67Wa6lWZsS3QXvJX88TQzKWI7LpE3h4K2wuc9m1lizbIpug6GIek7EWT5W8+Bhpa1DA0jhZ+yI1cft+ACcypoK7z6Efo+pnV7VwHutJPAZZQ3d7AL6UZXz9rYe0Xvr/ww5/XJ1HAaMlXsA9+bKhq4ZrVwHIu97Urq9d72GbQ89GEZZJtmK1v52G3VM61/2WgFbQIiwvrOvdTrvM2sasDWPvNen+zD82b3uzFMC+sWjz7vMhvrJWwrZS/11b7Fr2eFZfS86QeBldZMnVueSUAraJdQWKuw+/qbpPbrzXwG7XZLz9t9dFeQ4c9mfXsP0ZkEbiWmYfDGkz1KBSe1Old9ELZ0MryX7+KVcYplnerpLuMkLrxhcHWy12nIOhXSsmVn5k4CWR+gYWC5xbGH0JBEFQUqPsb+bJsk+7muNbJZJDJfjsSJZCPVtx88oVTZzKTuWYjCEDGMwTmNwpDa6TFbe7c3AfKisEyTYsGl3U3H1vD2KwGrfIQKxy3EgdD7MmxsUmDUGcFINhGUCMo0zk3RxKCbkvzKSmwX1ipsP3gtFwPMMAxGshlUjw9Y39xJJfyHVtCuIauYaGmz9NHtArw4s0RFCi+lSOZp2y+mje0UcqGcATTrVVDQ7bTxvR73Hr+GW8z3+VslwhQDSC9dxmTMSh+eEsJuI5WqxClev9GSqezcy4dyFgTUTg5RStHrtKls7VBIMT/o8o8wx/RazdSE0q13Elh6B25Gmpcn7/aGvtfNjXNAr9OhenJIt90kDHya9RqltfVld+tWEAQh3ZRmz+D2K2irTJpXJu/yC6DTqC27CwD0um1a9SrtRo0w8KmfHeMWS6kFz9UK2iWEgc/Jiy8tuxvTo4XnzVDpCaW8P8gsx8V23Avlfq9L4/R4oX1xCudxoUzLYn1ja6Ht31aUUrw8qtL1bm6ecV5nalXdUfKv+KwKhfJFL2WlFPXjV4TB4la9bGc0rt3Gzv1UFVy9xDmBKAw4fvqFdCVS5nHQlrzEuOqyR9I9dwM7teHEzONlY7+JSTEqDQFxCgxf2/kDUsZUtvculAWex+HTL6JUBCjWJ2yTBe1Wi+L6Bp1GDadQvNVpWxbJwXGNZjvduIHZDjzyoP3loQ/zkuazijiBPAxk1CCpPEL8T863Hdv1/INgChjJYLCf6jAIA9QNlt7d0hp2YXQZVinFyYt36bWbdFsN9l7/8oXMBPZ6XcRyMA0h8HoUyuk6LWhpOAkRQj+9kSesdq5Kt1Ck1+0gIsmPXuKbdXBQAkphW8ZoTJ1h5SS5seN861HcWxWNuIfH8n/oM3FCY+kLhhXB77Yn9lYN/bVsm2gs8O6kKzi+QHVToSOGgdftYLsFuu0mURhiWjbNs+NEOYPG2QlrW7sLEXDbu3uEYch6ZYMoilZieWUV8DNIIH1NEo+bV54Rtm3j+x4iMkgwP2ngFKoIwymciy3FwLZWRAZ9VCQyTCXe8/3yMQW2n5Bd+spOlgpuilVH3TbT+P66EqF67Znrl8p9vBsMMsUw6LWbRGFA4HkYloXf7QyCy/u9Lr12k0I5e1OJYqmEee8BoIjCCMNIN7SNVtAm0K6dLbsLM6OUujAjk0Ubi16+6ws2MbJ+cN8NxUBFEc3qCa3a2UAhG8cpFBenKInguAWiKKRQut0u84vC8wPanXRtDFfcBygRjWqK7ArpOVX0G1b0T12GJ3Ap4mvORm/Y126zntiHT75WIgbWBBOOLAiDAKdQxOu2KVc2Uq9f26CN0aqeUD/cz6Dm1VUA8iCX8+CxkypLPqmXCTeQTJc3vW6HMAyonx7T7bSxbDueTUw7LdEdxfMC3ntxlPr5zH4G+3ZnKQButKynGeVy+QWljS0sO734f8N43Q5KKVqNGo3qKaZlIyI4bjG1KADD6Bm0IZpnxxkpZ5D5EznDGY/so+VP04c89CIHZHwS1rZ2MgtzoZTicP8ZghCGAZbj8Oh9b6KUolDMv2t/3vGDkC8+O8jkJyIiq+skkBcNTUuwzDEtm43dB5nV36yd0W23CMMAlMIpFHELRZRIJvazWkEbwnaze0hkKiNEKFugpB9sUPpWncC57YgasQ9L/io5d2BUI3/OyYVcybgTC3/6zNeeGr2KqZNVfLQg8Al9n2jINmp9Y3tgo6O5OZZpYFkmfpC+/VnWSo5pCMULWQ+ma3TYanX4tuq/NURI16J4Xm5ZKoEc3rfF9Y1MVluUUvi9ODxRP3C9UyjGwYHJLjSJVtCG8DqtzOrO8v4REWieDm7/8Z/KZeXTEm5kNyKZlszlz4JlzbzHk2U37UvCb6RBs1alPhbCI0gMtzXpEIRRNsoZ2d8eURgSeJ1M6radbJa78kQuxtA5ILMBpu9x8Gw07FYYBigVIZJyztshbplhz80wzOz01Wxn0DK+jDm4+1OOgrF8cng8bmktM4Wp1z4f/KxvbbO+uZ3LEfgqY2boSJN9oO1Mh7AZ1p0PVslHICtMy8a0slHGe51zb1XDtLj/2vtJ4pFk0l4fPYM2TBR7ZAzT91oc/I2ic29GFU3t1Zj5DFqW5OFGzEMfckA/llAWzOuWHoYhoEby0Y0zvOxgWQ5rG5u5D+i7agRBQLlgj4Z+4Hz1XimIEtkVv+LPtx4tOzIiXyfWKZUxzPlms3yvd+XqwYj8sm1st8DD9705V1uzoBW0IbqnB0TXxGeZpDMrkgsoSRww6ccJ68cMi43cjeIa/bB/V+pUw0Jz8F6BUoP4Yf14YnEIimxn0PIgw7MdYS+DOY9HJJMLYjuFC4OTaei2W5wdHbD78AmTZGPg+zRqZwRDcQUN0xiKqadJi1azSXBNPsZJkiK2UZXza5L8VZwHKzXNCGUkNSSBTM8ZT8ot4+ZgidxSQwqjGlEYydDDcZViKN4Fsroa5Y3tmfdRSnH04iluscTGzkXvdaVUnMqpWR+UGYa5MNmlFbQhChs7tI5m9+IUSARMdFEoJZimjd9LJzaRqzxU4N8psZMHJTFNwjDEcgqYhoHfnS3Yo1g2fTEnIliWdWP7SYWaa/QpIiMBZqPkQdttNSmU16idHNEay5t31UybZn7W1tc5O5s9hmPflag/qzYJU5wbZ7DoUywU6Ha7COcKY3gXIlBk+FBfhnj0xcJc30Uas6WHM5onFI1EBogBhtA13BuHIZF5TH36gYSHLk0YhvjdDrbj4vs9zg5fjuyyyGwn2gZtCBVlY2Cb1J5iVbdMW7mjBL4/l9COwpAoDIjCgDDwB0rRjfri9WiczZ6H0y2WePS+LxsoaEopwsDn+NULDp+/RzR0T4lhsH3vIc1a9cb91VxkVeJsLd5fOh/yMmsrvkUTzSt7wgD8bvzy2tBNxzmvfvxqxEt8GsQw2Hv8OqX1UeeC2ukR++++Q68z6riysXsfr9cdeHJmjVbQhrBL2aWGSPX20QpaBtyl+cjJ1I5ecbL/dOYHvRjGYFRpmuZgf6/XpdNqDrZTUYTtuGzu3Uuv05oBtuNgZGTusMrL0avcd8309NpNDp9+AW/GFQmI7cr6xDH/4mds/fRoZLsw8Nl9+CRTh8JhtII2hF3MMNVMqhNoix0p50Id1ErpQug063Tbzes3vILTo1eXfnfw/N2R/Kua9BARXHcxKW5Wirz83LSimDmh79EYU6pmpVWv4o/lSe7TrJ7Sqte0DdqiiaKQ+v6Xrt9w3vo7DcQqpKJcmYUyYedmD1HNohmyc5Bz42tBMEzrwg2vomhkefBqFKYVjwCjMLzxb6x6+JLQ93HLa1PHReuPOEWE6BKDIrdQJAh87b2ZEfVajU4nm1hivV6Pglugd8mDayb09V9NEnk1EGQiiGEiljNwjBugFMqf3mbRNC2UoVAqmnmZcpxus07t+BVusTxT6KB+0OzLTJ3EMAaBaReFVtCAKAioPv1dgjmmRmcipZkvtWAj60VPs272TiEKRm54da3n6+Tifo7nfqKovjfZhd07xxQuNKAG2zYrTwiC4Mp+O8US3c7YA2xSp4ccc4OeBxgX+l90C0RTGv77QzngXLcw1xT/MGHgUz16CUfgrm+y9/DJtfv0haDf68VpUCZ8b5gm9x88HllO0KTD6ckJ1WrGtn0pTRp4/mLj+i96QFAxfNzqC6CfpUX6HblCWVAXP8rY32QbmSS/JmQYibeLZViw8ZDuNUGMTcelFVkX5eNVgjcEnK0LxY4JBf/gyvaGsRuH8ZtShU4KT5zm6RFNjhDLYffxGzhTzCzHDk8hve7kQY5l2VS29xaqpGkFDejWTrJXzkgERQpTo7d9oly6DSS6Whmaqb5L3s9bx6UobtUyhlMoYTjF85FlEhLhKjunevVkog3bxvYeVoY2UneZMAyzV85SZKBvLLLBBWKEAWY7X9dD1BSzUv2Lcotk2L0nb1Ct19l14xAaURRiGJd7q3vdLt3WxdUp23HZffw+gl537lhr86ClJVDY3MEurmXf0Ir+8PWCxJSXLsXLm4dfSqmywc7OzlRLBEopTg5e4nUnL4EZlklpbX2hwu2uYJom29uzx4CalbQmohZttK/joHEnhbhdKGI5Lru7w/HNLv8tdFoNaieHE78zTBPLsiiUF6AnDLe70NZyimFabLz+5dk2kmY6pjt4s60EaTqCzLtjis+i8ZGmiFw6A+Z7vdi49pJYWYUsHXA0bG5tUSqVsm0kPQ0tnXo0UyNTSJRchCNJsQuTZsouk19KKWrHh3iXLG8WSsuRX1MtcYrIu0CDeMU5UEq9JSLbwD8F3gDeBb5dKXUm8fDoR4FvBdrAn1NK/VpSz8eBv5ZU+zeUUp9Myj8E/ARQBH4e+ItKKXVZGzc64kvwM0yUDqvt6m3IWMLhfiTw8fdDn0ffnn8eGMcnBqcijAj+oT1zhdFXTvoG/oMMEfF7hcrFDJpSsS3cxVoUgdebyQB3FmeD08PJnpsiwr3Hry/V7uwuyC+lFL2UAmFfSkq/78VLQoUz9PsbkS5qWHZNkD7qotrSl+WSyAIZU22ku1gbu6lQCjGsRDfuC67RuUUxzPgOSaW9Oa9y4FEobZxXk/wVIFRcOgCc2IUZBhSt2tmldVe2d1nf3Jm6rjSZxQbt31NKDUey/H7gl5RSPyIi3598/qvAtwAfSF7fCHwC+MZEWP0A8Bbxef+siHwqEVifAP488CvEAu6bgV+4oo3UUdcYgN+Y1dXPMAKPaA7D3gtGq2OfFbBmBBhns2dvWDSW38IPL17EEcFsLd/43bvCy8627ZkUtGkjc7catUtHnuX1DdxixjM703Gr5Rf0c6JmyAp7Xw470tyU4bNgEbF98LnU6s4K22si1wSEjZwSlB+l0t7cjzu/i9TOZdhwPVZlD39Qev1vcdpJkSgMqZ1MDs1hmCbrWzuZp1O8jJu0+lHgk8n7TwLfNlT+kyrmM8CmiDwE/ijwaaXUaSLUPg18c/JdRSn1GRWrvD85VtekNlLHnDKcwPyssIaW5YzWqsj8VennFfiehxIjSa9igRgoBCUGlpN4JomBUyxhldZxS9PZW7Qbl+d/XNu46OGVE26V/AIyn6Vc1VsgS8k7zdKhJh1U/YhCt0qhe0bRb1HwGhR6NQrdKkX7fK7JKZRALDb2plM2u53WpSGNSusbVzoVZM20M2gK+FciooB/oJT6ceC+UqqfpOoVcD95/xh4NrTv86TsqvLnE8q5oo3UsUtrWIXSQrw5b87tEQqrIuBWeIX6nH46JhiEfDEsi3JlE8stIoZJuTybrUUYBlcGto1Toiw2dtAEbr38EhE2KhVOTk6yaiJFsXMbbqaY1TmSaS5evmXxyLkOz3NR25UdrM09LLFYr2zMbE7UHssVPNJMEAy82JfBtAraH1RKvRCRe8CnReTzw18m9haZXt2r2hCR7wW+F+D111+fu40sL4KI5P33vyRW5KSsSDdnwXIcHrz2ZTeavm+cnV5p69Fq1CmuZZdCbUq0/EqDtGzQVkeruZZVGWBO08tJ8dWybC8N1t/8/djl2GatOMf+Xq9Lp3n5CoDXbRMGPpbtXLpNlkwlmZVSL5K/h8DPAt8AHCTT+yR/+/6pL4DXhnZ/kpRdVf5kQjlXtDHevx9XSr2llHprb29v0ibXEoUBfpazZ2l6cS6Y1RBByycKPIquk7xsio6d/LXOX/Z00+WLOOdb9x7eSDlTStG+QrhtbO+y8yAdm5abcBfkl1IqsywCgzbSStG1cA3tFmmE8zLNpQt81s2IdTNk3QhZNwIqRkBFfCriUaHLBl0ce4p5nWzHOwC42w8Gytm8XKWc2W6BB6+/uTTlDKZQ0ESkLCLr/ffAR4B/C3wK+Hiy2ceBn0vefwr4mMR8GKgl0/y/CHxERLZEZCup5xeT7+oi8uHEg+pjY3VNaiN1GvvvZWoEm+rodoWNdS9ye4Rn5HUJ2vXk1SDoNJK/zcHLv8ZQt88izkqcFmq+31IURSilcAqXL19atrN07+W7Ir/a7TatVrae6KkNG26R/LoYv391ERXiHH8R5/hLOCdfwjl5F/vkXezT97BPn2KfPcc6e46ZkyF7dMO0cVEUYV+RYcA0zaXHbZxmifM+8LOJoLWAn1ZK/UsR+VXgn4nI9wDvAd+ebP/zxC7qXyB2U/9uAKXUqYj8MPCryXY/pJQ6Td5/H+du6r+QvAB+5JI2UscurdFrZOIBH3Ob5vVTRK1KGMmVDTFwOY2zU0prlZn3i8KQ6vEhpfUKlc0d2o36xO2CIBfhBu6E/CpcoSinRaTjoGmmZBE6uF8/IfQ9rDkc/BpnJyilqGzvUnNcggkhNpZtfwZTKGhKqS8CH5xQfgL8kQnlCvhPL6nrHwL/cEL528DXTttGFoRKIaX1JBnseflInJihmF6DmDlJvjMVhagwRE3IQxjvnOZFXuwIJh/jpbtF5udchK292WzWgyAgCgNqJ0d0Wk067RZ7jy7m6TRMk82de6xtbKbV27m5K/LLD0IMtxTLKBGkP7fTz2s92rPR+IRKgYqIomgwMzqJtJY4F/7Ay7C51ZGN6fU0L+p14d5rMylnYRgSeL041/BxnCfUdgsTg9eWK5ts7t5f+gqAzsWZoBC8m8bKERNMAzGM5KLHgVgFQVkW+OnEWhPDxHTHTCL7ARUHCXKT8LFKDcrU4H1+WBUj2zRNKvp5Lcdv/uHPgiTT64OwuIw8cZPrOUtcs2GK5TWcwvVmtYHvcXLwko2dPapjkbbDwKfTumjDsffoNdwp6takh2EYdHrTBxYeRQAzecX3pGEYGNIPxqpABOWnZKOrzgPHjt9WamTge142qTwPrIz8yqLOq+SXMRwXsh/Y92JPIn/O4MpiUNi93r5VRRG1k6NY5hJ7bIZDM/v106MLv6m1jW227j2Yr18poxU0EgPb5uWutjORPHzHg0baZoqnOgyQ1sVkvNPehIOfo8RxsAdRpQc3mIzM+PUyzQi2GgIuTQlnBp1rqxuPCzzpLFlOYS4FrVzZZGPn3rXb+V6PV0+/hFKKZq06MRht/fSEtcomvV4XP0kk7LhLD6tx56g10nNwUghhpJKg8mpQWpguPui1iIDROr1+wyuIu9HP5iGjg5fxaPlRdo+5vMwmLZLC4e/M5TE5CVXeRs1oCmG4RcpPvgLDutp4XynF0f5Tep02bqmM3+sRja1wed0OhdIa9vrGINzGovNtXoVW0IDA605cg06TdH0EbmaaOth3aLZNcw1qkKjqxqT2U5ijIsMw2dx7cGlOuj795Of90eVlsYKUUjTrVSzbwXYL7D54vPRlgbtIo5WtByekN/uexmz0IKKfGv57CW6Gzl+Z1Zwuac48pnrMc/SreP8N7PL1trOt2hm9Tjxw6bUvd6DptpuICIVSmUJpjWKOFLTVjf2QEkop6pdksE+XVbmVF0uasXcyJZeXb/ZOlSsb1ypnAK365embJhH4HsVSeTRnq2YhNFodel7GqerSZOH3UobxLfXodqGI5eBsbF+7XeD71E4np2+ahFIK3/Mo5yzzyZ1X0ALfo9uc7IWWV1ZGqblN4ZQOXAAAIABJREFU5PCUzzNRVVy/fuQZBsFMwq2PmYNcpHeR47PLYzmlxYr4WmtWjRmfZc7GzrU5gpVS1E+PZjb/MC1rqsHrIslXb5aAYRgLerCkGKVZzWsMrLlVzPiTcotlnHHnkgmcHr4kDGabkRHDYK1ys6CRmvlwnezll1bPJqPPywIxTAo71zsGdJoNWvWLNtrXsb65M0+vMuXOK2imZXPvjQ9QnmLaNC/oGTQNcGmC30k4hSJ7j1671j6s02rSaV2eW/MybNu5UVYCzfw83Nvk8f1tzCzPf4qaiFZqlkE+nxky7eSICJU3fz9moXTlZlEUDUJozMpVQWuXhZaoJIbT9x+z8/j9GRo4r3IctHze3HedWWa5Ktt7UylQ1TntMfPk+XTXEBEqa0XefP0+BTeb2TStVK04ORXhYkwXqd/Z2MUqXi9jWrWzkTAa02Ja9lJTOl2GVtCGKJTXMlbS0iFvsYDuAnn8RSgVTf1bLRSvHnn2MecJByPCeg6C0t51TNPgfY92M1HS0hWJOtD24snnWZhWfllT5tw05gxntb61ncvnvg6zMYZbKrNx7zHVg+fpVixGMgIYdg3vB5A9/zwcTLb/Xo25uMuCbdC0Pgh5FHBCPPsbXpa9or+dYUz1hPU8jzAIsGyb4ZhSw3Gm4j/D8fOgWCprB4GcYBgGTx7s8KXnh4RhenJCxMB1C0N3wXAw7PO//QDK/aDZkwLMphRObfq+Zzi8yp9UuIScCvFpvWBlitmtKIrothuj8oshJVCGfgtD78UwWMupiZNW0CZQ3thCRSH1kwNUlI6QixDa/rjNUPKgG0fG/sLgBhMRSlMm3F4FIsvFrOxd8u1QfKMLUn3sxhYZLRIZ2mosgOXojkAESiW2fQr613x4e8vBFudiXyam0oFLS8eLkvBqgeddCKJ4HZZlUexVCd01JAowAg/xOkgUQhQiUQBRCGs7EPpwTWDHk4OXeOMRcqdgc+ey66dZBrZl8vrDHfYPq/S8lPKhitDuzVrXuHxLbpoZbCfTQGWZ6smwiLYeX/btFRqcuvDRVMHQ6ZogpxJ5NKnKgexK0nYNFLK+cmLaGAUZVdT636n4HA3lLLnQ1nCOE5hwTkMf1Zo92Lv367+K8dr7wfNQzQbh6Qmq3UT1uqhel8jzENPC+c7HuJWrjfjbtVN6tYsBkK9TAUsbW7mcPQOtoF3K2tYuhmlx9urZsrsSM6RwiFq0VVh2rfmmS6/bveTb6W8as1TB6w5FU5/U5UsPw7jyI4Bl2PgzxAWbFctxZ1bQwjDEbJ1hH717dd2vfzVcYeuhlKLXrOH1LrsOl+O4rk7rlEMKrsMbj3d5571XRCnk0Bwf/8xZS/L/7bFqDU2HA+f6rBzXYaGotF4MlVxIgjVDbUMCLNnNMGwib/b7e1qm8Q6fhLW1Rf0f/4Or6/6qD1J+eDHnbx+lFFG9SnuCcjYN5a38DjC1gnYFg6WhNKaHU/URuC3iTdNnnp+HUgo1hZGtWVqfaIyrVIT/pc8RPP8d/L33g319jLRhxDDYvf8ot6PPu45IP31bGvIivWu88OCuqyAuV/4Wmu8kS+H6tHD2/ceXLnF2P/8bnPwPf5uo1cT6Sz84c/uVvYfYOU5NpxW0KyiuVbj/vq+gVTuheXa87O4MWPj82QoIuKx1hDSTpU8iy+rVBMPZ8OyA3m//awi8uXuwd/+RzhyQY0SED7zvAWe1Fie1JkEw/9Liou3GUmUlBhBZ9zHbq5dl7ebmNmGvizmkSEXdDkc/9tfx3vvC3A+o4voma9v5nT0DraBdi+U4bOw9xOt28Drz236N/4ZM08S2Rw2r1cgbdcl3CglLsZ2UUvTtp4ZfSg2VpXLr5F80Z65EZq0AZvgQEbtA1GtjuLEnZ9Rp0vut/ye2S0uY1TPYNE0Kpek8QzXLQ0TY3lzDMISXR7MH7zyv6GJRoeBe2OQyM9HhQaXCx3SLQ05SfdnVdzaY7Bx1q8n6ODM/jdnJL7NQJGq3BgqaCkNOfuJH8d59Z0Ifpj/QYs7SOk1CK2hTsr69h2k/otOoE/S6dJqzG0QOY5nGIJHrrJwZayNmBo7jEnWnTPcyEASCUyjge93YG3BEQZB4GUIEUTqsR9qUHBPD7w4p4gq6HVz6M3Wx58D4ee8rcUokmRUwEMcl2ryHGXio5sUHsPX+D4JpIUY806WUwn/3cyPKGcwe/Li0VtFLmytEqejyxuM9ep5Pz/OpNTqEMzhAjV9p13XpzWGzCBDP2Q7NvAogxvRyZshhSgyDRFSN+lYNvI8VlphASs4SGbFKEtY0beTZFyEKYye6KEIFHqKixKluyDli3CnBMOLrY5hgWUStFuWv/mqU6dL+3K9faKvwtR/C/dq3sDbPvSy7v/Vv6P7Gr97oGAzLwi2Wb1THItAK2pQUyusA2DuxFu91O9hugTDwEYRuq0HtaP9SITNe3PO81MYcMz0nhzZWSH/gSjTSQTXy97rcZ5rZMLw2RuP6JfPLLutwef8Re1n8MsOyMM0h+zMVEdVPLm44o4JW2cz/6FNzjmNbYEOxECtG25uxPOub2Pp+wKvj2qVen+PKePq6+Qy/vyGHKRVNXHAYqXPO0Fi3ijQH2RIGhL/+f6dXH8Dmg8lfRiFmeW3k99d7599OrmcGc/Hy5u5KZD7RP905cRLPtX704fLmNr0gotNqDEZvw6M6b8j+w7IsAt8br3LxTC1kV2l8twpkMfM0uU7vnV8jatexHn8F4pbw3/0cqjdh5nYGAW4YZhJrSLOq2JZ54fOjvQ1eHhzFM1MIGOdzUioa9TCe1eM4O5I4NavOCs1Gq0zO92T50/2tf8PR3/1hNv7Yd+B+4Gvw3vsCjV/++Ru3lmfHgGG0gpYipXKZWu16Ow+jP827ZKbtw2osca6SDUcGV/9SAa8IXrxD8OKdK4eYswQ/1rZntxPXdRAUwRSx8GZZHp2GWMbMfl+sgmSahuwdv1KcQctEP7u8f97vfZ6jH/3BeFk0pRh6bmk1UtNpBS1FTGu602mswNTqMKZlDylpChUljgh5Y1xypKlYpimUMplAm6LSq87HDIJvrnRQmtwjItiWRXBNjlfTNAjDMCeTPtN0QkaW+ZVSRCkrmDcl1k9H7YBJU8bmfYA5jay+RkZN2yvj/2/vXWMkya7DzO/EM1/VVdVV/e6elzgkdwTZkkxTJGQIBA1JpFbQ+Ie8Jm2sCS93CdgybIMLWOQa8BMEVv6xsgxLsgQTNg17l+JyZWtA0OJyRQr2epdDinpQ5mPEJjkUZzjTM93T3fXMR0Qc/4jIqqyqrHxE3siMyLwfOjsjb0bceyOz4uSJc8/DdStjsbSS1iC+77OxucmD+/dH7jdNketJKPpv7XTxWb9WLzRpax7ig11OZ/rSoE6So3DuUMou4GZEksn/JoMwHL+TpZJsb1/khRdfGmk1932fxHg1gLzXxHjL20G7d2ofwRMt1cqAIrzWOFmR4KLTgd1XjY1gikI+NRPfxYRdeGG9MgFO1TLlVICNzYtsbY/OrTLuDrX0lEeuzY+yW9Bm7XQKi4Jd4lxewjDk5o3r+P787t0X8WNpTjkrbu6lFbNFfF8GrIWTJkCuyvImWAXNOCLChfV1ti9dPhk9lxGGoVnTdTrqfI+rxs1HaaepRVx2M56s6BRLnBNUL7BUlzAMuHb1ypk8Z5DKt263s4BZnceCr/Iih6+IlccERs50QqXbGfK7XFbsEmdBrF24QGttjU67zeHhIYeHB3TabWPF1xdJZcRGWQVcIfOarU+dUEHzg7AS4emW2QiCgJs3rtPr9Tg4OKTd6bC7u0cYBrnzn52HuXJUk2JwrAKnXdr41ALklxGfZtWJ5hbUq7MCYBW0AhERavU6tXqdDd3k4GCf/d1dol7XsP/DfI3hSZIcpRk5M4OJzkv6/4xF5RTOsvugxZMJyLBuC6OvEr7vs77usw5srK+zt79HHEelcNOYpUxyeCrNwpkoysGXp/VH6T8J4gj0Cvos1JycKJO/3VCM+KCNV9DEcfCCaqTYAKugzQ0Rodls0Wy2ONzf59U7L5UukmhSIoNLHPWgwD/B0lrQFj2BIUxoQXPK+plaCicMA8LwIhvr67z08st0OmbkQH4LWn7XjoOOGaUqDArMB2gvtemYQMkTcSoTIADWB20h1JtNbj72BOubFyu1Hl4ERRYhN3sZlvwOdEYknkxBq1VoecBSDK7rcuP6da5evUoQBOMPGIPkvFKr9EO7cMqacqiPgflNkssxbJS/vNMgVkFbEI7jsLm1zfVbjy4uL5qVb1NQ8jDOWX+sJrCgiTjUKibgLMUgIjQbDW7euEGrOePfRN5YpSVX0IrJ2G+AIu5VTfigJeMnVlvbmH2cOWKXOBeM53lcvnadV176bv4lzwobd4qdekkFXBkI6uD5IA6IICIEQZAVmj5p0+iHrwe16uQPsswHEWF7e5sojmm3zQYPWFaIKX2RpVbHvXgJcVzwXMRxcTTBD2uZDBuQ/tqvKS1HNbWrglXQSkCt3qC1doGdCcpELR9V0S6rG9QxjN7GNdqnnJCd/deG7tufrbhbBc/KUkVc12V7a4sXXnwx1/F5lf75R39WmZIvcU65ihQ8/nq8t739VGsXd/eV8w8SAXly+rktELvEWRJMOdxWjULtMStk7JlW/Ob5aHqHe+WPBrMshFWVX6skY44oQgR0p7O+5kpXpUp8uDf9cQvEKmgl4cLGxgzLR9X90bS/94ti+g8+7nbo7j0sYC6WqlOv19Mk3DnIb0HLdZhRipyCUdFYdjk7rQ9azvRM7bv5rLyLwipoJaHZWiMI55yfpewXraU4cmrGB/dN1Qa0LBO+73NhrVr+PatFyYX9lNPTnHn4ug/vkvS6uY5dBNYHrUQEYUAnTxHyvMleS3AHWnrBMQ8yR/3UJJAlvxQn3c7adOD94+/t+AtUcZG1rbP9HZkZjrdVhEjcXB99d3+HJOrheAXmf7JUkrwpN3KvHGgawFLaaMdVwXHA8cBx0iojjguum74WFxwBccH3T1YhGZBRihDUNrPyCYK4/T7cdNtx0oc4iOsi65u5p9t9+Cq17RvjdywBVkErEXmMGrVajd7BTi4RFZXAbyT26kQ1Z0JloX+W/bzfihc06GR3REeqyEAUotTWsjZABcd10CQ+ilY8w8CXoAMpxHsIbnicA0yzNObSn4/2FSfNulBCUfy9u+lPiCokURoKrlEqsDQ5noPq0diD1v7+bHoXH6HdHXPn1zxrga2FAdHBEL+LGfTiuNe1CprlDHn8E33fp5czE38vigiCkE63l+t4E0Rxgj/JOtRpUZN9VJ7v0etFJ+65+rKpk7hEa1dThUUVEfAdB+m1R9xcZwWijiJ7+q+133G6j56YxtGLExUVFOT//RR6sIdGPeh20G4nrYQTRRDHExsH/Dd8H96f/IFz3xdgWI0Sqa/RM1y1Ip7S322RWAWtRKxvbLK3M5mPTxCGaOeQKKdyVhoE4gnLDJ3G8TwiVZIz+W/O/6Hw8IjnVKbGdxOks3f+9zMXJxrzY7h+Pl8jy3LT90ObJGBARHBdj263R0lM+bkQhGSGcnWO+GgSnyuxjlW37NnzYI5LdO6L3ybZnT27QN4a1EUEJblBdcrVWR+0EuEHAZevXp+oGLUDaDKjolFduQiAiEs8YQb842MKmswwDNbSk5IsBQfNCzieva+zDOfqlSsTBQs4jkOvN7tyVo6rIj/J1ArInIW2qSTqcV4rp+FvWAT/QnXSBVkFrWQ0Wi2u3bw1dj+tSpHxAlFN8EqsLFT9x+M0judz4doji56GpcR4nseN69epjVHSjF23FQ8DL33iZ0OlCPM69csE1QGmoXn9e3CD6qwAlPfXbYUJgpCrN27x4LW7tA+HBw3ECm4QEs9UuHwy4eB4Ho4c6/JnLpkj/4azfZ4eIUmSdElgBsHqeD6O5yHi0G4v3o9uPiz2hyhc22Dtyi1cf/bai5blRkS4evUq9+/fZ2d3d+gyVbfToVarZdUHilVSRATPyxSNcy6jNNTgnHkMNKsqSZzksHydJAxrxHFEp+zVF1wzKoJG+SxoppY4naBO49pjBBWynoFV0EpLrV7nyvWb7Dx8wOH+3hlFLYoiIiDwfJKcf/zj8MMaIkK3fVCAeiC5a5A6nken08M1JDyKYhnyGDmez/qNxwkaNoWCZXJc12V7e5tWq8XDnR329/dP/Ngq0Om0qddqHM5wkzXKAOU4DmHgEvd6xL3xitCkl5iQLj054uA6CXlsQ91uB8cxY50qkkncbSYit99vDMz2OdUvP0Lt8q3yWyuHMPGnLyKuiPyeiHwie/24iDwrIrdF5NdEJMjaw+z17ez9xwb6+GDW/pyI/PhA+zuyttsi8oGB9qFjrAoiwvrGJleu38Q/J4TdmcVhe8Tfq+sH9DqH9NoHxu9vs8QRuZZpHddLIyCBOJ6Ps39etAQ+aDOlIBBh89HXL4VyZuXXYqjValy5fJmLm8PTIsQzumqMuip8R+m2DwuRE0LqYpFHhgVhiKqWXn4BacoMA2gvpwUtb33qjPrVx6hfeaSSyhlM54P2N4GvDrz+OeDnVfV1wH3gvVn7e4H7WfvPZ/shIk8B7wK+F3gH8EuZ0HSBXwTeCTwFvDvbd9QYK4WIsHFxi2DI2nms+Y0r55r0OZmqolS4fvmXBTJUql2/06818II5J08uDiu/Fkiz2aTVap1pj6II38uvBIxaAdO5XDPTScmwVqNbgvRGE2PMgpZzlWfaCgOnqNqS5mkm+vRF5Cbw3wL/InstwNuBj2e7fAT4c9n209lrsvf/bLb/08BHVbWjqt8CbgNvzh63VfWbqtoFPgo8PWaMlaPZWuPKjZtn7gRkpJp1PrUwLLWD/TD8WjOL/KoIBi1oi3CGlgoswUyClV+Lx/d9rly+TLPZPNGuqsQ5rCS1wKceegsPEpjmChcRup3qZLEHsoTZs6N5U4PMGCQgJXeDGceks/8nwN8G+msdW8ADVe3baF8A+ql5bwDfAVDVSEQeZvvfAD430OfgMd851f5DY8ZYSVzX5dLVa9x75c5RegmJx//hu66H4zp4AoedLrUwxHn4MjgujdbFgfvMAXEjih+mKzKdXjSzqdkIRvSd+dkFTd7B5zfG5Z/DEgUEWPlVEra3tkiShMPMp7ZWq3F4ON6iFAYeHkonW1EM9u9CEuE1N6n5Az9jWTJWzf6XwKOnQrcEN3apD96sMmG+Cqk4huRl7txt+c9XXK+yS5t9xipoIvKTwCuq+kUReVvxU5oeEXkf8D6ARx5Z7jQAjWaL2iN1Xv7uCwgQHe6OVDlEHMK4Dfu7ANS9AOnupG8mMey8OlZlkeZWKVJGxFHF7j6NktcHLR/iurQuV1+fsPKrXHiex7WrV7n/4AEPHjyg3R5/TbcCF2f3DgANcdLIwiwHpOzfHyu/vAuXKIvkSCsnzKIszlvhMDTelPkqT4yeVW2Zlsb176m8BW0S++UPAz8lIs+Tmu/fDvwCsCEi/bO/CfTLxL8I3ALI3l8H7g22nzrmvPZ7I8Y4gar+qqq+SVXfdOnSpQlOqdo4mSWt3mgS1hp4IywdYeBDe/fotUTd/LU7F0zc6+H7VSoxVIIlzpzH1dY204CM6mPlV8kQETY3Nri4ucnaWosgCM61dLiug+zdPW7QBKa9USvD3WWGYyiv2NwwaYHK2ZfkWWZ1HIL1avufwQQKmqp+UFVvqupjpE6yn1HVvwR8FvjpbLf3AL+RbT+TvSZ7/zOa2nafAd6VRUk9DjwJfB74AvBkFvEUZGM8kx1z3hgrj+8HbGxtc/mRJ7j6+OvZunaL5vrFM6HbRjLQl0TACekyb1Uw66Sct1RKvtH8JYjcBCu/yoqIsLGxweVL2zxy6waP3LrBxsY6YXjyZlOQmR3Fi2Tay6s3rp5u2TAVJAD5lb0cCprfWM+n2JWMWc7gZ4H3i8htUn+LD2ftHwa2svb3Ax8AUNUvAx8DvgL8JvAzqhpnPhp/HfgUaZTVx7J9R41hOUV9bZ3NK9e58viTuJ6HOA4iDp0oRpfE2RsgiboE56QcKR0m02zM2Rl6iaI3z8PKrxLh+z7bWxe5eeM6a60mjuMgImkAQXN4io4yMK2Pk6oS1mrV8Y0yqaDlXFHI4wfnhNWptzmKqdYwVPW3gd/Otr9JGsF0ep828OfPOf5DwIeGtH8S+OSQ9qFjWM7HdT02L1zg4LvfMNdpiWRJ3Oui0kMcP1fggud5J6xwQ9We1MM43VDIsraluY/Iso6rcGwj0yznmR6ZrBQQg3f+eTNq57HiieNUqhzKpFj5VX5EhO2LGzj3XyhHYJJhVJVOu01Yq+VKF6Qi+PUB63ZfNmcyKxUTeuqNk1uO45AM+oSpntjrSNQo4JhzcxDJu54z/Q+QV2+O36kCLIWTieUk4dY1NIk5fPn5RU+lEEQVx3WJcgjwKI7ntsyQmIzizLnEmcu59uKV6vnKWJYG1w/YePQNPPjjr6MmkrkWeIM5d+8PhXZntjyQQVijN2GJQL8EFrRpl0adoEawcTnfWCXDKmhLiIhQu3ST6HCX3sN7Jno00IdZPNclKkHo/ChMrkpK3KPh+4jqUbj+0bYqogmqCZ36JvFAUsgTZVInxK81zE3cYsmBX29y4fqjPPzO7CsBRvNFGyLOGdVohGkEgkkFLQGnllm2VFPrfia/+nLMrTWJ9x+cnO6UCppXb1VnCXkMVkFbUkSE2tZ1ejuvzawplO1vXRx3hiK6JTuZCZEkQfbup9uD7ae2pXH6/Kb/nNxw6f3PLBUgaK7jBjXi7mwWo/lUFJgO13GJyHODaeJcppCBBoW/HuyiyTj/4WE3h1Na0JbE/wysgrbU+K0NalvXad8dGt1fWbywTrsi5Z5ksFaqpFVI0+30v+M7PTklh2TgGNKaeBNYDAMHvMDPXOhSS5sbBqklrZ8oczBhpiap9S1Jn8VxcWep72qxGEIch7Xrj/Lwj79eWn+0POqLHwR0ci5Tzl3VbDRx1jcBSa1pWfAZcuq1IyejLbMgD/r7CqgzvuqNIrjNDTSTe6JK4siR5U0HVw6yFlQhSdLtJMGvny0pVlWsgrbkNK4/Qdw5oLd7P3cfC66mcoYkb103mK8BTYQo0blGYHr799GBvHdHUxlz3JFLseNy0snYYlkcQWONtWuPsfPiN2fopVx/y71uF8/ziKLp/evmXSHZ+cG34j/x6PwGVCU62Bm9y7g+Frl8bJjqJwqxjCW8eG3GHsqlocW9Tm4n9lnEm+t505v8552LZ9bSLElM0j4wMxeLxQBB6wJOScuO5ZWMbs46yLMu1/pBSG+a63vu/i2z/9ZEBw8NzKMcWAVtBQjWt6hffSz38fn9vYrDd51cjqAznYlSPnPiKYzcYZf8HC2rheN6bD76BspmCQNyXyu9bjdXVZTZr++SX9smZM8SyS+roK0I/trFRU/BKFHnkMDPcRc6w7Wb7+51vj8qauCON2nvGZiJxWIONwjx8gavFGgFyhstmCQJvV5v6pWA2VWPEiq5J5j9DJP2voF5lAOroK0Ibq2BeDnrWJb0jiTqtudaVSCJY7xp73rnvERgIry8++oLBmZisZjFb17IdVxJxRdADivajBH5joM7ze/AvJc4DXxZ0c494s5yuGlYBW1FEBGCjXyFmIuUb7MtOSra6+JOVdR7trOJej2caXIDVbAeXLz/gOhULiKLZdHU1nOWfCqx0ahfVWBSZl3i7LYPcafy5yvxhzeC7p0/XvQUjFC9Xw9LbuqXb+U7sEANbdbLP0kiJJlSaZqZxeQRmnBAI73ojNnKLRbT+PUWQR4rWpF3mAYut067TRjOK7WNTvV5zL9Sgpl0Kkn30Eg/i8YqaCtEtJc3uqXEawSAJjFBkHP5dkIG63cmyRRh3BW0oAG4a+UtUG1ZTVQTeoc5/CMLvEcy1XW34PJzYb2BoAhCdxofrQoucQJ4reWQX9X89bDkIu62S1cWwJTqpwXnvskdyTrvz9vQeMvkaGtZDuJeVLqIcmPyS/XETaB59NRzOTH1/S6LD5pNVLtC3G9H9OqZH1p2IYicdCxPc93LUQE7yVqOsuDL6SSmgxeUAIomQ4SNHA+rSUKiSpIkmBJJvV4Hx/FJxmUcn7t8KpdCPClJ5wCWLPLXUm329vfZCzbSF0fyS87ck/TrdfTlWtKLcF134gCauB8MNExZEEnlV/YwGYHgef4ENTrnXeKuennQIJNfS4BV0FYEVaU3aEbPhNXw1F6DfgoDZYGmIJ6wNIupzNiiihcEdEpWAkpPV3CqCGLLPVlKRnew1NlI+ZVx4o3pLOy9CZccTWb2Twz5Xw2l3IazYwwpvI63HPLLLnGuCNEs5ZEKxZzkSKKOsb7OkHvpsHrqmVNr4l3YXvQ0LJYTTKo0zRWDLgx5k9dOglZFDBlSUoMrcyxPVSBWQVsVSua7UQRJr0c4LnltTkFVFflmQt8NLj9iJJ+axWKS5ZdgEEUR3sgyUPmuS8n56c37M9cJV15G4bY28HLmzCsbdolzRfCDEM/3iXols6SJ4Ho+Isc65KBuMHT5deApPWCgWRNcz8NJneuOi4Bnz67jIGOS2w5TTkTAHxCcCnQ6E1jsBvs68uOTdIz+62H7nphH35emPVbRnvVO2Qnr+FvXZ+vEYimAZqNBd5Jrbq7ocIVqUKCdPkLT487Dc730ipezC6iOgBOEIy13JzyKpT8dh6BWP/Fur9NGp7JYSd9p+eghR/7Jp0aXgf37xBEajbGC6mkf5+mpXf+emY4vE1ZBWxGSJCmfcgaQJNA9PCGuZr1rC8Ma0SlftP4lP4k4Gjb+mTZxqPlBmiw3SQAlcB2c3VezSIgYVE+KGh26OfE8nFqTpDMmv8+MltLwxuuRiqYGsSw3nTIucSKo4ZxbriRwuDv0vbz2pWEeeK4fAg5okt7Yun52LppGU6oSRx3i+lbOUU9SCz3ocL64AAAfeUlEQVTYuTN6J01A8oeOeeuXlibFBtglzpVBRGi01hY9jSGYX0pzHAdxigxZBzQh6bZJeh007qFxhIMicQ9JIuS0cmaCgpcdnfoa3nq+ahMWS9GsrZVRfpmnG0VTZvvPR9LrpI+oh8Zxak1LYkiSguTXhPvNcJNZu/ZE7mPLiFXQVgQRYfvKNYK5ZayejCJ0ju7hAapKUG+a73wExftrFKughVcfs75nltLSbDa5eLFsqV/MX/VJktCOQWot432PpDR+yvlkkHdhG7exXEq8VdBWCBHBKdqyNCWFiQRNSJIYx/Pwwhpn/CEKodr9u82NQvu3WGbFHelAv0SI4DiC4/m4QYgzVb3hfJRFPcsrp93WuuGJLJ4V+Wu3QFrou31YTAK/IKwNdcqH0WqFoHj11rFPrQjR4Z6Ru7koqyeZRNHRpLygRtQtV660SXE8H2leAM1k2FG6Oj2Ok1DFaawPBE4oSdAgjuP+S1RTH5P+/mlyYWgg9o7NUmp2d3YK6dfzPDzXPb5uTufiPkeICYJXaxzl8FZJ5U3Um91frt3pAk7qQKYOOAH1Wkh0MNw/rezEjoezcY3+4qnCQD67NGAqlUbOUZCBAuI4REm6T6KQZO0JciTHFOFisMHkZeergVXQVgjX8wqL5FRVYxFWoUhxJV2KtKIVbEDTqENyMP0PlAYter0x382pihIWSxmp1+u0C0hG7bouPUPlzcKwBgYUtBMMJOYtjIKXOBPxeaCnVnDOTTJ8vOm5AZ1z6x8fyyxdQvllb5hXCBFh6/K1QvquTCLE8hjyp6dgAVR0PVOLZVY2NjcLS+a66mjhsjFn/xMqjmPL/FUQq6CtGK5bzFcu1dHQCqRoE1pxXYvj4gblCiCxWE6T+tGW/GerqqKwNEECJ5l0VrWSBcCZoOR/6RbTjC/Gm5OqCiWDFC7ecn/G42cWtC4gZf/hs6w8qlqQDCuncnKGIqdZuAEtbzWD8ceJCPX6snmgWQVt5Xj42r1C+jVZNNhilkm+meYlWz3AUn72dneJ+kE/JqmIflYkxS9x5kMmUOw2NtZx3XJlKDCBVdBWjNa6TaVQXXIqwWMEXGPrKl5Yz9e3xTJHGs1m+X+Iy6nnjEeLVtFyWtDGHOb7Hpsby5diA2wU58rRbK3RvrBBt3M6Euo4lvwoQ4MAKogcJ3OQc45JKiuVzFHFT8ANQpqXrfXMUg1c12X70iXuvfbwbO3aYfkwRBGVLIjprATrJ3hwqrIAUPg8BwoblwQdU5/z8qVL5fdLzIlV0FYQ13WHKGiz4VsLTOHkD+I8X+A2tq7a2puWSuH7AYdds35ojcBeA5Cl2ikqWCCvD9oIBa1WC5fS96yP/atcQbpdM/nKCqNcN3DlIGduOHVcdETx4fDC8hQWtqwGna75PI4mkSoLsCJT+eSsYuMF56dVabXmXA5rzlgFbcVI4pjD/T3j/ZY9R6C4LkG9eVRdoJAxCusZnLBBsv9w6uOk1qLbPhzep+fPpYSMxWKSnT3z1VASgzpVERmHBAiDAC2wCoo4DnpuQtjZOSBf/rpoxJSCEcrbMmAVtBXDcV2uXL9FvVnmOw9z0tL1A0DQOKZ7aCZT+GIwn+TR8YKcc7FYFselrQ22Ni+Ut/KFSQOaKjVX8OIOyeEuSVSk9bDYWPy8fY/6mr0lv8Fc7rOzDKXWaFBrNOi027z26h3j/mizYlK+pbm9Krzk0CfvKWjCefdhzqoUnrYsFYHvsX1xnY31Fq/d3+X+w7LVpjSn5tRqIXFFa2+eIefHMkptLH1E74xYC9oKE9ZqXL1xi7C2nE6Wfr1R6JJmNThfswtbyxmablkNPNfl8vYGl7eWM3WQ53rLo5yRFpPPd+Dw5lotLKwyTlmwt9ArjjgOW1eu8d1vf2vRUzGOkyTU6Ndn0+PrXDV7pO9J9py2QVxr0pnCKUVE0ocWWAuugMiqYG05f9gsq8XGeovd/UMO22UJfjJzrSpKLayd299AmXBOZBhRaHenKNaeyS/HdaGAHMDH45jtrtlomO2whFgFzYLvBzSaLQ5mCR4w7HdhgiTq0fr8J6YfPmwQXtgGBEkiiDrQ6yJRD5I4fWQK3qDM0Tf8Kbp+MfXgcqeQVB0qGMX1cLzldrC1rAYiwuZGi8OX8ytoIuVzhIjjGKd7H6czve9sI2wdpzTTGJIkfRAf36BmHOl1InTC4m7aUp/BPJ/ycM0uCJbfh9YqaBYA1i9uzaaglQzXdWl8/Qu5jpXOAe6rfzz9gUmRFjSzfZfUvdpiyUWzXqMWBrQ7U1iOBihfelZouZpLOQOQTg5ZXtJi6efJqrLGiJhk7AKuiNRE5PMi8gci8mUR+QdZ++Mi8qyI3BaRXxORIGsPs9e3s/cfG+jrg1n7cyLy4wPt78jabovIBwbah45hMYuqsvPg/mx9GJrLrNSTiObBA9aefQbn7gvzHbxIAZe373OUxiSO6O5Nn7ajalj5tRocHHZyK2dlIgx8GoHLevsu3oMX5zp20fpO7v7POXBnd3kMCucxiYddB3i7qv5J4PuBd4jIW4CfA35eVV8H3Afem+3/XuB+1v7z2X6IyFPAu4DvBd4B/JKIuCLiAr8IvBN4Cnh3ti8jxrAYQlW5f/cV9nd3Zu3JyHxmoUZC7Xc+SfBf/iNSYD6f8ynuM8iTpDY98PzPoVfptCMTY+XXknNw2Oa7d+7N1kkJrDGu61J7+CLBvW8jveULbspr8TovirPdbueXixVh7BKnpp9AX1X1s4cCbwf+Ytb+EeDvA78MPJ1tA3wc+GeSLj4/DXxUVTvAt0TkNvDmbL/bqvpNABH5KPC0iHx1xBgWQzy49yq7Dx8Y7bNeC9GoQxLFiOviuD7qOPSloKoSxz1ENU2D4XjEUUQcj1eq0j8lQQeW/BzPw/V8av/5142ex9T0ujiXHgFO++zK2DWU02+5916C/R0k8x1Rkeyzyh5JAnHmDxfH6ecx6FuSJEgSE3V63P6XzxLv7hPvHRIftNFeRNLu4K01edtXPoXjL68vmpVfy81hu8MLL901+kPtBwF+0iXpHCKuD64HXoACibiAECdKHMe4roPrOCRRj2jCHGWO45IM3ECK4+C6Ho1oH4kWawV0mxfSjTPVlSTzg00FmejAa1E4ep3Soc79Th1VSDIVKzl0cdyB0AYVEoVYBVXpx2gdDRwrJCrs3RNevi/04vSRJBAnQpzA057yuhsl0K4LYiIftOwu8YvA60jvFr8BPFDVfszHC8CNbPsG8B0AVY1E5CGwlbV/bqDbwWO+c6r9h7JjzhvDYoA4itjdMbfMFYYhjsZEB8fWOE0ikl7qvOvXGucsQ3RBhDCs0+mc7+jrhXU6UYwmCeK4uJ6HiNDtdiHq0IoXWwJGej3ahsrQ1J//I/TFb5xQ3Kb9CVIgiR3u/9azQ99PlbYDnM3lTrdh5dfycv/BrhnlTMHzfXwHkv0HJNkNoCYxDFzSh/76CVNQPOBCUKs16LXPr3IgjoN4NbpRhDgujiO4rkuvF9GLoZUUGUI5GYftjhHnrodyka/dv2hgRtA+UF46x4awM7xIytIwkYKmqjHw/SKyAfw74I2FzmpKROR9wPsAHnnkkQXPpjrs7TxEZ3RsFxE8jUjae4gm5O5NlW77gDCs48YKrp/esyUxSRzjOC6dXpQqKZJWu4uiqD+JYuqrTItBr1XRuPBF48s/+Xb8JVfOwMqvZaUXRezuz/4LfcFPkP27oMlZw9EUtDtdPD9ExCGoN4FUwVNVom4H9UJ6mcxSSK1wJVDKikDnsGbsufDGWyWQ+wUyVZY3VX0AfBZ4K7AhIn0F7ybQ92h8EbgFkL2/DtwbbD91zHnt90aMcXpev6qqb1LVN126dGmaU1pZDvb3ePjgtZn6cF0XnwRNYiM5wATodQ5pRwntTofDTpd2LybCIXHcEni5jcGggjYP34qNP/19hY9RJqz8Wh66vYiX7swmvwDW/Rhn/y5yKmVOXqIoZr/T4yB7HPYS2r0Ev9YkGlVUsiwYkmHzUNA2W9AIV1xBE5FL2Z0nIlIHfhT4Kqmg++lst/cAv5FtP5O9Jnv/M5kfyDPAu7IoqceBJ4HPA18AnswingJSR9xnsmPOG8MyI6+9emcm65njOPiuQzKHZcUkSdJlzLIjBrNaz0Ebff4X/w3tl14pfqAFYuXXcvJwZ2/mxLTN0EX2pwwuyKPAiHBoyPWhUEyuABjr6fwA9lcfwpe/XWBqoxIwyS/KNeCzIvIlUmH0aVX9BPCzwPszZ9kt4MPZ/h8GtrL29wMfAFDVLwMfA74C/CbwM6oaZz4afx34FKng/Fi2LyPGsMyI4+RTJlzXI/Q9nLhL1F6JKMDJqdjNXPuFl/n6h35p0dMoGiu/lpC88gvgQqCsJw/wd18yOKNloGICDPjUF5V4iqovVWOSKM4vAT8wpP2bHEcxDba3gT9/Tl8fAj40pP2TwCcnHcMyO0kO65kfBEjcIx7hyL/SmExUa0hWjuvmtf+UL5lvVbDyazlJcvwou65D04mQvbsVVEXmgMFk2CZVplG6eLsHrzyAa2biEUqHrSSwokyioLmeh+M4uJI6u0btfaT8nmCLw6jfmJmfEJHR3/N/849/1sg4Fss8mUR+iQiB5+A7ihu1obuPLDjSu/SoGlnqnJcC/L2PCNcuLq+6bRW0FcVxnBMh4kP3SSK02z2qn1vqy6AUkzOnoImYUYVH9XLxR97MpR/9MwZGsVjmi+uOX+Js+eDuGV7GNKTALD/mZKEz4uN+259Y7u/CoFezpUr4IwrN+n5A4DnogpMmVg+TUZxmlhtkhHTr3Hk1TX5rsVSMwD/ftuC5DmsBOHt35jgjy0kMKk4jutpfvoILJ7DSeUWp1Rtn2kQcarUQ7eyRdCv2l1+GlVeTUVCGIkJHpevYf+5bvPzvP21kHItlnjTqtaHttcCl1X4Vd++VchjVq4YxGWZQII/o6nNfS3L5I1YFq6CtKLV6gzDwCESphQG1MMSJ20QHu4ueWnUx6IOmpgTlmCl98S/8DV76+H8wM5bFMic8z2Wj7rKhO6w7bdYCYUN3qe29NLL+rGUMhmTYvHyVv/YC/Pr/lyxtTU6roK0onu+TdA5J4h7R4R7R4a694ywTpnKqjftSk4Tn/v4vmBnLYpkjYXIIvTbS3sHduwO9Ja/7UyXm+GPyRy/CS7PnLC4lVkFbUVzXwwuHLxNUklJol2aDBIz1M6avzp0pk3VaLCUgaKzNf1AbIDARYtCiNclHvqy+aFZBW2E0Xu4szHPHpJndYFUC8dyR72sUER8uqYSzLC1JsoClzCVdSjONUT12gr668XJ+L1ZBW2Fcb5myrCzXna0ajK50An/0+74/U9kvi2URuN7ov+vKUQYRZsz31WCajQn2KcNHVwRWQVthamvry5NmoQxLD0bN+ubOx/FH/5BpFOHWQmPjWSzzwK83cYMlctMohZphKEhgTILs6foav48zKllahVmSX2dLHlobW1x59PX4YX3RU1kOTEZxOqOXJadh3BJntLvPg89/ydh4Fss88MI6W088RWPr6qKnsjSIqQomRv1xx+/z/MvLucS5TGtclhy4vs/6pavcfeFbi55K9THpn2LQsuk2anhrTZwwQHwPJ/BTy6kI4jqI6/DKb36WzbeeKVlpsZQaEWHt8g3aD++RRLaM08wIRoxoglkLmu+C64LrgOsoIulU+893HijtLtSC5bI5WQXNYiwp6qK596efPl4lUDnaHswpJqqpADpqGpRGaWOru0vtlW+ljvoiKFkk5ImHc3IbSMI6YRCcueVTTpZuUsc7NYesXfvyUYlvvg7ZvHxiTBUHHRjXFcU9eJDOXMjmOXBGkm790Ft+ZPyHJw7J/kOc5vr4fS2WkrEMMmy/tkXkZAnE9VhTOiuhOCM7Bo8RgfV49+RORzJJjndNVZzjZoXQC1Bx0t01lUV9d4u+yEocj0SdIXqcHE07UJcnt/dJRZKm80olFCKZDEZRcSFJQNL304Ltmil4CWSVtZ66Mv7z29ltUdu6MH7HCmEVNMvIckBVInZ9I75o3fXLSG8/x5EKO6+caZVTz73GFjouAi2sp48RiCu43Yf9kc+MN5VPiya0v/b/U/++tyFL5ddjWQWWwZdWRYh18Jodcf2ea+VK79AO4tPH6qnnc+iNT1QerV1lrzvOQtajLg+GvzUwBRGfKFPEhjKFCLt7f49a6HOhtTwuO1ZBszBP59Sg3ji+GPsWrn794YGLNLU6cfbCHZzqKStUp21TRcyCdg44/NJnCd/4FtzW5qKnY7GUjlot7JurU0RPWLvOIue/7Muv7FkM1d9dZV54+T6Xt2K2N1uLnooRrIJmwfOzZbk55PhptzuF9b0cdsApKODr0qhTjohYi2UKvFqdqFN8JYHu4WFhgiZcpsThCyRZopRB1bcLW2ZGHAffLmtVjwJ+KNwL27jNDfMdWywF4tcacxnH3ruUn4sbzUVPwRhWQbMAUGsuh0nYBFq0La58uSCP+4wj851aLAUTNJbAOdyg2KlK0gmT+R77JElVzn48VkGzANDc2KK1eWnR01gJTAklLUKPXB7ZZlkhvFqdjVuvo2hHh8pcHoWa+uab42xqKvMljccqaBYgLZtyYfsKm9du4bjVdE00pvgY6WUUhuZZwETdDaukW6pJ2Frn4uNvJGgugTVtRoqwTB31bbIvw/P0PRffN5fke9FYBc1yhIjQWNtg89ot3DHlgSz5MSWUEtMamjj4115ntk+LZY74tQbrN54oTkkr9O5tzhXG82KypJ2xnlK2NluFKqfzppqmEkuh1BotLj/6enqdQ7qHB+y99irJuLxdpcBMGuzCg1mNFSQe0nV9DQ3POskO7qr9E9ST7ziNNcS3NTkt1cZxXTYfeZLe4T5JHLH3yotzifCcHYOCp1AL2pAs27k7O9tPGNZOZm1TQbMxVdMtzVKbnJbVrcZyBbtZBc0yFMdxCOtNwnqT1uY23fYBnf09uu19uocHxz/ylqkpdCnWD+l18uWD82prM83HYikTfj29UQlb68S9Lp3dB3QP9+gd7M1QFsqgclIkRVqR1NxnMKyXB4en+x+WZPes9As8h2CJljfBKmiWCRCRI2UNUgtMqrDtsv/wNZK4LNa1FVMah53uDIqzY/MwWZYU1w9oXLxMg8sAxL0Onb0d2g/v0TvMUzWk5FR4mS9vSs7AXz6PLaugWaZmUGEL6k3uvfj8pEcWOa0VRJHWxUwYZ/U6HReinApzaRRti6VYXD+ksXmJ+sY2d7/+JZJSpJcxJx8N1Tw/p+8EU+7rjiOEYTBQY1Ro51yNXqb0Gn2sgmaZibDRwnG9CQVcNS6gwl3QDPpvdBI9VSIm79INRAc7WA80yyohItTWtzh47c6ip4JZyVPkEqe5rqJEeHBoZq6HnZgkUZwlqS0NNorTMiMiQjhxktuqXDgFq2hGg7XMXcI2Sa1lFQlaU0R8VmTpsNhk2yaDGcx1pRQQ2b5grIJmmZmNy9fxgiWyvRR9jZuUb469hC2WWQibF2huX5ts50IVAINLnBVRJKUiqyqLwkp3y8w4jkvjwuaip2GMwkVGSS1oVCKVisVintqFi4gzQQRgRTL0VwXTn+YS1UkHrIJmMURrc3u88Cr8pq4ad41GMfiDEXcObfoUy0rihTXCtY3RO1XEKgXMIc1GOen0lusm0wYJWIwgItSaa0TdTvYjr9k/PUqIKuKcWJI7V4QMEy6qaZLCOQiHKikp4vio00s/1/7nO/gsTvp5Zg/Joj2POzi5nUQ9XD+Y/4lYLAsmaLSylBuayhs4UkZUFRHBcZwjO9f5KpAMffNIfs1DvlTE0icCjqS/H46AOILT3xYZEF1paJVAFrB+fH6DZxrFy2VCswqaxRhb1x+dyzjnKWp9IXqqtf/vSDiqJlkf2ftwYhtN8LY2jwW0KgxESo5U4I6S+OjxsQwIZlUSv0ZNHMRxEcfJHm469yxlxtF5SH9bMsHkHEmpqviZWCxVoL6xTX1ju/Bxzr/RPCcB7ICyqKceJ+VX2ocqOBqfkEMnx9PBro8YFCfpoQOya3B8xyN0AhzHSZUpx8FxJFOoshj1o+csaj17lgHlqv/82PiPbGWxCpqlcojMQzmZNDLVYrFYJmc+8suyDFgfNIvFYrFYLJaSYRU0i8VisVgslpJhFTSLxWKxWCyWkmEVNIvFYrFYLJaSYRU0i8VisVgslpJhFTSLxWKxWCyWkmEVNIvFYrFYLJaSMVZBE5FbIvJZEfmKiHxZRP5m1n5RRD4tIl/PnjezdhGRfyoit0XkSyLygwN9vSfb/+si8p6B9j8lIn+YHfNPJUsSc94YFovFMglWflkslqoyiQUtAv5nVX0KeAvwMyLyFPAB4LdU9Ungt7LXAO8Enswe7wN+GVJhBfw94IeANwN/b0Bg/TLwPw0c946s/bwxLBaLZRKs/LJYLJVkrIKmqi+p6u9m27vAV4EbwNPAR7LdPgL8uWz7aeBfa8rngA0RuQb8OPBpVX1NVe8Dnwbekb13QVU/p2lNiX99qq9hY1gsFstYrPyyWCxVZSofNBF5DPgB4Fngiqq+lL31MnAl274BfGfgsBeytlHtLwxpZ8QYFovFMhVWflkslioxcS1OEWkB/xfwt1R1Z7CWmKqqiJgrcT+EUWOIyPtIlyMA9kTkuRxDbAN3886vBFR9/lD9c6j6/KGa5/DouB2s/KoEVT+Hqs8fqn8OVZ3/UBk2kYImIj6pcPu3qvrrWfMdEbmmqi9lZv5XsvYXgVsDh9/M2l4E3naq/bez9ptD9h81xglU9VeBX53kXM5DRH5HVd80Sx+LpOrzh+qfQ9XnD8txDqex8qsaVP0cqj5/qP45VH3+p5kkilOADwNfVdX/beCtZ4B+JNN7gN8YaP/LWTTUW4CHmZn/U8CPichm5lz7Y8Cnsvd2ROQt2Vh/+VRfw8awWCyWsVj5ZbFYqsokFrQfBv574A9F5Peztv8F+F+Bj4nIe4FvA/9d9t4ngZ8AbgMHwF8BUNXXROQfAV/I9vuHqvpatv3XgH8F1IH/kD0YMYbFYrFMgpVfFoulkkgaeGQRkfdlSw2VpOrzh+qfQ9XnD8txDqvIMnxvVT+Hqs8fqn8OVZ//aayCZrFYLBaLxVIybKkni8VisVgslpJReQVNRGoi8nkR+YOslMs/yNofF5Fns/IrvyYiQdYeZq9vZ+8/NtDXB7P250Tkxwfa35G13RaRDwy0Dx0j53m4IvJ7IvKJis7/eUnL3fy+iPxO1laZcjoisiEiHxeRr4nIV0XkrRWb/xuyz77/2BGRv1Wlc1hFxMqvssy/0vIr66uyMkys/BqOqlb6AQjQyrZ90iSUbwE+Brwra//nwF/Ntv8a8M+z7XcBv5ZtPwX8ARACjwPfANzs8Q3gCSDI9nkqO2boGDnP4/3A/w58YlTfJZ7/88D2qbZ/DHwg2/4A8HPZ9k+QOlJL9l09m7VfBL6ZPW9m25vZe5/P9pXs2HeOGiPH/D8C/I/ZdgBsVGn+p87FJU2M+mhVz2FVHlj5VZb5P0+F5Vd2/FLIMKz8Ov4sFj0BoycDDeB3Sevl3QW8rP2tpCHxkIbLvzXb9rL9BPgg8MGBvj6VHXd0bNb+wewh542RY943SWv1vR34xKi+yzj/7PjnOSvgngOuZdvXgOey7V8B3n16P+DdwK8MtP9K1nYN+NpA+9F+540x5dzXgW+R+WRWbf5DzufHgP9c5XNYxQdWfln5lfPaYYlkGFZ+HT0qv8QJR+b13ydNBPlp0juuB6oaZbsMll85KtmSvf8Q2GL6Ei9bI8aYln8C/G0gyV6P6ruM8wdQ4P8WkS9KmhkdqlNO53HgVeBfSrpM8y9EpFmh+Z/mXcD/Mab/sp/DymDl18LnD9WWX7BcMszKr4ylUNBUNVbV7ye9k3sz8MYFT2liROQngVdU9YuLnsuM/BlV/UHgncDPiMiPDL6p6e2JFjmBGcbwgB8EfllVfwDYJzV1m+h7YkyMkfnh/BTwfxbR/zjmMcayYeVXKaiy/IIlkWFWfp1kKRS0Pqr6APgsqbl7Q0T6iXgHy68clXLJ3l8H7jG6xMuw9nsjxpiGHwZ+SkSeBz5KukzwCxWaPwCq+mL2/Arw70h/aO5IWuIGmbycznntI8vpDBljGl4AXlDVZ7PXHycVdlWZ/yDvBH5XVe+M6b/M57CSWPll5Rf5r51lkWFWfg1QeQVNRC6JyEa2XQd+FPgqqaD76Wy39zC8/MpPA5/JtOZngHdJGmX0OPAkqVPhF4AnJY0YCkjNr89kx5w3xsSo6gdV9aaqPpb1/RlV/UtVmT+AiDRFZK2/TepD8F+oSDkdVX0Z+I6IvCFr+rPAV6oy/1O8m+PlgVH9l/kcVgYrv6z8OmeMqVgiGWbl1yBFObfN6wH8CeD3gC+RXlR/N2t/gvQCv01qLg2z9lr2+nb2/hMDff0dUv+P58giPLL2nwD+KHvv7wy0Dx1jhnN5G8dRUJWZf9bPH2SPL/fHIPUT+S3g68D/A1zM2gX4xWw+fwi8aaCv/yGbz23grwy0vyn7fr8B/DOOkywPHSPHOXw/8DvZ39G/J40Aqsz8s76apJaF9YG2Sp3Dqj2w8mvh82cJ5FfWV6VlGFZ+nXnYSgIWi8VisVgsJaPyS5wWi8VisVgsy4ZV0CwWi8VisVhKhlXQLBaLxWKxWEqGVdAsFovFYrFYSoZV0CwWi8VisVhKhlXQLBaLxWKxWEqGVdAsFovFYrFYSoZV0CwWi8VisVhKxn8FpejdGdfTq70AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(ncols=2, figsize=(10,5))\n",
    "wi.plot(column=wi[\"pc0\"], cmap=\"coolwarm\", figsize=(8,8), ax=axes[0], vmin=-cap, vmax=cap)\n",
    "wi.plot(column=wi[\"pc1\"], cmap=\"coolwarm\", figsize=(8,8), ax=axes[1], vmin=-cap, vmax=cap)\n",
    "axes[0].set_title(\"Component 0\")\n",
    "axes[1].set_title(\"Component 1\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although all 72 counties could in theory vary independently, we can take different combinations of just those two maps (added to a base map of the average per county) and explain 95% of all variance.\n",
    "\n",
    "Let's see what that base map looks like (similar to component 0, it turns out):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NAME\n",
       "Adams          4.484290\n",
       "Ashland        3.334858\n",
       "Barron        15.192472\n",
       "Bayfield       3.024574\n",
       "Brown         85.731733\n",
       "                ...    \n",
       "Waukesha     115.279972\n",
       "Waupaca       13.570511\n",
       "Waushara       5.961534\n",
       "Winnebago     48.406193\n",
       "Wood          18.998011\n",
       "Length: 72, dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean = pd.Series(pca.mean_, index=dft.columns)\n",
    "mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'mean')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVgAAAEeCAYAAADPbQ0xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO29e7Q0WVXg+dsRmRmZed/f+1VFoZTQ6FIe34Jyads0tFAgbTGOOmC3lDTL6l5it65xWkFnhmkfa+GaNb5aZKYW0IKjlgw2TS0txAJhnF4zhVTJGwQ+HsX3ftz3vfmO2PNHnLw3b958RWbkI+Ke3/riu5knIs45mRmxY5999t5HVBWLxWKxxI8z7Q5YLBZLWrEC1mKxWMaEFbAWi8UyJqyAtVgsljFhBazFYrGMicy0O2CxWI42L3TmdEv9yOddovphVb1/DF2KDStgLRbLVNlSn9/JPCPyea9ufOXEGLoTK1bAWiyW6SIgWYl+XiP+rsSNFbAWi2WqiAhOZggBmwDsJJclNkTkmyLy70XksyKyKyLvEpHTIvIhEdkWkY+IyIo59j4R+X9FZENEPiMiL2mp5w0i8iVzztdF5F+37HuJiFwRkV8QkVsicl1E3jCFj2uJCwHJOpG3JJCMXlqSxH8L/CDwHcA/Bz4E/DJwkvB6+3cich74S+DXgWPA/wD8uYicNHXcAl4NLAJvAH5bRF7Q0sYZYAk4D7wReHtTcFsSiICTkchbErAC1hI3/1FVb6rqVeD/AT6hqp9S1QrwAeD5wL8EHlPVx1Q1UNXHgSeBVwGo6l+q6tc05P8G/hr4xy1t1IFfVdW6qj4G7ADPntxHtMSKscFG3ZKAtcFa4uZmy+tyh/fzwDOAHxORf96yLwt8DEBEXgm8lVALdoAi8LmWY1dVtXWKo2TqtSSQNNtgrYC1TIPLwB+p6k+37xARD/hz4PXAB1W1LiL/BUjnHWgZ3osgAVgBa5kG/yfwSRF5BfARQu31PuASsAl4wG2gYbTZlwOfn1JfLePG2GDTiLXBWiaOql4GHiCc/LpNqNH+e8BR1W3g3wHvA9aBnwAenVJXLRNAAHEl8pYExCbctlgs0+Q5xaK+897nRD7vH3/2U0+p6sUxdCk2rInAYrFMGUGcZGikUbEmAovFYhkTVoO1WCzTRUDcdOp6VsBaLJapIoCTkEmrqFgBa7FYpouQWhts6gTsiRMn9J577pl2NyyWI8dTTz11R1VP9j+yHbEabFK45557ePLJJ6fdDYvlyCEiTw93Honxa41K6gSsxWJJHuLYSS6LxWKJH2uDtVgslnFhbbAWi8UyFsRqsBaLxTI+rA3WYrFYxoHVYC0Wi2VcWBusxWKxjAVrg7UcOer1Gtlsbu+9qqKqOC22MlWlVq2SyWRwM/ZSsgyPtcFajhQi4QUfBAEioXbRnpx9Y32NjfV1Tp85S9EKWMuwWA3WcpSo12pkcznK5RK+75PJZMnn8/h+A9d1946rVWsA5AuFaXXVkgpswm3LEUFVuXrlMoHv47oZCoUiuVzOmANqlEulA8d6+fyehmuxWA5iNdgjSrVaYXdnl5Vjx/aG/o7jICKcO38Bx3W5ff0ap0+fQUUol0sUCkUwxwZBQKFYZGFx0QpYy8ikVYO1AvaIsr21RbVa5ZgcB8D3G4gItWoVL59HVTl77jzVahW/0WBhYREIBWtzsmtpeXmaH8GSEkIvgnQOpq2APWIEQYDjOKwcO47jOARBwNrqKpVyibPnzuPl82ysr1Pa3eXMuXM06nXm5ufZ3dlhZ3ubSqXM0vIKyysr0/4olhRh/WAtqcBxHFSVG9evha+DgGq1CsDq6h0KxSKu61KtVlhfW+XY8RNcvXKZei2c0GqaBSyW2JD0TnJZAXvEqJTL3Llze09gtrK7s0O9XufkqdOcOnOGTCaDBgG5bI5jx0+Qy+XwG/ueBKpq7a+WWEiriSCdn8oCcECI1us1arUqq3fudBSuTWrVKrs728zNzVOr1qg3Gpw6c4ZisUgmk8HL5wmCgK2tTWo96rFYBqUZyRV1612n3CUiHxORL4rIF0Tk50z5MRF5XES+av6umHIRkd8TkUsi8lkReUEcn80K2BSiqqytriKOQ7VaoVqpsLG+ztXLl6nVqj3PdV2X+YVFtre2mF9YwPM8Sru7rK+t7R3TjN7yPG/cH8VyRIhbwAIN4BdU9bnAfcCbROS5wJuBj6rqvcBHzXuAVwL3mu0h4B1xfC5rIkgZoXC9w9bmJo1GnXKpRBAEA5+/uLTE1uYmxbli6FVQq3H71k2CIMDLexSLczawwBIzEruJQFWvA9fN620R+RJwHngAeIk57D3Ax4FfMuXv1dBn8QkRWRaRs6aeoen7qUTk2SLy6ZZtS0R+fhhVW0QeNMd/VUQebCl/oYh8zpzze2IMe93asBymXCpx9cplvvn1r7G1uQmENtUowlVEmJubZ3FpCc/Ls762yvraKidOnuKeb/t2isW5cXXfcpQZ3kRwQkSebNke6li9yD3A84FPAKdbhOYN4LR5fR643HLaFVM2En0FrKp+WVWfp6rPA14IlIAPEFHVFpFjwFuBFwMvAt7aIjDfAfx0y3n3m/JubVgMjXqdUmmXG9evUav2Hv734/iJk2RzORzH4fatm2ysr7O0tMzc/LydzLKMkVCDjboBd1T1Ysv28KGaReaBPwd+XlW3WvcZbVXbz4mTqHr5y4CvqerThCr1e0z5e4DXmNd7qraqPgEsi8hZ4BXA46q6pqrrwOPA/Wbfoqo+YT7we9vq6tTGkUZVqVar3L51k8vfepqb10caxQBw+sxZ5hcW2Nrc4PLT30QDZWFxEd/3Y+ixxdIHkehb3yolSyhc/1hV/7MpvmlkDubvLVN+Fbir5fQLpmwkotpgXwv8qXkdVdXuVX6lQ3mvNg5ghgYPAdx9993RPtGYqZR22d3eZHHlBNlcrv8JHajXauxubyBOBj9Qdra3YhV8c/PzFIqhzdX3fYpzc5w4eepAakKLZVyMIx+sMTO+C/iSqv5Wy65HgQeBt5m/H2wp/1kReYRwlL05qv0VIghYEckBPwy8pX2fqqqIjFXV7tWGGRo8DHDx4sWx9mNQfL+B32iwdusGjXqNSqnEmbuegZvJDnS+BgGVconS9ha726FNNVucp1IZzQzQiXyhsGcCWF45Zs0BlokzBj/Y7wN+EviciHzalP0yoWB9n4i8EXga+HGz7zHgVcAlQjPoG+LoRBQN9pXA36vqTfP+ZnOWbUBV+yr7s3fN8o+b8gsdju/VxsyzcfvWnmAE8Bt1bl39Fgsrx5lbWOoqxGqVCtuba5R2ttEIE1SjsLO1tZdroFat4rgu2exgDwKLZWTGEMmlqv8V6Fbpyzocr8CbYu0E0Wywr2PfPAD7qjYcVrVfb7wJ7mNf1f4w8HIRWTGTWy8HPmz2bYnIfUatf31bXZ3amHnmlw87PNRrNdZuXmf1xjUq5RI7Wxus3rzO7etXWL15nZuXv8mNy99gd2uzs3Adk25erVap12qICF4+T61WO5Rc22IZJ0NOcs08A2mwIjIH/CDwr1uKI6naqromIr8GfNIc96uq2vRe/xngD4EC8CGz9WpjJtnZXCeby5PJZrlzvbt9vLSzRWlnq+v+SbOwuHTAPlwsFmk0GmQyGWsusEyEI52LQFV3geNtZatEVLVV9d3AuzuUPwl8V4fyjm3MKjubG9SqlbHVPw6d8tSZsxTNBFeTne1tVu/cJl8ocOr0GTvZZbEMib1zYiSJw+p824oEtWqVjfU1VJVyqcSNa1etq5ZlrIwjF8GsYENlYyRpAnZhcWkvM1a5VGJjfZ1KpXzgmGq1yvWrVzh99pyd+LKMCYGUjpKsgI0R1fHO+sctwOfm5qjX69y+dZNqpbtpo16vc+3KZU6eOk2hWDQLIdpLxxIfabX127skRpKmwa7euU29Xh/o2CAIuHnjOtmcR61WY3l5meWVFUQktTeHZULYJWMsg5A0ATtIuGE7jUYDgI2NDTY3NykUi5w+fZogCGg0GjaFoWUIkmNTjYoVsDEybgEbe/0jVqeqlHZ3uXXrFrVqFVXlLhOqbLVay8AI1gZr6Y2q7i1pPTZiF1rx9Hd3ZwcI1/u6euUKx44fJ5vN2kkxy8BYDdbSk8SZBxiuz7000yAIqNVq3DDZvc5fuGBNBpa+CIJIOjXYdH6qKTARAZswId66zIzF0hUBHIm+JQCrwcbEuF20xsIQJocoD5JSqUS5XKZgl5ix9CGtXgTp/FRTQINkaZfQPdVQnKzeuTOBVixJx0ZyWXoyCQ02iXbeer2O7/t7EWMWyyHCWNlp92IspPNTTYEkCr9J6LCqyubGxtjbsSQbq8FaepJIATvENTrM59ze3mZpedlqsZbupNQGawVsXExAwMbexISeCb7vU6/XrYC1dCTN4dZWwMbEZLwIYpaIE7ymr129Sj6fp1AsUiwWyeVyqb2pLENgNVhLLyZhIoi7jWEEXPggGU4wVioVKpUK62trOI5D0QjbfKFgs3MdcZJiU42KvapjIoluWsMpxPHcCEEQsLOzQ6Ne587NMEtXoVikUCzi5Qt2FQVLKrACNiYmosGOvYXeiEj8dmAjr+u1KvVala2NdUSEfKFAvjhHoVAka80J6SbFblpWwMbERGywU/ZUUFXynseeVBSoVqqxC/7mcjXlUol1wHVdCsU58sUihUIR15oT0oc1EVh6kUwbbPRzqtVqWyXOaIJ/gFN932dne4ud7XAl3pznkS8UKRTn8PJ5a05IAWlN9mIFbEyMPRdsQuqMzBBCvlatUqvumxO8fMHYb+esOSGJNJO9pBArYGNCxSGTP7j8deslI4C2izRlv3T/v31hbf4qigaA45rJNI1FoMsk/bTGhKpSKZeolEusr97BcV0KhWJoTigWyWRsTtrZR1Kb7MUK2JgIgoBqrfv6Vl4+33NhwX6ICBph+ezQedvBcYwTtxH8YkR9XMt0eF6utVWqlcpUNePA99nd2WZ3ZxuAbDa3J2zzhaI1J8wqKR11WAEbE/00ytEvn6ZgHAxVRdUn6DH35nn5kXvV/tAQx0V7NTph6vUa9c0a25thPoS5+QVWTpywmu0sYZeMsfQj6OMHO+qIXmQMTgRjUBpmPSfD7s42pd0dFpdXWFo5ZjXamUCsBmvpjKqysb5OubTb+8BRr5+EXIChr2wEITsFeayqbK6vsbO1ycrxk8wtLNiJsSljbbCWjtRqVTbWB1gaZUTNLim3v+M4BDNkIuiF7/vcuXWDcnmXhcVl8nblhekg2EADyz7lconS7i5LyytUyoNNXOmoInIcGtYYtMfIvZyBJ0e93uD6tasU5+Y4cfLUntnAarWTIjlrbEXFCtiIqCp3bt+mUa+ztbk58HkyojQby+WXzmt6aEq7u1ytXKZQLOKIEGiAl8+Ty3nk8+GEYK1WC1fPrVbJZLMUi8Up9zr5CDbQ4Eijqqyt3iGbzaGqNOrd3bG6Mqo2lBBt6pCvb8Lw/cZexBjAzvY2Xj7PqdNncByHa1ev7plAcrkc9VqNXC5HoVik0Wjguq7VfKNiAw2OLqrKzvZ2JG21E41GY+Tz47ZvaqB4+XybqSD0kT34HlrdxHqJ0ND3NhQyrWfu1WZs0U0ZNI3VeDPZLK5rLn2BRr23f3Gj0eDyt74FHPSSqNVqrK6uAlAoFKgZYXvy1CkymUzLZ02n8IgPm+zlyFGv19ne2mJ3Z3tk4QgxjMZV0Zhv1Fqt2nO/53mHcw+MiOflqI0QcBEHrpuhEuFz+Y1GXxt6uVze+3v16tVQ4FarzM/P47ou2VwOz/OssO1GSr8XK2A7sL21yZ3bt+OtNI4LaMZ9TJPCuM0YfqPBznYYSba2tu9hUigUWDl2zAraTqTUTWugTyUiyyLyfhH5BxH5koh8r4gcE5HHReSr5u+KOVZE5PdE5JKIfFZEXtBSz4Pm+K+KyIMt5S8Ukc+Zc35PzNXXrY1xoarcunkjfuFKUueTxtHrGfgmpvScKpfL3LxxY+aDMSzxMehj43eBv1LV5wDfA3wJeDPwUVW9F/ioeQ/wSuBesz0EvANCYQm8FXgx8CLgrS0C8x3AT7ecd78p79bGWNja2mR3Z2dMtY8uWCZ/Y46hPQ0nh3I5j5y3v3kHtnzLX/M6n99/n8/vv8/vb3mztS8/kztwvjfV6K3mApCWFpoJt6NuCaCviUBEloAfAH4KQFVrQE1EHgBeYg57D/Bx4JeAB4D3aigNnjDa71lz7OOqumbqfRy4X0Q+Diyq6hOm/L3Aa4APmbo6tRE7qjpG4ZpaE1N0RGlUymNtIpMvQovdXIB6eT/SLpsfxrWqffJveFbv3OHsuXPWTNDKGLwIROTdwKuBW6r6XabsGPBnwD3AN4EfV9V1M2r+XeBVQAn4KVX9+1H7MMhj4JnAbeA/icinROSdIjIHnFbV6+aYG8Bp8/o8cLnl/CumrFf5lQ7l9GjjACLykIg8KSJP3h5yeL++tjZStqv+2JspZALfQxDg5XJ4uRz5vIf67ZOU0xuiZzIZTp8+bYVrO+PRYP+Q/dFwk0gj71EZpJcZ4AXAO1T1+cAubUN1o62O9art1YaqPqyqF1X14smTJ4eqf+zXu72fJkajVqFeLVOvlqmVSzTqtQP7p2kCPXP2rF3yphPNlJpRtj6o6t8C7XHsDxCOhjF/X9NS/l4NeQJojrxHYhABewW4oqqfMO/fTyhwbzY7YP7eMvuvAne1nH/BlPUqv9ChnB5txEqj0WBjfX0cVScc+1RoEldy8rXV1T1buqpSq07XZW0mEAm9CKJucKI5cjXbQwO0FnXkPRJ9Bayq3gAui8izTdHLgC8CjwJNT4AHgQ+a148CrzfeBPcBm+YDfRh4uYismMmtlwMfNvu2ROQ+Ywd5fVtdndqIlWZy5nEy8u2ZkiFlYj9FTN9/qVTi2tWrbKyvcvXrX+bm5W8QREiknlqG02DvNEeuZns4SpOTGHkPOlb5t8Afi0gO+DrwBkLh/D4ReSPwNPDj5tjHCA3FlwiNxW8AUNU1Efk14JPmuF9tTngBP0NoLykQTm59yJS/rUsbsRAEAZubG2ysDZANa2RGu0GjpduOicRKw35M102qWq0i6F5i8ltXn8bLFxFHyOY8igtLR89GOzmvgJsiclZVrw848h6JgQSsqn4auNhh18s6HKvAm7rU827g3R3KnwS+q0P5aqc24qBer3Hj2rWBo7TyXi4UciKgLbeohm7rrcM+VSUIgnjdqsaScXvyJPUTRM5z27/GvVf1aoV6i6kgk/Pw8kcodWLTRDAZmqPit3F45P2zIvIIoSvpZospYWiOrLW9UqlECoH1q6WBl0IRwCvMU6nWDhaOwFQ02JQyjJzM5bI9PbUEgWZehfYfy7zfD7dVhO7X0u2rTzO/tMLS8VNHR5Mdw+cUkT8ldPM8ISJXCP3wu42KO468R+XICthaxBj7QKNmdI35gjkqN9oEkCEkbD8XvoKXo16Ox49ag4Dt9VXcTJaF5WOx1DnzjMFEoKqv67Ir0sh7FI6sgK2UIzq7T3l4bsVrfIwjF0HcNTpuZiqZxqaDXZMrVagqfnWXDIIjoctHuIR1+ENL869JuRdGAtX61Dpu0nkBWjqTL86xuHJi2t2YDEJqk70cSQErIrjiEAQ+gQb0MIcNjRPUyfqVUECLEJTreCKoODgiqBHgTYGu0u5p2fZUFyHI9vi5dO+/w+ld28tby8xr7XBsWkX6ODTYuG2lpe1N5haWyM/Nx1rvLKIQeyrOWeFIClhVDQXrBJCwQVB/T44P4/XoZrJU1Y2vYwOg2exE20sy4zAg3b72Lc7ecy+Z1P8O6U24nc5P1QcRYX5xhUwuByLkxuISE/Mtl9In/FQYhzQcQ52O404189dEOarZtNLKyqkzB5b02F5fZePOzfgasD5Vs0sCfpviwhLZnIfjTnbUMi2siSCFtNrNFlaOU9rejC02PG47X1yx8LNAd3ul7Bl+D3xeOXhM+9n+oWxZ/Zh9C3N5Z5uFu45PuxuWETnSAradTM6LL/lG3FpSip7wQS/hpodedJq1O0A2m4220m8CNFjVgHq1Qs7LT7sr40esDfZIsHLyDBKTzSt2H8b0yNcxEO3LScrS4pXSbv+D0sIY0hXOAlbAtuC47tHxPRyAZFzCGB/m9DGeydcZZbh0hTOPNRG0sXjsBOI4bNy+MVI9cWuwod1ysppXMvQ8CB8Fsjfjru3mhb1cAGqS88x+JJdXmGNuYSnmWqPRTFw0fk8GSe0kVzIeAxMmjgtKg9iNsDHXN4stDkeYOkVoKDQUfJX9jf2/AU64qZko02Bvm6WkKl6hyMnzd0/Fg6Bc2sX3fUq7OwRBwOrNkRNK9UewblpHiUGzZvWuJDn6XzeS8gkkik2ueZz2nGqbOtMQ+KrK+p3b1Gs1QJlfXOLEmXOTaTshAjMqVsC2UauUWY/BHzZ24TRDGtYojGPp8ThqnPyS6N3p1BcNAur12li9ClSVRqNO8xvNefkJCfrkTFpFxQrYFhr1Oqs3r8aifeqMa0hTYxx5P0f8pgf5tTsJmtYyEfaG9KphRvZh7fDLJw4unqyq3L5+mVq5xMnzz8ArDLPseH9830eDADeTQYOA+cXJ2YCtBnsE8Bt1GrV4smYlQbjmslmCwN8TFOHf/Z4Hvh8mmjZlcuA/I5ZMgpggXNZhr/yg66pZAkIEZwbjVFs/tbbMiu1F+gH5bI56tdS1jkbjoP90NudRr0XLOQxhzolKaRcRoVou4bguldIuVeOytbV+h5OFuyPXOwiZTIbjp8/iiOD7/mTNFFaDTTeqyu7WRix1iVlWJl7ivwA1CPAjrOoQB5nMGCZuYvhqDmqb+4J10viNOltrt9lev9PRVJDNjddEkMt5BL5PcX5hbO0cIsWBBlbAAkHgs3H7ZmwCNilP46Q43B9FOglXN5NhYSXeFQ5Kuzs44rCztcHC8go5L08QBNRqNXK5XKxtdcOmK0w51VIpPuEKe4m6LYdJ7G0UsePjeHgtHT+F68Z3y6oqa7dv7o1isvk8Xr6A67q4k3YRsxpseonf3zB+MeKoT76+c3AtPcehmSBFMEK9mbh7z57aXJ0h7NdeYm+F6hRmzsfT4vjF9rQfmCJCvhhf8u0gCELh6u9nJ85mJ6OxdiLqindJwQpYiN0OORYTbLPu1jct/rqt7XVqu9N+8SZoZ0s40779MzkPNxPf7Vou7bK7vQVAoThHvV6bYrCFWC+CNCOOkDFP79DFJgynbIYKRnfbGseFOgb/0WmrZZaBmV9cjnxOEPg4TufRWWswTb5QZPn4yVgFeGSsgE0v5Y1VtM0F56DrTjgcF5zwOhCHA4sjwv7QWwQRh8DNdJjr0ua/A+9Dl6Dm++b6WE0BH4xREloJOzZi/GpFBK84F+mcjdU7OI7DYodJsUq5RLklU5fjuuQ8b+R+Do3YSa5Uk59fpLKz1XW/AAQBSsAgvuOZnEetMczKWy19yrg0apU9Id8oRch3OiBWvA5OveGTKczTKO8MdHyjXsPN7K+l5TjOUH6xEI6qWusahCDw91JvBkGABgHVSoV8ocDNq5cPHDtVzTXl2G+WMSiIMTyNJxK6aW0EA+P73YfbXc9p7D8U/RGvibWb1zh++tzA+YqPnTxNEDQf8kqtVuX2jasU21apPXn2PKWdbQoRNeQ40RTbYNP5qSLiFeMOPYzF8330OmaR2QvkGpwpjmLLO1tcf/prBBESETUfCI7jmgQuod/rgXp3d1g5cSq+jg6LTbidXkJZFt8PpoEfg7/i+KXGVBKcSEofHBPAb9QpbW1E+t2ax25vrnfcv7MV3zp0o6DiRN6SwJE3EVRLO6xd/SZxCjS/XiNTmBtiMb4Wsh7U48mL0A3HcSJpRFHJZbNkZH+BQRXFUcXN9bAnNtMctD/vevw8go+by7QW7B1faQQ9P2Mun++ZLKZ9QjKTnzvUHb9a6j/iUCXr5RFCm2hjyN92/fYNttZXKcwvsHT8VN/cxSISTpZ2yE+c8/LMLy3NwLpfYv1g00jgN1i78o3xaHIjVjmZy228rWQlwN2cQMLmHjgL53o/RBQqQ04+Ncky2M9dN5pidkSB5jfqiDhcu3aNCxcu9D1+c3310MM+k8kyt7iE5+UnsGJBf5KikUblSAtYcVy8uYWeHgTDkoyBcDJ6OQp9TXVxmMtHryIyC8srLBvPglBDDTpGJO5ub1Etlw+Vu5kMC4tLs7GSg5AYm2pU0vnYGBARYeXcM3AiusAMQhISqcxSkunx0efG1WQuxeO02PhFpKNwVVXW7tyiUj6cZjFfKM6GcAVCE4ETeUsCA2mwIvJNYBvwgYaqXhSRY8CfAfcA3wR+XFXXJfzVfhd4FVACfkpV/97U8yDwP5pqf11V32PKXwj8IVAAHgN+TlW1WxsjfeI2Ar9B0IjfxzQJXgCe5xkh2xpS0UEj2zum5bGxF29rEl6LsaS1fG7xpz95knEcyOX2hVq7UIlDxkT8qRXI5YttJwqBX6dRj+9a3FxfJfAP+2MvLC2zdOx4bO2Mis2mFfJPVfVOy/s3Ax9V1beJyJvN+18CXgnca7YXA+8AXmyE5VuBi4Tf6VMi8qgRmO8Afhr4BKGAvR/4UI82YiOW9bc6MvsXTL28G6uePZ8BdlZjrHF0pFEmaHT/jd3C5PMx1LvM2ufyRbKOuxfN14wMDHz/gE+tDOCP6/s+m+trh8pFhMWV4zOkvYak1QY7yqd6AHiPef0e4DUt5e/VkCeAZRE5C7wCeFxV14xQfRy43+xbVNUnNFSn3ttWV6c2YsPNZMmMJYnx7GuwcTNbt2zILPapG7VKiXq1Qr1WpVGrUq9VqVcrOJkMkslSXFph5dRZlk+d7SsgK6XdjqOobM4jM4ORW2o8CaJsSWDQb1qBvxYRBf4PVX0YOK2qzSniG0BzIaHzQGss3hVT1qv8SodyerQRG+I45ApFGrV4h7NHT7xCEj/1LN+mXqHI0onTePlCpPNUle3NzvmN67Uq1XIZrxCtzvGS3kiuQQXs96vqVRE5BTwuIv/QutPYS8d6d/VqQ0QeAh4CuPvu6OsVNUZ00+nE6BNIyRNWs0lvESoo2Vz2gC/sfkqVLXQAACAASURBVD7d8Hes9VunrcXvNi4cx+X4mQtD5Qmo12pUK4c9BwAWl4+RnWZily6k1QY70GNDVa+av7eADwAvAm6a4T3m7y1z+FXgrpbTL5iyXuUXOpTTo432/j2sqhdV9eLJkycH+Uh7lLc3qJV3+x8YlZEF7ASSSCdgIm7c1Ms7aGWXoLKzt/mVHfzyNn55G0f7J+0Zxy/luO5QieBVlaDbgoUiLK4cmwm/11bCpTLTaSLo+02LyJyILDRfAy8HPg88CjxoDnsQ+KB5/Sjwegm5D9g0w/wPAy8XkRURWTH1fNjs2xKR+4wHwuvb6urURmxEzVI0KKMKr2QuOjOLF/1o3+NgZ8f/uRv1GuXdwTJ37Z3TqLN+5xY5zyPfIXmLwEyExR5C5EiHyp4GPmCeiBngT1T1r0Tkk8D7ROSNwNPAj5vjHyN00bpE6Kb1BgBVXRORXwM+aY77VVVtTnP+DPtuWh8yG8DburQRG74f4Jjwx+aK1O3DxYORm+HzVlUhCEwqOP+QN8KoAjaJ4jWNDPSgG4OJwCvMUZgbbImYRqNBo15j/c4tatUqIIe01Gwux7GTp8kX4k5sFA9J0Uij0lfAqurXge/pUL4KvKxDuQJv6lLXu4F3dyh/EviuQduIE4X+Nra+uGDca6SZ6WeQxLE9cMQxvpIclPLmRj4QyKBtJdriz9ou6Fv3xe6hNouPhdFu3HAp68NrVcnef+Bq+Nt3i0Zq1KqRcz4s9HGl8hsN1m7fJOt5VCuV0GvAsLO1wdzC4oHjT529QCY7ntFaHCRFI43K7PlrTJg4V5OluZqsBiM/jzXw0d3RYir69iG32O+I5DPi3KtfqwK9J0H7WWkz2RxBMPhD/NSFe/B6aJqqyu2b16mWSziV8qFgAlWltLPD/OISO1ub5AvFmRau40JE7icMenKBd6rq2ybdh3Q+NgYk8P2OYYQWy7QozC/0FK4ApZ1tqua67RSpBeGKBjtbm+SLc5w6e77jMbNE3JNcIuICbycMfHou8DoRee4EPsoBjrQGu37n1lgiuSQWo9wsDrd7I5qEDAyzzdxC78UNw+W2OzrTdMRxnIFXQZgWY1rR4EXAJWPiREQeIQxc+mLcDfXiSAvYbk//kREZ3U3LSqpYkCkkcznMYD+mm8mQ7zOxtXrzestSMP1pt8XOKkNOcp0QkSdb3j9sgqCgc2DTi4fs3tAcaQF7/Mw5ttbvsLU2W/HzIeOWsLMgeCbB9J9UYe6A3klcHNfl1IV7ek5slUu7h5Z86UfUKLBpMWSgwR1VvRh3X+LkSAtYx3FYPn6K4twCt65ejqQZ9EQkBgvBmAXDUZGvM/A5B0mssrB8jEz2sLdCKxurd3rubyfnebhDBCtMA41/pNEtsGmizLZxZkLk8gVOXbg7NlvVrGUq6sRY+jiDH3s2lgDr/8XsueT1IGoE1uLysUjHT4+x5IP9JHCviDxTRHLAawkDlybKkdZgW8l5eY6fPsed61f6H9wX2Vu7CWkNW1Czt/lO2zTdppOrQmPcETdjkIYzIczamAGh77Q5HDe1SjHXhkgYhdWLcmkXRXsM+XVv0KMoruNSnJ98KsZhaIbKxlqnakNEfpYwgtQF3q2qX4i1kQGwAraF4vwCpy/cw+3rl0eaABNHqNRGS5y8SDBeeTUGDbaRKaDFDgmtey1Y2HK4k8niH0g4rXsThiaWwpTux9aFAsq8FjkUeaVulqx0mndsfbh1aqdLp1WRTIZ6dfAEQdnt2xR31qBexqnsHBIlsnSK3Nm7Op4bNqlsrN6JFOZaWF5JxEiqyTgiuVT1McLI0qlhBWwbXqGAly9S3t0euo5Y3LQSmIglIAzbHJasm6M+wvkd68x4NKqdM0sNX2c0p30/v0j++pe7HzC3hHiHNVMNAvxvfYF6tUzNXRq4PTeTSZB5IOTIhsoeRZaOn0BEKO9uD5dTQBLgB5sg7SbtSK6INsOsDf6tb9F4+nPo1ipBYQlODSZgRYRTZ88PleZweiQnO1ZUkvQrTIycl+fE2fNsrN5may3azG0nHMchm8sdziPQmkOgLX+A+C6SyYblapLLjJjf4AApjf1OIpLzCKolXGO3DzbvUP/837b83oM/bPPFOXIjLgs+DcbgRTATWAHbg+L8IvniHNXSLtVKOUyoMUAQQbvty3FdqpVok1abeJDZT4zsZTP7Q929pDIHM3+FjYdl2VyOeqPeYq6QPcU6zEfjwMhJbtKPiDAnPqHN1jwEKxt0TVndmnpNBHBwajWCE3fh+nV0/cbB+pdOIsfO4XihF4EGAfWv/N3Bh2mEaMO5hExstTKOSa5ZwQrYHuRM5vdmirdGvYabyaJBQKNRZ+3WDWpdMsfvIRLrSqHNOpvzMIeCU80OV4SGr+xrPwePczNWgx0Mwd24NnItPuBkOvi5Nuq4mez+Q7lWQrcPLlYoAyT9hvBBnhTPgXasgLXsOYKL65JzXRaPn+Lmtavh7HXLVg/2hVkum40hHSJEtsn2GXKN5XIe0a47m2bh8drCdXeD6if/ksxd/whZWKHxjc8eNgUNOA+Q87xEeQ60YgWs5RD5fH5Pm1TVzjdCTBd81Lm2/q2Oww82eZ4P/Yh1WZ1u10K1ROPSU91PCwYbAXkdPBGSgVgbrOUwjuNQKBQp90h5eMhGOiQS8Ubvd736fsNEDxmjoUBtyqkbgyAga8wxQpiMZxwLUiaOAX2yk5rzVYHRMyjPJlbAjsjJ06e5dfMGlXJnW6zvx+PXqVFjPvscrqpUWxzXZ2EhvEabI312gPDRSXBwyaDJM2jbs7oczCCk1UQw/bsq4biuy5mz5yh2WGQuk8mM5Hh/gPhtBG3Hz+AFPjNdiqkjI1QziG01Wb6vRwP7i8SAiHD67Fnq9Rql3V2qlSq7uzu4MQrY8Zs3RxcicXdxZuSr40AsmdaG/0Qi0tMenM0ld4Ir9IBLaN/7YAVsjGSzOZaWQ0+DarXK7s42jXodP4bE3uO+/DQIyBWKoZ/sYeevAZCB3YkG7tPMiNiYJirH+ImSkve1G7PzW8eLFbBjwvM8PM9jfn6BO7dvUY2QHKQTUWezo16uqkHkYIh2itl4c4/OzC03A5phP+1UnOn3cXisF4FlSHKex9nzF9jd3WF9bW3ooIOoAjZ9DlNTJDYBO4KJoM/+uflkLA3TCRvJZRkJEWF+foFcNse1q1fi9a3sxhQkbPxNpu0xMYbPI4LruntRh0nFarCWkcl5HqfOnOHm9euRz41sFZ3C9Rp/k2m76Qb/PJVzzyWQ/SxTrl/FBZP4J9gLbBHAyy0kd4LLEP/azrOBFbATxnGGs1OqRhxEpUKDnQ0if/fd6mFwEVsR9+CwuW3Fg9Z6GtXKoXSHSSOtGqz1g50wjgjZYSJuEhGGmoQ+DkNMN3+E3zBKXInfqFPd3RqiQ7OBmnywUbckYAXshMl5HgtLg2enH57kq7DDOIulhfalb/pRWh89b/E0UZXIWxKwJoIpkHGjfe2ZTAaNuMbXNIaLIoLT87PpASWucx+bx2hseRwGQ/adBUQQZC/nrpPzkCBjvAmax8ne+70TBVQyaPvnMu99BU6GJiJtOVdbjgFBRSLH5ld3twn8Rp/vf3ZJikYalWT+GgknSlKOXC6H1muRdTk/5rWtBmrTzVGt9wk26HAfefkilda8uuaYas0H12TnVyWbzZnVV2VPtoX7Di7R07pCRL3aAGc/D2tTeO+3dbBDK7vXoC1/RJQJGH/lPOV623ff+uPlVw7sctwM1ONJaBP4fjIFrEKQ0sFKAn+N5ON5HsW5OUq7uz2Py3se9fLOhHo1HdxMhnp9gHy5EsaXNRfW0dZc4oMuWwtGKHfXlkSiDs7biertEY9kcdwMbrZDQu8EkGY/WGuDnRKnTp9heWWl/4FDkpQJZVXFHVDrmoiSM+JaZZHFc0y2xPkTZxLvRWBtsJbYEBGWV44RBMr21mbX4IPhU+XN/gXouBkcx8GPJZFKXIz4vUV9Coy6kKUICyfOUlw+Plo9UyYRTjJDYAXsFBERjp84weLiIqurd6iUywcEbaVaJesV8aulWMVlJpvFdV3qlTL7EmHQFlrvhDCKSBxnKO0p/KyC3xhsAm8ij4xRtcCoIc0j+Nh680ssnbmAm0lmou19ok/qJYWBx0Mi4orIp0TkL8z7Z4rIJ0Tkkoj8mYjkTLln3l8y++9pqeMtpvzLIvKKlvL7TdklEXlzS3nHNtJGNpfjzNlznDh56tC+er0e6w3kui5+rUK9vBtqT82lbjQYcNMD5/iNOo1alXq1wqDqmxKm1wuCYDD7a8t5Y2fkYXZEATvk7E5+YYmV8/ekQLgaG2xKTQRRDE4/B3yp5f1vAr+tqs8C1oE3mvI3Auum/LfNcYjIc4HXAt8J3A/8gRHaLvB24JXAc4HXmWN7tZFKCsUiS8vLh8rd7BBx5l2uP5HJOj91IpfzYloIcgxMWIMd9rGRX1hOtM31qDCQgBWRC8APAe807wV4KfB+c8h7gNeY1w+Y95j9LzPHPwA8oqpVVf0GcAl4kdkuqerXVbUGPAI80KeNVOK6LseOn2Bufv7gjoh5VotZl0zX9HXTvymDtBrcgOheBDKkUJ/+7xgnBwZGA25JYFAb7O8Avwg0F10/DmyoatPh7wpw3rw+D1wGUNWGiGya488DT7TU2XrO5bbyF/dp4wAi8hDwEMDdd9894EeaXU6cPEU2m2VjfZ1sLketstvzdirmMtCootlCOIzfvgMIc24mdNdvdYwPQg1W3QzlYDpOJLOw/ld3RhNcURenhP6rFRw63nXJdViiKMmk1U2rr4AVkVcDt1T1KRF5yfi7FB1VfRh4GODixYsJebZ1x3EcVo4dR0SolEo9Hd0LXg62bgEg5e29ckHBr3e8bBUg64GTj7PbA+OkeGg7VCrKiN/H/LFTqbC97nHEAw2+D/hhEXkVkAcWgd8FlkUkYzTMC8BVc/xV4C7giohkgCVgtaW8Ses5ncpXe7RxJFheOUawtEyjXqNWrbC9vka9bRlrqXdezbY/0xNylUoZx3EJghlMUjei8JchEu+JOCiDm4G8uYX+ByWI5iRXGuk7VlPVt6jqBVW9h3CS6m9U9V8AHwN+1Bz2IPBB8/pR8x6z/280fKw/CrzWeBk8E7gX+Dvgk8C9xmMgZ9p41JzTrY0jg+M45Lw884vLnH3Gt3HunmcdWD00yCZvLSZBIoUL75EELWcYVSyiUM94yfvN+3HUbbCd+CXgERH5deBTwLtM+buAPxKRS8AaocBEVb8gIu8Dvgg0gDephrM3IvKzwIcBF3i3qn6hTxtHlkw2y/GVFXaufT2MOmokMZRWqVcreF6BajXKOmCKVyi0vg1ppiI4mJJg7xABcBw0CA4eqB1kdmNE+/AQgQNRxGvGy6fSeyCtfrCRBKyqfhz4uHn9dUIPgPZjKsCPdTn/N4Df6FD+GPBYh/KObRx1vKXjqF9n9/o3R6hl+irAMHKiUhkuMYqXz4+88ORAjNkGO3/8dPT6E0BSNNKo2EiuBCIiFI6fpb6zSW17fbhKZuCKjmqDHanHg37eUW2wQQNP6y3j2DA4o3WJl3phmaAlY1cUjTSbRvOAXVXWMosUT11IrIAVx0GiumuN1OUBb+BRBawGHdMPttZ6uAk5sFNa886KIBKGIotIYjNm9eSIexFYZpRMYZ75c98W2mMjcigp9KRRpRZ5yD78XTjomb43h9slo9ahOg7YfcO9AUCt9+fytEGg1T0tN2jskm3RcLvhesXoD6WEMAMDqrFgBWzCyR87TenONYJalMmi6ax40IqqksvnqVai9XsYMtnswHdwQx38SBNvQ+A3oCUHw6Ai06+V0cBHhlw4c5ZJa6BBOh+HR4zFu589RB7T6V/QGgQTEfSu41Lro1XuM/3vpSuqxhMiXSihiSDqlgSsgE0BmXyR7NxipHOmrcEC1GtVMplJDKJm7G4c4buPOlJJCmn1g7UCNiVk56OtVDt1G6yhXq+R88YbshuoDh7YMIHvZZSHW2X9Vow9mR0mLWBF5MdE5AsiEojIxbZ9kdKq9sIK2JTgLR5PpG1OgHqtQnag2fHhBFO9Vhs4wcysK0bVjdsEEXLoJgFVCFQibyPyeeBHgL9tLRwyrWpX7CRXSnBzHm6+SKO03f/gWUOVRqOG47oEfo+Y/A6RWv3IeR7o4AlmJmI6GakNTeSDtB+THvKr6peg4++9l1YV+IaJSG0GO10ywU+IyCPm2C/2ascK2JSgQRBRuI5BkIxyk6iSy2ap9BKwQ1Cv1SJluNoLrR0rw7cgbgZx0ydgh+SEiDzZ8v5hk1lvFKKmVe2JFbApoVatwsJxUKMgHRIqxhHGSBB1XHLS3y657+qpHaWPIHtLYY/qajNUqr/YmQ3bdDe0UcevVXFzQ6xyMcMM+dPfUdWL3XaKyEeAMx12/YqqTiRxlBWwKaFU2mXnwNqB7YKi/b0C8dryvPxoN32tWo2cfLovIrM35Tzhdb+SwDjcrlT1nw1xWtS0qj2xk1wpodFo9D9o3Ix8k2hoM42RWdRHR+lTdm4JNzedROnjopkPdkYWPYyUVrVfZVaDTQkz4YAewzVfq1TIeV7HMNpELNs9AKM8h/InzsXWj5lhCn6tIvLfAP8ROAn8pYh8WlVfMWRa1a5YAZsS8oUCpd3p5oYVccjlvN6SsEve1lYcx8HroMmKCJkDyV87VRy+FvO/tDVUbfgEvr+XPKW1bjE1OM1VfM3+/eOkpdlDxmjzR/CrYUhrN4YV4ZniItn5w6sOp4FJR2ap6geAD3TZFymtai+sgE0JjXq9/0HjJmjgV3ZHrkayOfwx+Xo6ZmsKuXw2g5Y2Dxwzqh9DJufhV7sv5TOsLCmeecZMRODFTWgimHYvxoO1waaEpZVjwy3DEicx3SSNeo1coUhmDDPlTS11vMTfQnZhhWwxXWtxtWJDZS0zjZvJsLxybLqdiFG7qpVLOBPw95RxiNsxVOktn4q/0hkirclerIkgRZTLpVjry2SyGA/YrjQFlKKIOOFy0kbQauD3jszqQ71aJZvLEwS+WehRqFXi/YzjmIwWJ4OTyxv34IPJtEEQxyFTmDM2YkCc0Gps/ImhdTAg4S+QpmW620mQRhoVK2BTRJzamIhQjygcG34NcPekQzGXIygPP/GmgU+9FvbBb9SRyCkZp0TQgOpu1wfToW+1uEi13tsLZG4mHc7iQYFZcIIZBwm5Yi2DsHLiZConQfYYy2h+HJVGq1MGUN9mwg1vjKTVBms12BThui7iOGgM8fyqGqtNdVaZift0AGmRyadvscNWkiIwo2IFbIpo1OsEfoyazswJ2WRMSEWttJ9syeaLZNK42KFBEzRpFRUrYFNEOMmV0it1bEz/ASIa0Mtat3gqhdFbbcxGop/4sQI2RcwvLFLa3aFea3PS1/A/7VWmiqruJVuJ5YKfvuxKBL2+6+LycXLF+Qn2ZjqkVL5aAZsmRITi3DyruzeHrqN5s2tY4Wgdiv2mSchdGPVr04AwvL29HmHh5Nk4ejTzpHUOz3oRpAxNqzFrbMxC4vHOfSgsLOO4VgdKMlbApoyd7c3+Bw2IDLiO1SQQxyU7lsUR430gOVkPv7wV6ZxGprOHQDblngNNhnHRSopJYXbuIEssHD95mvmFaEt4d2JWzKduJouIgwY+tZgj1ULiv1MH8Ws9eELnb9vJpNdzoB0bKmtJBF4+j5c/w9LKMe7cvEG1Whm+shlQExzXxW/MQKawQRniO+v2MHPTHB7bxgxcamPBarApJZvLcfr8heln2BqBXL5AfZQHxCDMwo3dQYN1MlmyheIUOjMdNNDIWxKwGmyKcRyHYydOcet636WDZpJ6rcrSra+FQ25VwqwgQfjeb4AqEvjG8yHcV/n2i/hOxiTaVkTVDNnDOpqpVEQD8ANww7T18RHPjZ8rzKU77LkFG2hgSSyFYpFsLnfYN3YARveFHe18DyV7/VKkc+Y+/deRjq995/fHKmBVR11bN8RNceRWJ9JqIrACNuUMKyTjud6jixoRIeO65HdWyX35if4njEio3caoKWo8Dp218u5e4MdRIEipCtvXBisieRH5OxH5jIh8QUT+gyl/poh8QkQuicifmZUWMasx/pkp/4SI3NNS11tM+ZdF5BUt5febsksi8uaW8o5tWAZDNeD2jetDaa/TYmF3laWn/hLvy0/MjCdDFIaxDXYSovXy7ki5dJOEkl43rUE02CrwUlXdEZEs8F9F5EPAfw/8tqo+IiL/O/BG4B3m77qqPktEXgv8JvDfichzCZe6/U7gHPAREfkO08bbgR8ErgCfFJFHVfWL5txObVgG4NaN65RLo6+RlcvlcNUn8BuI4yBOBhWzspVA4PuoBjiOQ4DQqDfQPpqc42YI/P3BuZvJkkXJffWTI/c3Ck6jRmbx5P5CjM2k1wqI0j7gF9UwSbcKiCJBQHbztrnjAxTBQaHpQxwoqG9ClXTfhaspJQKfK3/5FJufuYRfKhNUqgS1BlqrU/3JH+Gef/MTE/ompkiCBGZU+gpYDceYzazJWbMp8FKg+eu/B/hfCIXfA+Y1wPuB35fwEf0A8IiqVoFviMgl4EXmuEuq+nUAEXkEeEBEvtSjDUsfqpUy5d3hhWvGdRE3Q8aBenm7o6lBvCL1+mELpuO4SKb7pZXJz1Op1UCyYYpFcag26mRrU1gVt1YZyZXN9RsULj01Uhc2//az3Hr8sDmk8pLvHane5KAEKZWwA9lgRcQFngKeRahtfg3YUNXm3XUFOG9enwcuA6hqQ0Q2geOmvPUqaj3nclv5i8053dpo799DwEMAd9999yAfKfVsbWwMfa7nefjlbaiHM+xRh+pB4KOBTzafI5OfQwFHBA0aiJOhXDUmCxH8IACC6aVFHLHd9mXB4+TCgz8ytrpnjZhM1zPHQAJWVX3geSKyTLiW+HPG2quIqOrDwMMAFy9eTOejMAJrt2+xu7M91Lm5nEejvD2QUO2ldAhQqRzO6iXOrAUNjCrYY3gwdPkiNc7cvjNMaINN520bKdBAVTeAjwHfCyyLSFNAXwCazpZXgbsAzP4lYLW1vO2cbuWrPdqw9GCYIW8mkyWf9whquxFERtSQ0Nm7kXQGZum7LSv5qZ/4+dQvFQOAhibqqFsSGMSL4KTRXBGRAuFk1JcIBe2PmsMeBD5oXj9q3mP2/42x4z4KvNZ4GTwTuBf4O+CTwL3GYyBHOBH2qDmnWxuWGMnmcmhtl0ZpOzlXblzMgMDvllRn+/NfYeMTn5lwb6ZDMwdxlC0JDGIiOAu8x9hhHeB9qvoXIvJF4BER+XXgU8C7zPHvAv7ITGKtEQpMVPULIvI+4IuEpr03GdMDIvKzwIcJk2K+W1W/YOr6pS5tWHrQb9kYEcFxHLLZLAQ+jcrOkAPd6Wt/oyIjG/9Gv9G7Cdi5Zz+The+6d+T6Zx3lCEdyqepnged3KP86+14AreUV4Me61PUbwG90KH8MeGzQNiy9Eae34Ms5il8v06iXR2snbvmaRHkdw5cgmQ7JtoHv+J/+LZmF9K9mEEY5p1PC2mQvKSSb6xyPkc3lyGdd/PqMBh5M5R4bddWG0TvtuJ1vw8b26D7MluliBWwKyReKxgwQakaumwknsCo7NKqjaa2pY2QngtE12G72xK/9b++kfOXGyPUngbRGclkBm0I8z8PVOtIo42UEx6/SKA3mehWJhFzk4ySWr6CLtChdeponXvYv8SvVOFqZaYJAI29JwArYFJLNebiOiwB+rYoGRyOmfShGVoVieGz10IJLX7/M1mf+YfQ2ZphhPAjS5EVgSRhhZHIyLkALSBcb7N7+PpOWaeBIR3JZkoebyeI34k0lfYiY7/t4MqlGbXS0B5HGocB28SJoUr52k+XRm5lpjnQuAkvyyOWL1Cp2Qqsfo+YS0BgkrNNHwOZPnxy5jVknKUP+qFgbbEpZPnWWkxeeieP2vnlnimmErY54X8eiwfYxEdz52P83eiMzjGp6J7msBptivOIchfkldjfXpt2VgVAE8nP7BSKEOWeNFNvLpdrBYNc8trITcdg/yRtVIJMBJwOuC064Fc6XWH7xd+PkskjGxcllQJzQ9uoI1WvXUr+6QUoVWCtg006SJkjq8yvces4/GamOY6VbSKOxZx/WFgHdTOwS3ssmWbiTxRM9IMTVEaR5npslCE9uScqiqJrk2xqw84++r0X7FlTCdpvtzPkltLzdVUM//cBpTj/wwp6fq3b9a3jnnjXMV5II0hrJZQWsZWiaQqhVbBzUskJvhnb7mkJHlUVimOSq5YogEbKJqQ/VwxFTzd5Jfo6G3+fmd/ssjS4ysvmjdvUrSNYje+JC6jRZ1SOecNuSXNx+N/8ImIWxDw6yO77pIBA6CIluafsiMZYECTPgKwtUv/k53MIC7nz6fArSqsHaSa6Uk/W8aXdhoqTzNg1xF46nUrhCKGCjbqMgIv+riPyDiHxWRD7QTMlq9kVanLUXVsCmnFyh2DUdXiqJXYOdIa06SR4hUdAwXWHUbUQeB75LVb8b+ArwFoC2xVnvB/5ARFyTrvXtwCuB5wKvM8f25AjdeUcTx3E5fu7uZAjZWGRZzJ8zDttgXAI2pSHPyuQ1WFX965b1/p4gXDEFWhZnVdVvAM3FWV+EWZxVVWvAI+bYniTgrrOMSr44z+m7v5383GznFo3DpzTuJWBisQ3GJPQzS2kNOBg6F8EJEXmyZXtoyA78K+BD5vXeoq2G5mKr3cp7Yie5jgiZnMexMxfYuH2Tys4WwSxqQ7EYUOMVsEJAuNDGKJVE6JM4LV4H0lLskj15V/fzkowJNBiCO6p6sdtOEfkIcKbDrl9R1Q+aY36FcIWVPx6mA/2wAvYI4bgZjp0JH8bV8i67m+uUd7aHzrYV/4RSDMuvxCxgNfDDgIBRaBOwzsJx6o0GTeOjahCaInoI4sz8MjJGj5BpM45QWVX9Z732j3p8mgAACI5JREFUi8hPAa8GXqb7Hei2CCs9yrtiBewRxSvM4RXmUFUatSrVcolqeZfK7s4U0xvGkLxa4haxcdDeI0H9tkQ8fbRctzDb5p2kISL3A78I/BNVLbXsehT4ExH5LeAc+4uzCmZxVkLB+lrgJ/q1YwXsEUdEyHp5sl6e+eVjqCrrN69R2lof5Oyx9y8ysdtgg5FnKvaiycQBcYayNcuoWvQM05zkmjC/D3jA4yZw4wlV/TdDLs7aFStgLQcQERaOnRhQwMbd+ExUcZAYEpX62TyaLe4XDJHlzC9tj9yPmWUKix6qate446iLs/bCCljLIbI5Dzebxa/X+xw5e279GrsVNgZi8CII2k0KqSK9obLWTcvSkRPnnjHtLgzHDMbpxyM60imAmkzaD3ZSWA3W0pGslyc/N09ld2dyjc6km9ZsoP4MutXFhJLehNtWwFq6Ulxc6S1gY78p0nmTxUFQK6N+A3FTeMsO7wc786Tw17LEhZcvsnL6nJGjuvc3/KeouAR00/LC0vYRe7imfdASkWOOUZPqsFDoUJdpe++c4IDG0/o6Iz4SLOyXNSepdK+mtl7Kfrnq3kNjzzcVcLI5BEEcB3Hc8G9LMED4GU3qxr2yMN8sIriOgxTnTJkThi2LE4b1Nuva8zRofm9taR8l/AizolHHTVKG/FGxAtbSFTebZW7p2LS7YUk9yVmGOypWwFoslqmiavyNU4gVsBaLZepYG6zFYrGMCWsisFgslnGgyfFrjYoVsBaLZapMKRfBRLAC1mKxTJ0ghpwPs4gVsBaLZbpMIdnLpOibi0BE7hKRj4nIF0XkCyLyc6b8mIg8LiJfNX9XTLmIyO+ZlRc/KyIvaKnrQXP8V0XkwZbyF4rI58w5vyfGy7pbGxaLJT0o0fMQJEUgD5LspQH8gqo+F7gPeJNZTfHNwEdV9V7go+Y9hKsu3mu2h4B3QCgsgbcCLyZcQOytLQLzHcBPt5x3vynv1obFYrHMPH0FrKpeV9W/N6+3gS8RLvb1APAec9h7gNeY1w8A79WQJ4BlETkLvAJ4XFXXVHWdcNnc+82+RVV9wizb8N62ujq1YbFYUsSQix7OPJFssCJyD/B84BPAaVW9bnbdAE6b11FXZTxvXreX06ON9n49RKgtA+yIyJcjfKwTwJ0Ix08T29f4SUo/Yfb7OlyOS4XgqEdyicg88OfAz6vqVmsyClVVERnrI6VXG6r6MPDwMPWKyJO9VqacJWxf4ycp/YRk9TUqSbGpRmWghNsikiUUrn+sqv/ZFN80w3vM31umvNuqjL3KL3Qo79WGxWJJCUozQ1q0LQkM4kUgwLuAL6nqb7XsehRoegI8CHywpfz1xpvgPmDTDPM/DLxcRFbM5NbLgQ+bfVsicp9p6/VtdXVqw2KxpAU92isafB/wk8DnROTTpuyXgbcB7xORNwJPAz9u9j0GvAq4BJSANwCo6pqI/BrwSXPcr6rqmnn9M8AfAgXgQ2ajRxtxMpRpYUrYvsZPUvoJyeprJJIiMKMiSZmNs1gs6WTpxHP1vh/6k8jn/fV7n//UrNukbSSXxWKZKpriSC4rYC0Wy9RJa8LtxC7bLSJ5Efk7EfmMCeH9D6b8mSLyCRN2+2cikjPlnnl/yey/p6Wut5jyL4vIK1rK7zdll0TkzS3lHdsYoM+uiHxKRP5ilvsqIt80ocufFpEnTdnMhUaLyLKIvF9E/kFEviQi3zuj/Xy2+S6b25aI/Pws9nUqpHiSK7ECFqgCL1XV7wGeRxgVdh/wm8Bvq+qzgHXgjeb4NwLrpvy3zXFIGPb7WuA7CUN0/0BCQegCbycM/X0u8DpzLD3a6MfPEUbCNZnlvv5TVX1ei41rFkOjfxf4K1V9DvA9hN/tzPVTVb9svsvnAS8knPz9wCz2dTocYTetWcWE4jbXlM6aTYGXAu835e0hvM2w2/cDLzNP+QeAR1S1qqrfIPR+eJHZLqnq11W1BjwCPGDO6dZGV0TkAvBDwDvN+171TLWvXZip0GgRWQJ+gNCFEFWtqerGrPWzAy8DvqaqTyegrxNBCZeMibolgcQKWNgbcn+aMADhceBrwIaqNswhrWG3e6G6Zv8mcJzoob3He7TRi98BfhFoPnp71TPtvirw1yLylIRhyDBDodGGZwK3gf8kodnlnSIyN4P9bOe1wJ/2qWdW+joZzKKHUbckkGgBq6q+GXZdINTinjPlLnVERF4N3FLVp6bdlwH5flV9AeFQ9U0i8gOtO42WNPbQ6D5tZIAXAO9Q1ecDu7QNf2ekn3tIaP/+YeD/GqWeYZlEG8NxtNMVzjxmaPgx4HsJh1NN74jWsNu9UF2zfwlYJXpo72qPNrrxfcAPi8g3CYfvLyW0H85iX1HVq+bvLUJb4YuYvdDoK8AVVf2Eef9+QoE7a/1s5ZXA36vqzT71zEJfJ4q1wc4YInJSRJbN6wLwg4STHB8DftQc1h7C25x1/VHgb8wT/VHgtRLO3D+TcILg7wgjzu6VcBY+Rzi0e9Sc062NjqjqW1T1gqreY+r5G1X9F7PYVxGZE5GF5mvCkObPM2Oh0ap6A7gsIs82RS8Dvjhr/WzjdeybB3rVMwt9nRwp9iIYKg/jLGzAdwOfAj5LKAD+Z1P+bYRC5xLhUMwz5Xnz/pLZ/20tdf0Kof32y8ArW8pfBXzF7PuVlvKObQzY75cAfzGrfTXHf8ZsX2jWRWjP/SjwVeAjwDFTLoQeDF8DPgdcbKnrX5l2LwFvaCm/aH6zrwG/z35EYcc2evT1ecCT5hr4L8DKLPbTnDNHOKJYaimbyb5OeptferZ+/w//beQNeHLafe+32VBZi8UyVUTkrwhz3Ubljqre3/+w6WEFrMVisYyJxNpgLRaLZdaxAtZisVjGhBWwFovFMiasgLVYLJYxYQWsxWKxjAkrYC0Wi2VM/P/Cr3AdR8RRVwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wi[\"mean\"] = mean\n",
    "cap = mean.max()\n",
    "ax = wi.plot(column=wi[\"mean\"], cmap=\"coolwarm\",\n",
    "             figsize=(5,5), vmin=-cap, vmax=cap, legend=True)\n",
    "ax.set_title(\"mean\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How do we weight the combinations of component rows to approximately construct a given row in our original dataset?  We can do this with component weights, as given by the `PCA.transform` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DATE</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-03-16</th>\n",
       "      <td>-363.67</td>\n",
       "      <td>-2.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-17</th>\n",
       "      <td>-360.99</td>\n",
       "      <td>-3.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-18</th>\n",
       "      <td>-357.78</td>\n",
       "      <td>-5.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-19</th>\n",
       "      <td>-356.65</td>\n",
       "      <td>-5.82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-20</th>\n",
       "      <td>-355.23</td>\n",
       "      <td>-6.59</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0     1\n",
       "DATE                    \n",
       "2020-03-16 -363.67 -2.28\n",
       "2020-03-17 -360.99 -3.83\n",
       "2020-03-18 -357.78 -5.66\n",
       "2020-03-19 -356.65 -5.82\n",
       "2020-03-20 -355.23 -6.59"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pc_coef = pd.DataFrame(pca.transform(dft), index=dft.index).round(2)\n",
    "pc_coef.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Even though we have 2 number per day instead of the 72 per-county numbers per day, we can use those two numbers as weights on the component rows to approximately reconstruct those 72 numbers.  Let's try it for a specific date: Mar 1st, 2021."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0   -231.30\n",
       "1      4.62\n",
       "Name: 2021-03-01, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mar1 = \"2021-03-01\"\n",
    "pc_coef.loc[mar1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>NAME</th>\n",
       "      <th>Adams</th>\n",
       "      <th>Ashland</th>\n",
       "      <th>Barron</th>\n",
       "      <th>Bayfield</th>\n",
       "      <th>Brown</th>\n",
       "      <th>Buffalo</th>\n",
       "      <th>Burnett</th>\n",
       "      <th>Calumet</th>\n",
       "      <th>Chippewa</th>\n",
       "      <th>Clark</th>\n",
       "      <th>...</th>\n",
       "      <th>Vernon</th>\n",
       "      <th>Vilas</th>\n",
       "      <th>Walworth</th>\n",
       "      <th>Washburn</th>\n",
       "      <th>Washington</th>\n",
       "      <th>Waukesha</th>\n",
       "      <th>Waupaca</th>\n",
       "      <th>Waushara</th>\n",
       "      <th>Winnebago</th>\n",
       "      <th>Wood</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.014671</td>\n",
       "      <td>0.012164</td>\n",
       "      <td>0.063659</td>\n",
       "      <td>0.012197</td>\n",
       "      <td>0.186703</td>\n",
       "      <td>0.013653</td>\n",
       "      <td>0.012156</td>\n",
       "      <td>0.039235</td>\n",
       "      <td>0.081403</td>\n",
       "      <td>0.034584</td>\n",
       "      <td>...</td>\n",
       "      <td>0.017245</td>\n",
       "      <td>0.018460</td>\n",
       "      <td>0.065934</td>\n",
       "      <td>0.013056</td>\n",
       "      <td>0.120046</td>\n",
       "      <td>0.379391</td>\n",
       "      <td>0.034502</td>\n",
       "      <td>0.018779</td>\n",
       "      <td>0.124864</td>\n",
       "      <td>0.068892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.010148</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>-0.026789</td>\n",
       "      <td>-0.001165</td>\n",
       "      <td>0.488611</td>\n",
       "      <td>-0.003380</td>\n",
       "      <td>0.000764</td>\n",
       "      <td>0.125926</td>\n",
       "      <td>0.006220</td>\n",
       "      <td>0.003825</td>\n",
       "      <td>...</td>\n",
       "      <td>0.003234</td>\n",
       "      <td>0.010982</td>\n",
       "      <td>-0.023864</td>\n",
       "      <td>-0.011774</td>\n",
       "      <td>0.010618</td>\n",
       "      <td>-0.225074</td>\n",
       "      <td>0.099319</td>\n",
       "      <td>0.058002</td>\n",
       "      <td>0.452854</td>\n",
       "      <td>-0.009747</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 72 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "NAME     Adams   Ashland    Barron  Bayfield     Brown   Buffalo   Burnett  \\\n",
       "0     0.014671  0.012164  0.063659  0.012197  0.186703  0.013653  0.012156   \n",
       "1     0.010148  0.000019 -0.026789 -0.001165  0.488611 -0.003380  0.000764   \n",
       "\n",
       "NAME   Calumet  Chippewa     Clark  ...    Vernon     Vilas  Walworth  \\\n",
       "0     0.039235  0.081403  0.034584  ...  0.017245  0.018460  0.065934   \n",
       "1     0.125926  0.006220  0.003825  ...  0.003234  0.010982 -0.023864   \n",
       "\n",
       "NAME  Washburn  Washington  Waukesha   Waupaca  Waushara  Winnebago      Wood  \n",
       "0     0.013056    0.120046  0.379391  0.034502  0.018779   0.124864  0.068892  \n",
       "1    -0.011774    0.010618 -0.225074  0.099319  0.058002   0.452854 -0.009747  \n",
       "\n",
       "[2 rows x 72 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "components"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NAME\n",
       "Adams         1.137748\n",
       "Ashland       0.521406\n",
       "Barron        0.344396\n",
       "Bayfield      0.197941\n",
       "Brown        44.804757\n",
       "               ...    \n",
       "Waukesha     26.487082\n",
       "Waupaca       6.049110\n",
       "Waushara      1.885923\n",
       "Winnebago    21.617261\n",
       "Wood          3.018259\n",
       "Length: 72, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "-231.3 * components.loc[0] + 4.62 * components.loc[1] + mean"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Or, more concisely using the dot product:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NAME\n",
       "Adams         1.137748\n",
       "Ashland       0.521406\n",
       "Barron        0.344396\n",
       "Bayfield      0.197941\n",
       "Brown        44.804757\n",
       "               ...    \n",
       "Waukesha     26.487082\n",
       "Waupaca       6.049110\n",
       "Waushara      1.885923\n",
       "Winnebago    21.617261\n",
       "Wood          3.018259\n",
       "Length: 72, dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "approx = pc_coef.loc[\"2021-03-01\"] @ components + mean\n",
    "approx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Reconstructed from 2 PC weights')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVgAAAEeCAYAAADPbQ0xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9ebgsV1no/Xuret7j2WceEhIuufIElOkI4eKARCAgGq6PcMNFichH9BGu8HxXGURFERTvd5Xhivjlg0hAJXBRJCoYw6RyNSFhNgTIARJyxn3Onqce6/3+qNW9e/fuqbqru6trr9/z1N7dq1attaq66q13vetd7xJVxWKxWCzh44y6ARaLxRJXrIC1WCyWAWEFrMVisQwIK2AtFotlQFgBa7FYLAMiMeoGWCyWvc2TnAld1Urg405RuENVrxtAk0LDClhLaIjIZ4E/V9X3NNknwC3A84EHVPXJQ27eWCMivw48UlX/ry7y/jbwKFX92YE3LARWtcLbE48IfNzzyt86MIDmhMqeNxGIyIMisiUi6yJyXkTeJyKTo25XM0REReRRAyr7ClP+oF66PwQ8EzgxCuEqIr8mIv8uImsi8l0R+bU2eavXYt1sD4rI6+r2i4j8iilvQ0ROi8j/FpHvH1T7VfX3uhGu3WDO58fDKCsUBCQpgbdxYM8LWMNPquok8HjgCcDrR9yenhigcAyDRwAPqupGs51DaLsALwH2AdcBrxSRGzocM2vuixcBvyUi1e7oO4BXAb8CzAH/Efgb4CcG0fC4IyI4ieDbOGAFbB2qeh64A1/QAiAi14jIv4rIsoh8RUSeXrdvTkT+TETOisiSiPxN3b6Xi8gpEVkUkdtF5FjdPhWRXxKRB0y57zJdaETkUSLyTyKyIiKXRORDJv2fzeFfMVrVfxGRpxvt6bUich74MxH5eRH5XP151Wu+IpIVkT8UkYdMHZ8TkSxQLX/ZlP9Uk/8XROR+c353iMgj6sp9poh8w5Tzx/hCbBci8jLgPcBTTdm/06LtaRF5u7meZ83ntCmjmv81IjIvIudE5Pki8lwR+Za5zr/e5rf9H6r6RVUtq+o3gY8BT2uVv+HYfwPuAx4rIlcBrwBepKqfVtWCqm6q6l+o6lubnPuPicjX6r7fKSL31H3/FxF5vvl8TET+SkQuGi37V+ry/baI/Hnd95eY33BBRH6ziVaaEpH3i6+x3yciJ81xHwAuB/7W/BavEZGMiPy5KWtZRO4RkcPdXJtQEJCkE3gbB8ajlUNCRE4AzwFOme/Hgb8H3oyvqfwq8FcictAc8gEgBzwGOAS8zRz3DOD3gRcCR4GHgNsaqnse8IPAD5h8zzbpvwv8I76mdQL4XwCq+iNm/+NUdVJVP2S+HzFtewRwUxen+T+BJwH/yRz3GsADquXPmvL/TUSuB34d+GngIPAvwAfNOR4A/hr4DeAA8G1aCCxVfS/wS8C/mbLf2KLtbwCuwX/BPQ54simfuvwZ4DjwW8D/B/ysOZ8fBn5TRK7sdAHMy+yH8YVmx7wi8jT83/hLwLXAaVX9fKdjDXcBV4nIARFJ4v/ex0RkyrzYTgL/IiIO8LfAV8z5XQu8WkSe3VigiFwN/AnwYvz7a8YcU89P4d9zs8DtwB8DqOrPAd/D9NpU9X8AN5oyLgP24/9WW12eX/8IVoONOX8jImvAw8A8UBUAPwt8XFU/rqqeqt4J3As8V0SO4gvjX1LVJVUtqeo/meNeDNxiNKYCvsnhqSJyRV2db1XVZVX9HvAZtrXmEr7AOaaqeVXdoY02wQPeaDSptg+FeYh/AXiVqp5R1Yqq/qtpYzN+Cfh9Vb1fVcvA7wGPN1rsc4H7VPUjqloC3g6c79DWTm1/MfAmVZ1X1YvA7wA/V5e/BLzF1HcbvmB/h6quqep9wNfxBXMnfhv/3v+zDvkuAYv42vfrVPVT+ALoXLcnaM7rHvwX2JPwBej/wX8ZXYM/4LeA/7I9qKpvUtWiqn4H/wXSzIzxM8DfqurnVLWI/7JpDCryOXPfVvAVgXbXpWTO61HmnviCqq52e459E2MbbJRtdsPk+ar6SRH5UeAv8R/cZXxB9wIR+cm6vEl8gXgZsKiqS03KOwZ8sfpFVddFZAFfy3jQJNcLo02gOrD2Gnwt9vMisgT8oare0qbtF1U1391pcgBfA/x2l/kfAbxDRP6wLk3wz+MY/gsJAFVVEXmYYDS2/Ri+tl/lIZNWZcEIDNjWsC7U7d9i+zo2RUReiW+L/eE2L5YqB8yLpZ4FfK0xCP8EPB04bT4vAT8KFMx3MC9VEVmuO87F7zU00njtN839VU/j/ZURkUST8wFfAF8G3CYis8CfA28wL7KBIzI+GmlQrAZbh9FA34ffjQb/Jv6Aqs7WbRPG1vYwMGduyEbO4j8wAIjIBL6GcKaLNpxX1Zer6jHgF4E/kfaeA42aywa+2aJa95G6fZeAPPAfuigH/HP8xYbzz6rqv+JrcZfV1SP137uksc4d1w3fVng2YJktEZFfAF4HXKuqp3ss5lPAiapNs0uqAvZHzOd/whewP8q2gH0Y+G7DtZ5S1ec2Ke8cvvkI8O3q+PdXt+y47qb39TuqejW+6eh5+C+h4RBjDdYK2N28HXimiDwO/03+kyLybBFxzWDA00XkhKqeAz6BLwD3iUhSRKp2zA8CLxWRx5tBmt8D7lbVBztVLiIvMLZg8DUdxe9Kg6+tPbJDEV8BHmPqzuB3hwFQVQ/fF/WPzICKKyJPNW28aOqpL/9PgdeLyGNM22ZE5AVm39+ben5afA+AX8G3kfbDB4HfEJGDxsb7W/i/Qd+IyIvxf4dnmu53T6jqA/j2zw+aeyFl7osbpM6Vq4F/Bb4P36b8eWPOeATwFLYHFz8PrIk/6Jc1v81jReQHm5T3Efz78j+JSAr/Nw4icXbcR2Yg7vtFxAVW8U0GXquDQ8faYPcOxvb3fuC3VPVhoDrQcxFfy/g1tq/bz+HfjN/At92+2pTxSeA3gb/C1zb+A81tac34QeBuEVnHH5x4VZ1A+G3gVjPS+8IW7f8W8Cbgk8ADQKMN91eBr+HbBReBPwAcVd0E3gL8H1P+Nar6UbP/NhFZBf4d3+6Mql4CXgC8Fb/bfBW+bbEf3oxv4/6qaeMXTVoYvBlfy7tHtv1b/7THsn4Ff9DoXfimpG8D/xl/kGoXxjXti/g266JJ/jfgIVWdN3kq+Jrj44Hv4vc23oM/+NRY3n3Af8O3Q58D1vHvv04mjyq/j/8iWxaRX8V/MX4EX7jej69Vf6DLsvpGAHEl8DYOiNqA2xbLWCP+xJhl4CpV/e6o2xOUR+dy+p6rHh34uB/+6pe+oKpBTDVDxw5yWSxjiBl4/RS+Avg/8TX+B0fZpt4RxBkPjTQo1kRgsYwn1+MPAJ7FN8/coLY7GjmsBmuxjCHqxyUIJTbByBEQN566nhWwFotlpAjgjMmgVVCsgLVYLKNFiK0NNnYC9sCBA3rFFVeMuhkWy57jC1/4wiVVPdg5ZyNiNdhx4YorruDee+8ddTMslj2HiDzUOVez4xgbv9agxE7AWiyW8UMcO8hlsVgs4WNtsBaLxTIorA3WYrFYBoJYDdZisVgGh7XBWiwWyyCwGqzFYrEMCmuDtVgsloFgbbCWPUepVCKZTNa+qyqqilNnK1NVCoUCiUSCRMLeSpbesTZYy57CX2ILPM+rfW6Mhre0tMTy0hJHjhyxAtbSO1aDtewlisUiqVSKra0tKpUKiUSCTCZDpVLBdd3tfAV/hZJMNjuqplpigQ24bdkjqCpnTp/GM8I0m82SSqVQVYqFAlubmzvyZjKZmoZrsVh2YjXYPUqhUGBjfZ19c3O1rr/jOIgIx48fx3FdLp47x+HDh1ERtra2yGazYPJ6nkcul2NqetoKWEvfxFWDtQJ2j7K6ukqxUGDOCMdKpYKIUCgUyGQyqCpHjx2jUChQKZeZmpoCfMFaHeyamZ0d5SlYYoLvRRDPzrQVsHsMz/NwHIe5uTkcx8HzPBYXFtja2uLosWNkMhmWl5bY2Njg6LFjlEolJicnWV9fZ31tjXw+z+zsLLP79o36VCwxwvrBWmKB4zioKufPnasJ2IIZrFpYWCCXy+G6LoVCgcXFRfbv38+Z06cpFosAZI1ZwGIJDYnvIJcVsHuM/NYWFy9domQEZj0b6+uUSiUOHTrEYeN6pZ5HMplkbv9+UqkUlXK55kmgqtb+agmFuJoI4nlWFoAdQrRUKlEsFrnUQrhWKRYKrK+vMzExQaFQoFQuc/jIEXK5HIlEgnQmg+d5vg23TTkWS7dUZ3IF3dqXKZeJyGdE5Osicp+IvMqkz4nInSLygPm/z6SLiLxTRE6JyFdF5IlhnJsVsDFEVVlcWEAch0KhQD6fZ2lpidMPP9xRKLquy9TUFKurq0xNTZFOp9nc2GBpcbGWp2hmb6XT6UGfimWPELaABcrAf1fVq4FrgFeIyNXA64BPqepVwKfMd4DnAFeZ7Sbg3WGclzURxAxVZWFhgdWVFUrlMlubm3ie1/Xx0zMzrK6skJuYQEQoFovMz8/jeR7pTIZcLmcnFlhCRkI3EajqOeCc+bwmIvcDx4HrgaebbLcCnwVea9Lfr77P4l0iMisiR005PdPxrETk+0Tky3Xbqoi8uhdVW0RuNPkfEJEb69KfJCJfM8e8U4xhr1Udlt1sbm5y5vRpvvud77C6sgL4NtUgwlVEmJyYYHpmhnQ6zeLiIouLixw8eJArH/lIcrncoJpv2cv0biI4ICL31m03NS1e5ArgCcDdwOE6oXkeOGw+HwcerjvstEnri44CVlW/qaqPV9XHA08CNoGPElDVFpE54I3AU4AnA2+sE5jvBl5ed9x1Jr1VHRZDuVRic3OT8+fO1bwBeuXAgQMkUykcx2F+fp7lpSVmZ2aYmJy0g1mWAeJrsEE34JKqnqzbbt5Vssgk8FfAq1V1tX6f0Va18ZgwCaqXXwt8W1UfwlepbzXptwLPN59rqraq3gXMishR4NnAnaq6qKpLwJ3AdWbftKreZU74/Q1lNatjT1ONYjU/P8/3vvc9zp/rqxcDwOEjR5icmmJlZYXvPfQQ6nlMTU9TqVRCaLHF0gGR4FvHIiWJL1z/QlX/2iRfMDIH83/epJ8BLqs7/IRJ64ugNtgbgA+az0FV7Xbpp5ukt6tjB6ZrcBPA5ZdfHuyMBszW5ibr6+vMzs6STKV6KqNUKrG2toYjgud5rK2thSr4JiYnyeVyiAiVSoWJiQkOHDy4IzShxTIoBhEP1pgZ3wvcr6p/VLfrduBG4K3m/8fq0l8pIrfh97JX+rW/QgABKyIp4KeA1zfuU1UVkYGq2u3qMF2DmwFOnjw50HZ0S6VSoVIu+25RpRJbW1scO36867B+nueRz+dZX1tjfX0dgEwmQz6fD72t2Wy2ZgLYt2+fNQdYhs4A/GCfBvwc8DUR+bJJ+3V8wfphEXkZ8BDwQrPv48BzgVP4ZtCXhtGIIBrsc4AvquoF8/1CdZStS1X7DNujd9X0z5r0E03yt6sj8iwsLLC+tlb7Xi6XOXf2LLOzs0xOTbUUYoVCgdWVFdY3NtAAA1T9sGZcsqr1u667I9i2xTJQBjCTS1U/B7Qq9Nom+RV4RaiNIJgN9kVsmwdgW9WG3ar2S4w3wTVsq9p3AM8SkX1mcOtZwB1m36qIXGPU+pc0lNWsjsgz3WQ6aalU4uLFi8zPz7O1tcXa6ioX5+e5cP48F+fnOXPmDGdOn2ZtbW1owhV8oVosFhERMpkMxWJxV3Bti2WQ9DjIFXm60mBFZAJ4JvCLdcmBVG1VXRSR3wXuMfnepKpV7/VfBt4HZIFPmK1dHZFkdXWVVCpFMplk/sKFlvk21tfZMN3+IAxK5E1PT5Oqsw/ncjnK5TKJRMKaCyxDYU/HIlDVDWB/Q9oCAVVtVb0FuKVJ+r3AY5ukN60jqqytrvbtKtWWAWiV1Wmw9YJ0fW2NS5cukc1mOXT4sB3sslh6xD45ITKO3erGFQkKhQJLS0uoKpubm5w7e9a6alkGyiBiEUQFO1U2RMZNwE5PT9ciY21ubrK8tLTLS6FQKHD2zBmOHD1qB74sA0Igpr0kK2BDZNACNuzSJyYm/IG3+fm27l+lUokzp09z6NAhsrlcbSFEiyUs4mrrt09JiAxcgw25/KqPbjd4nsf58+dJpVIUi8XaqgYiEtuHwzIk7JIxlm7wxsxE0M10w0bK5TIAy8vLrKyskM3lOHz4MJ7nUS6XbQhDSw+Mj001KFbAhsigfVdD15D7LE9V2dzYYH5+nmKhgKpymZmqbLVaS9cI1gZrac9QBrgiKrSqPr2O43Dm9Gnm9u8nmUzaQTFL11gN1tKWcfMggN7a3E4z9TyPYrFYi+51/MQJazKwdEQQROKpwcbzrEbAUATsmAnx+mVmLJaWCOBI8G0MsBpsSIyjBtuLySHIeW5ubrK1tUXWLjFj6UBcvQjieVYjYBwF7DB0gIVLl4ZQi2XcsTO5LG0ZRvSrcRTipVKJSqVSmzFmsezCnys76lYMhHie1QgYR+E3DFSVleXlUTfDEnGsBmtpy1gK2AHbYKusra0xMztrtVhLa2Jqg7UCNiSGIWBDr2FIL4VKpUKpVLIC1tKUOE+3tgI2JMbRTauXm7rXFpw9c4ZMJkM2lyOXy5FKpWL7UFl6wGqwlnYMRYMNu45eBFwfbcjn8+TzeZYWF3Ech1wuR25igkwmY6Nz7XHGxaYaFHtXh8TYBXqBkU5c8DyP9fV1yuUy8xcukEqlyGazZHM5MpmMXUXBEgusgA2JsRzkCoiIDOw8i8UixWKRlZUVf/HFbJacEbjJZNKaE+JMjN20rIANibE0EfRQ/47YAiIU2gTq7qeerc1NtjY3YWEB13V92202SzabxbXmhPhhTQSWdoyjgO1FK2xc1HGQWm2VSqXC+toa62trAL45wQjctDUnxIK4BnuxAjYkRq1d7iVq5oTlZd+cYLwTstms9U4YR6rBXmKIFbAhIbD9cDc84PXfdojhBqGsJk3r9tXSTHp1Sq4V6D6qytbWFltbWwC4rrvDfmu9E8YBiW2wF3v3hUTFxEJtRTqT6cteGbQrXnXedhynpSN3GDd1Kp3e8QJpt3jiMKhUKmysr9eCgCeTyZo5IZPNWnNCVIlpr8MK2JDoJPyGffuo0Xq9NkFowgiG3fjScBynbZ3DplQqUVpZYXVlBYDJyUnm9u+3mm2UsEvGWDrRKZpWv136gQwmDUBriLrpYn19nY2NDWZmZ5mdnbUabSTYbVaLC1bA9omqsry0xObmZvuMfd5A4zJwE/RFMApxXP3N1lZXmdu/n8nJybG5vnHF2mAtTSkWiywtLXXOGHHNLiyCmghGKdYqlQoX5+fZ3NxkenrarrwwKgQ70cCyzdbWFpumm5k3o9eDZiAa1h4R+p0ol0qcO3uWiYkJDhw8WDMbWK12WIzPGltBsQI2IKrKxYsXKZdKrJiBk7HFChAfcx02NjbI5/PkcjlEBE+VTCZDKpUik8kAfo/F8zyKhQKJZJJcLjfKlscCwU402NOoKgsLC6SSSVSVcqkUuIy+taEBCEMrXndTqVRYMzPGANbX1khnMhw+fBjHcTh75kzNBJJKpSgVi7WZZeVyGdd1reYbFDvRYO+iqqyvrdXcfHqlXC73fXzYLlCe55HOZEIzFSjgGN/bZpMtarWo1vaPwusgmUzuCP7d6beplMs8/L3vATvbWywWWVhYACCbzVI0wvbgoUMkEolaXitwO2GDvew5SqUSa6urtZB6I6dOKIVFu4kR4PvJNsYe6Jd0Ok2hQ72DxnGcQBMiuvn9qzPJtra2OHPmjC9wCwUmJydxXZdkKkU6nbbCthUxvS5WwDZhdXWVSxcvhlpmGA9W1H1MLT6VcrkWmGZxcbGWns1m2Tc3ZwVtM2LqptXVWYnIrIh8RES+ISL3i8hTRWRORO4UkQfM/30mr4jIO0XklIh8VUSeWFfOjSb/AyJyY136k0Tka+aYd4q5+1rVMShUlfkLF0IXrhYL+NrthfPn7YtyD9Hta+MdwD+o6qOBxwH3A68DPqWqVwGfMt8BngNcZbabgHeDLyyBNwJPAZ4MvLFOYL4beHndcdeZ9FZ1DIRVYxKIKnF4MFWVVCpFynSZe9kyGX9Lp9OkM9XvmdrWOA02la7Ln06PdPZWdQFISx3VgNtBtzGgo4lARGaAHwF+HkBVi0BRRK4Hnm6y3Qp8FngtcD3wfvWlwV1G+z1q8t6pqoum3DuB60Tks8C0qt5l0t8PPB/4hCmrWR2ho6q1ACGDwHYJfUSEYmGwAWGSqZ0xFoSdMRNSIcRg6IeFS5c4euyYvSfqGYAXgYjcAjwPmFfVx5q0OeBDwBXAg8ALVXXJ9JrfATwX2AR+XlW/2G8bunkNXAlcBP5MRL4kIu8RkQngsKqeM3nOA4fN5+PAw3XHnzZp7dJPN0mnTR07EJGbROReEbn3Yo/d+6XFxZFHgrKEhZJOpUinU2QyadSrNO4eGYlEgsOHD1vh2shgNNj3sd0brhKo590v3bQyATwReLeqPgHYoKGrbrTVgd627epQ1ZtV9aSqnjx48GBvFQz6hrcP1NAoFYsUiwWKhQKFfH53l3yEP8WRo0ftkjfNqMZRDrJ1QFX/GVhsSL4evzeM+f/8uvT3q89dQLXn3RfdCNjTwGlVvdt8/wi+wL1QbYD5P2/2nwEuqzv+hElrl36iSTpt6giVcrnMcjfxBPrAitcIMUINdnFhoWZLV1WKIbvBjSUivhdB0A0OVHuuZrupi9qC9rz7oqOAVdXzwMMi8n0m6Vrg68DtQNUT4EbgY+bz7cBLjDfBNcCKOaE7gGeJyD4zuPUs4A6zb1VErjF2kJc0lNWsjlCJ8sBWjZhowFE4De1BwobVpd/c3OTsmTMsLS3xvQe/y5nTD+NVKp0PjDu9abCXqj1Xs90cpMph9Ly77av8N+AvRCQFfAd4Kb5w/rCIvAx4CHihyftxfEPxKXxj8UsBVHVRRH4XuMfke1N1wAv4ZXx7SRZ/cOsTJv2tLeoIBc/zWFlZYWmxsRcRPXbMhBpapYOQhhGQsCOmUCj4sQ7MrLxzZ8+SyWYQEVKpNBN7MXzi8LwCLojIUVU912XPuy+6ErCq+mXgZJNd1zbJq8ArWpRzC3BLk/R7gcc2SV9oVkcYlEwEpW5naaUzmZpoaLVmlv9xeyWBMN2qhrF66zCIxDn00ISwr3+9Fl0sFigWt00FyWTSn8K8V6iaCIZDtVf8Vnb3vF8pIrfhu5Ku1JkSembPWtvz+XygKbDFQiHQA7ZrDa4x1EjGr8WDI5lK1aYrV3sTO66PYF669a/hxitYfQm3v7bnz51lanqGfXNze0eTHUQwI5EP4rt5HhCR0/h++K16xU173v2yZwVs0Dn2I9e8RvCgRUDXHAi92GA7LViZ6XNRy3o8z2NleYlEwmV6ZjaUMiPPAEwEqvqiFrsC9bz7Yc8K2GEFyq7Sr3jcI3rMcBjImyPcQl3XHf1LfWjYNbliharWIkk5jrMdwd5xaoKs1jUzXcLAkxAaHo698qiMA4P5LcIVENlcjpnZgYbeiA52Vdl4ISZmqed5tW3QFPJ5HMdBRGpbtS3mw45HdJftTYREJ42mfuCtU3rdoFxtX0P58dQpYBxed+tra0xMTu2JFRMUUKvBxgetG/UfJv0I8kQiQWnI/pJxvTnG5VG+cO4slz3iil3Ba+JHfANux/OsOiAiTE1Pk0wmAT8ItGXvEH391afa49kT7NVoWnHlwIEDO5b0WF5eZtEs/xEGoWvIe+VBGwYD6b2EW+bE5BSpVGrH0jZxxpoIYki9djA7O8vG+nroS6SERZxuv3ZaWct9VZt1k12VgKaTwViHwv2FNjfWmZ090TmjJdLsaQHbSDKZDE3Ahv4Mx+gN77W7OC2lX+uDkolEwHXTom8kUFUKxcLIY9cOBbE22D3B/gMHQot2ryF7JsRHvIZPUDvluLiX5jeH66s9UgYQrjAKWAFbh+u6zM7ukZkzXTAet3Avka7Cl7CD8EpJZ/aA9lqlt3CFkceaCBqY3bcPcRwWLl3qqxw7yDVM/GvT2Puo+faK+AF5qt/HgEw2y+TU9EjbUHVnHPwaZhLbQa7xeA0MmTBuqLAFbDxvv3Dw8IVpxdMdm+cpnlL7r+rna2YDjpI7VCaT5cjRYyNZnHFzc5NKpcLGxgae53FxfiAx7nciWDetvUQY9tM4zCMflzPwhWO3AjI6grQdoxD4qsriwkJtGvnU9DSHDjddBi/8usdEYAbFCtgGCoUCCyH4w1oTQXMGIbTDuDJReiGq7n7Be55HuVQaqFeBqu7wxkin00MS9OMzaBUUK2DrKJfLzF+4EKmHzRINmgma+jQRcFx3R4yHXu+juf0HdnxXVS6cP0chn+fI0WNkstmeyu1EpVLB8zwSiQSe5zE1NTWQepphNdg9QLlU2r0CaYxJJ5O1+AjSxPXF8zxSqdR2QkNAmirKTrNKM8Giqn6QmzAaPgCqwrKVUEwnExQLrSOqFbd23jeJVLqneymRSLC1tYWIkDcBgvL5rVp4zZXl5YEJ2EQiwcFDhxARKpXKcM0UVoONN6rK2tpaKGWNy/IunucFdNDvn6hO/YzKoKS/wvEiK8tLTduUrH/hhYyqkk6lqHgek5OTA6tnFzGeaGAFLL6gWVhY2HMCdiSMraYy3HY3u39c12UmZD/tjY0NHMdhbXWV6ZkZ0uk0rudRLBZ39l4GiA1XGHO2trZYW10Nr8BBrC8UeomjGdiJ52M0HPbN7Q+1B6CqXLp0iYrpxaTSaTKZDK7rDr+nYTXY+OKG7G84EGEIiO5cTareprlzwGU7mLc25KsX/kGDpITBuOr1KkFbHrLJQYRsiMG3Pc/j0qVLeHX3QDV85yjQmL56rYAFymELmgFqsDtKrhO4vWijUbWHBmYI3ctRD88lU6lQA29vbm6ybkxi2VyOcqk0wskWYr0I4owjQjKZrE2l1LpplZ5q4OmVg9JgQy/T2okHRtiXdqqHabOe57WcDVbv9ZHNZsntD9f8EBgrYOPL2toq5VKx6T6BWjfb2bGelhjFafs/te+CW6dtNArcesFWPz9em/wf1XK8b70AACAASURBVPI2AyNO5zIkRIRsQNesxcVFHMdpGrxoa2uLrbpVlV3XHdqAVlPEDnLFmlxugs2NjZb7Bba12S5IJlMU+/SnTaVStSmL4C+aaBkdpXKFVCZHMb/ZXf5ScbtLL4IjTs/3hKrueGF3g+d5OEZoeZ6Heh75QoFsJsO5s2d35E3ExVQUQayAHQBhvIyHobXGSjMeMJVKJVDwFYHa6DxApc8u8KWL8xw8dLhrO+mBAwd2LLJZKBa5cP48uYmJHfkOHznC+sZGqANoQVFrg4036Uwm5BLj2d2JLHvgRbGxvk4hn+f4ZZd3Leir+RzHoWR6Q409tc3NTfbv3x9uY3shpiaCeL42Rkx1Pnc/WA3W0ki5XGZ9bS3Q71bNu9LCz3ttdTUS69CpOIG3cWDPa7BbW1vMnz8XapnlcolUOtPXNFTXdQc+jdVxnB3dyLBJJpO4rnkQtFqnkAkSqb9RljRRdBxpUqb6eYvFUttzbIwY1Rike2eZipvJGqEltfzlYr6jFq3qkUpnzGftOebFwqWLLC8vMTExwb65/R212eqswmYhONPpNFPT0xFYtl6sH2wcqVQqXDh31mpyA8J1HEpbrQcPh4GTzLQVsKratwbn0tmNToCSCRaTzPQXrKVSLiPicPbsWU6c6Lzy7PLS0q5JJYlEgsmpKdLp9EgCezcyLhppUPa0gHUch2wu19aDoFe0X8/VmNqkhk3HQaExvc7TMzPMGdupquJ5XlM/1vW1NfJNPFASiQTT09PRWMlBGNvfoRPxfG10iYhw6PARXHcA75m2a1N3Zhi3m9XcRzQcGcJlrxemItJUuKoqCwsLO3xeq2Sy2WgIV8A3ETiBt3GgK8kiIg8Ca0AFKKvqSRGZAz4EXAE8CLxQVZfE/9XeATwX2AR+XlW/aMq5EfgNU+ybVfVWk/4k4H1AFvg48CpV1VZ19HXGDXiVCpXKAGydfd67wxB96XS6eezW2ofmrag/pj7uwa5SmkTmHzau6yCSpP4HCVuuBH9RKZkGu6cClYpHuRxePOJmpgGA6elp9u3bF1o9/WKjafn8mKrWL7X6OuBTqvpWEXmd+f5a4DnAVWZ7CvBu4ClGWL4ROIl/Tb8gIrcbgflu4OXA3fgC9jrgE23qCI1uJw8Epd9ih3G7Nes69kM2neraEX9YaKVEudhaaCUzw/f/LLUI3J3O5nBcxxfYZqBN2B23txubaaVSYXl5eVe6iPgrJ0dMoMXVBtvPWV0P3Go+3wo8vy79/epzFzArIkeBZwN3quqiEap3AteZfdOqepf6qsD7G8pqVkdoJBKJgQYx3lNE65n10SGtKRUCha1NSvktyoU85WKBciFPqZDHdV0c12Vqepr9Bw8yd+BgRwG5tbnZVLNOhRw0JizUeBIE2caBbq+0Av8oIgr8v6p6M3BYVav+TeeB6vKTx4GH6449bdLapZ9ukk6bOkJDRMhkMjVH7LDYk/bNCJ5yxzGuUGoZzIlnsln27T8YeCKMqrLawu+1WCySz+fJhD65ph/sTK4fUtUzInIIuFNEvlG/09hLB/p4tatDRG4CbgK4/PLLA5cdtnD16e9yRFBWjSmdJWwq1cxGu71GV7HD/TGIFSwcx+Hg4aOBYxCAfz+3Mv/MzM6ONrBLC+Jqg+3qtaGqZ8z/eeCjwJOBC6Z7j/k/b7KfAS6rO/yESWuXfqJJOm3qaGzfzap6UlVPHjx4sJtTqrGxvh66LdJvU3/Hj+ftNn6vheLWpt8lL2zVtlJ+i1J+k1J+E2E0A3WOMQsERVWpeF7LVXBnZ2cj4fdajxJfE0HHKy0iEyIyVf0MPAv4d+B24EaT7UbgY+bz7cBLxOcaYMV08+8AniUi+0RknynnDrNvVUSuMR4IL2koq1kdodGLhtANg5whFVkiqYX0KfSHYsPdTblUYmszmH92uVxm4dIl0qlUy/CGUZgWuwuRPT1V9jDwUfNGTAB/qar/ICL3AB8WkZcBDwEvNPk/ju+idQrfTeulAKq6KCK/C9xj8r1JVRfN519m203rE2YDeGuLOkLD8zySKd9lRur+tJQVWv2nqKd46tXCwe3MZ00EUaB/+dj5lxiEiSCTzZHNTXTOiC9YS6USCwsLFAsFPzxig5aaTCY5ePDgwJb87pdx0UiD0lHAqup3gMc1SV8Arm2SrsArWpR1C3BLk/R7gcd2W0eYdGNj6w4/NLcg/hu5zwfOEfF9JZu5lzYmsFOeN/VtbZIWeqyDCL4V+l3qRVWb2iy37bS+23v18kotQnu1AD82bNAezcxse1eqqraaSqXIFwpsbW67x62tru5advvI0aMjXXOrE+OikQYlev4aQyaspbp9ZMcKBf2gWqGY3z0DJywUQPZAoOU+f4pWK10EIZlMBhKwR45f1lbTVFXmL1wgn8+Tz+d3TSZQVTY2NpianmZtdZVsNhtp4TooROQ6/ElPLvAeVX3rsNsQz9dGl1QqlabTCCPBiGx//RFJFXbkBLkquYnJjt34+oHZVisDe57nC9dcjsNHjgRowWgIe5BLRFzgXfgTn64GXiQiVw/hVHawpzXYxYWFpmHcokEEhZVl4ExOt1/csLrcdrc4jhM5r4FGBrSiwZOBU8bEiYjchj9x6ethV9SOPS1gKwMSroMY9LD0RgQU2K5xE4mOA1sX5+cDmRsabbFRpcdBrgMicm/d95vNJChoPrHpKT02r2f2tIA9dOgQy8vLLC+FGj8mFPoOd2iJDE4X2pnjuhw5flnbga2tzU02AobWjNaMrdb0ONHgkqqeDLstYbKnBazjOMzNzTExMcG5s2cj5bs62HlxRssebBWRIArnKE5n4TE9M9txIGpxcbHt/kZSqVTTMIZRRMMfc2g1sWmoRNs4MyTS6TTHjh0LzVYVRqSiKAiGONCvm1Y4dG5DuotVDoLen7Ozs4Hyj46BxIO9B7hKRK4UkRRwA/7EpaGypzXYelLpNAcPHeLC+fN9lyUiO9Y5alzzqc5FsqlLlx+ubrDadDQEzxAYdFegmyY0vC5d10VMKELE/y0qlfZxYDdNdKztLn/dnaQKstPX2XEcJsbG/hr+RANVLYvIK/FnkLrALap6X6iVdIEVsHVMTExw7Phxzp8715e5QET6npKYTo7hTyMubjJDTf9u+cw035FwHbxKuck7pzq7Qtn+W1eSiHmJbQstre13SNW97HaWrdv/pK6a+hoaGqOA67gUi93/vonCBjOlDaiUoFzYdfbuxAy5zPGmx/pNUJYXFwLdUzOzs5GL+dqOQczkUtWP488sHRlj+BQPlkwmQyab7WudrnBu7IEbYUOvQlUp9RGR33VSoUc2S2Wy/vTREHHTwR4bL5kmsXah5X5J53CSu2eLqXoU5h+mVCxQKHZvHnATCaZnxsU84LNnp8ruRfaZiO+bGxujc7catHwdbPG9Mfre/EhwkilUdceLubh8kcL896hsruEls5DZ31VZIsLhI0cjGVS7NeMTHSso4/QrDI10Os3hw4dZXFzszYWrQYN1HMdfNaFOWNfbX5t9FsdfyE7VuGyphivsq3Pmw6TfZySez1hHnEQSr1TATfn21fLGKpsPfb12TwTx98hmczvs/+PCALwIIoEVsG2YnJwkm82S39raFVCjHY23iuO6FALGnN0ZhkVIpdLbQWmMttPUFGHSUqmU6a5v55M6oeo4DuXIha6LngorImQTxrZbfcFV8iQT7DTV1n3yA/7USkAqJXRyDlc9vI2d62S5E9O4k/twkr5QVPXYOvNAY/Serts7LgNb9QxikCsqWAHbhmoUpWpszVKpRCKRQFUpl0pcvHix+cBDveAToVwKb6XQaplKi2Vp6jThcqX1QF00u5DhP2ShlLixHEj0N+ZVwAPcJlG5tFLGTSRqL0EtFalsrTdk6m7A1XHdsRSwEF8Ba/1gA5BMJhETazOVTrNvbq62r5qeSCR2hAFMhRTFKGxbsNipvEOn2RX38pusfeuL5Oe/R2ltkY2H7t/tudDlb5VOpcfKc6CeuK5oEEU1Zmyon4aoxka6y70rrBs+7AdnTB/EoPT7GgnzxdbK7K2lAvmz32l9nDaPmNVIOjN+tlcfsTZYy24cxyGbzbYNeRjabROyxlmuVEhlMjvaF9ROvIs+m+gppDK57e+Vct/xWKP02PYsQ7r87ROJ8Yz56ptQovRLhYcVsH1y6PBhLly4QL6FkG0VrzMoQTWpTrerqu6wH4cxTbjfV0DjyhKZdApCNl/3Qv3su/7orZRuj4rqcjDdMC5d/qBYG2yfuK7L0aNHyeVyu/Y12mOHSVBhF8btvUesDiOhG9tqNAcu9zb2FwkBEeHI0aOUSiU2NjYo5PNsbGyMVMAGJowANSGPm0VGXosz8NgQHZvQIcZwKpUa2wEu3817TNveAStgQySZTNYiGBUKBdbX1ymVSqGYCcI2ETTieR7pZrFDVZsL3x2r/DWkhUQYpYVRhrQanYoQTX+7MSKuJgIrYAdEOp0mnU4zOTnJpVb+sgEY9JRdVSXfZxszTfw8+yEUs0UIZURBl+6kncpYr8pqvQgsPZJOpzl2/DgbGxssLi6GP+mgBRFXuMaLkLre/bwjO7VgXJaGaYadyWXpCxFhcnKSVDLJmTNnetJGx2ONr7gaYSNAq2shQsJ1d4RkHEesBmvpm1Q6zeEjRzh/7lzgY8dBvIZNXB+6bijOHPUD/Ygxg3seCcf/XJ3UAooopDLjO4OrSnQWawoXK2CHTM/+pmOhwYZMVGRGaNe++3LylcbcjeF/tikVC7vCHY4bcX2ZjrNlfCxxRDoubmcxxOydEkROOwFOvlIus7XZe4D4UdNLHIJxsdlaATtkUuk00zMzo27GmBAzCRuAoNro2kr0lp4PgqoE3sYBayIYAYmASyn3MmFhFLefIDjtzq1RhZPdSy9WB/NG0eWt1WfaVY2565AC9fz91UW/avl3pvmhJKWxYEzYdMTZX5fmb9WronXlVCrBJujmNzepVCpjs0x3I+OikQbFCtgRkAhgIkilUj1NVAgrBkIQVIRKmxi0TY4A/Khk+SaBZjbzRRBzi6qSSCapmBdNfcDxZh4W1bStfBHE3ZVeo0GI5yqbqLfz2gVxrHMnZtgqNAaoqa8ztSPJTSRq59Qv6nkwjgJW/UA/ccQK2BGQTqeZmJhgo8PCiulMpv8IVxHHTSQodeMbXBWm5qu/tHkXT2WjFtxJKx72QFFIgsVxXdwxjUUQZz9Ya4MdEYcOH2Z23762eeJ5y+1EPS9S3dq+zRIBBaaGJGFn5w6MvReBtcFaQkNE2LdvH+p5rK6uhj+RYAweNtd1cRxnd5DyMSaowOz3dxcRZucOMDk93gOncfVCtAJ2hIgI+w8cYHp6mksLC+S3tnY8cPl8nmQy2V0XOgDJRALXdSkWWpsfBOkoLPxBLcfYQ53AGnf1XCMVcazPF1Pga9CH8TE7McncwcMxCFMosQ243bWJQERcEfmSiPyd+X6liNwtIqdE5EMikjLpafP9lNl/RV0Zrzfp3xSRZ9elX2fSTonI6+rSm9YRN5KpFEePHuXgwYO79pVKpZ5sa61uV9d1KZeKFPNb1WlBTTdVr+3+ap5KuUy5VKJUDBYoJpVK4Xle6C+Pftnt1xCMHiZB91RPbmKSg0eOxUC4GhtsTE0EQWywrwLur/v+B8DbVPVRwBLwMpP+MmDJpL/N5ENErgZuAB4DXAf8iRHaLvAu4DnA1cCLTN52dcSSbC7HjAl3WE8yxIdImrhGDZtkKrVr9YLI0OfFkSHFjc1NTo21zXWv0JWAFZETwE8A7zHfBXgG8BGT5Vbg+ebz9eY7Zv+1Jv/1wG2qWlDV7wKngCeb7ZSqfkdVi8BtwPUd6oglruuyf//+XZGRgtoos6kErhvh8cu4GtzoVR/tQVDGTLh26Cw13caBblWjtwOvAabM9/3AsqpWjWengePm83HgYQBVLYvIisl/HLirrsz6Yx5uSH9Khzp2ICI3ATcBXH755V2eUnQ5cPAgiWSS5aWlrrS9bDoFlTK4/s9Z3lpDgbTj7nSCR6DsBw1xXId8aTSDS2Gs/zU4+rTB9vDkiwScRuu4pDPju/5WM+LqptVRwIrI84B5Vf2CiDx98E0KjqreDNwMcPLkyTF5t7XGcRzm5uYQEba2tto6umczKcqba7vSBXY5zFdRAE0SviNYd5e+167tOIRs7KWFnZaDaWR6375Y2F5r7PGJBk8DfkpEngtkgGngHcCsiCSMhnkCOGPynwEuA06LSAKYARbq0qvUH9MsfaFNHXuCffv2MTMzQ6lUolgosLyyQqlRm43SCHyX5PP56Lpn9etF0JMG6xAkYF82NxG4jihTHeSKIx37aqr6elU9oapX4A9SfVpVXwx8BvgZk+1G4GPm8+3mO2b/p9V/Pd8O3GC8DK4ErgI+D9wDXGU8BlKmjtvNMa3q2DM4jkM6nWZqeprLLruMyy6/fIdXgbpR0mS6f0jiGlGsp4kDAWVLMjXewbWbsddtsM14LXCbiLwZ+BLwXpP+XuADInIKWMQXmKjqfSLyYeDr+MEtX6GqFQAReSVwB+ACt6jqfR3q2LMkk0kOzM2yOH8OQfAK46fBgr8oZDqdDrxWWaZhcb/6kCiNnzHfVbWpxtzYLVdA+o3h0It8DaA1J8d59dg2xNUPNpCAVdXPAp81n7+D7wHQmCcPvKDF8W8B3tIk/ePAx5ukN61jr5ObnMarVFi+eL6PUoJFaxoEvQiKZkFhuiGdyXQlzDN9z9rt5bp2n39m3/6AZY8H46KRBiVK/UtLl4gIU7Nz5DfXyW+s91RGFG7ooDbYvpo8rBP2PFwnCZhlXbTabv+7AE4iuSPaWZD3zLivvdUMtavKWqLIzNzBngXsqINZ14cb7JqhCMl+H3TFqzQ320iTT7u/i7k2258RqV2vIKEux4Y97kVgiSipTJZ9h46yNB98EcVRmweASM7m0kSKRM4f+60989Vevza+lup2VFMU6DD913EdFLcWcnHnNGNf0232LvHtr1H2Ie6dKPSoBoEVsGPO5Mw+1pYWKJeCCSsRGakSq6o9DXL1QjKZ7NrPtKJCaVfA7HBRT/F6GEwrFYt4nhfxiRq9EdeJBvH7pfYgB45dFry7HYH7eVjLwjiO07W2PBRTYB916JBiHQwTxTcRBN3GAStgY0AylSadDep8PnoJWywWozcjaQgPbj/hdsoRiz4WFnH1g7UCNiakc7lgBwxAvvbiZF8qlUgPeGRcVbue2DAUBbaPStZXV8JrSIQYtoAVkReIyH0i4onIyYZ9gcKqtsMK2JiQm5xGAtjmBiFIei2zWCwOdGZXsVjs+toMx0TQeyUba6uhLZIYFVTBUwm89cm/Az8N/HN9Yo9hVVsSsf6ZpVcSyRTJVNoPpN0NsnP0Oxx6D+JSLpdxXbf9arg9CKZUOg2qOF0eO/poue1RJdCLdFwYdpdfVe+HppNdamFVge+aGanVyU6nzOQnROQ2k/fr7eqxAjYmqOd1L1xhILbGfqJdVbvx7QRsL6KvVCxGLgpXP0LcdZ1YehH0yAERubfu+80msl4/BA2r2hYrYGNCsVAgmZ1ssqeVcHFIJ7t40GuunrstrL4vvNRl7XO5lYgJwsHR+3lWKhXKpVLsJhz0+NNfUtWTrXaKyCeBI012vUFVhxI4ygrYmLCxtcVmYbQjzP0OVhUKhcCxUTsRdnnhEG0zxCgYhNuVqv54D4cFDavaFtvXiAlxcd+J41z7XfQhXzPZXPy0V4jSooeBwqp2KsxqsDEhelpabxTyeVJRXhRxxEzPzo26CeEzAr9WEfnPwP8CDgJ/LyJfVtVn9xhWtSVWwMaETDbLxsbGSNvgOA6puniliiLIzjitXTxJrus2NTeICIlEYnsNK1Po9qDR7lCBO9e7UoqlMl6lsivYTPWziOAmkrVjQWrrmu0oecdaZ9tJCpSLxbYzrnrVvdKZLJmg/s5jwrBnZqnqR4GPttgXKKxqO6yAjQlRMBF4ntcQuKQ3NJGkXB7M+WzLWj+gSjqVoljI19Lz5f4151Qi0TY2RK+yZN+Bg7EMtu2bCEbdisFgbbAxYWZ2NjbLsJTLJdKZzEDORxj8EFMnISg9CJNsbiJ2K8nWY6fKWiJNIpFgdnZ2pG0IU3AV8nkct+/lBTozCI2wQ5G9jM9MTM/01pYxIa7BXqyJIEZsbQWYaNAF9YFYFMw0ojZu8iK4iYTZL3hepa+VY0vFIslkClWvttBjocclY4aJOK4Z6a/adWHblguOSM1bQky6Ghvv7giz5pg4Ty4YI400KFbAxogw7XMiQjngnPfG/JlMui+B6HkenleslT0u9kf1PCptbLCNVzWRylLoaEMfj3PvBQWiuIJ7GMT4tbj3mNu/f2yEUJwJPhW2s/oWxziw9cTVBms12Bjhui6O47QPmNIl0fSrDT9AzSBeR0FtrN2EeUyl4j0BI5K3WwhYARsjyuVyKMI1quz0aQ2HQTzXgYV2h0ak0pnYzd6qR8do0CooVsDGiK3NzVE3YeyIhkmlvXSZO3BgSO0YHdHsMfWPFbAxYnJqio2NDUql0s5HtuHm1Ya06vfqGllqPvdPFIRX9Gl3raemZ8hk4zl7q56YylcrYOOEiJCbnOTixYu9F1J3p/cvHsN9asbnIQxqhG1+YiLC7P74a69gvQgsY4JG6k4NW4MNX8JGQma3uEy5yUncYUy2sAwMK2Bjxtr6emhl9e/cHp74chxnIKEMJWS1OJFIUMwHC7qj0rwjmUpnwmhS5OnFRWtcejNWwMaMgwcOMDXZbGWDgERi8AcTPUvwPG8gs7gi4UXQ4ojILWk+QOI6VdYK2JiRTqc5dOgQl5040fcKA1EY2XUcNxLt6JaeWtpCIrvu3hGwVoO1jBWpVIpjR4+OdYStVDpNMYTwh+0IXVHv4cFv1gTXdUln9oaJAEA9DbyNA3vnFbkHcRyH/fv3c/78+VE3pSdKpRJT3oYvgHxfMj98t/pTR3376U51pjR1iIpuC86q8JIGySfql1UJfQpqOA9+OpuNiI/u4LETDSxjSy6b7XkJllF3zdMJB3d1uem+VqIntfhQsEr2nQiWvwOqPayt20SQJhLj2/PohXHp8gfFCtiY07OQHNEdLyIkXJc0RdygwrIHwj7LngRsEwr5rdrEj72AF1MVtqMNVkQyIvJ5EfmKiNwnIr9j0q8UkbtF5JSIfMistIhZjfFDJv1uEbmirqzXm/Rvisiz69KvM2mnROR1delN67B0h6pyYX5+rBYQnHQqTCw/RGL53FDmgYVugu3hxdRMiBbyeTwvvnEl6lHiO8jVjQZbAJ6hqusikgQ+JyKfAP5v4G2qepuI/CnwMuDd5v+Sqj5KRG4A/gD4LyJyNf5St48BjgGfFJH/aOp4F/BM4DRwj4jcrqpfN8c2q8PSBecvXGAzhPgEqZT/XlOvgjgOjuOyHRraj9tam2aLvz5YJ0Hjuu6OwDRuIuGbBC59t+/2BsFRr8VSLNvn11LPVfwg45WiH09bfQHr1OkB/sKP1SJ2Tk8WVRQlvfAgsnIJqZTBKyOeh3oVPN3AveoJIZxlxBkjgRmUjgJW/Sel6r2eNJsCzwD+q0m/FfhtfOF3vfkM8BHgj8V/RV8P3KaqBeC7InIKeLLJd0pVvwMgIrcB14vI/W3qsHQgn8/3JVzdRAJXBNdxKBQKNBMyiWSKUpNA0Y7jtJ2BlEpnTJm+G5bjOP46XCN4yhyvRKHUe72uI7j5FWD7CjUOm7UqvZqeWjoP803MIVurPbdrvFC8mErYrmywIuICXwAeha9tfhtYVtVqcPbTwHHz+TjwMICqlkVkBdhv0u+qK7b+mIcb0p9ijmlVR2P7bgJuArj88su7OaXYs7La+8OZTqcp5bcoszv6fj2ttFR/JQKPRMI1s5GMdut5iOOQz1ddr6SWd/c4/7Doz0gQPLh297hXfv/Ayo4acY0n3pWAVdUK8HgRmcVfS/zRA21VQFT1ZuBmgJMnT8bzVRiASwsLrPc4ZTadSlHMb3UlNjrl2RakVRSRaLle932zDNJQHFOtrhHfBhvPcw10t6vqMvAZ4KnArEhtEvUJ4Iz5fAa4DMDsnwEW6tMbjmmVvtCmDksbeplSmkgkyKTTlIvFrmVGD8M5sX2Q+qP5NSn96+1743qpH00r6DYOdONFcNBorohIFn8w6n58QfszJtuNwMfM59vNd8z+Txs77u3ADcbL4ErgKuDzwD3AVcZjIIU/EHa7OaZVHZYQSaWSVEpFivmtQGs/7Q0HoiHQwhVLVy6iC2eH3JjRUI1BHGQbB7oxERwFbjV2WAf4sKr+nYh8HbhNRN4MfAl4r8n/XuADZhBrEV9goqr3iciHga/jm/ZeYUwPiMgrgTsAF7hFVe8zZb22RR2WNlQ6vN5FBMdxSCYT4CnFQj4awnIED00kzrtFK2RqDpmJfzxYZQ/P5FLVrwK7fEXMqP+Tm6TngRe0KOstwFuapH8c+Hi3dVja0ynMYMJ1qJRKFMudloreC/T5ZIchGFrYpROPfRqSjPdih4DvphVTCRutEQdLKLQK8JJKJkknk1SauFb1QjwfiYCEoAJrKxNBaXwmiFiaYwVsDMmaQCFVTdZ1XTLpNKVigdKAo1P1xQimhfY9sXWAZo3KN+5GN9cGVn6UiOtMLitgY0g6lcJRRStlUgkXvErXrldBiIb9sj/6PYdQLAQtStH1ZYqf/iBaaeeNHA88TwNv44AVsDEklUrhug6CmbY6Lj4tI2H0D2o7LVo3ltHl+SG2Zvj04kEQJy8Cy5jhL7096lb0wBg2OpQWdzSNxKGv0J49PZPLMn4kEi6VPdC17Jt+JWQYL4UOs9u8rTUcjvZfT4TZ07EILONHOrMdUGVQtBr97plRxD7td4wrlDa0F7BOJoRFLCPOuHT5g2JtsDFl/4GDHDl2PISlt2NOvw92KKNc7aW81yzSVoxQje8gl9VgSbyRtgAACz9JREFUY0w2m2Vicoq11ZVRN6UrVEFqS6VUF9VqFD5aJ9R0V14tBdXa+1Nhg2peIg44jh9kW8Rfkjw9gbvvKDj+PhWnblExobS+ghvz1Q1iqsBaARt3xumhLCfSLGUO9VXGdK6CeBVU8GWxyA6BrCqoSfCnaArphFA7QMCPnu2HIhQngadenQAwH9SU4nmUnEmqglprizJSqyeTcCgX8jtfFlotSilMHoDJ9lNipxbnmdp/uI8rE23iOpPLClhLz/iyaGdE1GYCvVHL8+XK7gcqjFdBvgJaCTgkXWm9NIubTFMqdzNY2EFA9PmiW7t0AcdNkpvZN1YvzW5Q3eMBty3jSyLRemWBUKhqcrWv/TwoYYzI919E+OWF06iVC6dJpjOksrlQyosScdVg7QhIzEmmxihYSCjyNVwJGyVdMZWbjKVwBV/ABt36QUT+HxH5hoh8VUQ+Wg3JavYFWpy1HVbAxpxMJjM+ngRhSLOQu8+h9FxDalLcTAM11A9XGHTrkzuBx6rqDwDfAl4P0LA463XAn4iIa8K1vgt4DnA18CKTty1j8uRZesVxHA4dPjIWD2coPvvhL8QdQhnhtCmuvqLK8DVYVf3HuvX+7sJfMQXqFmdV1e8C1cVZn4xZnFVVi8BtJm9brIDdA2RzOY5fdhnZXDy7l/Vo6PK1f6EW1sstMzEVSjnRo+dYBAdE5N667aYeG/ALwCfM59qirYbqYqut0ttiB7n2CMlkioOHDrO4sMDmxrpZyTVahKGhhS9fhzzwJuJr4VV3sWqy45Cbmeu/LVHETDTogUuqerLVThH5JHCkya43qOrHTJ434K+w8he9NKATVsDuIVzX5eChQ8Ah8ltbrK2usrm5ERlhG86Afdg22PBNBMlszo9ypgrqmf9aa7vWJlNs151OZXHcAXuEjJBBmD9U9cfb7ReRnweeB1yr2w1otQgrbdJbYgXsHiWTzZLJZlFVSsUi+XyefH6Lrc3NrgVuNC2CgxCwYXsmCF5jIJ4OL4ZkJhtqG/Y6InId8BrgR1V1s27X7cBfisgfAcfYXpxVMIuz4gvWG4D/2qkeK2D3OCJCKp0mlU4zPTODqnLp4kXW11Y7HxvFQZfQx/JCErC1rn+r8NodDh8XT5AeqA5yDZk/BtLAncZGfpeq/lKPi7O2xApYyw5EhJnZ2a4E7AAq77+MkJ9TVe1bvnpmzQI1c2iL+c2OxzRSzG/114goM4JFD1X1UW32BVqctR1WwFp2kUqlSCQSlLuaIhotIumOFkKbtM103vEnvlNl49vvsPTF4aPHhl5nGI9YPB/T7cAxcWXYfrDDwmqwlqakUimyuRxbm627s6Hf4hF00zIxtkIopT/ivK6aEt9JFFbAWloyNTXdVsCGTRTdtKJCuVTE8yo4TgxdtXr3g408VsBaWpLOZDhw8GAtvGB9TNTq51aPRU3MNQo8E5quNhun6v+piiPSwhZXrc8c52ltwGhnmFYl4To4mq3lb4z2Vd+OZm2ralK1tgHiJvHjYzs4joNUA2YjpggTP7Yubfuz4LgO5LKI+Mf5wbYdxPE/V70LGsvbvpbS/mLHgHHp8gfFClhLSxKJBFPTM6NuhiX2jM8y3EGxAtZisYwUf0JbPG3MVsBaLJaRY22wFovFMiCsicBisVgGgY6PX2tQrIC1WCwjZUSxCIaCFbAWi2XkeGoHuSwWiyV8RhDsZVh0jEUgIpeJyGdE5Osicp+IvMqkz4nInSLygPm/z6SLiLzTrLz4VRF5Yl1ZN5r8D4jIjXXpTxKRr5lj3ikmYkerOiwWS3xQgschGBeB3E2wlzLw31X1auAa4BVmNcXXAZ9S1auAT5nv4K+6eJXZbgLeDb6wBN4IPAV/AbE31gnMdwMvrzvuOpPeqg6LxWKJPB0FrKqeU9Uvms9rwP34i31dD9xqst0KPN98vh54v/rcBcyKyFHg2cCdqrqoqkv4y+ZeZ/ZNq+pdZtmG9zeU1awOi8USI3pc9DDyBLLBisgVwBOAu4HDqnrO7DoPHDafg67KeNx8bkynTR2N7boJX1sGWBeRbwY4rQPApQD5R4lta/iMSzsh+m19RE9HKZFZFy5suhawIjIJ/BXwalVdrQ9srKoqIgN9pbSrQ1VvBm7upVwRubfdypRRwrY1fMalnTBebQ3KuNhUg9JVwG0RSeIL179Q1b82yRdM9x7zf96kt1qVsV36iSbp7eqwWCwxQVFUvcDbONCNF4EA7wXuV9U/qtt1O1D1BLgR+Fhd+kuMN8E1wIrp5t8BPEtE9pnBrWcBd5h9qyJyjanrJQ1lNavDYrHEBd3bKxo8Dfg54Gsi8mWT9uvAW4EPi8jLgIeAF5p9HweeC5wCNoGXAqjqooj8LnCPyfcmVV00n38ZeB+QBT5hNtrUESY9mRZGhG1r+IxLO2G82hqIcRGYQZFxGY2zWCzxZObA1XrNT/xl4OP+8f1P+ELUbdJ2JpfFYhkpGuOZXFbAWiyWkRPXgNtju2y3iGRE5PMi8hUzhfd3TPqVInK3mXb7IRFJmfS0+X7K7L+irqzXm/Rvisiz69KvM2mnROR1delN6+iiza6IfElE/i7KbRWRB83U5S+LyL0mLXJTo0VkVkQ+IiLfEJH7ReSpEW3n95lrWd1WReTVUWzrSIjxINfYCligADxDVR8HPB5/Vtg1wB8Ab1PVRwFLwMtM/pcBSyb9bSYf4k/7vQF4DP4U3T8RXxC6wLvwp/5eDbzI5KVNHZ14Ff5MuCpRbuuPqerj62xcUZwa/Q7gH1T10cDj8K9t5Nqpqt801/LxwJPwB38/GsW2joY97KYVVcxU3HXzNWk2BZ4BfMSkN07hrU67/QhwrXnLXw/cpqoFVf0uvvfDk812SlW/o6pF4DbgenNMqzpaIiIngJ8A3mO+tytnpG1tQaSmRovIDPAj+C6EqGpRVZej1s4mXAt8W1UfGoO2DgXFXzIm6DYOjK2AhVqX+8v4ExDuBL4NLKtq2WSpn3Zbm6pr9q8A+wk+tXd/mzra8XbgNUD11duunFG3VYF/FJEviD8NGSI0NdpwJXAR+DPxzS7vEZGJCLazkRuAD3YoJyptHQ5m0cOg2zgw1gJWVSum23UCX4t79Iib1BQReR4wr6pfGHVbuuSHVPWJ+F3VV4jIj9TvNFrSwKdGd6gjATwReLeqPgHYoKH7G5F21hDf/v1TwP/up5xeGUYdvbG3wxVGHtM1/AzwVPzuVNU7on7abW2qrtk/AywQfGrvQps6WvE04KdE5EH87vsz8O2HUWwrqnrG/J/HtxU+mehNjT4NnFbVu833j+AL3Ki1s57nAF9U1QsdyolCW4eKtcFGDBE5KCKz5nMWeCb+IMdngJ8x2Rqn8FZHXX8G+LR5o98O3CD+yP2V+AMEn8efcXaV+KPwKfyu3e3mmFZ1NEVVX6+qJ1T1ClPOp1X1xVFsq4hMiMhU9TP+lOZ/J2JTo1X1PPCwiHyfSboW+HrU2tnAi9g2D7QrJwptHR4x9iLoKQ5jFDbgB4AvAV/FFwC/ZdIfiS90TuF3xdImPWO+nzL7H1lX1hvw7bffBJ5Tl/5c4Ftm3xvq0pvW0WW7nw78XVTbavJ/xWz3VcvCt+d+CngA+CQwZ9IF34Ph28DXgJN1Zf2CqfcU8NK69JPmN/s28Mdszyj8/9u5YxSEYSgMwC+zheIpXD2AB3DyrI7Ozj2Kd4hDO3QQS4dnInwfvKUk9A3hhwSSj//40us5IqZlDdwj4thjn8ucQ8w7inH1rctef13DeKqX23N3RcTUuvetclUWaKqU8oj5rdu9XrXW6/awdgQsQJK/PYMF6J2ABUgiYAGSCFiAJAIWIImABUjyBlbaLSRijrj9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wi[\"approx\"] = approx\n",
    "ax = wi.plot(column=wi[\"approx\"], cmap=\"coolwarm\",\n",
    "             figsize=(5,5), vmin=-cap, vmax=cap, legend=True)\n",
    "ax.set_title(\"Reconstructed from 2 PC weights\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How does that compare to what the state *actually* looked like on March 1st:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Actual')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wi[\"actual\"] = dft.loc[mar1]\n",
    "ax = wi.plot(column=wi[\"actual\"], cmap=\"coolwarm\",\n",
    "             figsize=(5,5), vmin=-cap, vmax=cap, legend=True)\n",
    "ax.set_title(\"Actual\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Not bad!\n",
    "\n",
    "## Conclusion\n",
    "\n",
    "In these examples, we weren't trying to predict anything.  We were looking for patterns and simplicity underlying our data; we were applying unsupervised learning.  For the two clustering approaches, we tried to group similar counties.  For the PCA, we tried to understand how counties vary together."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}