analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## L.A. homeless arrests analysis\n",
    "\n",
    "By [Christine Zhang](mailto:ychristinezhang@gmail.com)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Los Angeles Times analyzed daily arrest logs between January 1, 2011 and December 31, 2016 from the LAPD to determine yearly trends in arrests of homeless people.\n",
    "\n",
    "The results were reported in a February 4, 2018, Los Angeles Times story titled [\"L.A. leaders oppose 'criminalizing' homeless people. But thousands are jailed for minor offenses\"](http://www.latimes.com/local/politics/la-me-homeless-arrests-20180204-story.html).\n",
    "\n",
    "Here are the key findings of the data analysis, which is documented below with code written in the programming language R:\n",
    "\n",
    "   * The LAPD made 14,000 arrests of homeless people last year, a 30% increase over 2011\n",
    "   * LAPD arrests overall went down 15% from 2011 to 2016\n",
    "   * Two-thirds of those arrested were black or Latino\n",
    "   * In 2011, one in 10 people arrests citywide were of homeless people; in 2016, it was 1 in 6\n",
    "   * The 14,000 arrests of homeless people in 2016 included more than 500 unique charges\n",
    "   * By far the most common was failing to appear in court on an unpaid infraction or misdemeanor citation\n",
    "   * The top five charges were for non-violent or minor offenses\n",
    "   \n",
    "Read more about the methodology [here](../README.md)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How we did it"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import R data analysis libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "library('dplyr')\n",
    "library('feather')\n",
    "library('ggplot2')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The file `arrests.feather.zip` must first be unzipped to run this notebook. This is a file that has been prepared outside of this notebook as part of an unpublished processing script. The raw LAPD data includes names and other identifying information about arrestees that The Times has chosen to withhold.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "unzip(\"arrests.zip\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding: The LAPD made 14,000 arrests of homeless people last year, a 30% increase over 2011"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read in the data for analysis. Note that the data has been pre-processed to remove the name, address and date of birth fields for privacy purposes. Age was calculated based on the individual's date of birth at the time of arrest."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "data <- read_feather('arrests.feather')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<ol class=list-inline>\n",
       "\t<li>'booking_num'</li>\n",
       "\t<li>'homeless'</li>\n",
       "\t<li>'arrest_year'</li>\n",
       "\t<li>'arrest_ymd'</li>\n",
       "\t<li>'booking_ymd'</li>\n",
       "\t<li>'gender'</li>\n",
       "\t<li>'race'</li>\n",
       "\t<li>'age'</li>\n",
       "\t<li>'occupation'</li>\n",
       "\t<li>'charge_code'</li>\n",
       "\t<li>'charge_desc'</li>\n",
       "</ol>\n"
      ],
      "text/latex": [
       "\\begin{enumerate*}\n",
       "\\item 'booking\\_num'\n",
       "\\item 'homeless'\n",
       "\\item 'arrest\\_year'\n",
       "\\item 'arrest\\_ymd'\n",
       "\\item 'booking\\_ymd'\n",
       "\\item 'gender'\n",
       "\\item 'race'\n",
       "\\item 'age'\n",
       "\\item 'occupation'\n",
       "\\item 'charge\\_code'\n",
       "\\item 'charge\\_desc'\n",
       "\\end{enumerate*}\n"
      ],
      "text/markdown": [
       "1. 'booking_num'\n",
       "2. 'homeless'\n",
       "3. 'arrest_year'\n",
       "4. 'arrest_ymd'\n",
       "5. 'booking_ymd'\n",
       "6. 'gender'\n",
       "7. 'race'\n",
       "8. 'age'\n",
       "9. 'occupation'\n",
       "10. 'charge_code'\n",
       "11. 'charge_desc'\n",
       "\n",
       "\n"
      ],
      "text/plain": [
       " [1] \"booking_num\" \"homeless\"    \"arrest_year\" \"arrest_ymd\"  \"booking_ymd\"\n",
       " [6] \"gender\"      \"race\"        \"age\"         \"occupation\"  \"charge_code\"\n",
       "[11] \"charge_desc\""
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "names(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>booking_num</th><th scope=col>homeless</th><th scope=col>arrest_year</th><th scope=col>arrest_ymd</th><th scope=col>booking_ymd</th><th scope=col>gender</th><th scope=col>race</th><th scope=col>age</th><th scope=col>occupation</th><th scope=col>charge_code</th><th scope=col>charge_desc</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>2497688                                </td><td>0                                      </td><td>2011                                   </td><td>2011-01-01                             </td><td>2011-01-01                             </td><td>M                                      </td><td>W                                      </td><td>25                                     </td><td>ENGINEER                               </td><td>23152(A)VC                             </td><td>drunk-driving-alcoholdrugs             </td></tr>\n",
       "\t<tr><td>2497689                                </td><td>0                                      </td><td>2011                                   </td><td>2011-01-01                             </td><td>2011-01-01                             </td><td>M                                      </td><td>H                                      </td><td>38                                     </td><td>AGENT                                  </td><td>23152(A)VC                             </td><td>drunk-driving-alcoholdrugs             </td></tr>\n",
       "\t<tr><td>2497690                                </td><td>0                                      </td><td>2011                                   </td><td>2011-01-01                             </td><td>2011-01-01                             </td><td>M                                      </td><td>W                                      </td><td>25                                     </td><td>BUSINESS DEV                           </td><td>23152(A)VC                             </td><td>drunk-driving-alcoholdrugs             </td></tr>\n",
       "\t<tr><td>2497697                                </td><td>0                                      </td><td>2011                                   </td><td>2011-01-01                             </td><td>2011-01-01                             </td><td>F                                      </td><td>W                                      </td><td>40                                     </td><td>NONE                                   </td><td>273.5(A)PC                             </td><td>corporal-injury-on-spousecohabitantet  </td></tr>\n",
       "\t<tr><td>2497698                                </td><td>0                                      </td><td>2011                                   </td><td>2011-01-01                             </td><td>2011-01-01                             </td><td>M                                      </td><td>H                                      </td><td>34                                     </td><td>LABORER                                </td><td>12021(A)PC                             </td><td>poss-firearm-by-convictd-felonaddicte-1</td></tr>\n",
       "\t<tr><td>2497699                                </td><td>0                                      </td><td>2011                                   </td><td>2011-01-01                             </td><td>2011-01-01                             </td><td>F                                      </td><td>H                                      </td><td>24                                     </td><td>TARGET                                 </td><td>23152(A)VC                             </td><td>drunk-driving-alcoholdrugs             </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|lllllllllll}\n",
       " booking\\_num & homeless & arrest\\_year & arrest\\_ymd & booking\\_ymd & gender & race & age & occupation & charge\\_code & charge\\_desc\\\\\n",
       "\\hline\n",
       "\t 2497688                                 & 0                                       & 2011                                    & 2011-01-01                              & 2011-01-01                              & M                                       & W                                       & 25                                      & ENGINEER                                & 23152(A)VC                              & drunk-driving-alcoholdrugs             \\\\\n",
       "\t 2497689                                 & 0                                       & 2011                                    & 2011-01-01                              & 2011-01-01                              & M                                       & H                                       & 38                                      & AGENT                                   & 23152(A)VC                              & drunk-driving-alcoholdrugs             \\\\\n",
       "\t 2497690                                 & 0                                       & 2011                                    & 2011-01-01                              & 2011-01-01                              & M                                       & W                                       & 25                                      & BUSINESS DEV                            & 23152(A)VC                              & drunk-driving-alcoholdrugs             \\\\\n",
       "\t 2497697                                 & 0                                       & 2011                                    & 2011-01-01                              & 2011-01-01                              & F                                       & W                                       & 40                                      & NONE                                    & 273.5(A)PC                              & corporal-injury-on-spousecohabitantet  \\\\\n",
       "\t 2497698                                 & 0                                       & 2011                                    & 2011-01-01                              & 2011-01-01                              & M                                       & H                                       & 34                                      & LABORER                                 & 12021(A)PC                              & poss-firearm-by-convictd-felonaddicte-1\\\\\n",
       "\t 2497699                                 & 0                                       & 2011                                    & 2011-01-01                              & 2011-01-01                              & F                                       & H                                       & 24                                      & TARGET                                  & 23152(A)VC                              & drunk-driving-alcoholdrugs             \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "booking_num | homeless | arrest_year | arrest_ymd | booking_ymd | gender | race | age | occupation | charge_code | charge_desc | \n",
       "|---|---|---|---|---|---|\n",
       "| 2497688                                 | 0                                       | 2011                                    | 2011-01-01                              | 2011-01-01                              | M                                       | W                                       | 25                                      | ENGINEER                                | 23152(A)VC                              | drunk-driving-alcoholdrugs              | \n",
       "| 2497689                                 | 0                                       | 2011                                    | 2011-01-01                              | 2011-01-01                              | M                                       | H                                       | 38                                      | AGENT                                   | 23152(A)VC                              | drunk-driving-alcoholdrugs              | \n",
       "| 2497690                                 | 0                                       | 2011                                    | 2011-01-01                              | 2011-01-01                              | M                                       | W                                       | 25                                      | BUSINESS DEV                            | 23152(A)VC                              | drunk-driving-alcoholdrugs              | \n",
       "| 2497697                                 | 0                                       | 2011                                    | 2011-01-01                              | 2011-01-01                              | F                                       | W                                       | 40                                      | NONE                                    | 273.5(A)PC                              | corporal-injury-on-spousecohabitantet   | \n",
       "| 2497698                                 | 0                                       | 2011                                    | 2011-01-01                              | 2011-01-01                              | M                                       | H                                       | 34                                      | LABORER                                 | 12021(A)PC                              | poss-firearm-by-convictd-felonaddicte-1 | \n",
       "| 2497699                                 | 0                                       | 2011                                    | 2011-01-01                              | 2011-01-01                              | F                                       | H                                       | 24                                      | TARGET                                  | 23152(A)VC                              | drunk-driving-alcoholdrugs              | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  booking_num homeless arrest_year arrest_ymd booking_ymd gender race age\n",
       "1 2497688     0        2011        2011-01-01 2011-01-01  M      W    25 \n",
       "2 2497689     0        2011        2011-01-01 2011-01-01  M      H    38 \n",
       "3 2497690     0        2011        2011-01-01 2011-01-01  M      W    25 \n",
       "4 2497697     0        2011        2011-01-01 2011-01-01  F      W    40 \n",
       "5 2497698     0        2011        2011-01-01 2011-01-01  M      H    34 \n",
       "6 2497699     0        2011        2011-01-01 2011-01-01  F      H    24 \n",
       "  occupation   charge_code charge_desc                            \n",
       "1 ENGINEER     23152(A)VC  drunk-driving-alcoholdrugs             \n",
       "2 AGENT        23152(A)VC  drunk-driving-alcoholdrugs             \n",
       "3 BUSINESS DEV 23152(A)VC  drunk-driving-alcoholdrugs             \n",
       "4 NONE         273.5(A)PC  corporal-injury-on-spousecohabitantet  \n",
       "5 LABORER      12021(A)PC  poss-firearm-by-convictd-felonaddicte-1\n",
       "6 TARGET       23152(A)VC  drunk-driving-alcoholdrugs             "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "head(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Group the data by arrest year and homeless and sum the total number of arrests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "arrest.totals <- data %>% \n",
    "  group_by(arrest_year, homeless) %>% \n",
    "  distinct(booking_num) %>% \n",
    "  summarize(arrests_number = n())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Filter to homeless arrests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>arrest_year</th><th scope=col>homeless</th><th scope=col>arrests_number</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>2011 </td><td>1    </td><td>10496</td></tr>\n",
       "\t<tr><td>2012 </td><td>1    </td><td>11837</td></tr>\n",
       "\t<tr><td>2013 </td><td>1    </td><td>12237</td></tr>\n",
       "\t<tr><td>2014 </td><td>1    </td><td>12622</td></tr>\n",
       "\t<tr><td>2015 </td><td>1    </td><td>13418</td></tr>\n",
       "\t<tr><td>2016 </td><td>1    </td><td>14011</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|lll}\n",
       " arrest\\_year & homeless & arrests\\_number\\\\\n",
       "\\hline\n",
       "\t 2011  & 1     & 10496\\\\\n",
       "\t 2012  & 1     & 11837\\\\\n",
       "\t 2013  & 1     & 12237\\\\\n",
       "\t 2014  & 1     & 12622\\\\\n",
       "\t 2015  & 1     & 13418\\\\\n",
       "\t 2016  & 1     & 14011\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "arrest_year | homeless | arrests_number | \n",
       "|---|---|---|---|---|---|\n",
       "| 2011  | 1     | 10496 | \n",
       "| 2012  | 1     | 11837 | \n",
       "| 2013  | 1     | 12237 | \n",
       "| 2014  | 1     | 12622 | \n",
       "| 2015  | 1     | 13418 | \n",
       "| 2016  | 1     | 14011 | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  arrest_year homeless arrests_number\n",
       "1 2011        1        10496         \n",
       "2 2012        1        11837         \n",
       "3 2013        1        12237         \n",
       "4 2014        1        12622         \n",
       "5 2015        1        13418         \n",
       "6 2016        1        14011         "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "homeless.totals <- arrest.totals %>% filter(homeless == 1)\n",
    "homeless.totals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"The *raw* increase in homeless arrests between 2011 and 2016 is 33%\"\n"
     ]
    }
   ],
   "source": [
    "print(paste0(\"The *raw* increase in homeless arrests between 2011 and 2016 is \", \n",
    "            round((homeless.totals[homeless.totals$arrest_year == 2016,]$arrests_number / \n",
    "            homeless.totals[homeless.totals$arrest_year == 2011,]$arrests_number - 1) * 100), \"%\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "However, there are some missing dates in the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# extract the booking dates as a vector\n",
    "booking.dates <- data %>% select(booking_ymd)\n",
    "booking.dates <- booking.dates %>% distinct(booking_ymd) %>% select(booking_ymd) %>% arrange(booking_ymd)\n",
    "booking.dates$has.data = 1\n",
    "\n",
    "# get the time period (minimum date and maximum date) of the data set\n",
    "time.min <- booking.dates$booking_ymd[1]\n",
    "time.max <- booking.dates$booking_ymd[length(booking.dates$booking_ymd) - 1] \n",
    "\n",
    "# create a dataframe of all the days spanning that time period\n",
    "all.dates.frame <- data.frame(list(booking_ymd = seq(time.min, time.max, by=\"day\")))\n",
    "\n",
    "# merge this dataframe with the vector of booking dates to find the missing dates\n",
    "merged.data <- merge(all.dates.frame, booking.dates , all=T)\n",
    "missing.dates <- merged.data %>% filter(is.na(has.data) == T)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There were six days in 2011 for which arrests data were not available"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>booking_ymd</th><th scope=col>has.data</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>2011-02-20</td><td>NA        </td></tr>\n",
       "\t<tr><td>2011-03-17</td><td>NA        </td></tr>\n",
       "\t<tr><td>2011-04-17</td><td>NA        </td></tr>\n",
       "\t<tr><td>2011-05-18</td><td>NA        </td></tr>\n",
       "\t<tr><td>2011-06-11</td><td>NA        </td></tr>\n",
       "\t<tr><td>2011-08-04</td><td>NA        </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ll}\n",
       " booking\\_ymd & has.data\\\\\n",
       "\\hline\n",
       "\t 2011-02-20 & NA        \\\\\n",
       "\t 2011-03-17 & NA        \\\\\n",
       "\t 2011-04-17 & NA        \\\\\n",
       "\t 2011-05-18 & NA        \\\\\n",
       "\t 2011-06-11 & NA        \\\\\n",
       "\t 2011-08-04 & NA        \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "booking_ymd | has.data | \n",
       "|---|---|---|---|---|---|\n",
       "| 2011-02-20 | NA         | \n",
       "| 2011-03-17 | NA         | \n",
       "| 2011-04-17 | NA         | \n",
       "| 2011-05-18 | NA         | \n",
       "| 2011-06-11 | NA         | \n",
       "| 2011-08-04 | NA         | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  booking_ymd has.data\n",
       "1 2011-02-20  NA      \n",
       "2 2011-03-17  NA      \n",
       "3 2011-04-17  NA      \n",
       "4 2011-05-18  NA      \n",
       "5 2011-06-11  NA      \n",
       "6 2011-08-04  NA      "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "missing.dates"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pro-rate the 2011 figure to account for the missing six days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "10671.4206128134"
      ],
      "text/latex": [
       "10671.4206128134"
      ],
      "text/markdown": [
       "10671.4206128134"
      ],
      "text/plain": [
       "[1] 10671.42"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "prorated.homeless.2011 <- homeless.totals[homeless.totals$arrest_year == 2011,]$arrests_number/(365 - 6) * 365\n",
    "prorated.homeless.2011"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"The *prorated* change in homeless arrests between 2011 and 2016 is 31% (rounded down to 30% in the story)\"\n"
     ]
    }
   ],
   "source": [
    "print(paste0(\"The *prorated* change in homeless arrests between 2011 and 2016 is \", \n",
    "            round((homeless.totals[homeless.totals$arrest_year == 2016,]$arrests_number / \n",
    "            prorated.homeless.2011 - 1) * 100), \"% (rounded down to 30% in the story)\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding: LAPD arrests overall went down 15% from 2011 to 2016"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Group `arrest.totals` by arrest year and sum the total number of arrests to get the overall arrest numbers for each year."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "all.totals <- arrest.totals %>% \n",
    "  group_by(arrest_year) %>% \n",
    "  summarize(arrests_number = sum(arrests_number))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>arrest_year</th><th scope=col>arrests_number</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>2011  </td><td> 96701</td></tr>\n",
       "\t<tr><td>2012  </td><td>101234</td></tr>\n",
       "\t<tr><td>2013  </td><td> 98126</td></tr>\n",
       "\t<tr><td>2014  </td><td> 94077</td></tr>\n",
       "\t<tr><td>2015  </td><td> 87067</td></tr>\n",
       "\t<tr><td>2016  </td><td> 83608</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ll}\n",
       " arrest\\_year & arrests\\_number\\\\\n",
       "\\hline\n",
       "\t 2011   &  96701\\\\\n",
       "\t 2012   & 101234\\\\\n",
       "\t 2013   &  98126\\\\\n",
       "\t 2014   &  94077\\\\\n",
       "\t 2015   &  87067\\\\\n",
       "\t 2016   &  83608\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "arrest_year | arrests_number | \n",
       "|---|---|---|---|---|---|\n",
       "| 2011   |  96701 | \n",
       "| 2012   | 101234 | \n",
       "| 2013   |  98126 | \n",
       "| 2014   |  94077 | \n",
       "| 2015   |  87067 | \n",
       "| 2016   |  83608 | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  arrest_year arrests_number\n",
       "1 2011         96701        \n",
       "2 2012        101234        \n",
       "3 2013         98126        \n",
       "4 2014         94077        \n",
       "5 2015         87067        \n",
       "6 2016         83608        "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "all.totals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"The *raw* change in overall arrests between 2011 and 2016 is -14%\"\n"
     ]
    }
   ],
   "source": [
    "print(paste0(\"The *raw* change in overall arrests between 2011 and 2016 is \", \n",
    "            round((all.totals[all.totals$arrest_year == 2016,]$arrests_number / \n",
    "            all.totals[all.totals$arrest_year == 2011,]$arrests_number - 1) * 100), \"%\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Again, we need to pro-rate to take into account the six missing days in 2011."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "98317.1727019499"
      ],
      "text/latex": [
       "98317.1727019499"
      ],
      "text/markdown": [
       "98317.1727019499"
      ],
      "text/plain": [
       "[1] 98317.17"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "prorated.arrests.2011 <- all.totals[all.totals$arrest_year == 2011,]$arrests_number/(365 - 6) * 365\n",
    "prorated.arrests.2011"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"The *prorated* change between 2011 and 2016 is -15%\"\n"
     ]
    }
   ],
   "source": [
    "print(paste0(\"The *prorated* change between 2011 and 2016 is \", \n",
    "            round((all.totals[all.totals$arrest_year == 2016,]$arrests_number / \n",
    "            prorated.arrests.2011 - 1) * 100), \"%\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding: Two-thirds of those arrested were black or Latino"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Group the data by arrest year, homeless, and race/ethnicity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "arrests.race <- data %>% \n",
    "  group_by(arrest_year, homeless, race)  %>%\n",
    "  distinct(booking_num)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create a variable, `race.grp` to represent racial/ethnic grouping, where W = white, B = black, H = Latino (Hispanic), etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\n",
       "     A      B      C      F      H      I      J      K      O      P      W \n",
       "    41 171726    431    541 253265     81    102      4  32248    104 102270 "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "table(arrests.race$race)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "arrests.race$race.grp <- ifelse(arrests.race$race == 'W', \"White\",\n",
    "                                ifelse(arrests.race$race == 'B', \"Black\",\n",
    "                                       ifelse(arrests.race$race == 'H', \"Latino\",\n",
    "                                              ifelse(arrests.race$race == 'A' | arrests.race$race == 'C' | arrests.race$race == 'J'| arrests.race$race == 'K'| arrests.race$race == 'F', \"Asian\",\n",
    "                                                     'Other'))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Group by `race.grp` and calculate the total number and percentage of homeless arrests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>arrest_year</th><th scope=col>homeless</th><th scope=col>race.grp</th><th scope=col>arrests_number</th><th scope=col>arrests_percent</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>2016      </td><td>1         </td><td>Black     </td><td>4997      </td><td>35.6648348</td></tr>\n",
       "\t<tr><td>2016      </td><td>1         </td><td>Latino    </td><td>4681      </td><td>33.4094640</td></tr>\n",
       "\t<tr><td>2016      </td><td>1         </td><td>White     </td><td>3833      </td><td>27.3570766</td></tr>\n",
       "\t<tr><td>2016      </td><td>1         </td><td>Other     </td><td> 481      </td><td> 3.4330169</td></tr>\n",
       "\t<tr><td>2016      </td><td>1         </td><td>Asian     </td><td>  19      </td><td> 0.1356077</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|lllll}\n",
       " arrest\\_year & homeless & race.grp & arrests\\_number & arrests\\_percent\\\\\n",
       "\\hline\n",
       "\t 2016       & 1          & Black      & 4997       & 35.6648348\\\\\n",
       "\t 2016       & 1          & Latino     & 4681       & 33.4094640\\\\\n",
       "\t 2016       & 1          & White      & 3833       & 27.3570766\\\\\n",
       "\t 2016       & 1          & Other      &  481       &  3.4330169\\\\\n",
       "\t 2016       & 1          & Asian      &   19       &  0.1356077\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "arrest_year | homeless | race.grp | arrests_number | arrests_percent | \n",
       "|---|---|---|---|---|\n",
       "| 2016       | 1          | Black      | 4997       | 35.6648348 | \n",
       "| 2016       | 1          | Latino     | 4681       | 33.4094640 | \n",
       "| 2016       | 1          | White      | 3833       | 27.3570766 | \n",
       "| 2016       | 1          | Other      |  481       |  3.4330169 | \n",
       "| 2016       | 1          | Asian      |   19       |  0.1356077 | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  arrest_year homeless race.grp arrests_number arrests_percent\n",
       "1 2016        1        Black    4997           35.6648348     \n",
       "2 2016        1        Latino   4681           33.4094640     \n",
       "3 2016        1        White    3833           27.3570766     \n",
       "4 2016        1        Other     481            3.4330169     \n",
       "5 2016        1        Asian      19            0.1356077     "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "arrests.race.yr <- arrests.race %>%\n",
    "  group_by(arrest_year, homeless, race.grp) %>%\n",
    "  summarize(arrests_number = n()) %>%\n",
    "  mutate(arrests_percent = arrests_number / sum(arrests_number) * 100) \n",
    "\n",
    "arrests.race.yr %>% filter(homeless == 1 & arrest_year == 2016) %>% arrange(desc(arrests_percent))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {},
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAYAAAD958/bAAAEDWlDQ1BJQ0MgUHJvZmlsZQAA\nOI2NVV1oHFUUPrtzZyMkzlNsNIV0qD8NJQ2TVjShtLp/3d02bpZJNtoi6GT27s6Yyc44M7v9\noU9FUHwx6psUxL+3gCAo9Q/bPrQvlQol2tQgKD60+INQ6Ium65k7M5lpurHeZe58853vnnvu\nuWfvBei5qliWkRQBFpquLRcy4nOHj4g9K5CEh6AXBqFXUR0rXalMAjZPC3e1W99Dwntf2dXd\n/p+tt0YdFSBxH2Kz5qgLiI8B8KdVy3YBevqRHz/qWh72Yui3MUDEL3q44WPXw3M+fo1pZuQs\n4tOIBVVTaoiXEI/MxfhGDPsxsNZfoE1q66ro5aJim3XdoLFw72H+n23BaIXzbcOnz5mfPoTv\nYVz7KzUl5+FRxEuqkp9G/Ajia219thzg25abkRE/BpDc3pqvphHvRFys2weqvp+krbWKIX7n\nhDbzLOItiM8358pTwdirqpPFnMF2xLc1WvLyOwTAibpbmvHHcvttU57y5+XqNZrLe3lE/Pq8\neUj2fXKfOe3pfOjzhJYtB/yll5SDFcSDiH+hRkH25+L+sdxKEAMZahrlSX8ukqMOWy/jXW2m\n6M9LDBc31B9LFuv6gVKg/0Szi3KAr1kGq1GMjU/aLbnq6/lRxc4XfJ98hTargX++DbMJBSiY\nMIe9Ck1YAxFkKEAG3xbYaKmDDgYyFK0UGYpfoWYXG+fAPPI6tJnNwb7ClP7IyF+D+bjOtCpk\nhz6CFrIa/I6sFtNl8auFXGMTP34sNwI/JhkgEtmDz14ySfaRcTIBInmKPE32kxyyE2Tv+thK\nbEVePDfW/byMM1Kmm0XdObS7oGD/MypMXFPXrCwOtoYjyyn7BV29/MZfsVzpLDdRtuIZnbpX\nzvlf+ev8MvYr/Gqk4H/kV/G3csdazLuyTMPsbFhzd1UabQbjFvDRmcWJxR3zcfHkVw9GfpbJ\nmeev9F08WW8uDkaslwX6avlWGU6NRKz0g/SHtCy9J30o/ca9zX3Kfc19zn3BXQKRO8ud477h\nLnAfc1/G9mrzGlrfexZ5GLdn6ZZrrEohI2wVHhZywjbhUWEy8icMCGNCUdiBlq3r+xafL549\nHQ5jH+an+1y+LlYBifuxAvRN/lVVVOlwlCkdVm9NOL5BE4wkQ2SMlDZU97hX86EilU/lUmkQ\nUztTE6mx1EEPh7OmdqBtAvv8HdWpbrJS6tJj3n0CWdM6busNzRV3S9KTYhqvNiqWmuroiKgY\nhshMjmhTh9ptWhsF7970j/SbMrsPE1suR5z7DMC+P/Hs+y7ijrQAlhyAgccjbhjPygfeBTjz\nhNqy28EdkUh8C+DU9+z2v/oyeH791OncxHOs5y2AtTc7nb/f73TWPkD/qwBnjX8BoJ98VVBg\n/m8AAEAASURBVHgB7N0JnFxVlfjxU91dve97ZyEkZF8IYScsCRAWRRxFHVfGfdRBx9HRURlH\nRcZt9O/OOIPMgAoyKApuiCCEJQQIeyCQFbKn9yXpJL3X/5zb/aqrqiudXqq7qrp+N5+Xqnrr\nfd/3uvuduved5wtoEQoCCCCAAAIIIIAAAggggICkYYAAAggggAACCCCAAAIIINAvQIDEmYAA\nAggggAACCCCAAAIIDAgQIHEqIIAAAggggAACCCCAAAIDAgRInAoIIIAAAggggAACCCCAwIAA\nARKnAgIIIIAAAggggAACCCAwIECAxKmAAAIIIIAAAggggAACCAwIECBxKiCAAAIIIIAAAggg\ngAACAwIESJwKCCCAAAIIIIAAAggggMCAQEoESM8884xceOGFUYeLLrpIrrzySnnve98rN9xw\ngxw9enRST47//u//dvX605/+NObttrS0uHXYPoyk/O3f/m1Ui7e97W3yla98RX71q19Jd3f3\nSFY1qfPEwsoq/NRTT7n9/9rXvjap9U+kjd16661y/vnnS05OjlRWVsqDDz54zOr95Cc/cV5/\n/vOfjzlPsk743ve+5/Zt7dq1yboLSV/vQ4cOSXNzc9LvBzuAAAIIIDB1BDKmzq4ce0/sj+9D\nDz3kZkhLC48JA4GA2GDl5z//ufziF7+Q3/3ud1JVVeXGTfR/r776qqvbu9/97jFvqqury61j\n4cKFI1rH448/Lnv37o0675133unGW0D529/+VoqLi6POF4+RsbCyeltAaefDtGnT4rEbcd+m\nHf/3ve990tvbKzNmzJCKigqZNWvWMeu1Y8cO53X11Vcfc55knbB161a3bx/96EeTdReSut6/\n//3vxezt987KlSuTel+oPAIIIIDA1BEIjxamzn5F3ZPly5e7i0K7MPSGvr4+aWxslHvuuUcW\nL14sTz75pLzrXe+KuvxEjLz88svl61//upx++ukTsfph12ktAnv27HHDrl275JVXXpG7775b\n5s2bJ/aN+khbpIbdCBMTTsBai+z8t+Nrx//ZZ5+Vk046KeHqSYWmvsD//M//yIEDB6b+jrKH\nCCCAAAJJJZASLUjHOyJlZWXyute9zgVIc+bMkUcffVSs20dBQcHxFh33dK/r37hXNIYVWCuZ\ntSCEFmuFsmDNxt93331y5MgRyc3NDZ2F90ku4F2Q2rlHQQABBBBAAAEEEAgXIEAK8bBuRtbt\nyrqf2X0qdn9SaNm+fbs8/PDDsnnzZtfqZMHUsmXL5G/+5m/E5/OFzureW3Cxfv16d3+HtVLN\nnz9f/u7v/s7d8+HN/MQTT8jzzz8vq1atkkWLFnmj3etotxe28Dg+TJ8+XU444QTZvXu3a2FY\nsGCBW9tjjz0mL774orz97W93rQ533XWXWKuc7b/dx+KVDRs2yCOPPOJapGbOnClnnHGGvP71\nr49qZC0Z1nr38ssvi3V3KiwsdC1YFrDOnj3bW+Wwr7W1ta7ly2ay+8ms/l6x7pX33nuvq4/5\n2/ThSmtrq/zf//2fq09PT487vlb30C5ot912mwug3/Oe90h+fn5wdXb/2s9+9jO3n3//938f\ntr91dXViXnPnzpU1a9aIdXN74YUXxO77slZMa82zc8H2/+yzzz7mORXcWJQ3dm6au3VFtPpa\noHvFFVcE62HH06ztGFpZt26du+fu5JNPHnH3JvP8zW9+I3ZfnwXY1i3K9ic9PX1IjUZiaQvZ\n/W7WkmDH+7LLLnNd3qwF05a3nwsbl5eXJ52dnc7Jukdat1ibdtVVVw3Zro0YzTkYdQU60r4k\nse629vN5+PBhd67buWA/G5HFftbtvLFjavPaz/JZZ50l5513XuSs7kuHkc47ZOGQESP9/WBu\nN998s2sltHPixhtvdL+/3vjGN7r62e+o4/1cj8bCqvjcc885O2uZtvPEWuff8IY3SGlpqdsD\nM7LuzDt37nSfreV648aN7ndLSUmJG3e8dbiZ+A8BBBBAAIGJENALjSlftCXEbjIK6MX8sPuq\nF5BuPpt3y5YtwXn1AjbwzW9+M5CVleWm631MAQ2IgvOee+65gfb29uD89kYv+AN6MeDmsXkz\nMjLce72gDvzhD38Izvsv//IvbvxPf/rT4LjRbk8DBLcObf0JrmO4N9o65ObXrlVRZ9MA0e2f\nzRdaPvGJT7jlzCJ0/3/0ox+52TRACHz84x8PTvO8zFO7EgasnqFFA83AmWee6dZp83hG9l5b\nrQK//OUvQ2cPRLPSwCOgF6NuHTY9tGhAEtALdzfNW7deaAe8/dCulKGzB379618HysvL3fy2\nf96ytowmKgjOqwGim8fWH1o0+HDjrf56oRw6KfCDH/zATbNXK//8z//sPuuFckAvuIPL2bI2\n6MVkQC9sw9ZxrA9tbW0BvdgNrsPv9wffaytRQC9S3aJ//OMfg+O97djrpz71qWOt2o336vqx\nj30sal01UAxooBu2jpFa2kJ68e3q9aY3vSlwzTXXDKmjBiVuHg3EhkyzOoWW0Z6Dev+LW6cd\nh9CiwWNAAzY3zc6FzMxM914D2IDeqxg6a0Av8gMalAfrFur//ve/P+w4jmbesI2EfBjt74em\npiZXtze/+c0BDb6D9dTAJaDBafDn4Vg/16OxsGr+67/+a3AboRb65VNAg1u3J9q1MzhP6Ln4\n0ksvuekjWYebkf8QQAABBBCYAAH7JnbKl5EESBbQnHLKKe6Ptl1E2EWIVzTDnBtvAYO912+2\n3cX+//7v/wa0hcRN84IEW2bfvn3uQtIChB/+8IcBu4C1ixTNmOXmtSDJ5rES7aJ/tNsba4Bk\ngZq2NrhBv40O6P1XAW0dcReGdkF4xx13uDp6/3mBRXZ2dkBb1wL/+Z//GfjgBz8Y0NYxN4t3\nIW0Bo34b7Az3798f+NCHPuT2+9JLL/VW5V61ZcmN/+QnPxmwCyO7WLOLsQ984ANufE1NTUAT\nUASXibSqr68PLFmyxM37pS99KTifvbHjqV0kA/ptdEC/nXYXqWauiQbc/HZRFhog6f1XAc3o\nFrB9s/2yY6ytAi5o8gJdbU1w27DAzZaPvDi3QMO72LNjHVouueQSN80ukK14VuZsQcztt98e\nsP3RxBgBuwi39WimudBVHPO9XYTb/JqVLqAtAS5YseOqrVNuvLYauHEdHR0B7V4X0HuP3Hht\nSXCfDx48eMx12wSvrraNiy++2F3kWmBqQaMmeHDrCg0aRmNp6/cCJLuYtmNwyy23BOy8sfPP\n+/myc8GCaTtHtRUroK0PLqC24GXbtm22Gle8uo70HIwWINm27dyxn18LaK1+FqxqS2TA6mEO\n2prqbTJgAZyNu+6661wwauestoAFtLXQjbdg0SujmddbJvJ1tL8fvADJzm07t774xS+6L3x+\n/OMfu1UP93M9WgttuXP7rK217mfZzjkLhq699lo33n5erVhAbeei/U4wO03W4D7b74CRrsOt\niP8QQAABBBCYAIGUCpDs4uu0004LG+wPtrUQ2B9pG6wVJrT1yMztotouxOyiLLLYhbktZxed\nXtHuVW7cV7/6VW9U8PWtb32rm/bd737XjYu86LeRo93eWAMkb5+jvWpXsWCdvTfehZRd/Ng3\n9aFFu8cF7ALXvknXLHGhk9x7u0i37VgrhhULiOx4mLddFIUWa42zaTb/a6+9FpwUamVBmXYN\nc/Nouu7gPN4b7ZblpkULMrzALDRAWr16tZvfu2j01mOvf/nLX9w07bbmAjbbP2uR0i6WobO5\nYE2THQSshVG78gWnWQBigdCKFSuC47wLedv/0CDQZrBzw/b9He94R3D+Y72xb+RtXgveI4+J\nLXPOOee46RYMecUCO1tG07l7o4Z9Da2rBQuh5R//8R/duj772c8GR4/G0hbyAiSrk33pEFq8\nY24titayGVq8nyULKq2M9hy0ZaIFSJrhz+3Tt7/9bZslrHitcHY+e0UzPboWz8jz2C76rSUw\n9GdpNPN66498He3vBy9AMl/tyhi5umALUrSf69FaeC2lmngmbDsWEOnjBdzvNgtwveK1fIYG\nnKNdh7cuXhFAAAEEEIiVQEoFSHaBEG2w4EfvZQj813/9V9SLTMO21oTQViUbZ5/tgs7Wact7\nxbsA91pWvPH2ahcr1pXPW5d3ARjaxc7mG832xhogWUuZtWx4w6mnnhpsEbBWLmv9Cu065QVI\n1g0qslhgYQ52wRyt2MWmTbfWIq+Yge1nZLEARO/tcvNbS5RXPCvrCuS19kVeiHnzWsuGfVuu\n9xF5o4KvdsFqdfECJJvHWgus1eBY3dq8YGzTpk1uPdaCZuuwljcr1jplny2YWLp0adi29Z4d\nN81aGLziBR22T5FF70dy81trw/HKv//7v7t5v/GNb0Sd1YIHq5e1JnllrAGSdZ+MLNaqYuu3\nC10rY7EMDZAiu2FaN0ZbvwV6keXTn/60m2Y/t1bGcg5GC5Csm6X9TrCf1chiway1SlqdvC8C\nLrjgAvfZul5aC1foz0zk8qOZN3LZ0M+j+f0QGiA1NDSErsa9H+7nerQWmiHRWdjP30033RSI\ntr3QCkQLkEa7jtD18R4BBBBAAIFYCKRUkgZLKGA3/lrRb3vdzeza190970cvzFwmO+2G4qZH\n/qctGi7hgqUB9xIK6MWy6MWHm1UvioKvdsNzUVGRWHa8yGI3KXs3KkdOC/080u2FLjPa99qV\nTLRVY8hilgbansukwY7L5GfPzAktlgY8smg3JzfKEhFYMofIohfBbpQ3n33Qi1CXJMCet2SW\nlvxCW+/cqwZPbn7P1X0Y+E9b7UQvVN0nS34QWSxLm16YiQYqUZMHRNZfW6lcAgBLRqEtPZGr\nc5+1pcfdRG51tBvOLTGFOWnrkvzDP/yD3H///W4+7YLm9slsn376aXejvrY6uGl6j82QdUe7\n4d87P6Lte+QKLDW7FS+RRuR0q7cVq/d4iyW5iCx2A74VbfVzr2OxdAvqf5aIwVtf6Dh7H5p4\nw5vmZVfUX4RulHdujeYc9NblvVpiCEuoYkkntPuXNzrs1X53WLHtWQISe+CwXuiLdgl0g/3c\nW2IJSwhi47162jKjmdfmP1YZy+8He6aZBjzHWqVLjhI6cSwW9vDhd77znaJdRkW71oolK9FW\ne5ekxRKSaIt96Caivo/FOqKumJEIIIAAAgiMUCClnoNkJnZRboNdCNvFo3Yzkre85S0u+LEL\nGu9CL9TPLsItI5Xe1yD6rbXLCGUX4PoNv2hLQOisoveRiPa7dwFS2IRRfBjN9kax2hHPatn7\nvvCFL7j5LWNZZIkW+HmBirbEiD2MN3KwgNEuJr0MVbZO/cbfZQSzbHDaAuIyBFo2vC9/+ctD\nMvqF1sEuULX7oluX3vvl0rKHTvfqoi1IoaOD7yMvEu2YWRnuobjeuuyi0YqXDc9SoVuxAEm7\nGIq2EIgFSVYsgLKLd8scZxnaLFtcZLEL3fGU49U9st7j2ZYFMMcrx6uPLX+sOnmBYbRt2Pl0\nvOId99Gcg5Hr9NZhvx8iz2Hvs2Wos3NZW3Hc4papzrL6feYzn3FBhn1povepuUDBgmnvSxmb\neTTzRtbN+2x1HOnvI28Ze432czvc9LFYaNdTsSyPNlgmSjsWlhFUW0/dFxb2+/N4gX8s1hG6\nX7xHAAEEEEBgtAIp1YIUDce+KbYUw5YW2NLMWhpua9EILXqvg5tuKZPtIkhvFg9+K2zflFrx\nvsW2b8DtosBapGycBWOhxVo+7AKqurp6yDRvvtFsz1sm1q+2j1YsFfRIyoknnuhm0+538k//\n9E/HXeSvf/2ra6Gyi25rbbFv3C1tulfsAsuK5+qNt1dr9fu3f/s396wmTeggmqTAHTvvm3q7\ngLVgxR6CGq3YsQkt3naPNb/NaymyrWjXIfdqAY+leLcgyAI22x9Lz237Y88Xsotpm2bBkl1o\nagIHt1ys/zte3SPrHevtR67vePWx+SeqTqM9ByPrbp+tRc9+Zu1n2H4njLTY+aDdSN1gadYt\ntbx2m3UpwvWeIdF77oKrGs28wYVC3kzW74exWpifdl91g31ZZKnn9X4s0S53oslLXGp424fh\nSizWMdz6mYYAAggggMBwAsf/Wna4pafINGvd0MxZ7sLIuufYs2y8Yi1FFiTYxbe1pqxevToY\nHNk81t3OivetqF0Y27NuLBCy56dEFgvG7FlLFohFK6PdXrR1xGKcdQ+zYkHASIrXxcuCgmjF\nnidjwZP5WrFXC34+97nPid4TExYc2XNbduzY4ebzXN2Hgf/sGT9WLDCyZ/DYvJ///OcHpoq7\nuLV6631BYl2+Iot9ox1aLFi1Vg27cI82v7UqWguBldBWIOs+ZV0H7ZhaEOS1HFkrmXUrsq6G\nd955p1suWvc6N2Gc/3nu9gykaEWTOLjR1r10MspYLWNRN89ipOdgtG1ai54FBtZSqGnwh8xi\nP9fWpdKCdPuiw56ZZueeBexesSDR5rHWRfu94XXFHc283roiXyfz98NoLayu9sWGZrYM/u6z\nLsvWVdFai63LrhULmIYrsVjHcOtnGgIIIIAAAscTIEAaELJuZdZn3op1m7MLESvevS520W73\ntoQWu4CyLl5WbLpX7ILfil002cNGvWL31dg3qFbs4jpaGcv2oq1nrOMsINH036Jped0q7OGO\nIynWTVFTILtlLRgKLdZiYy091lLkdUvy9nPnwIMivfktaLL7FrzAKNTVmyf01R56aa02dgEW\nGiToM1/cA1g1CUJwXbacdQH7zne+E7oK994u3mybduy9e0y8meyeJ7sYtnPEC85smt2HZOX6\n6693rxasecXea1Y5ueGGG9x9H9Y9cyKKtU5YQKZptt2DZkO3YcGe7asF7ZplMXTShL4fi2Us\nKjTac/BY29SkBW6S7Yd3vnrzfuUrXxFNb+4CX+sSaF+uaNY10WQZ7p4zbz57tXPNzgF7WLJ1\nbxvNvKHrCX3v/dyM5vdR6PKjfT8aC1u3fWGkiWvCAkZvm/bQWCuhwbp3z6f3+9amj3YdtgwF\nAQQQQACBmAroBemULyN5DpIhWFYq/Qbc7vgO6I3GQRfv4Yr2oFnLxmbPwbFMYHqBFNBvrd38\nlnXNK3qB7VKJ23psWXsmjqUD1641bl5L/TtcFrvRbk8DELdeSxk9kuI9KNaeMWPPavEG/eY8\nYNnrrN42WHa70CxwXraraOnObbuWNtoerqoX5AF7No+lltbudsFn2djDYs3GipepzdJff/jD\nH3bPvPnWt74VsCxfli5cv4V3dbAMcF7xsthFZvwzX6uvLWMpwq1osBPQAMaNt+cDaaASsHTg\ntu/2wEqb38tiZ/MfPnw4YGmObbylcNbAwmVF857TYim6Q9MT2zJ2DL3n4pibXrzaaFceeOAB\nty5bn1lEFi+LXbS0y5YJzZazVOUjKeahXZJcKnE7RrZOvYcsYCmlbXxk6uyxZrGLVle9v8bV\n1bL6eWW0ll4WOzsfI4v3s2spoiOLPc/HnEIf4juac9DWFy2LnR1Hy+ho67ZzwjIl2nPO7Py1\n89vOWe1SGayO94wfy26nyU1cNsyPfOQjwYfH2rnkldHM6y0T+Tra3w8a3Lt9sRT00cpwP9ej\ntbBnHul9hG57ls3TMjf+x3/8R8Ae8mvnov0O1GAoWA3v58B+juxBwZYlcrTrCK6MNwgggAAC\nCMRIwLo5TfniXWRZgHO8Yg91tAsjG+yBjFb0m08XLHjj7dUuhuyC3i74LcCwP/76jb2b3/6z\nNLx24aHdVILrswt/S01sF4ReiXbRP9rtjTVACt0fe6/dgdy+2MWMPT/IC2a8ug53IeXN88QT\nT7gU3ObhrV9beAJ675YLQrz57FVblNxFvDefBVbahdE9I8m2b+PtOSxeiWZl0ywY0pvW3fyh\nKcgtWLJnUnlBr9Vp1apVgccff9zNGxog2XrsuNgFsz1Q06uTBVP23JloKdttGe+ZV5Epue0B\nmd6x9x4wa/N7xbswjBZ0jDZAsnVaQOYF61Z3O9dWrlwZ9WGzEx0gWX1GYxnLAMm2PZpzMFqA\nZOuw42fPdgr9wsBc7aJ/rT4ENrRYoKytiO4LE++8sVf7wkG7i4XO6oLqkc4btmDIh9H+fhhP\ngGSbHY2Fza9di92DpEMt7GdBW82H/Bxp11Rn6v2+8J4ZNZp12DYpCCCAAAIIxFLAZyvTP2SU\nEQjYfSbWJczuIbJuMyMp1m3L0gFblxi7R8LrUjKSZceyvZGsdzLmsft2LLW0XmC6DG5203u0\nYi52U7t1RbLkCnbfQ6yLneJ2k7ylih4uU5q3XTtm+nwjd09atNTW3nyJ+KqtoO6eLLO0rofx\nLvG0HOk5OJyRnTv2M29dwCwJhGVZPFbRLxRcYhDrWmep5IfLGjeaeY+1vcn+/TAaC6uztriK\ndavzMoZaQpxjFX2YsvsdYL4aLAVnG806ggvxBgEEEEAAgXEKECCNE5DFEUAAAQQQQAABBBBA\nYOoIkKRh6hxL9gQBBBBAAAEEEEAAAQTGKUCANE5AFkcAAQQQQAABBBBAAIGpI0CANHWOJXuC\nAAIIIIAAAggggAAC4xQgQBonIIsjgAACCCCAAAIIIIDA1BEgQJo6x5I9QQABBBBAAAEEEEAA\ngXEKECCNE5DFEUAAAQQQQAABBBBAYOoIECBNnWPJniCAAAIIIIAAAggggMA4BQiQxgnI4ggg\ngAACCCCAAAIIIDB1BAiQps6xZE8QQAABBBBAAAEEEEBgnAIZ41w+4Rdva2uTzs7OuNczMzNT\ncnNz5ciRI9LV1RX3+iRSBbKzs6W3t1e6u7sTqVpxr4u52HDo0CHnE/cKJVAF8vLypKOjA5eI\nY2Iufr9fWltbI6bwsaCgwP0sIREuUFhY6EYcPHgwfAKfhHMm+klQXFzs/l4fPnw4+gwpOjY9\nPd39zcYl/AQwF/tZsr/ZNsS7WH3KysqOW40pHyDZhbcNiVAsSLIAKVHqkwgmVoe0tDTp6+vD\nJeKAmIudM4FAAJsIG/sFxzkTgaIfMzIy3DljNnbeUAYFLHDkd++gh/fOXKxg44kMvnLODFp4\n73w+H3+XPIyIV/ubbX+b+FkKh/GuZaxxIJls6GIXfhz5hAACCCCAAAIIIIAAAiksQICUwgef\nXUcAAQQQQAABBBBAAIFwAQKkcA8+IYAAAggggAACCCCAQAoLECCl8MFn1xFAAAEEEEAAAQQQ\nQCBcgAAp3INPCCCAAAIIIIAAAgggkMICBEgpfPDZdQQQQAABBBBAAAEEEAgXIEAK9+ATAggg\ngAACCCCAAAIIpLAAAVIKH3x2HQEEEEAAAQQQQAABBMIFCJDCPfiEAAIIIIAAAggggAACKSxA\ngJTCB59dRwABBBBAAAEEEEAAgXABAqRwDz4hgAACCCCAAAIIIIBACgsQIKXwwWfXEUAAAQQQ\nQAABBBBAIFyAACncg08IIIAAAggggAACCCCQwgIESCl88Nl1BBBAAAEEEEAAAQQQCBcgQAr3\n4BMCCCCAAAIIIIAAAgiksAABUgoffHYdAQQQQAABBBBAAAEEwgUIkMI9+IQAAggggAACCCCA\nAAIpLECAlMIHn11HAAEEEEAAAQQQQACBcAECpHAPPiGAAAIIIIAAAggggEAKCxAgpfDBZ9cR\nQAABBBBAAAEEEEAgXIAAKdyDTwgggAACCCCAAAIIIJDCAgRIKXzw2XUEEEAAAQQQQAABBBAI\nFyBACvfgEwIIIIAAAggggAACCKSwAAFSCh98dh0BBBBAAAEEEEAAAQTCBQiQwj34hAACCCCA\nAAIIIIAAAiksQICUwgefXUcAAQQQQAABBBBAAIFwAQKkcA8+IYAAAggggAACCCCAQAoLZCTC\nvm/fvl1effXVsKqUlpbK6aefHhy3e/duWb9+vdj4lStXSn5+fnAabxBAAAEEEEAAAQQQQACB\nWAgkRIB0++23y7p166SgoCC4T8uWLQsGSL/4xS/kpptuklWrVsn+/fvFPv/whz+UkpKS4Py8\nQQABBBBAAAEEEEAAAQTGK5AQAdLWrVvlwx/+sLz1rW8dsj/WcnTzzTfLD37wAznllFOkp6dH\nPvrRj8odd9zhXocswAgEEEAAAQQQQAABBBBAYIwCcb8HqbOzUywIWrBgQdRd2LBhg0ybNs0F\nRzZDRkaGXH755XL//fdHnZ+RCCCAAAIIIIAAAggggMBYBeLegvTaa69JX1+fPPHEE/L9739f\n2tvb5cILL5T3v//9kpWVJQcOHJDp06eH7Z8FTI2NjW65tLTBGG/v3r3y+9//PmzeCy64QGbN\nmhU2Lh4fLLCzkp2dLenp6fGoQsJu0+/3OxNcwg9RZmamG5GTk+N+FsKnpvYnO1dyc3Pd74DU\nlgjfe+9nKC8vL3wCn8Tn83HvapTzwFyscF/vUBzOmaEm3hj7XcM542n0v9r1KC7hJvbJu063\na5pEOGcCgcDQSkYZE/cAadu2ba5a1pJ0zTXXyNNPPy133XWXNDc3y7XXXiu1tbVSWFgYVnW7\nV8mCqra2trD7kKwlyrrihZbZs2fL0qVLQ0fF9b0FSDZQEBipABe70aUssKZEFwi9nzP6HKk5\nFpdjH3dsotvgEt3FvvTFJroNf5uiu1iA5H3xG32OyRnb1dU1og3FPUC69NJLXTKGmpoaV+FT\nTz3VReC33HKLfPzjHxc70ey+o9DifbZvkEPLokWL5MYbbwwdJTNnznTBVtjIOHzwIufDhw+L\nBYOUQQFrIent7ZWRnrSDS07td+Ziw8GDB4f8DEztPT/+3tm3UEePHnXnzfHnTp05zMV+19gX\nTJRwgaKiIvelWvhYPhUXFzuE1tZWMCIEOGciQPSjtapZgiz7e209fiiDAtZ6ZH+zcRk0sXcW\nTFtDh/3NtiERimXEPl6Je4Bk3ei84Mir7Nlnny0WIFnrUXl5uezcudOb5F7tgtF+QG3Z0GLj\nLNNdaLELhUQISLwmxu7u7oSoT6hRvN/bBZ0FvYlwnOJtEbp975sW+0Nk5w1lUMC+HDEX78uS\nwSmp/c5rbTSbkXYjSBUx8+B3zNCj7Z0n2ES3wSXcxeuSyc9TuIt9si/07bqUcybcxnp8WbEv\nwhPBxuuKHl7LoZ8Gb+AZOm1Sxtx5553yuc99LmxbL7zwgvuWwgIn6yK3efPmsAuhTZs2Dbkv\nKWwFfEAAAQQQQAABBBBAAAEExiAQ9wDJHvr65JNPyu9+9zsXBD3zzDPuvWWqs/6ta9ascbt1\n2223ufuO7IGy99xzj1x99dVj2F0WQQABBBBAAAEEEEAAAQSOLRD3LnaWkc6SM/z4xz92D3+1\nJrjLLrtMPv3pT7taW3Pl9ddfL9ddd51YkGT9O6+66iqxwIqCAAIIIIAAAggggAACCMRSIO4B\nku3M2972Nnnzm98s9fX17p4j794Lb0dXrFghd999t9TV1UlFRUUwZaA3nVcEEEAAAQQQQAAB\nBBBAIBYCCREg2Y5YlgtrTRquVFVVDTeZaQgggAACCCCAAAIIIIDAuATifg/SuGrPwggggAAC\nCCCAAAIIIIBADAUIkGKIyaoQQAABBBBAAAEEEEAguQUIkJL7+FF7BBBAAAEEEEAAAQQQiKEA\nAVIMMVkVAggggAACCCCAAAIIJLcAAVJyHz9qjwACCCCAAAIIIIAAAjEUIECKISarQgABBBBA\nAAEEEEAAgeQWIEBK7uNH7RFAAAEEEEAAAQQQQCCGAgRIMcRkVQgggAACCCCAAAIIIJDcAgRI\nyX38qD0CCCCAAAIIIIAAAgjEUIAAKYaYrAoBBBBAAAEEEEAAAQSSW4AAKbmPH7VHAAEEEEAA\nAQQQQACBGAoQIMUQk1UhgAACCCCAAAIIIIBAcgsQICX38aP2CCCAAAIIIIAAAgggEEMBAqQY\nYrIqBBBAAAEEEEAAAQQQSG4BAqTkPn7UHgEEEEAAAQQQQAABBGIoQIAUQ0xWhQACCCCAAAII\nIIAAAsktQICU3MeP2iOAAAIIIIAAAggggEAMBQiQYojJqhBAAAEEEEAAAQQQQCC5BQiQkvv4\nUXsEEEAAAQQQQAABBBCIoQABUgwxWRUCCCCAAAIIIIAAAggktwABUnIfP2qPAAIIIIAAAggg\ngAACMRQgQIohJqtCAAEEEEAAAQQQQACB5BYgQEru40ftEUAAAQQQQAABBBBAIIYCBEgxxGRV\nCCCAAAIIIIAAAgggkNwCBEjJffyoPQIIIIAAAggggAACCMRQgAAphpisCgEEEEAAAQQQQAAB\nBJJbgAApuY8ftUcAAQQQQAABBBBAAIEYChAgxRCTVSGAAAIIIIAAAggggEByCxAgJffxo/YI\nIIAAAggggAACCCAQQwECpBhisioEEEAAAQQQQAABBBBIbgECpOQ+ftQeAQQQQAABBBBAAAEE\nYihAgBRDTFaFAAIIIIAAAggggAACyS1AgJTcx4/aI4AAAggggAACCCCAQAwFCJBiiMmqEEAA\nAQQQQAABBBBAILkFCJCS+/hRewQQQAABBBBAAAEEEIihAAFSDDFZFQIIIIAAAggggAACCCS3\nAAFSch8/ao8AAggggAACCCCAAAIxFCBAiiEmq0IAAQQQQAABBBBAAIHkFiBASu7jR+0RQAAB\nBBBAAAEEEEAghgIESDHEZFUIIIAAAggggAACCCCQ3AIESMl9/Kg9AggggAACCCCAAAIIxFCA\nACmGmKwKAQQQQAABBBBAAAEEkluAACm5jx+1RwABBBBAAAEEEEAAgRgKECDFEJNVIYAAAggg\ngAACCCCAQHILECAl9/Gj9ggggAACCCCAAAIIIBBDAQKkGGKyKgQQQAABBBBAAAEEEEhuAQKk\n5D5+1B4BBBBAAAEEEEAAAQRiKECAFENMVoUAAggggAACCCCAAALJLUCAlNzHj9ojgAACCCCA\nAAIIIIBADAUIkGKIyaoQQAABBBBAAAEEEEAguQUIkJL7+FF7BBBAAAEEEEAAAQQQiKEAAVIM\nMVkVAggggAACCCCAAAIIJLcAAVJyHz9qjwACCCCAAAIIIIAAAjEUIECKISarQgABBBBAAAEE\nEEAAgeQWIEBK7uNH7RFAAAEEEEAAAQQQQCCGAgRIMcRkVQgggAACCCCAAAIIIJDcAgRIyX38\nqD0CCCCAAAIIIIAAAgjEUIAAKYaYrAoBBBBAAAEEEEAAAQSSW4AAKbmPH7VHAAEEEEAAAQQQ\nQACBGAoQIMUQk1UhgAACCCCAAAIIIIBAcgtkJHf1j1/7jIwMyczMPP6MEzyH1cNKdna2eO8n\neJNJs3o7PmaCS/gh887b3Nxc6evrC5+Y4p/sXMnLy8Ml4jxIT093Y/Lz8yOm8DEtLU0KCgqA\niBDw+XxuDDYRMPqRc2aoiTfGftdwznga/a92vtjfJlyGutgYu6ZJBJuRXk9N+QApEAhIT09P\n+NGKwyf7wbFiByYR6hMHgmNu0n6h9Pb24hIh5AWMZmMDZVDAfq5xGfTw3pmLFX7HeCKDr4ny\nt2CwRon1jnNm6PHgnBlq4o3BxpMYfLWgEZdBD++d98Vdsl3/TvkAyS6iOjs7veMU11drCejq\n6pKjR4/GtR6JtnELBOyPMy7hR8YLkOz87e7uDp+Y4p+sJbajo4NAIOI8yMnJcWPMxv5QUwYF\nrFWN3zGDHt477xtdbDyRwVfOmUEL753X4mgXu5wznkr/q9/vFxtwGepiv2cS5TrPC9jCazn0\n05QPkIbuMmMQiK1AIGCtX0ekp/eI9PYc1dej2rqhn917HRfls81v44OvwWUHxrvlD4vPlyZV\nledLTdUlOlysTdRFsa08a0MAAQQQQAABBBAIEyBACuPgw1QUsG/S+wMWC2IsKDkcDGB6+zpC\nApX+oMYFLi5A8YKb/kDGBUC2vAU8LiCy8e36TX3sur+lpWVJRnqOpGfkSl7uNFe3vfvuERt8\nvnQpLztLplVfItNqLtXpM6bi4WKfEEAAAQQQQACBuAoQIMWVn42bQCBg92Ud0a6QzWGtMP0t\nMhakaFeqYMBiQY4GNQMtNv1BS8hnbbXxAhybp1uDoUAgdt3TfD6/3oSZ64KYrMxSF6Ska0CT\nkZ6rQc3Aq33WebzxGTo+Xad7gY97tc8h4906M/LCTghrkrYuHtu2Pyy79/xJ9h24Txoa17vh\nhZeuk+KiJQPB0mX6fnHYsnxAAAEEEEAAAQQQGJsAAdLY3FJuqf4gZbDlxOse5rWq9Hcts+BF\nA5qwbmZelzFv2YHPwe5n/QFQrECtlcUFIxp8+LU7Wk5OTUjwMhCoBIOX/pYaC276gxUvkBn8\nHBr4ZGgAY13eJruUFC+R/Lz5snjhp+TIkX0aKP1F9tfeJ41NG6S1bZO8vOX7kpszQ1uVrGXp\nMqkoO9O1Nk12PdkeAggggAACCCAwFQQIkKbCUdR96Ovr0lYY6/7lBSBeYKKvxxhv87uuZyFd\nxrx7Z7xWmP57ZI7EUMnSYPYHIv6MfMnOqpCsrALX2uKTTNe1rD9gsRYYTYluLTMugAn5PNCC\n41po3HudpuvMSM/TtKxT+5TOzZ0u8076gBu6utrkQN1fZf+B+6W2/iHZ/urNbvD7i6Sm+mKZ\nXn2Z3r90gWvNiuEBZFUIIIAAAggggMCUFpjaV5MJdugsIDlytFYOt7f0dxkbQWtLfxAzGMhE\nu6nfxmmIFKO99blgpb8bWI5k5ZaEBCgWhAy2uvR3IfO6k3ldyGweC1gGPg90N+sPdLJd0BNZ\n0UTKbhJZt0T+bAkbZs18ixusha+uYZ0Llixo2r3nt26we5qqKs5zLUsWNGVnlSfyLlE3BBBA\nAAEEEEAg7gIESJN4CJ5+7j/kqWe+Nu4tuhv5B7qJ5firgy0yg8GL11XMC2a8IMYLXgZe3b0z\n3jzWQmPz9acJHnclWcGkClhr27TqNW6we7qamp/Rbnj3a8B0n7YyPeAGEZ+UlZ7mEjxY61J+\n/omTWkc2hgACCCCAAAIIJIMAAdIkHqWy0iUyc8YVepmqmcpcNzMvYAlvbfFu7h/sQuYFMf2B\nj/ccgkmsOptKIgG7T6q87Aw3nLzkWjl4aJsLlOy+pabmp93w4qavS2HBPA2oLnWtSyXFJ+t9\nS74k2kuqigACCCCAAAIITIwAAdLEuEZd67yT3qo30K/hIWJRdRg5UQIWCNmwcP41crSjznXD\n26+JHuobH5fN225wQ3Z2ZTBYqiw/W+/lypyo6rBeBBBAAAEEEEAgoQUIkBL68FA5BGIrkJNd\nJSfNfo8burvbNbnD2oFueGvl1Z23uiFDk2dUV10o0zUjXnXlan0yeEFsK8HaEEAAAQQQQACB\nBBYgQErgg0PVEJhIAb8/X2ZOv9INfX3drkXJ7lmyrnh79/3BDfbcp8ryc1w3PHtAbU5O1URW\niXUjgAACCCCAAAJxFyBAivshoAIIxF8gLc2vrUUXuGHFyddLS+tGFyhZV7y6hkfc8NzGf5WS\n4uUDSR4ulcLC+fGvODVAAAEEEEAAAQRiLECAFGNQVodAsgtYsobSkuVuWLros9J+eJd2w/uL\ne0CtZcdraX1BNr3ybcnLm+WetTSt5lKXHS8eD9FNdmvqjwACCCCAAAKJJ0CAlHjHhBohkFAC\n+RoIzZ/7927o7GzSliV7OK21LK2TrTtudENWZpk+nFbTjGuwZM9dsrTjFAQQQAABBBBAIBkF\nCJCS8ahRZwTiJJCVVSazZ73dDfYQ47r6h11XvAO1D8jO3Xe4wdLTV1eucsFSTdXFkplZHKfa\nslkEEEAAAQQQQGD0AgRIozdjCQQQUAF7ltf0aZe7IRDolYamDS4jnj2gdt+Be93g86Xr85jO\nHEghfqnk5c7ADgEEEEAAAQQQSGgBAqSEPjxUDoHkELBAyLLd2XDKsi9La9vLA8HSfdKgz1uy\n4YWXrpOiwsX9SR60K15x0ZLk2DlqiQACCCCAAAIpJUCAlFKHm51FYHIEiosWawC0WBYv/Cc5\ncmSf64a3T1OINzY9KW0HX5ZXtnxfcnOmu2BpWvWlrpUpLY1fR5NzdNgKAggggAACCAwnwBXJ\ncDpMQwCBcQvk5k6XuXPe74aurjY5UPeAa12q1fuXtr96sxv8/iKx+5UsyYPdv5SRkTvu7bIC\nBBBAAAEEEEBgLAIESGNRYxkEEBiTQGZmkcyaeZUbens7pV4z4dk9Szbs3vtbN6SlZblMeBYs\nWWa87KzyMW2LhRBAAAEEEEAAgbEIECCNRY1lEEBg3ALp6VkaAF3shlMDX5emlme1ZUmDJU0h\nbq1MNoj43DOWLFiyrngF+bPHvV1WgAACCCCAAAIIDCdAgDScDtMQQGBSBOwhs+Wlp7vh5CVf\nkIOHtvUHS7V/kabmp93w4qavS0HBPA2ULpGF89+qLUvzJ6VubAQBBBBAAAEEUkuAACm1jjd7\ni0BSCBRqIGTDwvn/IEc76uSAPpx2n7Ys1Tesly3b/tMN2dlVGizZw2kvc9nz0tIyk2LfqCQC\nCCCAAAIIJLYAAVJiHx9qh0DKC+RoIDTnxHe7obu7XWrrH9JseGtl956/yKs7b3NDRka+JndY\nLdM1WKquWi1+f2HKuwGAAAIIIIAAAmMTIEAamxtLIYBAHAT8/nyZOf0NcvLSq6WtrVlblizJ\nw30uK97e/X8UG3y+DNeiZC1L1h0vJ6c6DjVlkwgggAACCCCQrAIESMl65Kg3AikukJbml6rK\n892w4uTrpbll40Cw9Bepa3jUDc9t/KKUFJ/suuFZkoeiQu5bSvHTht1HAAEEEEDguAIESMcl\nYgYEEEgGgdKSk8WGpYs+I+2Hd7lWJcuI19j8jLS0bpRNr3xb8vJmuWx41hWvrPQ0bW1KS4Zd\no44IIIAAAgggMIkCBEiTiM2mEEBgcgTyNRCaP/fDbujstK54muRBM+LV1T8q23b81A2ZmaXB\nJA9VFedJenr25FSOrSCAAAIIIIBAQgsQICX04aFyCCAwXoGsrFI5cdbfuqGn56h2vXuk/1lL\ntQ/Izt2/coMFR1WVq2R6tSZ5qL5IsjJLxrtZlkcAAQQQQACBJBUgQErSA0e1EUBg9AIZGTku\n0511sQsEejUb3lPuvqV9ByzRw1/cIJImFWVn9t+3VHOJ5OXOHP2GWAIBBBBAAAEEklaAAClp\nDx0VRwCB8Qj4fOlSUX62G5Yv/ZK0tr0SzIjX0PSE2PDCS9dpYodFA0keLtGED0vHs0mWRQAB\nBBBAAIEkECBASoKDRBURQGDiBYqLFokNixd8Uo4c3T+Q5OE+Fyi1HXxFXtnyfU0ZPm0gycOl\nUl52lqSl8St04o8MW0AAAQQQQGByBfjrPrnebA0BBJJAIFcDoblz3ueGru42qa19UJM83Ce1\ndQ/JjtducYPfXyQ1VRdp69Kl+pDaVZKRkZcEe0YVEUAAAQQQQOB4AgRIxxNiOgIIpLRApgZC\nJ8x8sxt6ezulvvGx/tYlzYy3e+9dbkhLy5LKinP1/qZLNWhaI9nZFSltxs4jgAACCCCQzAIE\nSMl89Kg7AghMqkB6epZrNbKWo1MDAX047bMaLN3vUojX1j2oLUwPan18UlZyqmtZstalgvw5\nk1pHNoYAAggggAAC4xMgQBqfH0sjgECKCvh8Ggjpw2ZtWLbk83Lw0HZN8nC/y4TX1PKM2PDi\ny9/QAGluf7BUfak+yPYUfTitL0XF2G0EEEAAAQSSQ4AAKTmOE7VEAIEEFygsmCs2LJz3Meno\nqNdg6a8uWKpvXC9btv2nG7KzKqWmeo1LNV5ZsVKTPGQm+F5RPQQQQAABBFJPgAAp9Y45e4wA\nAhMskJ1dKXNOfJcburvbpa7+Ydmnz1mqrVsrr+36pRssqUN15WqXQrym6kLx+wsnuFasHgEE\nEEAAAQRGIkCANBIl5kEAAQTGKOD358uM6Ve4oa+vWxoan9TWJXso7f2yd/+f3ODzZbjnMdkD\nbGuqL5HcnJoxbo3FEEAAAQQQQGC8AgRI4xVkeQQQQGCEAmlpfqmqPM8NK06+XlpaX9RA6T7X\nulTfsE5seG7jv+kDaZfp85Yuc/cuFRUuGOHamQ0BBBBAAAEEYiFAgBQLRdaBAAIIjEHAAiEb\nliz6Z2k/vNsFS9a61Nj0tAueNm3+juTlnuC64VlGvPLS0zXJQ9oYtsQiCCCAAAIIIDBSAQKk\nkUoxHwIIIDCBAvl5J8j8uR9yQ2dnsxyoe8AleahreFS27fipGzIzS7UL3sUyXVuXqirPl/T0\n7AmsEatGAAEEEEAgNQUIkFLzuLPXCCCQwAJZWaVy4glvc0Nvb4fU1j/igiULmnbt/rUbLDiq\nqrigP8mDBk1ZmSUJvEdUDQEEEEAAgeQRIEBKnmNFTRFAIAUFLBCart3rbAgEel33Oy/Jw/7a\n+zThw32qkiblZWfIvJPeLPPmvlk/56SgFLuMAAIIIIBAbAQIkGLjyFoQQACBCRfw+dI1291Z\nbli+9EvSdnDzQJKH+zRwetINj2/4vOTnzdIueKt0vpVSUXaWWIsUBQEEEEAAAQRGJkCANDIn\n5kIAAQQSTqCocKHYsGjBP8qRo/ulte0x2bP3Hm1Velx2vPZzN1ili4uWuaDKgqby0tPEnsFE\nQQABBBBAAIHoAgRI0V0YiwACCCSVQG7ONH3W0kfktBX/KPv379HnLW2Q+sbHNHX4Yy4jXmvb\ni5ro4SbNgpcuZZoNr7LiXKnUFqbSkuWSlpaZVPtKZRFAAAEEEJhIAQKkidRl3QgggEAcBOx5\nS5UVK90giz4rPT1HNFha756zVN+wPtgd72X5rsuEV1F2Tn/ApEGTtUiRSjwOB41NIoAAAggk\njAABUsIcCiqCAAIITIxARkauPnh2jRtsC51dLVJf/6jUaeuStTLV1q91g03z+4s0O955GjDp\noC1M+fkn2mgKAggggAACKSNAgJQyh5odRQABBPoFLCX4zBlvdIONsfuX6uofdgGTJXvYu/9P\nbrBp2dlVmvDhAqkqP1fKy8+W3JwaG01BAAEEEEBgygoQIE3ZQ8uOIYAAAiMTsPuXZs96pxts\nibaDW9y9S/UN67Q73lPBZy/ZtIKCedqypPcvaXe8irIzJTOz2EZTEEAAAQQQmDICBEhT5lCy\nIwgggEBsBIoKF+i9SAv0uUof0Gcv9Ulzy/P9AZN2x2tqflZ2HNqmGfJu0Y35pKR4Wf/9S+Xn\nafIHy5DHM5hicxRYCwIIIIBAvAQIkOIlz3YRQACBJBCwhA1lpae6YdGCT0hvb6c0Nj810ML0\nmKYWf0mz5G2ULdt+oskd/PrAWs2QN9DCZMGTJYygIIAAAgggkEwCBEjJdLSoKwIIIBBngfT0\nLJfEwRI5WOnubh/IkNef8KGh8XFNMf64bNr8Hc2Ql6vB0mCGvMKCBRpE+eK8B2weAQQQQACB\n4QUIkIb3YSoCCCCAwDACfn++TK+51A02W2dnkyZ7eDTYwnSg7gGxwUqmJoew7HhV7hlM50pe\n3gluPP8hgAACCCCQSAIESIl0NKgLAgggkOQCWVllcsKMN7nBduXwkT2aIc8CpnXSYBny9v3B\nDTYtR5NDVFWc35/wQTPk5WjGPAoCCCCAAALxFiBAivcRYPsIIIDAFBbIy50pc058lxtsN1vb\nXgkmfGhs2iA7d9/hBptWWDC/P+GDtjKVl50hmfpMJgoCCCCAAAKTLUCANNnibA8BBBBIYYHi\nokViw/y5H9IMeb3S1PJcsDueZcs7eGirbH/1ZhXySWnJ8mDCB8uQl56encJy7DoCCCCAwGQJ\nECBNljTbQQABBBAIE/D50qW89HQ3LF7wSc2Q1+Geu1TfYAkf1ml68Y0uxfjmbTdoNrxMne+M\ngRamczXIWqrj+BMWBsoHBBBAAIGYCPDXJSaMrAQBBBBAYLwC1kJUVXm+G2xd3d0HtXVpvQZL\nGjC5oKk/U568Ivq8pTypsAx5LqX4edo9bx4Z8sZ7AFgeAQQQQMAJECBxIiCAAAIIJKSA318o\n06dd7garYEdHvWbIWxfskneg9q9ig5WszDKXIa/SMuTpkJc7w43nPwQQQAABBEYrkHAB0tNP\nPy2tra2yZs2asH3ZvXu3rF+/XkpLS2XlypWSn58fNp0PCCCAAAJTWyA7u1JmzbzKDban7Yd3\n9WfI0xamhsYnZM++37nBpuXmzHAtURYsVZSdrWMqbDQFAQQQQACB4wokVIBUV1cnX/ziF2X5\n8uVhAdIvfvELuemmm2TVqlWyf/9+sc8//OEPpaSk5Lg7yAwIIIAAAlNTID9vluTPniUnzX6P\nJnwIaIa8TRooaZc87Y7X2PyUvLbrdjfY3peWLpWykrNcl7zysjPF7y+YmijsFQIIIIDAuAUS\nJkDq6+uT66+/fkgfcms5uvnmm+UHP/iBnHLKKdLT0yMf/ehH5Y477nCv4xZgBQgggAACSS/g\n8/mkpHipG+bP/Xvp6+vRBA+WIa+/S15z6/PS3PySbNvxP7qvaRosrZCKipWuO15ZyamaIS8r\n6Q3YAQQQQACB2AgkTIB0++23u+DooosuksbGxuDebdiwQaZNm+aCIxuZkZEhl19+udj8FihR\nEEAAAQQQiBSwDHf2LCUbFi/8lPY4yJdXNv9REz70tzA1tTyrKcafkc1bf6TZ8LLcfJX6/CVL\n+lBSvET/HqVHrpLPCCCAAAIpIpAQAdKWLVtcwGPd6G699dYw+gMHDsj06dPDxlnAZEGUtTql\npaUFp9m89957b/CzvTnrrLNkxoz436zr9/tdvbKyssLqHFbZFP1gNunp6bhEHH/vnMnJyZHM\nzMyIqan90c4Xc7HfAZRBAXOxkpubOziSd07A78+Vk+Zc4QYb0dnVKrW1j8qBukc00cMjAy1N\n6wbmLZCaqvOluuoCqam+QFOKz3fjp+J/1vJmJS8vbyru3rj2yWxwiU5ov2uwCbex61Fcwk3s\nk/d3ya5pEuGcse7YIylxD5A6Oztd17prrrlGqqurh9S5trZWCgsLw8YXFBS4C6O2traw+5Be\ne+01+eY3vxk27/e//31ZvHhx2Lh4frCLOhsoQwVwGWpiYxLhF0r0msV3rBdAxrcWibn1yN+Z\niVnLya9VuEuhpgl/tyxb+m5XkfbD+2XP3r/K3n0PuWH33nvEBiu5OVUyc8bFMmP6hTqs1pTi\ns9z4qfRfuM1U2rPx7Qsu0f2sNw820W342xTdxRoIbIh36erqGlEV4h4g3XDDDTJr1ix53ete\nF7XCdqLZfUehxfsc+S3pggUL5Hvf+17orDJv3jxpaWkJGxePD9YCYBe6hw8flpEenHjUMx7b\ntMCot7cXlwh8c8nOzpaDBw86n4jJKf3RfpY6OjpwiTgLLLun/c5MhN95EVWL+0e7mLOfpWOX\nHKmuvNINp68QaTu4XWrrHtXhEU0t/rhs2fZLN9jyBfmzXetSddV5minvXMnOKjv2ahN8SlFR\nkauhfeFICRc4/jkTPn8qfLJWteLiYn1GWbe0t7enwi6PeB+tpcT+Ztt1HmVQwFzsZ8n+Zh89\nenRwQhzfjaRXTlwDJMtad9ddd8myZcvkc5/7nKPasWOHu1C2z1/4whekvLxcdu7cGcZof+Qs\ng11kJFpWViavf/3rw+Ztbm52ByVsZBw+eN0Y7JeKnSSUQQEvCMZl0MTeed9CWUBt5w1lUMCC\nR2t99r4sGZyS2u+8L43MZqTdCFJFzHoejOZ3TFbmDE0n/k439GfIe2mgG956lyFv246facKH\nnzm+osLFLtlDf0rxM91DbJPF1WsFGI1NsuzbeOs52nNmvNtLhuW9axnr3sw5E37E7G+2XXjj\nMtTFxtjf60Sw8br8hddy6Ke4Bkh2kfOhD30orFYW0Fj0bd3i7GSbPXu2u6/IYK1J18qmTZuG\n3JcUthI+IIAAAgggECMBuygsKV7mhgXzPqZdvLulqflZTfjwmEsp3tzyvLY4vawB009dcodS\nzZDXn/BhpZSVrtD7K7mHMEaHgtUggAACkyIQ1wDJvrl673vfG7ajDQ0NYoM33h4Y+5Of/ERu\nu+02ufrqq11r0j333CPXXntt2HJ8QAABBBBAYDIE0tJQpGbGAABAAElEQVT8ev/SWW5YsvDT\n+s3oEWloesIFS/UN6zV4etoNr2z5vt6gnC3lpWcGW5iKiyxD3mByocmoL9tAAAEEEBidQFwD\npJFU1brR2fORrrvuOhckWavTVVddJStXrhzJ4syDAAIIIIDAhApkZORq1ruL3GAb6uxq0S/6\n1uu9S/oMJm1lqmuw+5gecXXw+4s0lXj/85fstaDgpAmtGytHAAEEEBi9QMIFSJ/5zGeG7MWK\nFSvk7rvvFrtnqaKignTQQ4QYgQACCCCQKAJZmSWa7e4KN1idjhzdL/X166ROg6UGfQ7TvgN/\ndoNNy86ucq1LVeXnuQfX5ubU2GgKAggggEAcBRIuQBrOoqqqarjJTEMAAQQQQCDhBHJzpsmJ\ns/7WDVa5g4e2DXTH04Cp6UnZvee3brBp+flz3MNqLeFDZfk5etN3sY2mIIAAAghMokBSBUiT\n6MKmEEAAAQQQmBCBwoJ5+iyleTJ3zvs022CftLS+2B8waQtTY9NT8urOX7hBRFMq6z1LLljS\ngKm89AxNVsRDgCfkoLBSBBBAIESAACkEg7cIIIAAAghMpoAlbCgtWe6GhfP/QTPkdWka8WeC\nLUwtrS9Ia9tLsnX7f2tyhwzNindqsIWptOQU7XLun8zqsi0EEEAgJQQIkFLiMLOTCCCAAALJ\nIGApwa1rnQ2y6DP6DLT2kAx51sK0wQ0vb/meZsjLkYqys4ItTPY8Ju85Ncmwr9QRAQQQSFQB\nAqREPTLUCwEEEEAg5QX8/nyZVr3GDYbR2dmkmfHWB1uYausfEhusZPqLg8FSZfm5ej/TiW48\n/yGAAAIIjE6AAGl0XsyNAAIIIIBA3ASysspk5vQr3WCVOHxkrwZLmk68QR9a2/i47N3/JzfY\ntJzsmrCAKSeHREfmQkEAAQSOJ0CAdDwhpiOAAAIIIJCgAnm5M2T2rHe4warYdnBrf8DkUoo/\nKbv23OkGm1aQPzcYMFWUn60tTkU2moIAAgggECFAgBQBwkcEEEAAAQSSVaCocL7YMO+kD2iG\nvF5pbtnoHlZrD65tbH5adrz2MzdYhryS4qWaevxSKStZos9qCkh6WpakpWe513R9TdPP6XpP\nVPi4zGSlod4IIIDAiAUIkEZMxYwIIIAAAggkj4DPl65Z71a4YdH8j0tvb4c0eRnytIXJgidL\nMT7a4vP5XbpxF1BpABUMpgaCKy+g6n/VAMsCLTctOyLYCg/AvOVC19cfpPUv763HMv9REEAA\ngYkUIECaSF3WjQACCCCAQIIIpKdnB7vYWZW6uw9KV88WaWvbrl3zmqSvt1N6Nc14X5++uvcd\nA+M6B191ev80b1yHdHW3uWUsRfnkFJ8L0KK3cFkwlR1sDbOsgJYK3YI5y/rXH6gNtpTZ9Mjg\nzQvU0tKr5VD74eC6+tdhLWu0ok3OcWYrCMRPgAApfvZsGQEEEEAAgbgJ+P2FMn36FW779fX1\nMalHX1+PBlBHIoKo/oDLBV4WgLngywuwbJoGYgNBmQVZvcH3Ict5ywwEaH19HcFtdHe1ScfA\nMiJ9MdmP462kvzVLW8QswBpoIfNauFzLmteaNhCADU6LbEXrX34wKLNWtZAWM10+bSDgC21Z\noxXteEeI6QiMT4AAaXx+LI0AAggggAACAwJpaRkaNBSKP07Prw0EAho4HXVBlheIWYtXfwBm\ngdhIA7BOycz0Sfvh1oiALrSFzQvyQlvRulUiMOHng3WftJYyL0DzAqzBIGqgFc0FWBFBlxe8\nudfoXSC99Q22sPW3umXoNnt7Syd8/9gAAvEWIECK9xFg+wgggAACCCAQEwF7UG5GRq5kSO64\n11dRUSENDQ2jXo8LwrRVLLwrotca1h+gecGb9xpsRQtpYRsM7LwAz15DA7zBbXitaDbdknNM\ndCkqnCdVFRdKddVqvcftNBesTfQ2WT8CkylAgDSZ2mwLAQQQQAABBKa0gLW62BC/VrTeYwRn\nGqRpV8TBoCw08AoP4EKDs17tzugtYwGYz9clDY3P6X1rN8rWHTfqvmZJVeX5Ul15oQZN5/GA\n4il9dqfOzhEgpc6xZk8RQAABBBBAYIoLWPe7WLWiRVJZC111dbUc1q6Hm7f8XmrrH5Lauofk\nQO1f3WDz5+XNkmnVazRoWi0VZWfSuhSJyOekECBASorDRCURQAABBBBAAIHEEMjIyJFpNZe4\nwWrUfniXC5Bq69bq87aekm07/scN1rpUWXGuti6t0u54F0q+Bk8UBJJBgAApGY4SdUQAAQQQ\nQAABBBJUwAKfeSd90A3WPa9en7NlLUv9LUwP6vsHRV78suTmzpSaqovdvUsVZWe5lq4E3SWq\nleICBEgpfgKw+wgggAACCCCAQKwELJNeTdVFbrB1Hj6yR1uXHnABU0PTk7LjtVvcYA8criw/\nx7UsWQtTQcFJsaoC60Fg3AIESOMmZAUIIIAAAggggAAC0QTytNVo7pz3ucGSPNQ3PqHB0lrX\nulTX8IjY8IIumJNdIzXV1rp0od67dLYmuciPtjrGITApAgRIk8LMRhBAAAEEEEAAgdQWsOx+\n1ZUXuEHky3Lk6P7+1iVN9tCggdOrO291g8+XoUHSWa4rXrUmeygsnJ/acOz9pAsQIE06ORtE\nAAEEEEAAAQQQyM2ZJifNvtoNfX3dYl3w6uoe1talte4+JruXaeOmr0l2dqV22Vvjkj1UVqzU\n1qVC8BCYUAECpAnlZeUIIIAAAggggAACxxNIS/O75yjZs5ROln+Vo0fr5ECd3rtkrUsNj8tr\nu37pBktjXl56hlTpQ2qtdamocKE+m8l3vNUzHYFRCRAgjYqLmRFAAAEEEEAAAQQmWiAnp0rm\nnPguNwQCvdLY9FQwM15D0xPa2vSEvPTyNyUrq9wlhLBgyVKKZ2YWT3TVWH8KCBAgpcBBZhcR\nQAABBBBAAIFkFbBWo4rys92wbMnnpaOjQVuXHpS6+oc1ycM62bn7V24Q8UlZ6WnBzHjFRUtp\nXUrWgx7nehMgxfkAsHkEEEAAAQQQQACBkQtkZ1fI7Flvd0Mg0CdNzc8MPHPpIfe+qflp2fTK\nt7U1qVRbly50XfEqteteVlbpyDfCnCktQICU0oefnUcAAQQQQAABBJJXwOdLk/KyM9ywdNFn\npbOz2aURP2CJHhoek117fuMGa10qLTnFBUuWSrykeJm2LqUl745T8wkVIECaUF5WjgACCCCA\nAAIIIDBZAtZKNOuEt7ghEAhIc8tz2rqkmfH02UvNLc+7zy9v+Z5mwityXfFq3L1L52mmvIrJ\nqiLbSQIBAqQkOEhUEQEEEEAAAQQQQGB0Apbdrqz0VDcsWfgp6epqDXbFq2t4VPbsvdsNttaS\n4pMHWpdWu5Ymu++JkroCBEipe+zZcwQQQAABBBBAIGUELMPdCTPe5AZrXWppfVEDprXu2UvN\nrc/r543yytYfij+jUKrsgbbaFa+q4nyxjHqU1BIgQEqt483eIoAAAggggAACKS9grUulJSe7\nYfGCT0p398GBrngPaWa8R2Tv/j+6waCKi5a41iV79lJZyQqxZzZRprYAAdLUPr7sHQIIIIAA\nAggggMBxBPz+Qpk5/Uo32KwtrS+5NOJ271KT3sfU2rZJNm+7QTIy8vtblypXudfcnGnHWTOT\nk1GAACkZjxp1RgABBBBAAAEEEJgwgZLipXpf0lJZOP8abV1qd61KtXXaulT/iOzbf48bbONF\nhQs1UFotM6at0a54qyesPqx4cgUIkCbXm60hgAACCCCAAAIIJJGA35+vAdDr3WDVbju42WXF\ns+x4jU1Pu89bt/+Xti7lSWX5uf33LlWeL3m5M5NoL6lqqAABUqgG7xFAAAEEEEAAAQQQGEbA\nWo1sWDDvY9LTc0Rbl9bpM5ce1uFR2V97nxts8YKCef2Z8bSFyZ7VlJ6eNcxamZRIAgRIiXQ0\nqAsCCCCAAAIIIIBA0ghkZOTK9JpL5cQTrpD8/HzZtXuDti495NKJNzY9Jdt2/NQN6enZA61L\nq13QlJd3QtLsYypWlAApFY86+4wAAggggAACCCAQc4FCbTWyYf7cD0tvb4drXapzD6p9SA7U\nPeAG22he3iypqbrIdcerKDtLW5eyY14XVjh2AQKksduxJAIIIIAAAggggAACUQUs6JlWvcYN\nNkN7+04NkB50rUsNjU/I9ldvdkNaWpa2Lp2jyR5WuYCpIH921PUxcvIECJAmz5otIYAAAggg\ngAACCKSoQH7+iTIv/wMy76QPuNYlC5Jq67U73kCXPHv/wkvXSa4md6jRh9RW671LFRo4WTc+\nyuQKECBNrjdbQwABBBBAAAEEEEhxAWtdqtYHz9ogy0QOH9kjB2ofdM9eqm9cLzte+7kbfD6/\nBklnabCkAVPVKtd9L8XpJmX3CZAmhZmNIIAAAggggAACCCAQXcBSgs+d81439PV1SUPjk8HW\npXqXJW+dbNx0vT5raZprXbJnL1WWrxRLQU6JvQABUuxNWSMCCCCAAAIIIIAAAmMSSEvL1PuR\nznfD8qX/JkeO7u9/7lLdWrHWpVd33uYGny9D04efqa1L/fcuFRUuGNP2WGioAAHSUBPGIIAA\nAggggAACCCCQEAK52mo058R3u6Gvr1sfTvtUsHWpQQMmG158+RuSnV010BVvtVRVnKutS4UJ\nUf9krAQBUjIeNeqMAAIIIIAAAgggkHICaWl+qaxY6YaTl1wrR4/WabC01iV6qG94THbu/j83\n+HzpUlZ6WvDepaLCxeLz+VLOa6w7TIA0VjmWQwABBBBAAAEEEEAgjgI5OVUye9Y73NDX1yNN\nLc8OdMd7SFuaNrjhpVe+JVlZFcGueFUV50tmZlEca534myZASvxjRA0RQAABBBBAAAEEEBhW\nIC0tQyr0niQbli3+nHR0NAS74tU1PCq79tzpBpE0KStZMZBF70IpLlpK61KELAFSBAgfEUAA\nAQQQQAABBBBIdoHs7Ao58YS3uSEQ6NXWpeekru5h1yXPWpqaWp6RTZv/n7YmlWrrkqUcX6X3\nLl2grU2lyb7r464/AdK4CVkBAggggAACCCCAAAKJK2D3JJWXnu6GJYv+WTo7mzRQ0mBJH1Jb\n1/CI7N77WzeI+KS0ZLlm0NOASQd77/OlJe6OTVDNCJAmCJbVIoAAAggggAACCCCQiAJZWWUy\na+ZVbggE+qSldWP/vUsaNDW3PO+GV7Z8XzL9xRosXeC641nrkrVKpUIhQEqFo8w+IoAAAggg\ngAACCCAQRcBaiEpLTnHD4oWfkq6u1v7WpXptXap/RPbs+70bbNHiomX99y7ps5fKSk/V1qX0\nKGtM/lEESMl/DNkDBBBAAAEEEEAAAQRiIpCZWSwnzPgbNwQCAWlte8l1xavVgKmp+Vn9/KJs\n3voj8WcUuofZWle8Kr1/KUefwzRVCgHSVDmS7AcCCCCAAAIIIIAAAjEUsGcnlRQvc8OiBZ+Q\nru42qa9fJwf02Ut12h1v7/4/ucE2WVS4aKB16UL3DCbLqpesJXlrnqzi1BsBBBBAAAEEEEAA\ngSQUyPQXyYzpV7jBqt/a9vJAsoe12rr0jLQdfEW2bPuJZGTka0a882T6tIv1vqW36JzFSbW3\nBEhJdbioLAIIIIAAAggggAACiSFQXLRY70taLAvnfUy6uw9JfcNjGjA95Lrk7Ttwr9iwd9/d\nctmaPyRGhUdYCwKkEUIxGwIIIIAAAggggAACySpw5OgBF8SE1j8jI0dyc6ZHTeXdfniXzuqT\n/LwTQhc55nu/v0BbjC53g83UdnCrNDQ+ovcnzR+yTE/PYTl8ZJ9uu0ZsuUQrBEiJdkSoDwII\nIIAAAggggAACMRZ4cdPXg9noQledlpalrUBL5PQV35bCgrnBSeuf/LCkp2fKxav+GBw3mjdF\nhfOlvGyJlJeXS3t7e9iijc1Py7rH/07OOv0GmTn9DWHTEuEDAVIiHAXqgAACCCCAAAIIIIDA\nJAgsX/old4+Qbaqvr0uspWjHaz+XtY9eJWtW/0nycmdOQi0SexMESIl9fKgdAggggAACCCCA\nAAIxE5g5/Y1DHvhaWrJcnnz647Jv/70yf+6HY7atZF0RAVKyHjnqjQACCCCAAAIIIIBADARK\nipe7tRztOHDctVkQ1dD0hHab26n3DxW6bnknzX6vZGYWhS3b3LJR53tUDh/eLCX6INqyklVh\nXfjCZtYPXV1tsmX7f0l6WqYs0KQP6enZkbNM2ue0SdsSG0IAAQQQQAABBBBAAIGEE9i15zeu\nTtVVFw1bt8c3fFQef+ojmq1uvUuuYKm9N23+f3L/2kulp+docNm9+++RBx95o+zcdaebb8vW\nm+W+By+Wnbt/HZwn9I09X+mR9e/Wrn4/k4qKc+MaHFm9aEEKPTq8RwABBBBAAAEEEEBgCgs8\n9+KXtJWmv3Wmp+eQ3oO0Ww61b5dTl3/DPbvoWLte17BO03b/WRbM/ZgsW/L54GwvvPTvsm3H\nT+VA3V814cKVmr1us2x45pMyreZSOf+cG6WyskYOHTokf7r3Cnnplf+QGdOuCC5rbyw9+KPr\nr9aWpl06/636kNlTw6bH4wMBUjzU2SYCCCCAAAIIIIAAAnEQaNDWH/H1dyLr7T0ivb0dWguf\nHDy0zbUCWervaKUgf46cfcZPNIi6IGyyBUIWIHV2NrvxdfWPuuQPyxZ/QdLS/G6cz+eTFSdf\nL5a9zlJ8e8UCtEcfv1qDtJ1ywbm/lJLiZd6kuL5O+QApIyNDM3XEfzf9/v4TJCsrS08WejaG\nnvVmk56ejksoir73zpmcnBzt15sZMTW1P9r5Yi59fX2pDRGx9+ZiJTc3N2IKH+33bl5eHhAR\nAnbRYgWbCBj9yDkz1MQbY79rOGc8jf5XO18S3cW7Hn7TG9br39DK4A50drZogHOrPPvCv+uz\niXbIpRf1d7ezfUpLGzzWeXnzpKJ8nnave0pbi57XlqJt0ta2VRqbnnXr8vv7f8+2H97qushV\nVy11JjbRrmlqala4wT53dL5qL7JRW5+6e9rd85AqKxbr9c7E/p4OBAJuu8f7L/6Rw/FqOM7p\nI4UY52aOu7hXD3v13h93oRSZwTPBJfoB93yiT03dsbgMPfbez5D3OnSO1B3D+TL8seecGerD\nOTPUxBuDjScx9DWRf5a80CDy+GVmFsuSRR/XoGedHKhdK0ePNmiWu/KBnRu8bu3qOigPPPwO\nffjrUxr45GjShZM1+cJS7Up3oQZX17vrW1u3PZA2IyMv+NlTCrXRtbrR6dpatXzZ5+Tp5/5N\nnnr2i7LyrB94s0/Ia2gdhtvAlA+Qent7tcmvcziDSZlm33bbt7pdXV164g3exDYpG0/wjdg3\nLj09PbhEHCdzsRbHjo4O7Z/bHTE1tT96LnbeUAYFsrP7+5Tb75iR/hEYXHpqv7Nvu48cOTK1\nd3IMe5efn++WwmYoHufMUBNrcSwqKnKt95wz4T7WQmItNIns0jvwN7P/b8TQ34dZmVX6t6NP\nW4S2SmlJrjvOPl9fcJ+e2/hVFxyddsq3ZNbMt2rrUn8Ysf/A/Q6jq6vTzZudNV2vvR/V+45a\nNNDKF/s9Y9cxLS179EG1f5DKivOC1+bLFl+r67pK1/u8bH/1l1JVeanUVF0YjhvDT3ZtNZJC\nX6+RKDEPAggggAACCCCAAAJTVKCjo94lWUhLy5LiosVR97KldaMGRVkuoPGCI5vxQN0Dbv5A\noNe9lpee5l737v+je/X+2/7qLfL8i1/WFqZ93ihdX/8tBPbw2qzMMnnm+c+JZbSLd5nyLUjx\nBmb7CCCAAAIIIIAAAggkisCmzd/V1q7+e1UtqOnsapHaurXaytMmy5d+ORi0RNa3uGiJNLc8\nJxs3fUNOOvE9LpDZvfdu8VKEd2vCBSuzTniLbNWkDc9t/LJYC5T4VsvW7b/XFqJbtGXqFKks\nP1fqGx8LW7118zvl5K/qw2qvkRde/Kqccer/C5s+2R8IkCZbnO0hgAACCCCAAAIIIBAngdd2\n/TK4ZZ8vQ7vBVUpJ0VKZM/tqTcH9uuC0yDdLF/2LdsHrlT17f6fBzv/qZJ9UV67WpA4PyCOP\n6b1JDY+LLNKxvnTNSHe73pd0rTz1zL9oyu/+hErTqi+RFcu/FuyaF7n+mdPfoOu2gOtOrcfr\npab64shZJu2zT/upe/dsTdpGJ3NDzc3NwX6Ok7ndyG3ZPUjFxcXS2trKvTYROAUFBdyDFGFi\nH83F+u02NjZyD1KET0lJiXumAvcghcOUlpa6+9Zqa2u5BymcRh88WKF/vBsixvKxsrI/k1V9\nfT0YEQKcMxEg+tHuQaqurnbXVXZ9RRkUsHuQ7G92S0vL4Mgp+s7uUzrU/qrk5c5w2eqG201f\nWo9msDuosxRLV2f87+yxe5C833vD1Tv+NR2udkxDAAEEEEAAAQQQQACBhBHw6TOUCgvmHjc4\nsgpnWLa70iV6f1FRwtR/JBUhQBqJEvMggAACCCCAAAIIIIBASggQIKXEYWYnEUAAAQQQQAAB\nBBBAYCQCBEgjUWIeBBBAAAEEEEAAAQQQSAkBAqSUOMzsJAIIIIAAAggggAACCIxEgABpJErM\ngwACCCCAAAIIIIAAAikhQICUEoeZnUQAAQQQQAABBBBAAIGRCBAgjUSJeRBAAAEEEEAAAQQQ\nQCAlBAiQUuIws5MIIIAAAggggAACCCAwEgECpJEoMQ8CCCCAAAIIIIAAAgikhAABUkocZnYS\nAQQQQAABBBBAAAEERiJAgDQSJeZBAAEEEEAAAQQQQACBlBAgQEqJw8xOIoAAAggggAACCCCA\nwEgECJBGosQ8CCCAAAIIIIAAAgggkBICBEgpcZjZSQQQQAABBBBAAAEEJl+gL9A3+Rsd5xYz\nxrk8iyOAAAIIIIAAAggggECCCXT39cruzmYpzMiWCn/BhNVuX2eL9EpATsgqDW7DtvtQ6xZ5\n6OBWObVsjnxh1uuD05LhDQFSMhwl6ogAAggggAACCCCAwCgE9ne1yts2/7e8texUufaE2AQo\nnX09cnPdY/KW8lODQde1O++WQ70d8pVZV8qjbdvkr62vyC4NkKz49N/M/IpR1DoxZiVASozj\nQC0QQAABBBBAAAEEEEhogV/UPyE31j4qf1N2ilgL1bOHd0t3oFdquw7K+7be4uru96XLOQVz\nZHXxArmsfKksqDlR2tvbE3q/IitHgBQpwmcEEEAAAQQQQAABBBAYItDZ1+3GfWfvX2TDoZ1y\nuK/Lfc5Pz5JLChbJhcUL5YLCeZKbnunG+/3+IetIhhEESMlwlKgjAggggAACCCCAAAITKPBg\n62Z5pn2Xu2+pID1bZmeXy9vLT5cubSFad3C73FH/lGzpqHM1WNu2VSwouqpshaRLmlRnFcr7\nq851025veEoKdXlrRbq7/gXZurNBStJyZFXeXDm7cE7YHhzt7ZJfNT4jLx/ZLz2azGF+TpW8\nrfw0KfXnhc032R8IkCZbnO0hgAACCCCAAAIIIJBAAp999U55oG2zzM2ukJNyKuXZ9t1yb8sm\nual2netC51U1w5fmApkFOdVydsFs+eT0i+W9W26Wp9t3BgOke5pf1JDJJzceeETaeo/KqSWz\n5Z6mjXJH7Qa57oQr5cqy5W51ddot771bb5aWniNuXZlpGXJb/ZNyhwZY35nzNjk1/wRvs5P+\nSoA06eRsEAEEEEAAAQQQQACBxBB48tBrLji6svRkqcoslAe0Jam++5CrnN1fND2zWN6k9xyt\nKV4k97W8LD+pfVi+qwFMTWbRMXfgxSP75H1VK+UTMy+WaZXVsrVxj1zxzPfk1oYngwHSV3b/\nQdp6jsrN898ni3Nr3LosscQHt/5cvrLrD/LrRR+RLA2a4lF4DlI81NkmAggggAACCCCAAAJx\nFLDg5ym9j8hafKxL3B+aN7oWoz2age4sbR16d8WZrnbvqTxLPlh9nszKLhtxbTO0Delj1avE\nWoWsTMsukUU5NWKtRlbs1QKzN5evCAZHNn6aBmMfqD5X9na1yBOHXrVRcSnxCcvisqtsFAEE\nEEAAAQQQQACB1BU4ovf8WOCxtm2LPKz3EbX3djqMvLRMOT1/llT7C8Wfli579NlGv2l6zk2z\ne4NGWyq1JcrWE1pKM3KlYyDJw2sdjW6S13IUOt+S3Gnu486OJll17Eaq0EVi/p4AKeakrBAB\nBBBAAAEEEEAAgcQQ6NBnF/2+6QUXFK0/uCN4T1F5Rr6sKVvk7v+5rX6D3ke0S7J9GbJQu7st\n0GQJ5xSeJD/a/+CYdiI7Wtc4ny+4Lrs3yUpeWlZwnPcmV4M1K31jCMy8dYz3lQBpvIIsjwAC\nCCCAAAIIIIBAAglYNzkLiqz8sWWjG+z9zMwS93yiC4sWyPK8GeLToOVbe+4Vu2foSzOvkDdo\nAgVLxGDlIW1hstInAfcay/9mZJW41R3oahuyWrsPycr83Koh0yZrBAHSZEmzHQQQQAABBBBA\nAAEEJkhg65E6eVAz0Vn3uW1H64NbKU7PkXdWnumSLFjq7sjy8pEDkqkPd71CkzR4wZHN82jb\nNjdrX2AwQEobaAWy+5fGU+Zklbv7nn7X9Ly8s+IMF6h567troGvfQs2UF69CgBQvebaLAAII\nIIAAAggggMAYBezeoBfa97iAyDLP1XX3J0CwBAln5p8oJ2sL0U1166TCXyCHejrkrsb+e4pC\nN/exaatlQW61a0H6wf4H5K36DKKD2v3tz80vyR81aYOV9t6O4CKFGTnu/c21j8l5RfPkYn0w\n7FhKjj5I9pppF8o39vxZPvXqr1zGO0vocLfW0fblI9UXSJk/fyyrjskyBEgxYWQlCCCAAAII\nIIAAAghMrMBRTXLw5MH+JAvWBe7QQPBi9+1cpN3mLtSA5QINXOxBr7s0yYEFSNs66t0QrWaW\nMe7jNaulV4OtP7e8JL9s2KBPMBJZqfcf/WbRR+Uj226Vp/TeJK9Yqm8LtH7X/ILY/UxjDZBs\nffZAWLvn6XsamH1g28/cJqwL4Cen6bOVqs7xNhmXV19AS1y2PEkbbW5uls7O/gwdk7TJqJvJ\nycmR4uJiaW1tlaNH+29MizpjCo4sKCiQnp4eXCKOvbnk5+dLY2OjdHd3R0xN7Y8lJSVy6NAh\nd96ktkT43peWlkpWVpbU1tbKFP/VHr7jI/hUUVEhDQ0NI5gztWaprKx0O1xfP9gdJ7UEjr23\nnDNDbex+lerqanddZddXlEEBv9/v/ma3tLQMjozRO3uQ6mNt24NJFjoDPW7NpRl5cn7hXHdP\n0TkFc4Iptce6WetKt6uzyT3fKDvNP+xqrEUqQ7PU5RxnPnMpLy+X9vZ293f7WCu1tN8WmFn2\nu4ks6enp4v3eG247tCANp8M0BBBAAAEEEEAAAQQmWWCfptl+RO8BsvuJnm3fHUyUYC0sq4rm\nu6DolLyZ4t0TFIvq2bqi3aMUbd0FGdnRRo95nD2gNpEKAVIiHQ3qggACCCCAAAIIIJCSApZY\nYa3ef2NB0ZajdUEDS7ltWedWFy+Q+fqeMvECBEgTb8wWEEAAAQQQQAABBBAIE7D7fl44vFce\nat3iEhMc6O5PeZ2unc3soa0WFF2k9xQlWutK2E5M0Q8ESFP0wLJbCCCAAAIIIIAAAokl0GFJ\nFg695oIiS7LgPTDV7vlZrV3nLipa6LrQxboLW2IpJH5tCJAS/xhRQwQQQAABBBBAAIEkFWjr\nOSrrDm53QdFj+toxkGShOCNXrtRnD1lL0TmaNS5L01xTEkOAI5EYx4FaIIAAAggggAACCEwR\ngf1drS7JgnWfe1rTZPdJf9Lo6ZnFroXIgqIV+SfENMnCFKFLiN0gQEqIw0AlEEAAAQQQQAAB\nBJJZYGt7rdxdu8G1FL1ytDa4K/NyKl0rkQVF9lBWSuILECAl/jGihggggAACCCCAAAIJJmDP\nDdqoSRYs65wNezU1txVLsnCqtg55SRZqMosSrOZU53gCBEjHE2I6AggggAACCCCAAAIq0NnX\nIxs0yYIFRJZkoVUf4mrFkiysKV8i5+XOcV3oijJy3Hj+S04BAqTkPG7UGgEEEEAAAQQQQGAS\nBA5qkoXHDu5wQZElW7BMdFaK0nPkitJlrqVoVdlCKS8skZaW/lakSagWm5hAAQKkCcRl1Qgg\ngAACCCCAAALJJ1DXdVAe1hYiayl6+tBO6R1IsmDd5VZpOm7rPmfd6NJ9aW7n/NqCRJk6AgRI\nU+dYsicIIIAAAggggAACYxR4taNR1rZudkHRy0cOBNcyN7tCVhcvcEHRotya4HjeTF0BAqSp\ne2zZMwQQQAABBBBAAIFjCFiShReP7HNZ5x7UwGhPV3/3uDRNsrAib6YLii7WB7dOyyo+xhoY\nPVUFCJCm6pFlvxBAAAEEEEAAAQTCBLo0ycJT2mXOS7LQ3HPYTc/yZcj5hXPlwuKFrgtdiT7E\nlZK6AgRIqXvs2XMEEEAAAQQQQGDKCxzq7XBJFuyhrY8e3CZHB5IsFKZny+tLlrqWonM1OMrh\nPqIpfy6MdAcJkEYqxXwIIIAAAggggAACSSFQ333IJVmwoMhajHqkz9W72l8obyxd7oKi0/Jn\nScZAkoWk2CkqOWkCBEiTRs2GEEAAAQQQQAABBCZKYGdHUzDJwktH9gc3Mye73CVYsEQLS3Kn\nBcfzBoFjCRAgHUuG8QgggAACCCDw/9m7Ezg5yjqP/7/unp57JjfkDuEKNwmnhFVYUQGPdQFX\n/au4KKgouv5XX+66iC8XURcUFdg/yx4cusgKrLvyUhY8QEEQESEgcuaAEI4kJIQkM5l7uv+/\n31P9dHd1dzLTSfd01/SnNFTXXfWunu769lPPUwggULcCaW1kwYKQ1SeyRhbWDW5x+xrT/x7Z\nMd+Foj/XULSgZXrdHgM7Vp8CBKT6PC/sFQIIIIAAAggggECBwHBqVP7Qu9a1PHePPqdo80iv\nm6M5lpATu/eTN2urc/acounJjoIlGURg/AIEpPFbMScCCCCAAAIIIIDABAv0jg7KA9vXuJKi\n+7etkh2pIbcHnYkWOW3aoa6kyBpZaE80T/CesbnJKkBAmqxnluNCAAEEEEAAAQQiKrB5uNc1\nsmC3zz3U87yMpINGFvZKdsnbpx/uQtHRXYskqSVHdAhUWoCAVGlR1ocAAggggAACCCBQtsAL\n1siCBiJree5xfYCr7xa3zHCtzv35lKCRhVjMahnRIVA9AQJS9WxZMwIIIIAAAggggMBOBKyR\nhaf61rtQZMHo+YHNbk6LP0e0z3OhyOoULWylkYWdEDK6SgIEpCrBsloEEEAAAQQQQACBsMBw\nelQe6XkhKCnSRhY26fOKrLNb5ZZ37SfW6pw1sjAz2RlekCEEJlCAgDSB2GwKAQQQQAABBBBo\nNIG+0aFsIwv3aSMLvalBR9AZb5FTpx7iSor+TBtZ6NBGF+gQqAcBAlI9nAX2AQEEEEAAAQQQ\nmEQCW4Z3ZBtZ+L02smAlR9bNaup0Lc/ZQ1uP7dxHknEaWZhEp33SHAoBadKcSg4EAQQQQAAB\nBBConYA9qNUaWLD6RI/veEnSmV1ZpA9qtQYWLBQdrnWLaGShdueILY9PoG4C0po1a+TBBx+U\nefPmyfHHHy9tbW2hI1i3bp088MADMn36dFm+fLl0dnJvagiIAQQQQAABBBBAYIIFrJEFH4rW\nDGzKbv2w9rnZULS4dWZ2PC8QiIJAXQSkf/zHf5RHHnnEBZ/bb79dbrzxRvnOd74jU6ZMcYY2\nfO2118pJJ50kr7zyipt+1VVXybRp06JgzD4igAACCCCAAAKTQmA4NSq/3bJSfvLiw3KPNrKw\ncXi7O66mWFze0LVvtpEFe14RHQJRFah5QPrTn/4k99xzj/zwhz+UOXPmyNDQkJxxxhly5513\nyvvf/36xkqMbbrhBrrzySlm6dKmMjIzI+eefL7fccovrRxWe/UYAAQQQQACBnQuktAnolN6k\nldIHhAb98Q6Xnm/U1qLrHNV1Bv1Sw6LTg/GdqVdk67Zt2eFSywX7Zfsnu7GfhcdVer93dvy5\n/dnV8ezqeIuXyx2PuZfen/wz1h5vlrdOPdiVFJ04ZX/pSrTmT+Y1ApEVqHlAmjlzplx22WUu\nHJliU1OTdHd3y5YtWxzqQw89JHPnznXhyE8/7bTTXKCyoESHAAIIIIAAAuUJWEB4su8V+cOq\nh+Tl/i3SN9C/0wvinV0oF164+2CRu3C3oJELHOHxu7pwt8t0uj0ViEtM3D99qGrQj+90OKGl\nP0n9F56/9HKtzS1yQPvesrx1Hzm2ax9pjtf8UnJPqVgegSKBmr+rrdTI/lm3evVqueOOO2Sb\n/mJz6qmnunHr16939ZLcQOY/Fpg2b94sqVRK4vF4dtLGjRvl3nvvzQ7biyOPPFL23nvv0Lha\nDDQ3N7vN+n4t9qFet2mhOP881ut+TvR+mYt1LS0t7oeDid5+PW8vkUhIa2urjI4GrSLV875O\n5L75vyOzoQsLWKXwwrqt4Tkm/9BQakQe3LZG7t7ytPx6yzPZ58/s6ZEn9LI6bhfhBRfiCR22\n/9nFt+/H4/6i2/rBcvnz7fRCXteVsLXouqwfbCu83fztBNvz28rMl1nO9tfeDzZ/OBCEh9tb\n29xdLdnj89vNLJe/39njy86Td3w6LrTfBdstud/OsmC/dTm/naLjy9uuP749Pa87W37q1Kky\nPDwsO3bs2NksDTnevpfsX6N/zhSefDOxzq5pomRT84DkITdt2iQXXHCB9PX1yTvf+U5ZsGCB\nm7RhwwZXouTns35XV5cLRxak8ushWUMPX/7yl/NnlSuuuEKWLFkSGlfLgfb2drF/dMUCuBSb\n2Bh7v9MVCySTyeKRjHECdgFDVyzQiC6vD+2QuzY9Ib/Y+Ce5e9OT0q/Po7Guu6lN/nLOMXLa\n3kfIEVMWBGEhe+GfuTAfY9gu7OkaU8A+fxvx72k8ZxuX0kr2w109/HhnVXnG09VNQJo1a5b8\n/Oc/d6VIl1xyiVx00UXyzW9+U+yP0Ood5Xd+uPCCev/995dvfOMb+bPKfvvtJ1u3bg2Nq8WA\nlRzZ/loAHO/JqcV+1mKb9gdjpYG4hPX9h0lPTw8lJWEa6ejokP5+vSVI3zd0OQFzsc/MevjM\ny+1VfbyyHxrsb6kRuhcHtsivtJTISooe2b7W1d+x457bMlXOnHWUnDL9EDl2yj7SFEtkf4Dc\nvj2oaB/42A1u9i/4+3L1a4IJDfXfRnrPlHNiLQBQglQs5u9soGQtbGMu9rc0MDDg/oWn1mZo\nPHdz1U1A8kQWct773vfKt771LVd8a3WU1q5d6ye7vn2QW8mR3XqU3+21115y1lln5Y9ydZns\nQqoeOgtIFgLqZX/qwcT2wYpdLfTiEj4j/ha7wcFB92UUntrYQxYezcX/WNLYGrmj97cv2BeR\n1TGhywnYoyEm62eMnWvX1LI+e+ZebVVsdV5Tywe1zZaTpxwoJ+m/Je2zsyDDA0MyrEO+hHqy\n2mQPeDdeTOb3zG5wuEX884vsxyneM2FF+3HK/uFS7GKfM/Vynedv+QvvZfFQzQOStUZnzz/6\n7ne/m907S9/2x2d/iIsXL5af/exnDtZfMD755JNF9ZKyC/MCAQQQQACBSS5g9Yn+0LtW7t26\n0oWiTSO97oi1Rqcc37U4G4pmNwePy5jkHBweAgggUFGBmgekk08+Wa655hr56U9/KqeffrpY\n+Pmf//kfsfFW4vKWt7zFTb/pppvk7LPPdqVJ1pDDhRdeWFEIVoYAAggggEA9C2wb6Zf7t6/W\nZ888Kw9sXyP9KSsDEulMtMip0w51oejE7v3dcD0fB/uGAAII1LtAzQOStTD32c9+VuzBr9ag\ngt2C9ta3vlU+97nPOTu7jc7qJF188cViIcluITnzzDPdQ2XrHZf9QwABBBBAYE8EXhp83ZUQ\n2a1zj/auc8/wsfXNTnbLX0w/0oWio7oWaRPNQUtRe7ItlkUAAQQQCARqHpBsN+zBsO9617vE\nWqybMWNGUTOAy5Ytk9tuu02sGW9rzME3ZctJRAABBBBAYDIJuPpE/evdrXNWUpRfn2hJ294a\niJa4UJRfn2gyHT/HggACCNSDQF0EJIOw+kXz58/fpUk9PM9olzvIRAQQQAABBMoUoD5RmWDM\njgACCFRZoG4CUpWPk9UjgAACCCBQNwLbs/WJVrr6RH2p4NkcnXGtTzT1EDlp6hI5sXs/6Urw\n0N+6OWnsCAIINIwAAalhTjUHigACCCBQS4GXXX2iVa6RhcL6RO+afgT1iWp5ctg2AgggkCdA\nQMrD4CUCCCCAAAKVEsivT2SNLKwaeDW7aqtPZM8msjpFB+U9nyg7Ay8QQAABBGomQECqGT0b\nRgABBBCYbALDqVH3fKJ7NBBZKNo03OMO0Z5PdFze84nm8HyiyXbqOR4EEJhEAgSkSXQyORQE\nEEAAgYkXoD7RxJuzRQQQQKCaAgSkauqybgQQQACBSSnwyuBWV0JkJUUrel/IPp9ob30+0Tsz\n9YmO5vlEk/Lcc1AIIDD5BQhIk/8cc4QIIIAAAnsoYPWJnu7f4BpYuHdruD7Rge75RNQn2kNi\nFkcAAQTqRoCAVDengh1BAAEEdi6QHtFpO0TSvTFJ98REfF+ruPhxmwe0qejBYRlpaRZpT0tM\n/0m7BP0OHW7TdVg/f7y2Ih2L73y7jTxlrPpEQSMLBwr1iRr5XcKxI4DAZBQgIE3Gs8oxIYBA\n3QukB3QXdwRhJ231+DOBx8KOaACycdnXNk7nHasbkpRoWwAiqcRYs4anF4YpN6yz+PEdmZBV\nOF4DV0yz2GTqrD7Rb7evcSVF1vfPJ+qIN8vb9PlEJ2vLcydO2Z/nE02mk86xIIAAAgUCBKQC\nEAYRQACBcgXSmkukLxxoXOixoONLenwAsjC0XcPOyNiBR2JaAmQlPF3an52SWKeW/nTp8jps\n42w4eK3j9PWMxVOlpaNFNqzZKCndHxfArN+n+5Ht63ZtXy1w9et4LZUKpuvw6zp+tMzipITu\ng+2XllS50qlSpVQWrDRkSZv1M/PqOGmpj9Ir6hPpuaFDAAEEEMgKEJCyFLxAAAEEAoH0sPbt\ndrZMwAlKdIKwI/klO9npOn96HIHHwoSVxuytIaFLU5ULPPbaQkYmCGWCjxunYaic299iyZjE\nYvpP1xXXICKz9F+ZnTt2C3M+ULkglQlUFqw0bPnAlX1tVlt0/HpdTsbhkN2njIeFJX8rYGEp\nlY23Yyk1fg++wZ7qW59pZOFZWdX/anaPrD6Rv3Xu4PY52fG8QAABBBBoHIE9+HppHCSOFAEE\noi1gt7O5ejsu0Oix5JfsZMaF6vVoycq4uuZMuFmkfQ0lpUp2cqU8Oo+VstR5F0vqDk7TfZ1m\nO1pewMqWpJUqncqGq0zI0nnS/boJ61vo2qLjyy29SlrAyrhagLIgZbf9uUClfRe8gum9c4bk\n0a0vy/0jq+TuoadlnbymM4i45xN17iMnTV2iwegAmds81Q6cDgEEEECggQUISA188jl0BKIo\nYBfhqe1pSa3X/tZ4UMqjIUd6gxCUrbfjw5Ddzjaq/8bq7HY2K62Yov15ejubu41NFwqV8ujF\nth/fra/5BA2putIuKwlTs3LDla3I1csqvB3QSqfcrYB6DgtDVmbe9GadNqTvhdDehAdel2E5\nWA5w/z4mb5dULCWjbaPS1BHX0jYLUzq/BqrhTD/UkEV+CLPQZSGMcx8GZggBBBCYRAK7/RE/\nOjoqiURQEXhkZER+85vfyPr16+X000+X6dOnTyIiDgUBBKop4G7pyi/Fca+1VCE/9LjXuXEb\nRVtrkzFaB7Db2SzMzLFQo6mq8Na1zLAv9XG3uI0jR1XTotHXHdP6VtKq58t9hewq7hRLudIr\nDVMbt/XIY6++LE9v3iAbtm6XjsFW6Rpsk9kjU2S/1F6yYHSGTBvqkISGq0R/wtXFSmthUjpV\n5slv0f2zUqv8QGVBK1PPKnfLoIUvnddClq+fpXWv6BBAAAEE6ldgtwLSd7/7Xbnssstk7dq1\n0traKueee67853/+pzvKzs5OefDBB+XQQw+t36NmzxBAoGoCVncl/3a1tJbshOvwaNDx46x0\nZ2icF6Z24WwlE1qvpnl6Un/9H5Z0R1DSkws/ui1XwqPz2sU2XUMIPK31ieyBrfdsy9QnsrfU\nLJEDFu6lrc4t0X/7y5sWHSGbNm3KeFjADnfZVgX1/esbrXD1sNytgLpCK8myWwFDJVw6bpOO\nG6P0KrwlHYpbuMoLTvkhy4+3vrtlUN/LGryyLQra+DLb0SjaPiMQQAABBHYpUHZAuu++++Tz\nn/+8HHbYYdLf3y9PPvmkC0dvetOb5NOf/rR89atflQ996EPy6KOP7nLDTEQAgfoXSI/qPtqt\nawUtsfmGCnL1enKlO+O6nc0uEO3X9hna19Idd9uahR9rlKCgZCdbtyfv02ratHbp6RkQK72m\nazwBez7Rw71rtZGFVS4UvTpsTQOKJLRS0XFWn0ib4j5p6oFl1SfKll7NsDXpe7GMzj2jKnv7\nn/4taJgKWgi0QFUQrPIDl74WDVjp8TTwkb8/VnpVcNtfrhVBm6Z/R1ZqlV/CZSHLSrcovcqX\n5DUCCCBQUiDvkqPk9KKRd9xxh8yZM0cee+wxicfjctttt7l5Lr/8cjn22GNleHjYBaSenh7p\n6tKrHToEEKgbgfSg7kp+c9P+tfZLP3vHdl2njdVZZXmrqzPPAk6J1tmsvo4FIJtHA5C7gBvH\nasfaLNMbR6BnZECfT7TalRRZf0cqKAWy5xO9derBrqToz7r1+URNE1906Oojdet7X9/nZYcr\nW8Qaq8g0ue5KqfKbX7cQpdNds+yh0isNVht1ueFd173SOcKd3XqaLaVKy6vT+iXeFpOh0eag\nXpVNtysDu4M+04/5cX68Dsf89LxxNn92vF/G5tN/zqhJ150Z9n1tdJEOAQQQqDsB+2grq1u5\ncqUsX77chSNb8M4775RZs2bJMccc49Zjt9al02l3+93hhx9e1rqZGQEExi+gf2bBr9NWPydb\nb0fHZV/reB+A7Ad2m294PFcjumKrhG5BZnYQbLLP3vEBx93GpuvMDPOrtFrQVVzglaGtrpTo\n3q3PyoredTJiD8LVbq9kl7x9+uGupOhYLTFKxu2qO5qdCwgaWFxpz+6UXlmT9LsqpSq8LdAH\nLru99dWYDDxvxcTWlX05ECy2p/+1IGWnzzZv/3z4KhiXDVlufG4ZF8j8OLe8rcOCms6TXWcw\nTjSg5QKcbS+zHps/O29unL3f0tv8spm+zcstjopBh8DkFrCPhLI6a4Dh97//vVvGGmVYsWKF\nfOADH3DP3rCRv/rVr9w0K2WiQ2CyCbhQYtcTdp2W/8/G2W0y2neVxW2aG5fpWwXwvPndrWt5\nw9qkVjA9b939zSldZFAG18dldFvS3eYWlPLoL8euxCezTe3tsrMLBbvlZi8LO7pRF270tQWg\nUs/esXF2wUGHQA0Enunb4G6bszpFK/utiCToDmjdy902Z3WKDuH5RJ5FXLPsU/Rv1lpfLPfW\nQP04mNU5Sx8qnJbNG18LPrPsrlVr9VH77nPKPpPyxrnPODcu+Lxz09x0mz9Yzn322Tj9XHO3\nH2am568nt+68ZXS9Nn96MDxOMs2/2xFOZLfeFe1Zei3orMVL+4z0ocoCk4UvG/bjMwEtF+CC\nedx0C1luPl1PaBkbXzguL9T5efPX7cblLWPr3tk43QU6BBAYn4D9GZXVnXbaaXLdddfJBRdc\n4OofWWnRBz/4QbFW7azxhksvvVSOP/54mTlzZlnrZeb6EHABwF+4512sBxf3+qWl04ov7nXf\n7YvRL2fz2OvQ8nnTM+N9kLAgMKq/go4M6tsxfxl97e7NLxjnrgHcuMz+5G03t3x4e35b2ePI\nX2dm+ew+Z4JO9nhsXv3+GVfdGp2tUp02QaDVf/QBPv4b1a94Z8/ecYHHgo9+QdrtbD4Ilfh+\n96uij0CtBYb1A+WRnhfcrXP3aijaOLzd7ZLVJ7LSoZOtPpH+m9vC84kqfa6sJCTeGQv+hcKV\nfeDVV+e+dyxoZcOWhi/7bLZ/eeNsODuvD2yZ6cH4zHeDlby5eXU4u7xfX7DulqYWGdyh9yWP\n6DyZbbnA57ZZOE6XsVuY/bps2/pdUleSFqyyAU731Zeo+XE+WFnIyo7LLePHbe7ql3TLqLLo\nAna3gdV3s/pt+tr17TtHh0Xru3ELpVrQRVKg7IB0xhlnyGc+8xm5+uqr3W12X/jCF1zT3haQ\nLrroIjnllFNcUIqkRpV3euCZERlYoZ+fA8qevSjXD9GiMKE74sYF08IX9zrNPpx9iYNfj43T\nz6NgWmZ6wbjcejIf7Jll/b64ZcdT30RXX8mu1+247tNYzTZXZKOZD3vbnH0B2C9/9s9eZ34V\njFkDAn6877t5FSwz7G6xyF/eL5M3zs3j5tf12V+abdOvz/o6b8z/EunH2/KZfWttb5G27lYN\nST2uxbbsbW72IE86BCIs0DOq9Ym2rdbb51bK/Xn1idoz9YksEL2x+4Ca1CeKMOuk3nV3cW6f\nj9lGJvSCvOiI7Uuwct3MWVO05cOgAZDdWWv2RzcfmlxfQ5Mf9kFLh8NhT48sM0821Ll5dS+y\n43We0Lgg1PnpQT9YT27dufXasjZ+d0rretxjFkxkjEct2PdbNkDp7C5EhcdZs/jZxkR8yLLG\nRWy8LTvGJmwv6BCohkDZAckaZrjyyivla1/7mtsf3xCDPRPJmvdeunRpNfZzUqyz/9ER6b9J\nP6DG+lDZo6PVDxV/sZ25CM8NB9NiWgKRG6cby86XG5+9uM9esOemBWFCP7hsmtuWTsvO58cF\n/WzYyM6bN91+zdLl29pbNaOlZGhEK11n58vbns6T25Zfvnib2dBSGDrcPuo6yn6367Zq2LV1\ntUlnZ7MMbE5LeliPlw6BCAusH9rmApHdOrdCS4x8faJZWp/odK1PZCVFUa9PFOHTw65XQSD7\nPRr6Uasw2NXfZ3s4wCmMlqD5kBXTkrEZU2bK4LZB2f6qNnFq9d+0UZFs3xoY0XGu0RHf+IjV\ng9tq6ymzQRE7J3ad4EOWtr+SbZ0xU1qVnRYKV7qcDbuSreh999th09VeoOxLRnve0RNPPCHf\n/OY3i/bewpG1avfZz35WnnnmGWlrs3cwnRfoODEpw9MHtLUgDQKWkzIBwUKCCwBunI53F/T6\noZC5sHd9DQ7ZD9sSpRV+XjeP32BE+q1dcW2uWW+z67efw+gQQGCyCIxVn8hKig5tnztZDpfj\nQGBSCLgfE0NXh7lQZ7fMtczWC5LBhCS2lPed7YKXb1DEglQ2QCmbG9aV5wWu0DhrXESryaUz\n9dHKgrbGOPS2v2xpVSZcuabx84OUL7XK3DLowpcfZz/e0jWUQOhPYGdHbg/XGxoKmlS15xs9\n9NBD8vLLLxfNbvNYM+Dr1q2TgYEBAlKBUPPChDRrU7AEgQIYBhFAYFIIWH0iKx0KHtoark90\nTOci1xS3lRRRn2hSnG4OAoGyBFzw2s3m8P2GXD2vwlIqa/5ex7nm8K3vQ5Y1j2+lV5n505mW\nG8t+7pht3J49ZmHKwpW7BVD7BeEqrT/29k8bldF0PFcfKzO/q49lP3rTRUZgXAHphhtukL//\n+78PHdT8+fNDw/kDVpI0bdq0/FG8RgABBBCYhAKuPtH2NWJNcf9W+70pq6mu1xBan+gt7vlE\nB4o9n6i7iTsKJuHp55AQmFAB90gJDSsx12aLhpYyO9cQlbV9ZKHJSrBcuPL9TMhyoSpz62Am\nXPn501t0nvW6bIkacFY9bNBV/ir1LDYLWDqDC1WZfnY4mJatd5UpwcqVcGXmz9a/02G6qguM\nKyD97d/+rXtivT0E9te//rW88MILcs455xTtXFNTkwtGf/VXf1U0jREIIIAAApNDwNcnskYW\nrAW6/PpEp00/TFudO0CO61wc6ecTTY4zxVEggEC+gGtVzwcTN2E3QpY1cGWlU1ZalQ1ZMUkM\n6V1Co63S91pfruQqxUEW4wAAQABJREFUO92WCUq6dusBz7avVr/a3yqox+BbDwyNs1sC80u4\n7FjdcGbZUH04WyndzgTGFZCSyaRceOGFbh0HHXSQPPXUU/KVr3xlZ+tkPAIIIIDAJBN41j2f\naKV7RtGzec8n2r91lrt17qSpB8ohbXOyz8SbZIfP4SCAAAJOwNX1tmcLdmjocF3QTybj0tGZ\nkKHXrSxp152rj+VDli/FyivVygYwP833XYmWBjNt9CKtjV6U3Vnd92zIKrhlMC9IZW8nLLiN\n0NXLGldyKHvP6m6Bsg/zfe97nzsIa9bbWq6zbmRkRH7zm9/Ien1w7Omnny72MFk6BBBAAIHo\nCuTXJ7KSog15zyfy9YmspGheC7dTR/css+cIIFALAVcfy55ZaM8v3M2nZaXtWV6+vlUoQOn4\nTGmVC1o6zZV0+XBlwcwavdisJVr2yJhyO2sJ2UqwrLRqV7cMulsF7RhjMryfNuihxxulruyA\nZAdnD4S97LLLZO3atdLa2irnnnuuWOt21nV2drrmvg899FA3zH8QQAABBKIhYPWJHtB6RNbI\ngj2nKL8+0SlTD3IlRW+kPlE0TiZ7iQACk1ogZrfLTdEAMmXXIcsatkhrGLIHyMd0/vzOpqVe\n0ObXNWi5Z05Zw4R5QSrlWw/Upt6tifeUPcdbp4uGs/Q27W/UkKUPRN5VZzluy/IB6fjsruaq\nv2llB6T77rtPPv/5z8thhx0m/f398uSTT7pw9KY3vUk+/elPy1e/+lX50Ic+JNbaHR0CCCCA\nQH0LbMh7PlGoPlFTp5w27VCxW+fs+UTN8bK/Lur7wNk7BBBAoBEEtMRo6CttEj96RJo/GbRI\nnX/Yw9/W1h80/DSdNSRNp4dvDxx9Ji7Dl7dK0xk67R0jMvh1nVdvHmv5urV0oSHJspmuPyix\nyvQ1bI3c3OweQtx0yrDEtUn4tv1bZcd/D0r6DRrEXAMbOm+dd2V/41kz3nPmzJHHHntM7KGx\n9twj6y6//HI59thjxRpysIDU09Mj/iGydW7A7iGAAAINJeDrE9mtc8/0b8gee7Y+kTbFfUg7\n9YmyMLxAAAEEIipggSQ2NyWp1cUPc0ppALJwJFq6lHpCpxcEpNSzwTLxQ61liuLONXqhdZpc\nQxBusiUmLZVapUVRmsUsVCWTWvJ0V1r6fpSWlmOttCmYx81Yx/8pOyCtXLlSli9f7sKRHded\nd94ps2bNkmOOOcYdpt1al9ZIabffHX744XV86OwaAggg0BgCrj5R7zptijtoZCG/PtHR7vlE\nB+rtcwdSn6gx3g4cJQIINJhA/GB9PtPdSUltikl8Vi6guFDUnZbE8hEZ/WWTpLVgKJbXSnl6\npQYouzVvUemAtDPG5Hvsxrpc50qacoOReFV2QLIGGH7/+9+7g7NGGVasWCEf+MAHsi0X/epX\nv3LTrJSJDgEEEECgNgK9o4P6XKLVQX0i7duwdW3xpPj6RPZ8oik8n6g2J4itIoAAAhMkED84\npQFJy25Wa+CZZRWNgs4CkoWn+KEaoH6mAerphCSWBdOtpb3Uc3GJLx3Va3y/RNBP9+ptdBq4\n0uu0btKMtJsncUguRI38SuOFZqSmU0dk+LcxGXokuHVv5I4miS1MS9NJuVv5UmvjMvpQQtLr\n9QG701MSPzwlCd1mrbuyA9Jpp50m1113nVxwwQWu/pGVFn3wgx8Ua9XOGm+49NJL5fjjj5eZ\nM2fW+tjYPgIIINBQAuv6X5P7N6yWn6z7gzzcszb3fCKtT3TqjENdKdGxXdQnaqg3BQeLAAIN\nLxA/SANHXG+j04CUOCEIHylrYGFTXJr+QusJHaDhRlunSz2h031Ael7D1LCWOB0WDivp7TEZ\nvLhVrLnz2N5pGb03IaO/1jjx8SFJHBvMO/qg3pqnjTlYQEppAxGpbUGpVXqDLtSWF6TubpKR\nW7S1CZ09fsSopF7WsHSvBrU36q15f11cX2oiT2TZAemMM86Qz3zmM3L11Ve72+y+8IUvuKa9\nLSBddNFFcsopp7igNJEHwbYQQACBRhR4fmCzrNBb5x7pfUH/rZNNwz1Zhv3c84ns1rkl1CfK\nqvACAQQQaDwBu20utljrIa2yOkXB7W/u9jqtDxQ/REuINA3El2gp058SkvTT7fY67fJLhtyI\nXn0o7ikaYP6fYD3p1zQwfbFVrNTIByQ3X+Y/Le/WEqH2pGz9waAkPzrkSpxsUmpDTEZuTUr8\nwJQkPzWoD74NFrBSppH/bXaByYe1zKomtFd2QLKGGa688kr52te+5nbUN8Rgz0R68MEHZenS\npRN6AGwMAQQQaAQBK61fPbDJhSEfil4f0eaCMl1XolVOmXmInDTrEFkWny3zmiPSVJA/APoI\nIIAAAlUTiOstcKM/1XpGOzQs6YNuLSDFFuSa/rbb7FJ/anbBJT5bS5NW6nRt3CE2LVdnye1c\nTG+Ry6tjZLfYxfZLiSsdKmPvR+/VCKINRCROH86GI1s88eYRGflpUkuSNHBlSrPKWG3FZi07\nIPktWzB6/PHHxRptsNennnqqTJvGAwO9D30EEEBgTwRG0yl5tn9jNhA9qiVE2/U5Rb6b1tQu\np0w5SI7qWiRHdyyUA9r2khkzZkhLS4ts2LDBNZbj56WPAAIIINDYAglrqEGDR2qN1ivSOkmp\nZ/V2Oi0J8p2/lc5uw4tpPSV3O96bctP9fKKByT2DKTtCA5c2/JB+Lm/EOF4GgUpv0dOSp9Ff\nFyyg6SStJUy17HYrID311FNy/vnniz0Tybr3ve99LiAdeeSR8jd/8zfypS99yX1J1/LA2DYC\nCCAQJQFrae6pPm34JnPL3B97X5Qdqdw92LOS+kNU935yVOdCsZbn9m2lnmeUzi/7igACCNRS\nILZvpp6RNfdtd9oNaf0iLTXynZUaiTaSkF6jDSbM19eDOl1vvyvs8lu5y04rKGTKjt/FC2vo\nQSyF5LWa52e3+kh2218tu7I3v337dnn729/unndkD4x94IEH3P5bHSRrwOGSSy6Rl19+2TXk\nUMsDY9sIIIBAPQsMpkbkiR0vZwPR4ztekgFrNijTzWmeIid3LnFh6GgNRQtapvtJ9BFAAAEE\nEChLwNczSmurcSkLNC1a/2j/XIMJtrLEYVqypCVMsXla/6hJp2u9pGp1rtTpeW3I4R3aSMTc\ncMKyr8LIBaR///d/l23btskf//hHWbhwobz3ve91dlYH6eabb5Z58+bJVVdd5f51dOhNjnQI\nIIAAAtI/OiQWgqwxBSsl+lPfy2KlRr5bqAHIlw5ZIJqtAYkOAQQQQACBSglYk94jtyclpk99\nsJbtCkOIa+77voSrn2Qt28WaK7Nla/HOOhd8gpeu5bzUQ1oN6XdNEj8r99yk1CvaLPhXW11r\ne7Vsya7sEqRHH31UTj75ZBeOMscY6r3//e+X73znO+5BsfbQWDoEEECgEQV6tL7QH3stEL3g\nAtFTfa/IaN4TxO0WObtVzv5ZMJqZ7GxEJo4ZAQQQQGCCBCwgya3Nehud1j/6YC6U+M276Tpg\nzX3nN8Tgp+9uP94V1CcavUOfnaS3zyWO1n9av8maBx+9S5+NpA+jtXCW1ma+R/5Pm/3W/ze9\nO3eL+e5ud0+WKzsgtbe3y8MPP7zTbfb1Ba0qWWVhOgQQQKBRBLZqi3KPar2hFZkmt5/t14YS\nMgdvXw0Htu2dDUMWiKZqIwt0CCCAAAIITJRAzOoWdem/nnD9I799a2rb6ipZPaT8+kl++u72\nO05MyrY7BmTktxqI7FlLGpBiWg+q+QsDMvxfzTLyP5qI/jsororN1Ga/P67Nfte4IdayA9Jx\nxx0n1157rfz4xz8WeyZSfmf1ky6++GKZO3euzJ49O38SrxFAAIFJJbB5uNeVDPlGFdZoE9y+\n08ZR5dD2udlb5pZ2LhBrhpsOAQQQQACBWgnE9Ne61u/273LzLf+g99/tpGv5Uulpzedpac95\nuYVaLgzPl5gSl2mXJmT7Rn1WnzUQkeliXRqSPjHkbr1LvxoT1wCEtZJX2wbs3N6VHZA+8pGP\niNVDOvPMM+WEE04QC0VtbW3ywQ9+0IWm/v5+ueWWW/yx00cAAQQmhcCGoW0uEAV1iF6QFwa3\nZI8rqT+FLetYkA1ER3bMl7ZEhW7ezm6FFwgggAACCERXwD8MtvAIrC5UrKChhsJ5Jnq47IDU\n1NQkd9xxh3zxi1+U733ve5JKBS1c2G13c+bMceHJN9ww0QfD9hBAAIFKCbyoASgoHQoaVXhl\naGt21a36aX5s5z56y1zQ5PZhHfOkJV72x2l2fbxAAAEEEEAAgfoRKPsbfdOmTTI0NOSa8f72\nt78tq1atks2bN8u+++7r/iWTeh8hHQIIIBAxgecHNmcbVLBSok3DeitApmuPN8vyrv1cILL6\nQ4d0zBUrNaJDAAEEEEAAgcknUHZAuu6669yDYO1hsUuWLJFjjz128qlwRAggMKkF0um0rBp4\nNdPkdtDK3OvayILvurW+0EndB+gtc9rCXNdCOahttiR8O6V+JvoIIIAAAgggMCkFyg5ITz/9\ntINYsGDBpAThoBBAYPIJjKZT8oy2Kme3zK3o0UC040WxZrh9N01blDtl6kEuENltcwe07qWV\nROuglqjfQfoIIIAAAgggMGECZQekCy64QO6880758pe/LF//+teltZWWmSbsbLEhBBAYl4A9\ngPWpHa9k6xA9poGoL5V7psKsZJcs7w5umbPnEC3WZxLRIYAAAggggAACJlB2QHrxxRfFHgBr\nD4O94oorxEqSSj3z6JFHHkEYAQQQmBCBwdSIPLHj5Wwdosd3vCQD9sjuTDe3eaq8ufOgbB2i\nBS3T/ST6CCCAAAIIIIBASKDsgGSNNGzdulWWLl2aXZFvyS47ghcIIIBAFQX6Rgfld9vWyEPb\nnnOlRE/0vSJWauS7hRqArGTIGlSwW+ZmN0/xk+gjgAACCCCAAAK7FCg7IJ111lnyrne9S+bN\nm7fLFTMRAQQQqJSA1Rd6rPfFzC1zL8jTWp/I6hX5br/WWaFANCPZ6SfRRwABBBBAAAEEyhIo\nOyAVtmJX1taYGQEEEBiHwFZtUc41qKCNKjzS+4Ks7N8o6cxycYnJwZ1z3YNZl7UvkGWdC2Sq\nNrJAhwACCCCAAAIIVEKg7IBEK3aVYGcdCCCQL7B5uDdbOmTBaM3ApuzkJonLYe3zMrfLLZKl\nGogWzJwtPT09MjKSq2eUXYAXCCCAAAIIIIDAHgiUHZBoxW4PtFkUAQScwPqhbaFAtG5wS1bG\nHsC6rGNB9pa5IzvmS1uiOTudFwgggAACCCCAQDUFyg5ItGJXzdPBuhGYnAIvagB6xJ5BlLll\nzgKS71pjTXJc5z7ZQHRYxzxpiZf90eRXRx8BBBBAAAEEENgjgbKvQmjFbo+8WRiBhhB4bmBz\ntsltezDrppHe7HF3xJuDZxB1LJSjuhbJIe1zxEqN6BBAAAEEEEAAgXoQKDsgnX/++WL/6BBA\nAAETSKXTsrr/VXlkh5YQaRiyvjWy4LvuRKucNOXAbB2iJW17SyIW95PpI4AAAggggAACdSVQ\ndkCqq71nZxBAYMIFrHntZ7SZ7RU9QQtzj+54UawZbt9Nb+qQU6baQ1mD5xAd0LqXxGIxP5k+\nAggggAACCCBQ1wIVD0hp/TX5/vvvlze+8Y11feDsHAIIjE9gODUqT+mDWH0dosc0EPWlhrIL\n75XskhO798sGosWtM7PTeIEAAggggAACCERNYLcC0vXXXy9XX321vPrqqzI8POyO2YKRNblr\nTe/aOBuuhy6RSEh7e+2fkZJMJh1Hc3Mzv6YXvDGamrQh57g+3YZShpCMf8+0traKfx2aoUoD\ng6lheaznRfnDtuflD9uf19frZDCVa057Xss0OXXKYXJM92I5bspiWdA6vUp7svPV2t+1uaRS\nuYfF7nzuxplif0fWtbW1Nc5Bj/NI7fOlHr4Lxrm7Ezab/9zFppic90yxiR9jnzW8Z7xG0Lfv\npXq55gzvWW2HzMQ6u46J0num7IB03333yXnnnefeBMcff7z89re/laOPPloGBgZk1apV7kL3\nmmuuqe3ZyNu6//DPG1WTl34/rO9f12RH6nCj3gSX0ifH+5Seuudjd4wOyqPbX9BAtFYD0XPy\neM9LMpweza54cdtMOXbKvnJs9z4aiPaV2S1TstNq+aLaLrU8tt3dtv8b8v3dXc9kXI73y67P\nKu+ZYh/eM8Umfgw2XqK4z99S2CTfI/91eK6JGxpvAU7ZAen22293Iej555+X+fPny6GHHirv\nfe975e/+7u9k9erVcsopp7jwNHGHuustWanW4ODgrmeagKn2i679s33p7++fgC1GZxP2S5Sd\nJ1zC58xcWlpanIsvqQ3PsXtDPSMDYrfJPdKrDSpos9vP9K2XUcmV+O7fOksbVFikt8xpK3P6\nb0ayM7chLUjaMbIjN1yjV1YSa+8XHhQbPgH2frES2b6+vropxQ/vYe2G7JfLHTtq/96tnUDp\nLXd0dLgJ2BT78J4pNrEL3O7ubhkdHeXvqYDHSkjs85e/pTCMudjnjF3H1IONL9EK72XxUNkB\nac2aNXLCCSe4cGSrW7ZsmTz44INuzfvvv79cdtll8tnPflY+9rGPFW+twcf8f2t+IVes/pnY\nc19a4kn3rJe2WND3w6063k9v1WfB2HgbtvH2bBg3nJ3Hjw+m5eZLSrM2m1wPSb3BT3ldHP7r\n2qLco5nnD9lziFb2b8zGIb2xUaxVOQtC1qjCMu1PaeL2rLo4cewEAggggAACCNREoOyANG3a\nNNm+fXt2Z5csWSJWJ8l3y5cvd3WTXnrppWyI8tMavd+lF557N3dL/+iQDGg9D2sK2ep75P96\nX0kjH7QsWPmA1aqBzL/2421cEL6C+fxwbr6ktGVCXTKudbr0OTZu2bx10WxzJc/cnq1r03BP\n5oGs9mDWF8SeSeQ7re0lh7XPy5QOLZKlnQukM9HiJ9NHAAEEEEAAAQQaXqDsgHTQQQfJzTff\nLBs3bpS9995bDjnkEFm7dq2sW7dOFi5cKE8++aS7BW8iK5VH5Sz+9aI3yrunHF50K5k9R8Za\nBbOwZJXhLTwNpPXWQOv7cTocvA7msfn60/Y6t4xbVscN6LT88bYuC2O2vNUtyd1MVTk5K4lo\ns5ItV9IVBK780rFdlYZN2d4hSUlIfCTtlg+CnQ91mdKxbKkZpWOFZ2390LbsQ1nttrkXB1/P\nzmIPYD3KPZBVS4i0f4QGIjtPdAgggAACCCCAAAKlBcoOSB/+8IfdbXQHHHCA/PSnP5U3v/nN\n7t7Cs846S8444wy57rrr3C14Fp7oxicQ13t67Vf8ifolPxu6fLhyoSoXvGz6gAtfPpRpv2Ae\nH96CIOfnC4Jdv4a9iSody5VyaaDKK9EqLB3z8wXz5G5xtPFjlbTVW+nYusEtmRKiF9yziNYP\nb8u+0exYjuta7MLQUV0L5XAtLWrWEkQ6BBBAAAEEEEAAgfEJlH3lNGvWLPnxj38sF154oWu5\nzm65s1brPvrRj8rDDz/smvG79NJLx7d15qqJgAsFGgymSPXrmthDRftdEAuXhvmAZaVesRat\n1Dg8ID2DdsthELZyJWi50rCgxCw3HAS5XOnYUF7La5WEtdvSfMAK6oAFYcyHsOD2xPHVFbMA\nU7guX8/M+rauwm5N/6vy+63PaRjSQKR1iDaN9GZn6dDbHe0ZRL5RhYPb54iVGtEhgAACCCCA\nAAII7J5A8dXYONZz4oknyr333pttJenss8+Wt73tbfLoo4+6Vu0WLFgwjrUwSyMIWOnLWKVj\nXV1dFWvFzoWmTMnYYOa2xFIlZj6I5W5LDJeguQCXva3R3+6otzVOYOlYW1OzC4yvD+da3upO\ntMpJUw7MPpTVGliotxKuRnhfc4wIIIAAAgggMHkFdisgeY78VtLslrrTTjvNT6KPQE0EfOnY\nRGx8REvHgvAVLh3z43wIKywNKxy2+fJLx3ygsxKx9kSzHNu1jyxrn+9KiawJ7vy/u4k4TraB\nAAIIIIAAAgg0ksAeBaRGguJYESgUaBpH6VjhMuUMW8laZ2enbN682T0/oJxlmRcBBBBAAAEE\nEEBg9wTiu7cYSyGAAAIIIIAAAggggAACk0+AgDT5zilHhAACCCCAAAIIIIAAArspMGZAevrp\np+W8886T9evXu00MDQ0VPcdnN7fNYggggAACCCCAAAIIIIBAXQmMGZBWrlzpnm306quvuh3/\n+te/Lm9/+9vr6iDYGQQQQAABBBBAAAEEEECgEgJjNtKw3377ue1861vfko9//OOuJKm3t1dW\nrFixy+0fddRRu5zORAQQQAABBBBAAAEEEECg3gTGDEiHHnqovPGNb5SbbrrJ/fMHcPTRR/uX\nJfvpdLrkeEYigAACCCCAAAIIIIAAAvUqMGZAsmeu3H333XLXXXfJyy+/LD/+8Y9l9erV8rd/\n+7f1ekzsFwIIIIAAAggggAACCCCwWwJjBiRbazKZlNNPP91tYHBwUB555BE5//zzd2uDLIQA\nAggggAACCCCAAAII1KvAuAJS/s5fcMEFbnBkZETuueceefbZZ91DLJcuXSr2b+rUqfmz8xoB\nBBBAAAEEEEAAAQQQiIxA2QHJjsxKkM455xx54oknig70G9/4hvzDP/xD0XhGIIAAAggggAAC\nCCCAAAL1LlB2QNq6dau8+93vFitB+s53viPHH3+8dHZ2ytq1a+X666+XCy+8UFpbW6mjVO9n\nnv1DAAEEEEAAAQQQQACBIoGyA9J//Md/iIUka+b7wAMPzK7wiCOOkL/4i7+QT3ziE3LNNdcQ\nkLIyvEAAAQQQQAABBBBAAIGoCIz5oNjCA/njH/8oJ598cigc5c9jz0patWqVvPLKK/mjeY0A\nAggggAACCCCAAAII1L1A2QEpkUjI0NDQTg/MTxsdHd3pPExAAAEEEEAAAQQQQAABBOpRoOyA\ndMwxx8i9994rDz30UNHx2MNhv/nNb8rMmTNlwYIFRdMZgQACCCCAAAIIIIAAAgjUs0DZdZDO\nO+881ziD3Wb3sY99TI477jjp7u52jTR873vfc3WTrLEGOgQQQAABBBBAAAEEEEAgagJlB6S2\ntjb57W9/K+eee65cddVVoeOdNm2aXH311fKRj3wkNJ4BBBBAAAEEEEAAAQQQQCAKAmUHJDuo\nuXPnyp133ikvvfSSPP300/Laa6/JfvvtJwcffLBr8jsKB84+IoAAAggggAACCCCAAAKFArsV\nkPxK5s+fL/aPDgEEEEAAAQQQQAABBBCYDAJlN9IwGQ6aY0AAAQQQQAABBBBAAAEESgkQkEqp\nMA4BBBBAAAEEEEAAAQQaUoCA1JCnnYNGAAEEEEAAAQQQQACBUgIEpFIqjEMAAQQQQAABBBBA\nAIGGFNijRhq8mD0gdv369WJ9a+EuFov5SfQRQAABBBBAAAEEEEAAgcgI7HEJ0s033yxz5syR\nefPmuRbt7PUPf/jDyACwowgggAACCCCAAAIIIICAF9ijEiR7BtLZZ58tH//4x+Xkk0+W7u5u\neeCBB+TDH/6wHHbYYXL44Yf77dBHAAEEEEAAAQQQQAABBOpeYMyAtGHDBpk6daq0trYWHcz9\n998vS5culauvvjo77dRTT5W7775b7rnnHgJSVoUXCCCAAAIIIIAAAgggEAWBMW+xu/fee2X/\n/feXf/3Xf5Xh4eHQMS1cuFAef/xxsXlGR0dl27Zt8n//93/yyCOPyOLFi0PzMoAAAggggAAC\nCCCAAAII1LvAmAHpPe95j3z1q1+VSy+9VJYsWSLf//73XRiyA3vb297mbrGzfjKZlGnTpsk7\n3/lOsWXe8Y531Puxs38IIIAAAggggAACCCCAQEhgzICUSCTkox/9qKxcuVI+97nPyRe/+EVX\nv+jWW291K7r22mtl8+bN8qMf/UhuuOEGWbFihdx44420ZBdiZgABBBBAAAEEEEAAAQSiIDBm\nQPIH0dzcLJ/+9Kflueeec4HpU5/6lCxbtkx++tOfSldXl5x55pny13/9126cX4Y+AggggAAC\nCCCAAAIIIBAlgXEHJH9QbW1t8oUvfMEFpTPOOEM+9KEPyRve8Ab55S9/6WehjwACCCCAAAII\nIIAAAghEUmBcAckaYPje974nn/zkJ+UjH/mIXH755e5gv/KVr7igdNJJJ8lf/uVfuqa+rWU7\nOgQQQAABBBBAAAEEEEAgigJjBqR0Ou0aXrBgZM13r169Wi677DI58sgjZcuWLTJjxgw3vGbN\nGtes9ymnnCKnn3662DOS6BBAAAEEEEAAAQQQQACBKAmMGZD+8Ic/yK9//Wt59tlnXUMN9913\nn7z88svS3t4u119/ffZYZ8+eLf/8z/8sq1atkrlz58pPfvKT7DReIIAAAggggAACCCCAAAJR\nEBjzQbH2bCNrhGHevHnZ47EGG+wZSFu3bs2O8y9s/HXXXZdtCtyPp48AAggggAACCCCAAAII\n1LvAmAHp+OOPF2uYweoZvfWtb5WRkRH53e9+J1ay9E//9E87PT5rHpwOAQQQQAABBBBAAAEE\nEIiSwJgBqbu7W+y2uh/84Ady++23y4YNG+Tggw+Wu+66S5YuXRqlY2VfEUAAAQQQQAABBBBA\nAIFdCowZkGzpRYsWyZe+9CX3b5drYyICCCCAAAIIIIAAAgggEGGBMRtpiPCxsesIIIAAAggg\ngAACCCCAQFkCBKSyuJgZAQQQQAABBBBAAAEEJrMAAWkyn12ODQEEEEAAAQQQQAABBMoSICCV\nxcXMCCCAAAIIIIAAAgggMJkFCEiT+exybAgggAACCCCAAAIIIFCWAAGpLC5mRgABBBBAAAEE\nEEAAgcksQECazGeXY0MAAQQQQAABBBBAAIGyBAhIZXExMwIIIIAAAggggAACCExmAQLSZD67\nHBsCCCCAAAIIIIAAAgiUJUBAKouLmRFAAAEEEEAAAQQQQGAyCxCQJvPZ5dgQQAABBBBAAAEE\nEECgLAECUllczIwAAggggAACCCCAAAKTWYCANJnPLseGAAIIIIAAAggggAACZQkQkMriYmYE\nEEAAAQQQQAABBBCYzAJN9XJwr7zyitx3332SSCRk+fLlMnfu3NCurVu3Th544AGZPn26m97Z\n2RmazgACCCCAAAIIIIAAAgggsKcCdVGC9OUvf1nOOeccWblypdxxxx1y9tlny+9+97vssd14\n441u3FNPPSW33nqrfPKTn5TXX389O50XCCCAAAIIIIAAAggggEAlBGpegvTss8/Kb37zG/nv\n//5v2WuvvdwxXXzxxXLVVVfJCSecIFZydMMNN8iVV14pS5culZGRETn//PPllltucf1KILAO\nBBBAAAEEEEAAAQQQQMAEal6CZCVB5557bjYc2U4tW7ZMNmzYIOl0Wh566CF3u52FI+uamprk\ntNNOk1/+8pdumP8ggAACCCCAAAIIIIAAApUSqHkJ0hve8Aaxf/nd3XffLQcffLDEYjFZv369\nzJs3L3+yC0ybN2+WVCol8Xgu49m4Bx98MDTvQQcdJDNmzAiNq8VAMpl0m7W+BT+6nICFXjvX\nuORM7JW5WNfc3Ozq5rkB/uME7O++paUlawRLIOA/D82GLixgnzGtra3hkQy5z15jwKb4zcB7\nprSJjbXPGt4zYR+rQ49L2MSGzMU6u6aJ0num5gHJqeX9x26d++Mf/yj/9m//5sZaSVJ3d3fe\nHCJdXV0uHG3btk2mTZuWnWZ1mD7/+c9nh+3FFVdcIfvvv39oXC0HOjo6xP7RFQvgUmxiYwrf\n/6XnaryxFhzpSgvkfy6WnqMxx+Ky8/OOTWkbXEq72I+92JS24buptIuFo3oISENDQ6V3sGBs\nXQWk66+/Xm666Sb5+te/LkuWLHG7an+EVu8ov/PD7e3t+aNl3333lYsuuig0bp999hELUrXu\n7Dhsf/v6+mR4eLjWu1NX27c/mNHRUVwKzoq5WElAb2+v8ymY3NCD9rc0MDDgfihpaIiCgzcX\n+6yph8+8gl2r+aC1fGp/S3RhAfvB0bqenp7wBIaE90zpN8GUKVPc97Vdz9DlBKykxL6zccmZ\n2Ctzsb+lwcFB970dnlqbofGE2LoISHar3Le//W2566675PLLL3d1kDzZzJkzZe3atX7Q9bdv\n3+5+uSi8jWT27Nmutbv8mbds2VIXb9a2tjYXkCy59vf35+9iw7+2Px4LvbiE3wr+w9aCAKE6\nbGN/++bifywJT23cIf/rnP0tcctq+H1gJdRcuIRNbMg/MgObYhveM8UmdtuhBSS7buM9E/ax\nH6fsNjJcil3sc8auY+rBxq6txtPlKvCMZ+4qzXPJJZe4Zr2vueaaUDiyzS1evFieeeaZ0IXQ\nk08+WVQvqUq7xmoRQAABBBBAAAEEEECggQRqHpDuvPNOV3J0jj4HyYr4rf6R/2e3Xb3lLW9x\np8NuvbNfLJ577rnss5Ia6DxxqAgggAACCCCAAAIIIDABAjW/xe5HP/qRO8xvfetbRYf785//\n3N2WZiVM9mwkC0l2q9qZZ54py5cvL5qfEQgggAACCCCAAAIIIIDAngjUPCBdd911Y+6/PRfp\ntttuk40bN8qsWbNCTXuPuTAzIIAAAggggAACCCCAAALjFKh5QBrnfrrZ9t5773JmZ14EEEAA\nAQQQQAABBBBAoCyBmtdBKmtvmRkBBBBAAAEEEEAAAQQQqKIAAamKuKwaAQQQQAABBBBAAAEE\noiVAQIrW+WJvEUAAAQQQQAABBBBAoIoCBKQq4rJqBBBAAAEEEEAAAQQQiJYAASla54u9RQAB\nBBBAAAEEEEAAgSoKEJCqiMuqEUAAAQQQQAABBBBAIFoCBKRonS/2FgEEEEAAAQQQQAABBKoo\nQECqIi6rRgABBBBAAAEEEEAAgWgJEJCidb7YWwQQQAABBBBAAAEEEKiiAAGpirisGgEEEEAA\nAQQQQAABBKIlQECK1vlibxFAAAEEEEAAAQQQQKCKAgSkKuKyagQQQAABBBBAAAEEEIiWAAEp\nWueLvUUAAQQQQAABBBBAAIEqChCQqojLqhFAAAEEEEAAAQQQQCBaAgSkaJ0v9hYBBBBAAAEE\nEEAAAQSqKEBAqiIuq0YAAQQQQAABBBBAAIFoCRCQonW+2FsEEEAAAQQQQAABBBCoogABqYq4\nrBoBBBBAAAEEEEAAAQSiJUBAitb5Ym8RQAABBBBAAAEEEECgigIEpCrismoEEEAAAQQQQAAB\nBBCIlgABKVrni71FAAEEEEAAAQQQQACBKgoQkKqIy6oRQAABBBBAAAEEEEAgWgIEpGidL/YW\nAQQQQAABBBBAAAEEqihAQKoiLqtGAAEEEEAAAQQQQACBaAkQkKJ1vthbBBBAAAEEEEAAAQQQ\nqKIAAamKuKwaAQQQQAABBBBAAAEEoiVAQIrW+WJvEUAAAQQQQAABBBBAoIoCBKQq4rJqBBBA\nAAEEEEAAAQQQiJYAASla54u9RQABBBBAAAEEEEAAgSoKEJCqiMuqEUAAAQQQQAABBBBAIFoC\nBKRonS/2FgEEEEAAAQQQQAABBKooQECqIi6rRgABBBBAAAEEEEAAgWgJEJCidb7YWwQQQAAB\nBBBAAAEEEKiiAAGpirisGgEEEEAAAQQQQAABBKIlQECK1vlibxFAAAEEEEAAAQQQQKCKAgSk\nKuKyagQQQAABBBBAAAEEEIiWAAEpWueLvUUAAQQQQAABBBBAAIEqChCQqojLqhFAAAEEEEAA\nAQQQQCBaAgSkaJ0v9hYBBBBAAAEEEEAAAQSqKEBAqiIuq0YAAQQQQAABBBBAAIFoCRCQonW+\n2FsEEEAAAQQQQAABBBCoogABqYq4rBoBBBBAAAEEEEAAAQSiJUBAitb5Ym8RQAABBBBAAAEE\nEECgigIEpCrismoEEEAAAQQQQAABBBCIlgABKVrni71FAAEEEEAAAQQQQACBKgoQkKqIy6oR\nQAABBBBAAAEEEEAgWgIEpGidL/YWAQQQQAABBBBAAAEEqihAQKoiLqtGAAEEEEAAAQQQQACB\naAkQkKJ1vthbBBBAAAEEEEAAAQQQqKIAAamKuKwaAQQQQAABBBBAAAEEoiVAQIrW+WJvEUAA\nAQQQQAABBBBAoIoCBKQq4rJqBBBAAAEEEEAAAQQQiJYAASla54u9RQABBBBAAAEEEEAAgSoK\nEJCqiMuqEUAAAQQQQAABBBBAIFoCBKRonS/2FgEEEEAAAQQQQAABBKooQECqIi6rRgABBBBA\nAAEEEEAAgWgJEJCidb7YWwQQQAABBBBAAAEEEKiiAAGpirisGgEEEEAAAQQQQAABBKIlQECK\n1vlibxFAAAEEEEAAAQQQQKCKAgSkKuKyagQQQAABBBBAAAEEEIiWQFO0drf8vU0kEtLW1lb+\nghVeorm52a3R9yu8+kivrqmpSWKxWKSPoRo7by7WtbS0iH9dje1EcZ32d93a2iqjo6NR3P2q\n7XM8HvzmZTZ0YQH7jKmH74LwXtV+yH/2YlN8LnjPFJv4MfZZw3vGawR9+16ql2vO8J7VdshM\nrLPrmCi9ZyZ9QLIPOH/RUMu3iP8Ssn3xb5Za7k89bdufI1zCZ8W/b30/PJUhXIrfA/5zhr+l\n0ja4FLv4Mdh4iVzf/p5wyXnkv8ImXyN47b+TeM+EbbyL9evBJp1Oh3dwJ0OTPiCNjIzI4ODg\nTg5/4kZbarZ/AwMD0t/fP3EbjsCWurq6xM4TLuGTZV9AVuJoLsPDw+GJDT6UTCalr6/PvW8a\nnCJ0+PZ+sV/pduzYIeP9EgitYBIP2Odvb2/vJD7C3Tu09vZ2tyA2xX68Z4pN7HvJvrOt9J73\nTNjHvpcsAOBS7NLR0SFDQ0N1YWPnqLu7O7yTJYaog1QChVEIIIAAAggggAACCCDQmAIEpMY8\n7xw1AggggAACCCCAAAIIlBAgIJVAYRQCCCCAAAIIIIAAAgg0pgABqTHPO0eNAAIIIIAAAggg\ngAACJQQISCVQGIUAAggggAACCCCAAAKNKUBAaszzzlEjgAACCCCAAAIIIIBACQECUgkURiGA\nAAIIIIAAAggggEBjChCQGvO8c9QIIIAAAggggAACCCBQQoCAVAKFUQgggAACCCCAAAIIINCY\nAgSkxjzvHDUCCCCAAAIIIIAAAgiUECAglUBhFAIIIIAAAggggAACCDSmAAGpMc87R40AAggg\ngAACCCCAAAIlBAhIJVAYhQACCCCAAAIIIIAAAo0pQEBqzPPOUSOAAAIIIIAAAggggEAJAQJS\nCRRGIYAAAggggAACCCCAQGMKEJAa87xz1AgggAACCCCAAAIIIFBCgIBUAoVRCCCAAAIIIIAA\nAggg0JgCBKTGPO8cNQIIIIAAAggggAACCJQQICCVQGEUAggggAACCCCAAAIINKYAAakxzztH\njQACCCCAAAIIIIAAAiUECEglUBiFAAIIIIAAAggggAACjSlAQGrM885RI4AAAggggAACCCCA\nQAkBAlIJFEYhgAACCCCAAAIIIIBAYwoQkBrzvHPUCCCAAAIIIIAAAgggUEKAgFQChVEIIIAA\nAggggAACCCDQmAIEpMY87xw1AggggAACCCCAAAIIlBAgIJVAYRQCCCCAAAIIIIAAAgg0pgAB\nqTHPO0eNAAIIIIAAAggggAACJQQISCVQGIUAAggggAACCCCAAAKNKUBAaszzzlEjgAACCCCA\nAAIIIIBACQECUgkURiGAAAIIIIAAAggggEBjChCQGvO8c9QIIIAAAggggAACCCBQQoCAVAKF\nUQgggAACCCCAAAIIINCYAgSkxjzvHDUCCCCAAAIIIIAAAgiUECAglUBhFAIIIIAAAggggAAC\nCDSmAAGpMc87R40AAggggAACCCCAAAIlBAhIJVAYhQACCCCAAAIIIIAAAo0pQEBqzPPOUSOA\nAAIIIIAAAggggEAJAQJSCRRGIYAAAggggAACCCCAQGMKEJAa87xz1AgggAACCCCAAAIIIFBC\ngIBUAoVRCCCAAAIIIIAAAggg0JgCBKTGPO8cNQIIIIAAAggggAACCJQQICCVQGEUAggggAAC\nCCCAAAIINKYAAakxzztHjQACCCCAAAIIIIAAAiUECEglUBiFAAIIIIAAAggggAACjSlAQGrM\n885RI4AAAggggAACCCCAQAkBAlIJFEYhgAACCCCAAAIIIIBAYwoQkBrzvHPUCCCAAAIIIIAA\nAgggUEKAgFQChVEIIIAAAggggAACCCDQmAIEpMY87xw1AggggAACCCCAAAIIlBAgIJVAYRQC\nCCCAAAIIIIAAAgg0pgABqTHPO0eNAAIIIIAAAggggAACJQQISCVQGIUAAggggAACCCCAAAKN\nKUBAaszzzlEjgAACCCCAAAIIIIBACQECUgkURiGAAAIIIIAAAggggEBjChCQGvO8c9QIIIAA\nAggggAACCCBQQoCAVAKFUQgggAACCCCAAAIIINCYAgSkxjzvHDUCCCCAAAIIIIAAAgiUECAg\nlUBhFAIIIIAAAggggAACCDSmAAGpMc87R40AAggggAACCCCAAAIlBAhIJVAYhQACCCCAAAII\nIIAAAo0pUFcBaXR0VL7//e/L9u3bi87GunXr5Oabb5Zf/OIX0tvbWzSdEQgggAACCCCAAAII\nIIDAngrUVUD6l3/5F7n22muLAtCNN94oZ599tjz11FNy6623yic/+Ul5/fXX9/TYWR4BBBBA\nAAEEEEAAAQQQCAk0hYZqNLBx40a5/PLLZcWKFUV7YCVHN9xwg1x55ZWydOlSGRkZkfPPP19u\nueUW1y9agBEIIIAAAggggAACCCCAwG4K1EUJ0qWXXirpdFouu+yyosN46KGHZO7cuS4c2cSm\npiY57bTT5Je//GXRvIxAAAEEEEAAAQQQQAABBPZEoC5KkL74xS/K3nvvLS+88ELRsaxfv17m\nzZsXGm+BafPmzZJKpSQez2W8LVu2yGOPPRaad9GiRTJt2rTQuFoMWLCzLplMuv2uxT7U6zYT\niYTbtZaWlnrdxZrsl3dpbm4Ovc9rsjN1tlH7uzcXb1Rnu1ez3YnFYm7bZkMXFjAbPmPCJjbk\n3zPYlLbBJezi3y/8PYVdbMiu8+y7ifdM2MZf/9r3dZRs6iIgWTjaWbdhwwbp7u4OTe7q6nIh\nY9u2baHw88wzz7j6SfkzX3HFFXL66afnj6rp646ODrF/dAiMV6Dw/T/e5Sb7fISAnZ/h6dOn\n73xiA0/BZecnH5vSNriUdrHPX2xK2+BS2qWtrU3sX627oaGhce1CXQSkXe2plbhYvaP8zg+3\nt7fnjxYrLfrc5z4XGrdw4cKSreKFZpqAATsOe2P09/fL8PDwBGwxOpuwXxSsNBCX8DkzF/u3\nY8cOsRYe6XIC9rc0ODhIaWyOxL2yz0T7ta5US6AFszbcoP0wZX9LdGGBzs5ON4LWYcMuNsR7\nptjExtiPdnYd1tfXV3qGBh1rJSQWHO06jy4nYC72t2Tf2fav1p1V6RnPD6x1H5Bmzpwpa9eu\nDXnal7/dNldYVGe34n3iE58IzWu33dXDl6JPzvbm4I8ndIpckbR92OJS7GLvcUJ12MWG/JeQ\n/7GkeI7GHGPvFwtIduFiXwJ0OQELj/XwXZDbo/p45e9owKb4fPCeKTaxW+ssINmPdrxnwj72\nQ7iFAVyKXexzxn4ErwcbO0fj6XIVeMYzdw3mWbx4sditc/kXQk8++WRRvaQa7BqbRAABBBBA\nAAEEEEAAgUkmUPcB6S1veYsjv+mmm9ztNM8995zccccd7rlIk+xccDgIIIAAAggggAACCCBQ\nY4G6v8XObhm55JJL5OKLLxYLSXar2plnninLly+vMR2bRwABBBBAAAEEEEAAgckmUFcByRpZ\nuO+++4qMly1bJrfddpvYA2VnzZpFk8dFQoxAAAEEEEAAAQQQQACBSgjUVUAa64B21Rz4WMsy\nHQEEEEAAAQQQQAABBBAYS6Du6yCNdQBMRwABBBBAAAEEEEAAAQQqJUBAqpQk60EAAQQQQAAB\nBBBAAIHICxCQIn8KOQAEEEAAAQQQQAABBBColAABqVKSrAcBBBBAAAEEEEAAAQQiL0BAivwp\n5AAQQAABBBBAAAEEEECgUgIEpEpJsh4EEEAAAQQQQAABBBCIvAABKfKnkANAAAEEEEAAAQQQ\nQACBSgkQkColyXoQQAABBBBAAAEEEEAg8gIEpMifQg4AAQQQQAABBBBAAAEEKiVAQKqUJOtB\nAAEEEEAAAQQQQACByAsQkCJ/CjkABBBAAAEEEEAAAQQQqJQAAalSkqwHAQQQQAABBBBAAAEE\nIi9AQIr8KeQAEEAAAQQQQAABBBBAoFICBKRKSbIeBBBAAAEEEEAAAQQQiLwAASnyp5ADQAAB\nBBBAAAEEEEAAgUoJEJAqJcl6EEAAAQQQQAABBBBAIPICBKTIn0IOAAEEEEAAAQQQQAABBCol\nQECqlCTrQQABBBBAAAEEEEAAgcgLEJAifwo5AAQQQAABBBBAAAEEEKiUAAGpUpKsBwEEEEAA\nAQQQQAABBCIvQECK/CnkABBAAAEEEEAAAQQQQKBSAgSkSkmyHgQQQAABBBBAAAEEEIi8AAEp\n8qeQA0AAAQQQQAABBBBAAIFKCRCQKiXJehBAAAEEEEAAAQQQQCDyAgSkyJ9CDgABBBBAAAEE\nEEAAAQQqJUBAqpQk60EAAQQQQAABBBBAAIHICxCQIn8KOQAEEEAAAQQQQAABBBColAABqVKS\nrAcBBBBAAAEEEEAAAQQiL0BAivwp5AAQQAABBBBAAAEEEECgUgIEpEpJsh4EEEAAAQQQQAAB\nBBCIvAABKfKnkANAAAEEEEAAAQQQQACBSgkQkColyXoQQAABBBBAAAEEEEAg8gIEpMifQg4A\nAQQQQAABBBBAAAEEKiVAQKqUJOtBAAEEEEAAAQQQQACByAsQkCJ/CjkABBBAAAEEEEAAAQQQ\nqJQAAalSkqwHAQQQQAABBBBAAAEEIi9AQIr8KeQAEEAAAQQQQAABBBBAoFICBKRKSbIeBBBA\nAAEEEEAAAQQQiLwAASnyp5ADQAABBBBAAAEEEEAAgUoJEJAqJcl6EEAAAQQQQAABBBBAIPIC\nBKTIn0IOAAEEEEAAAQQQQAABBColQECqlCTrQQABBBBAAAEEEEAAgcgLEJAifwo5AAQQQAAB\nBBBAAAEEEKiUAAGpUpKsBwEEEEAAAQQQQAABBCIvQECK/CnkABBAAAEEEEAAAQQQQKBSAgSk\nSkmyHgQQQAABBBBAAAEEEIi8AAEp8qeQA0AAAQQQQAABBBBAAIFKCRCQKiXJehBAAAEEEEAA\nAQQQQCDyAgSkyJ9CDgABBBBAAAEEEEAAAQQqJUBAqpQk60EAAQQQQAABBBBAAIHICxCQIn8K\nOQAEEEAAAQQQQAABBBColAABqVKSrAcBBBBAAAEEEEAAAQQiL0BAivwp5AAQQAABBBBAAAEE\nEECgUgIEpEpJsh4EEEAAAQQQQAABBBCIvAABKfKnkANAAAEEEEAAAQQQQACBSgkQkColyXoQ\nQAABBBBAAAEEEEAg8gIEpMifQg4AAQQQQAABBBBAAAEEKiVAQKqUJOtBAAEEEEAAAQQQQACB\nyAsQkCJ/CjkABBBAAAEEEEAAAQQQqJQAAalSkqwHAQQQQAABBBBAAAEEIi/QFPkjGOMA4vG4\ntLa2jjFX9Scnk0m3Eeun0+nqbzBCW0gkEhKLxXApOGdNTcGfZ3Nzs5gRXU7A/q5bWlrEG+Wm\nNPYrc7HObOjCAvYZUw/fBeG9qv2QuViHTfG54D1T2sTG1su1VfEe1m6MfU/jUuzvr1/s+zpK\nnzMNEZD8RUPxaZu4Mf4NYn0fliZu6/W9JTOxLyJcwufJv2/tQ8W/Ds/RuEPmgUvx+fcXu/wt\nlbbBpdjFj8HGS+T6fC/lLApfYVMoEoRG+27ibyls469f6sVmvIUUkz4gjYyMyODgYPhs1WCo\nra3NJeeBgQHp7++vwR7U7ya7urrEzhMu4XNkLlZ61NfXJ8PDw+GJDT5k4WjHjh3ufdPgFKHD\nty9ms+nt7aVENiQTlJD09PQUjGXQvpusw6b4vWC/duMSdrFg1NnZKaOjo9iEaVwwshDAeyYM\nY99L7e3tMjQ0VBc29qO8XV+N1VEHaSwhpiOAAAIIIIAAAggggEDDCBCQGuZUc6AIIIAAAggg\ngAACCCAwlgABaSwhpiOAAAIIIIAAAggggEDDCBCQGuZUc6AIIIAAAggggAACCCAwlgABaSwh\npiOAAAIIIIAAAggggEDDCBCQGuZUc6AIIIAAAggggAACCCAwlgABaSwhpiOAAAIIIIAAAggg\ngEDDCBCQGuZUc6AIIIAAAggggAACCCAwlgABaSwhpiOAAAIIIIAAAggggEDDCBCQGuZUc6AI\nIIAAAggggAACCCAwlgABaSwhpiOAAAIIIIAAAggggEDDCBCQGuZUc6AIIIAAAggggAACCCAw\nlgABaSwhpiOAAAIIIIAAAggggEDDCBCQGuZUc6AIIIAAAggggAACCCAwlgABaSwhpiOAAAII\nIIAAAggggEDDCBCQGuZUc6AIIIAAAggggAACCCAwlgABaSwhpiOAAAIIIIAAAggggEDDCBCQ\nGuZUc6AIIIAAAggggAACCCAwlkDTWDMwHQEEEEAAAQQQQAABBBAoFBhJiQxl/8Uyr31fJBWP\nycL4qCxqLlyyvocJSPV9ftg7BBBAAAEEEEAAAQSyAmOFkqFULqDkXofHbR0S2TgQk56RuLQk\n0tKhiSCVDsLO4GhMhn3oScdkcFR0OFh+0MbrdNfXaePp/nKfYfnuMeOZs37mISDVz7lgTxBA\nAAEEEEAAAQTqUGCsUDKqJSWJnpRs2Z7QUhQfRnzfgkdMBjRojLgQ4sNGLojkh5LBzPL5ocS/\nHtBpE9nFJC1NusmE/utoSkuzVs5JxNLSMxiX+W2jsldLMC6p45vjaQ1bIu61zmOvW5vicvz8\nFt3lkYnc7T3eFgFpjwlZAQIIIIAAAggggEClBcYKJbkgEgSQ4FavoHQjKAEJBxSbbuHDl47s\nLJS40hGdNxdK7MjGE0x0IWnfYwYLJRZEWvRfUkOHve7ScOJft2TGWRCx0h+bnh9KLKi4aTo+\n//XKnpjc+mKzCzynzxmWg7tT0qHLP62h7ndbErKmNyGf2m9Q3r9wyK2zTQNOTA972S865e1z\nRuSSwwbcsd23KSEXrGiX//fAIXnb7F0Hn2QyKTNnJqW3d3CPXSZyBQSkidRmWwgggAACCCCA\nQEQELEj0a6lHv95SZaUfg6m4rI2Nyvb+lLy2LVxSYqHCh5L817sTSnxASY8rlFQGc6xQ0qpB\nIqlhoTCU+LBiJSWdrUlJDQ2EQokFFAswuf7OXgfztWoo0cKoinfbh0WuWtUh05vT8oM39Mns\nVi3KynTvmjfibqM79w/t8i9rmmVBe0reMTcXfOzWu0brCEiNdsY5XgQQQAABBBCIvIBdtFpo\nceFFw8mAhhgLM75v4/OHbfyAztfnwk7xtPwgZMv26bpKB5T+jN2el5TEMyUlQQnI+EtKfCgp\nDB35pSblhBK7FcxuIduTzkpKOjtb5fXXtXJPHXb//lyLbBmKyz8d3h8KR35XzeCqo/rljPvb\n5V/XtLiAtHkwJv+5tlnPksiT2+LynWdb5L0Lwsd3z6sJ+dWrSenRAHbE1FE5a/6wdCf9WoP+\n0Ghavr9a5A8bW9376uCulLxv4bBM07Dmux+8kJRpybTMa0vJD15oln06UrqtYdkrL8j5eSei\nT0CaCGW2gQACCCCAAAINJZBf+uKDig80/XqbV7/+QG/1SXwwsYrwPtS40OPDjx+fDUG+NGcP\nr+jzzoaVntjtXFZKYrdVdSVTrt+WGbbxrTq9Xa8aZ3a3S4uWIo0O9u+kVMSXmOy6pMQCjBa6\n0E2QwMN6C12rlmbZrXI766x0aem0Ufn1q01iJU5Wkreq105STIdj7rW9N313/WQbdPwAAC5s\nSURBVPPNsqonLst0mZf743K3BqVfbkzKdcf2ufePzbdZ76x7z+398uTrIvt1xOUADUf/tS4p\nt7yYlKs1kB0yRTei3U9eTuoyaXluR0J26C5qppL3aNiqVUdAqpU820UAAQQQQACBmgjkl764\n8KJBxYcX32/aPiqvbk2Gx7tAUzyvX8YHoZ2Xvuze4TZphXe79couILv1V/a99Vd1F1gsuOj4\nINjk5rGQY/P6ZWxeH37yQ4/N58ePZ89iWiFl9uxpMjg4KFu2hEsSxrM889ROYO2OuCudsTpF\nu+oO0XpJv341Jo9vTcifzRqVa47ulyN/3iknzBzN1kHaMKBvHO3W98fkf0/cIYs60pLWQPPV\np1rkf15qlj/pssfN0GSv3befTmo4Ssl/LBc5vqvPjbOSqQ8+2C4XP9kqPzyhL3tL4aNbm+Sc\nfYbkMwcMuhb2ZrflSpjcghP4HwLSBGKzKQQQQAABBBAYW8DqrVjoCG4H83VgcsHEl7rYLWO+\ntGVXt5a5khm3zqDExir3j93Zr9etY87mS198IJmipS8+mOQHFfv1vk2vuqzfnhdMXOmMhhk/\nrwst+fPqMnt6+9eYB8EMk1rASoPsb8mC9VjdfL3FzbrnNFBZQNpVd6aW8Fg4ss6C16naYIMF\npOd1WQtItt3bXorLMbPi8pa5KenpCdY2U1u+e9fcYbHb/p7QW/eOmBps027q/NT+g66e1/z2\nsfc1WFt1/ktAqo5rybW+siMla7SIcVTvAU3orznWTKJvOtE+/Oy1/UoUTPPDwZuu5AoZiQAC\nCCCAwAQLFJa++IDi+pkQYiUoPrAE/Uy4sRKYvFvLwvPkbjErXfdl9w7Uvld9ScmUTOlLu5Wu\n6PewDzW+FMX3bfysKZ0y0t8TLKuBxUKPW077fjnf3709YykEJkbA6gTNbE7JJi25GavbqM13\nW2d/K2N1i7SeUH7nw9U2vR3PunV9tq6YvD6Ylo8/oA19j+R+cHgts50XdB4fkFzJaFA45Zav\n5X8ISBOo/1+rRuSyx4x8d9iDD3MXpHYWrvQD3Kb7sOUCmJs3CFvBtCCU2X2/uXmD5VxY0/Eu\npGXWo4Ou9RVblw9u1oqLX5fbVnYbeeuxeTLrKgqBBfs5Yk9X1r+xYf1C9fNWowWXCTzVbAoB\nBBComIA1dWzPTnH/3OugmeJgWF9nplmdlxF9cGN2fvtcLRi26Ta/n7dlw7ArqXmtp6VEiY0v\nncnUk8nchja+0pfxHb6VvrhbxKzERC+MrPTFhxnrWwDxw9lSGf1u8cEk6Afr8PPmT7Nl7Xtq\nd7pZs5pk06ad19fYnXWyDAK1ErB6Qr/Z1OT+9q0xi511z7k6R1pfqFM/QMborCR0V93WoeCP\nr0svHO3HBfvs8d18bSnP/uUHsanjCGV++Wr3d+dKvdr7NGnXf7QWMX5ocUoGh0dc5bNR++LK\nfLFZZbTRzJebvbbxQT8WzOum+fHBOJvHVQJ189q6MvNmhqMFaW/FrtAu+9ZtXBjLhC0XyDIB\nLQhTQXCz19lh/9otY1+OBSV1eety63bz6zw63g8H28mN88HNrys3b27dbp6CdYf2K7P+IGAG\nYdJtJ2+7IQAGEEBgjwTsnnj/ObltKK0tOOXCgw8I2TDhPk+Dz+Rgmi6rgcCFj8xns/u8znwW\n2/NR/LrdeAseNi37WR1sK1g+mNemuWE/r+2fmz9/Xtvn/HXrB1RVOx8A7Jeq4s5+MPO3g9nF\nS5vVffHBxUKNfnTbLWN+nO/7cOOHXT8TgoJSmkwwGuMCq3iPGIMAArsjYAHJGlC4SVuLO2dx\n6cYPXuqLyZ3rm/QBsCk5LNN4wu5syy9jAci6RV1xueIYu8UueI6SjbPPPrsGGqtOlM1bi46A\nNIHqfz6vSZZpcWR//8Q9LMvegHYvt32B+0BmFw02zoWyzBexm54Z57+cbbqbV+ex8OaDm1+X\nW17H2xe+X1d2Hjd/LsgF0+0PIvjizw7rshJP6oVFSoZ1Z4N1BxcIQWAMr9u2ZRchA26btq78\n9VX7QqLab5Zc2PLBqTmxQ2LpFi2gbnaVGDVLBX09VHsds5I9+4DJjM9/beMs0FlpXPA6+CAq\nXEc8M0/+eL8+W9aCatAPtm0fZnZN49brXuvNMG6+YJyf182XGV+4n34e1y+YJzgu2254fX4/\ndXaZohUPBgZ0z0YT2Q/YYH7d19C+ZY5ZF7J9tv3w0wv3wdab75W/b/mv6/XDXHe/op3dRmV/\nj7mL/vywYOODv73sBb6f15bLCw+hgOCmBX/fuRCSGc5MKyzxKBz2f/O2fGjdblj3Sfu23/5z\nKodin7uducEav7K/K/sV1/2I4vrBDzId+q3cFE+5563kTyuaV9+wbpwum9T3tZ/XvS5cr59X\n+00aUuxHGvuRx/ozpk1x4Wag53VXUuNLYfytZvaZQocAAtEX+CutL3T3xibXhPdB2hDDGzKN\nKPgjsx+QvvxEq/tuvzjzQFg/zT4H7DO33G6+NrIwS+sb/ezFEdl+RHAt4tdh2/qVtpZ36wlB\nIw9+fL30CUj1ciaqtB/2JWj/gk6vGoq6UuOKZqrqiK6upN6XasHRP1th9zdngS64aMpdIPnQ\n5vr6B+6DYjakuQuq/NI3W0cuFGbn82EscwEWjM9cJOp2cyEyP7QFF3+5aeH9svG2v35d1ncX\ndzo+HUvYnYfu1kO/vM1rF64pjTx2rNa3ZYLXmWluvBlO5isbk2mxg6xJ50OjCbtQqi/szyz3\nOgiNNmzjfVi0+YOwtfOg5uf32ygd5Gy9uW349ba1jEhML4B7+1v1yyztgkr+34MLF0oXBJ9c\nwHHz6Ps7F4asqqyttR67TBjQ3QtCQxAs7HW7HruVeLjPPZ1uocGePG/9jtZmSQ0PuYDgwkTe\nND+/nzcbRvLWlb+t/DBSOK8PHzvbDz9/vQTtvfYKinBeTdjfFB0CCExWAXvO0ZXL+uUjD7XL\nxx9uk3dqc99HaamStYr4zPa4/K82s21NeV+l8xw7Pdw4g81jzYT/UEufTtrLlzqPLWWfrX93\nyLB84dFm+ZjWQfr4PgltKj4tP9+QlP9bn5SP7DOYbeRh7LVN7BwEpIn1ZmtVFrCLjmb7Z1eZ\nrtO0UNSVGlc0U81HdHV16UPnOmTz5s0yPFy6OHysnbTw5AJVJjRZkHKBShf0QSt/un9tQkHw\nirmQtrMQ5tbnw1pmG34ddrlVuI3cenWan64X5vba71v+a5s/WF8wj193S2ur9A8OaShN5QVE\nncf2oWA//DZ9qAwdf968/lhseb+Mjcvtp73ObcPmsXnD6wvvpz8WK9WwBnH9NtJ2zAXrtvUF\n69I38G53thb78ir90W71PXIX9/bjSVCaYLdDuQt6DQTuAl7/foJhm8deB0Ejd/EfTPcX+8E8\nQTDJzZMrqfDL+237dedvyy8X9HP76Zbx+6X7YdvanW7WrE6tT7JtdxZlGQQQQGBSCFhjDf92\nTL/8y+pmuf2VpNyuIcV3y6aOyCf3HyoqWbLpn9h3SL67skX+6ZlW/Y4Y0Nbw7NttfN0756Vk\n+pRWufDBATnv4eDhwvYDn7ViZ9ur1670t2i97i37hQACZQlYCYb923lnF9S76saavqtlqzdt\n2rQOvZd5REsex/9LVvX2pvJr9mHR9ItDWC6o2fT88No1Zaq0NDfL669t0ohkISMILUFACUqv\nKr+3rBEBBBD4/9s7Eyi7ijKPf6+3rEA2lizsGHYkCAMEZQBBFhE8iBwYBMkwyqIoMHoEVGYg\nBwdEBGRzw+ggBpDdEQQVRwhbgLAcSDKgCGELkIWQTjrp7U396726fd/r1/1ep/u9vrfzq5zO\nrapb9dV3f1Xv3vpuLRcCaSGgLbYv3Hmtnb9j7ltD2v57ojN43GSeHt0JW7bZcVu0mTZd0Mdk\n9TL6hUPze3bHcmlr7lLxR23VYAdNMHttSbMtd6NUk1x5+uhw3N02PfeNpHjcYPqL1BtMVSgb\nAhCAAAREQA8fGTU9u9KG67jRGRs2rM6GrdJH+0qn6VkmZyAAAQhAYH0hoBH83LeGKntW6GXr\neGdc9ccpf39l9Kf8vuRdx8kKfSmCtBCAAAQgAAEIQAACEIAABNJBAAMpHfWElhCAAAQgAAEI\nQAACEIBADQhgINUAMkVAAAIQgAAEIAABCEAAAukggIGUjnpCSwhAAAIQgAAEIAABCECgBgQw\nkGoAmSIgAAEIQAACEIAABCAAgXQQwEBKRz2hJQQgAAEIQAACEIAABCBQAwIYSDWATBEQgAAE\nIAABCEAAAhCAQDoIYCClo57QEgIQgAAEIAABCEAAAhCoAQEMpBpApggIQAACEIAABCAAAQhA\nIB0EMJDSUU9oCQEIQAACEIAABCAAAQjUgAAGUg0gUwQEIAABCEAAAhCAAAQgkA4CGEjpqCe0\nhAAEIAABCEAAAhCAAARqQAADqQaQKQICEIAABCAAAQhAAAIQSAcBDKR01BNaQgACEIAABCAA\nAQhAAAI1IICBVAPIFAEBCEAAAhCAAAQgAAEIpIMABlI66gktIQABCEAAAhCAAAQgAIEaEMBA\nqgFkioAABCAAAQhAAAIQgAAE0kEAAykd9YSWEIAABCAAAQhAAAIQgEANCGAg1QAyRUAAAhCA\nAAQgAAEIQAAC6SCAgZSOekJLCEAAAhCAAAQgAAEIQKAGBDCQagCZIiAAAQhAAAIQgAAEIACB\ndBDAQEpHPaElBCAAAQhAAAIQgAAEIFADAhhINYBMERCAAAQgAAEIQAACEIBAOghgIKWjntAS\nAhCAAAQgAAEIQAACEKgBAQykGkCmCAhAAAIQgAAEIAABCEAgHQQwkNJRT2gJAQhAAAIQgAAE\nIAABCNSAQEMNyhiQIhYtWmSPPfaYjRs3zqZPn26jR48eELkIgQAEIAABCEAAAhCAAAQgEAik\nYgTppptuspNOOsnmz59vt912m51xxhm2fPnycA0cIQABCEAAAhCAAAQgAAEIDAiBxI8gaeRo\n1qxZdvXVV9vuu+9u7e3tdvrpp9utt97qjwNCoUZCWu+5y+rvvtP82Fcm01VqN787F53O+7ul\nyWd38Vl5o/Mhb16AP4Q4nzAnu6/p8wplI3l5+aHsKBjKykf4g+Ly4ZBeRznF19dbvfOO7Oz0\nUVnFRcnljwJd/lJxQV78KEHK3kN6X1ap9EEBX3RehtL1KM8nzJUV0sTSZ7vJy6f3IkvIV/qm\nRls7bJg1tLZZRmzqlCefNpN/t+FZKV6ndC6eJheO2kdg0EO6nAwn14tTXrMcn5y/S7bC+Tiv\nT7HfhZ3L1WPpc126uoSRPs5fl7uugnqJ9O7SLdvcbJnVq62usyOXXwW6dAX6FusW5+fT5+V5\n+dIz5M/J8hyic4rL6Vagr86HP8nEQQACEIAABEoRyPqnce5M3B/SVhqn9GXT5suKFekyhZJi\n3lJx5eXnnvLF6UrJcnENjZYdMaKr7JT4Em8gzZ071yZNmuSNIzFtaGiwww47zGbPnp06Aykz\nZozZ5CnWqc6ub9z5xuQP7r980Hvkj/8AYukzPaUPAkqed5E+Pi+3U4F8hD+4//JB73HldfsB\nFMcVpHfigoCgdzj6U1FilyzvD0d3XmXJSMIVEmhzQf1IE/9DLVS76iFnFtnwqpfS9wJ8y44b\nTAX+LgMvZ+w6+fnzhflkdKns/DHI0DEel/d3GYUZa3P3xw73wmFE/mWDpPgyvCf2n5cVC8tb\nFNdlpMbTecW6pfUpivIXy+t7mlBWrPwKyoj0DtnyeVqbhtnw1tZcbF605xmlK/L0lqZYD2WN\n4or1jgTFiitOo7tncVwIx/JHKoZzIcIrkAtEyStL0zpypM83bPWqXP7oVh3u04qO+4vCsfu4\nFxCFu/LkX+NFYiKPT9KVzuePysrH++J6SlOkS06BAjG+rEin3KlMPBz3e3FdZbU2NdmI1rVO\nREyXeJpif7lwJKerjAJdKs6vhM55eU6WP4aK75KdT+QRdKXPxXaFc+lz/QrF9pI/f26Va6tZ\n93JqlO/LxOSFvDExkbyCuFggYhL7efjTeT3kDy6WNkQV1E1BluLrUI5YgnKyQgGl0nlRMVku\nHOqx3fk3CHk5RgTW/tM+Zl+cEYXT4El8v+udd96xyZMnF7CUwbRkyRJvaNTl3zgrwYoVK2zB\nggUFaSdOnJiI9Uoy7Br/+UBb+7G9rH2tu+HiIgIj3QO6o6PD1gYuoXMXvzF5f+yG1+2cExfF\nuXQ+aW/pQxqpEfzF6WPhKE2p9Ln8hZ2AIDOkz8vywXAuH+cPisuH82mGDx9uwxobrdm16w43\nchqdV7qQNitj22VwYf94jM7l0+TP5dIrLpe2W/5wfaXy+3NF8lycf6CqfDmfL3+Mp4/O5fPH\nw/IX13W4HsnwRrxLo2uTfC83F25sbLD2tnb/kFZ017k8j0iHcC5fvhfjM7gs+RcVxfl9lnyZ\n0Tnn8TqoqLw/HPPpdcjJ1LFE+pj+OVlBjjoG8sfyF+vfIc4hjU8YlVH4wiSbuy+631Pkgt4h\nIhYOXapwaigfRbBxKF/gOl5baClN65h/KGdTmxmsTpJ+5d5FhneJcDinfpDul/pBh7h88tLh\n/C8/ugHEwj3mj6Vx/RmNqIdB9VzBQb+o4OApOB0F4uVEeuguFysnkhBLEHkjjxMZ84c88biC\n0/lAqTjljecrJatUnJOl+sq4f3XuBZX6MyXlKO86yS9QNqdBKTkVyc/LWuNmYLgZWnKdW23t\n3jjmXjn22O58SvdfVG5eTvNKq1/0unVO2dyyG22kBLmU+YMC6qfXbbutmyxUb03upcNgu0x0\nDb1rMli//d61ip1dvHixbbjhhrEYZ51vsIHvBMggGjt2bHTupZdeshkzZkRhea666io7/PDD\nC+IGM6DNJdhgonQNwKU0F91ycBBIEoGsDPa4C0ZXOJY657sQ7kT0BJY/BPLHEFT+6JwCzkXh\nWKLgzZ/LxoUXnYsKDvGlZMbPBVmh3Ohc5IlE9qpbj3K8AvovurZQVD4yf8iXFxUbeaJ8uYRd\ncnw4Shbyh2OUWhniAed3vZrQsYk6EfmIbvFKHkXm5ChYFNcVjKWNvJGnK39QIGQMSaL47mWE\nUzn98xm65euKV0fWu1JlhLhciq7ricsLaeJxSh/i4/54XIh3HcZKO2nKgoPAQBNYc+PPrKNl\njWVXrrSmrbex4af86zoV0fH6a9Z615024ohPW/3U7cvKGJGAqXatYSZBGW0TbyA1ujfoWncU\ndyGskYe423zzze20006LR/nRp2a3XmGwnUaQNCKwZs2abtcz2LoNdvl6o6Bph6FeB1ufpJQv\nLvpb7dba+GmZSVEsAXrot6QRx2xhjzIBmg2uCuKie00S7nnVIxH1Sn1/2JdT3AktUbieF/ot\n4QoJhOcobPJcYvcU2kxhWwkhvczU81r9GVwXAY2U6JmdeC6arfPk42b77mfupmjtj82x5iOP\nikaRuq6oAt/4CWb/9mVrUdIe+triot+SDJNKjZMKSl7nJOpPqZ7KucQbSBMmTLDXXnut4Do+\n/PBDP3I0zC1gjzsZSOeee248ypYtW2YrnYU82E5WczCQWlp8UxpslRJTvkYEdbOFS2GViEsw\nkNratBoJFwjICFCHDqM6EMkd9UIpGEgYj4VsdP9NwrOgUKvBD4U3urDpXhe0me5MNPIlA0nT\nyGgzhXx0/5UxkHQuDU/PtRHOSFq9zbZuELnTRs59wtY88ldr22d64QW5UGbpUmt0xlT9e+9a\n1vW5OzedaK377Ot21coNUGTef9+annrS2qbtYZ0TJ0X5M66f3ujKqXv/PatzI1Vrp0y2tqk7\n2MottozSND46x2/e0DF1qivjCat/8w3r3GBDa991N+v4yNQo3UB7NNWveGZaqTLcxNVku623\n3toWLlxY0BHSVLridUnJvgq0gwAEIAABCEAAAhCAwOASkOHSOcoZudtuZx3bTbVOZ/A2PuFG\nlIpcxhk3o6650hqfecqyzqjQzrFND/3RRl11hckAkqtbvsyaHv5fbwiF7PWv/t1GXXm5j89o\nQMBt6NH2l4fMrr/GGh99JCSzxmefsabHH7WRN1xrTXMeNtfRt8YXnrMRN/7UGlyZg+0SbyAd\nfPDBntHNN9/spxm9+uqrdt999/nvIg02PMqHAAQgAAEIQAACEIBAGghoxKfBzcpqdyM++sSK\n/tqnfczqF79jdW49Udw1PTXX3Jw4W/21c23NSadYy6luKt0XTra6D1dYw/PPxpMW+JsevN8Z\nRZ226hvfspZTTrW20860UVdf53Y8cZuVPfFYQdr6NxZZmxsxav72f3j5q87+dz/VzxtMBSlr\nH0j8FDtNo5s5c6ZddNFFJiNJ0wGOOeYYmz69+1Bg7fFRIgQgAAEIQAACEIAABJJPoPEZZ/Q4\n1/axPSNl5W9yU+yanPGyZsutovismy6obzDWv/JyZFB17LCTrbzw4miKXZQ45mk95DBveGVH\nd214nnFLBmybbS3z5huxlG6Gnyuj9eBDvfGkE9kxY61j8hSrf/utgnSDEUi8gSQo06ZNs7vv\nvtveffdd23jjjf0cz8GARZkQgAAEIAABCEAAAhBIHQFn7DQ+87Rf91PvRov0F1zWrSlqeOF5\nyxx5tGVHjfLRbXvtbY3PzbMRt99q2Xvvtvbt3JQ8ZyC17bJryFbyqKl7tmqVG2V6zureXWwN\nS5fYKrcjtTl/xq0FjbvsRu77oG5kKe6ybvqfJWDddaFWcQ0T6N90000TqBUqQQACEIAABCAA\nAQhAILkE6v9vodXlNy0bfvedJRXVBg5t7pudctnx4231Wef4zRbqF7xkDQsXWOP8l2zY/f9j\nLSee7NYvfaQHGU/Z8HvutIwzcjrHjbfslCnWeOBB1qp1Tm4qX9xl3cYW3Vxso9Ju52oYkSoD\nqYZcKAoCEIAABCAAAQhAAAJDgoA2Z8i6XQhXnfdty7rd4gqc20xh9H/NtCa3m1zb/gdE3/TK\nuu2wW2Uw6W9NizW8+KINv+M2ZyT93hlPZxeIUCDjDDCd75w02VpOPsV9PHaMaXe/jdyO1K1z\n3AYN4QPw3XImLyLxmzQkDxkaQQACEIAABCAAAQhAIB0EMs0rrWHBfL9znZ/W5tb+uPUqXX9u\nWl37zrtY3bKlfs2Rrmr4zf9to5zR5D46mLvI4SOsfc+9rHOziSZ5pVzdO29bxn1LrH2nnb1x\nFNJ0vvdebvTIbSueFoeBlJaaQk8IQAACEIAABCAAAQj0kUDDvGf8hgttbse6npzWHMmFneY0\ncqStvUfccrPVu6l1dYtet6Y/3Gf1zghq3233kmI6N9nUbwmu0ar6hfMt47YBr3t2nrVcdolP\nr2l37iNaJfMmLZIpdkmrEfSBAAQgAAEIQAACEIDAABFofNp9y8hNdWvvZYMFba7QOXasH2nK\nrFhhnVM2t9ZDD7cm9w2jkW70SS7bNMzWuil4rYcdUVKz7Jgxtub4f/HfOxrxq1l+NEmbPjQd\ne5ytbXeGkRuV0neSqvkh2JKKrUOkGwlzY2FD2C1btsyNDuaHBwfxOrU9+RjXcD744ANr0Yez\ncBGBDdz2j+3uA2FwiZB4j7joi+VLlixxG7q4ty64iMBYdxPX18rVbnBdBMaNG2f6NMJit2PQ\nEL+1d110hT7tgPq++wYIrpDAJpts4iPe0xQYXAEB2kwBDh/IuDUsm222me9XqX+F6yKgtTZ6\nZi9fvrwrcij43HM2s+IDZx3lNm4w1wYqcX4anjOKGty9d4Jbg9Tc3Oyf25XkrWaa+vp6C/e9\n3sphBKk3OpyDAAQgAAEIQAACEIDA+krAbcOdHT+hz1cf/w5SnzMnIANrkBJQCagAAQhAAAIQ\ngAAEIAABCCSDAAZSMuoBLSAAAQhAAAIQgAAEIACBBBDAQEpAJaACBCAAAQhAAAIQgAAEIJAM\nAhhIyagHtIAABCAAAQhAAAIQgAAEEkAAAykBlYAKEIAABCAAAQhAAAIQgEAyCGAgJaMe0AIC\nEIAABCAAAQhAAAIQSAABDKQEVAIqQAACEIAABCAAAQhAAALJIICBlIx6QAsIQAACEIAABCAA\nAQhAIAEEMJASUAmoAAEIQAACEIAABCAAAQgkgwAGUjLqAS0gAAEIQAACEIAABCAAgQQQwEBK\nQCWgAgQgAAEIQAACEIAABCCQDAIYSMmoB7SAAAQgAAEIQAACEIAABBJAAAMpAZWAChCAAAQg\nAAEIQAACEIBAMghgICWjHtACAhCAAAQgAAEIQAACEEgAAQykBFQCKkAAAhCAAAQgAAEIQAAC\nySCAgZSMekALCEAAAhCAAAQgAAEIQCABBDCQElAJqAABCEAAAhCAAAQgAAEIJIMABlIy6gEt\nIAABCEAAAhCAAAQgAIEEEMBASkAloAIEIAABCEAAAhCAAAQgkAwCGEjJqAe0gAAEIAABCEAA\nAhCAAAQSQKAhATpUVYW6ujrT32C7lpYWa25utoaGhkToM9g84uV3dnZaNpuFSxyK869cudI+\n+OAD2kwRFwU7Ojosk8nQZorYLF261PR74j5TBMYF29vbaS/dsdjixYt9rNoMrpAAbaaQh0J6\nVr/55pv+t1RfX989wXoeo2dTEvqcSaoG/Y7UZtReksCmUh0yrrFnkwRyqOpyxx132AUXXGAz\nZ8604447bqheJtc1gAQuvfRSmzVrls2ePdv22GOPAZSMqKFK4NRTT7U5c+bYM888Y6NHjx6q\nl8l1DSCB/fff30t7+OGHB1AqooYqAb2423PPPe3jH/+43XjjjUP1MrmuASQwb948O+GEE2zG\njBl23nnnDaDk6ooa/KGV6l4f0iEAAQhAAAIQgAAEIAABCFRMAAOpYlQkhAAEIAABCEAAAhCA\nAASGOgEMpKFew1wfBCAAAQhAAAIQgAAEIFAxgfr/dK7i1CRcZwJaADtlyhTba6+9bMKECess\nh4zrD4HGxkbbbrvt/Hxv1pOsP/XenysdNmyY7brrrn7NGguo+0Ny/ck7fPhw/1xSu8FBoBwB\nbY6j59G+++7rn0/l0nMeAtoUQf3evffe2/eD00KETRrSUlPoCQEIQAACEIAABCAAAQhUnQBT\n7KqOmAIgAAEIQAACEIAABCAAgbQQwEBKS02hJwQgAAEIQAACEIAABCBQdQKsQVoHxKtXr7a/\n/vWv/nsj+ijYxIkTC6Qo7tlnn7U///nP1traapMnTy44HwJvvfWWPfDAA7bzzjuHqIJjufMF\niQkkmkC5NqNvS/zlL3/x7WaDDTawjTbaqOT19NQm9HHQF154wbcnffhx88039x8LLSmEyFQQ\nePvtt+2+++6zhQsX2pgxY0ztIu4qbTP6vs2yZcu63af0EeIHH3zQy1d7K5YfLwt/OgiUazOL\nFi2y+++/35Ru0003taamppIX1lObCYmXLFlit9xyi+22226J+PBj0Itj3whU8tyopM2oz3PT\nTTfZNttsY1oHGdzf/vY3e/rpp+3vf/979Ld8+XKbNGlSSMIxZQTKtZmB6P+q3fzxj3/0H5dV\n/7qn+1S10WEg9ZHwH/7wB/vqV79qK1as8H8///nPTQ+L6dOne0lqHKeffrr97ne/s7Fjx9qv\nf/1r/6VyLWiMu+bmZjvnnHPstddes6OOOip+yvvLne+WgYjEEijXZv7xj3/4j6i98847tmbN\nGrv22mtt6tSp3RYz9tQm1P5OPPFEe/zxx23kyJF25513+o71pz71qYKHVWIBoVg3At/97nft\nuuuu84uhn3rqKf/BYLUJGb5ylbaZ5557zr71rW/ZFlts4TuzoaCHHnrIzjrrLNN3wvUw+tnP\nfmY77rgjHZcAKIXHcm1GHVilGTVqlD3xxBN2zz332IEHHmgjRowouNqe2kxIpDZz4YUX2u9/\n/3s76aSTjM1AApl0HSt5blTaZnSvUtqjjz664EWL4m+99VZ78cUX/cer9QFr9Z0OOOCAdMFC\nW0+gXJsZiP6v+i8XXXSR78uovag/pP71+PHja18L7maHq5CAq/zs8ccfn73tttuiHG4kKeu+\nKJ195ZVXfNxvfvMbn8Z1Zn3YGUDZT3ziE1n3FjjK4x5O2WOOOSZ70EEHZd2X76P44Cl3PqTj\nmHwClbSZL33pS9krr7wy697M+Av65S9/mT3uuOOisCJ7axM33HBD9owzzohguNGq7GGHHZb9\n6U9/GsXhSQ8B3Sv233//7Lvvvhsp7TYb9feVEFGuzbS1tWV/8YtfZF0HOOs6I1n3oiZkzbpR\n7eznP//57OzZs6O4733ve9kvf/nLURhPugiUazOvv/66bwtuZoO/MLUPPXt07wiutzYT0uio\n55/uL3rurV27Nn4Kf4oIlHtuVNJm3GyF7De+8Q3fl1F7cDMcCgh84QtfyP72t78tiCOQXgLl\n2kx/+79upoNvS25mQwRJz6bzzz8/CtfSwxqkPtikmqaibboPOeSQKNe0adO8X1MW5ObMmePP\n6y2d3JZbbmm77LKLHy5UWNNiLrjgAjv88MP9qIHi4q7c+Xha/MknUK7NLF261BYsWODfvGn7\nVLkjjzzST4GZP3++D5drExo1Ovnkk31a/ac3wjvssIOXEUXiSQ0BTUFxnVfbZJNNIp11n9HU\nSfdwsErajKbm6Q2/e7hEo05BmN7yaRQ8PnKt0W61VVw6CZRrM3PnzvWjg7vvvru/QH12whk5\n0XNJkb21mUBFI5e/+tWvzL2QCVEcU0qg3HOjkjZz6aWX+nvSZZdd1o2CM55N0/O23377bueI\nSCeBcm2mv/1f3YP0OZx4H/trX/uanXvuuYMCrGFQSk1podrHvbiitM5IUwzCTUDTpIrn1yr8\n3nvv+atW59W9gfPDhW6koBuJcue7ZSAi0QTKtRl1euXibUZDyZpzqzaj9Wnl2kTcOJIsdXS1\nBu4rX/mKgriUEdhnn31Mf3Gn+4ymwMmIrqTN7LfffnbEEUf4dWjXX399XJTpuzduhMrHydhS\nR+iuu+7yRllBQgKpIVCuzei5VLwWVvccTZnRmgJ9p6S3NiMQboTJT31xI43dZKUGFIpGBMo9\nNyppM+edd55fy+ZGmyK5wSNjWm1L0zmvuuoq0xRxTemcMWMGU78DpJQdK2kz8b6MLq8v/d83\n3njDDyo8+uij/oWNlhx88pOf9M+ywUDFCFI/qGvh4U9+8hO//kMLXtvb2/0DZ8MNNyyQqnB4\nO6s3d73NpSx3vkAwgdQRKG4zeghpUWt8YasuSgvm9VZYri9tQpuCuOlY/ibz2c9+1ufnv3QT\n0Bz+559/3r7+9a/7C6mkzegeo3ZTzl188cX2/e9/33/Ez00FLpec8ykhUNxmZFQXP5d0j1EH\nVmtC5Mq1Ga1T06hmfOQxJThQswyBUs+NStqM+j09ObfswJ/SSJJe1qmjq3VvV1xxRU9ZiE8R\ngeI2MxD93/fff9+vib3mmmv8C0HNbLj88svt5ptvHhQy5Z+gg6JW8gvVjmF6e+LWEUVvXjWS\npDdxaihxp3CYchePx79+ESjVZhobG7u1F1HRNCgNZ/fFffjhh+bm6pqObk2TSTYu3QTcOiL/\ncLjkkkuiUeqBbDNXX321aTc7dX614P6OO+7ocQfFdJNcf7SvtM2E51Ql95l58+b53e80vQ43\ntAj09NwodZ/pS5vRJkF77rlntHvmHnvs4WfbaOaMpvgWG+xDi+rQvppSbUbTv/vb/1W/5803\n3zS3bs2PTIqiXuTovnPCCSfUfMdMRpDWoR1rnqV2oNOOLd/85jejStP0l3Hjxvl1RnGxakyb\nbbZZPAr/ekagpzajKXi6KWgb8LhTmynePj5+vtivqTJnnnmmN7a064vk4tJLQG/29eZMIwE/\n+MEP/PSncDUD1WaCPG0hrmlTaofaCRGXTgLl2ozWMsad7jF6Q1s8eh1PE/yaKSFDSmtNtCui\nDGq573znO/bII4+EZBxTRqC354buM/1pM2pXxc+wMHU4TBNOGS7UdQR6ajMD0f/deOON/chR\nfGTSbf5hLS0t0SysWlYCBlIfaetbNdriVAvHTjvttG659R2Al156qSBei+2L538XJCAwpAn0\n1ma0IFFToeJtRps2qLNTPJe3J0hutzNvHGkL6B/96EeMAPQEKkXxM2fO9MaK2zXIwkYwQf3+\nthl9WuBzn/tcwSYemustA0lvAXHpJNBbm9l66639967CCICuUPecSp9Ln/70p/06gJ122sn0\np82H5LQZjF4K4tJHoNxzo79t5vbbb/fGdJyMpgqrI11sOMXT4E8ugXJtpr/9X+VXGfHnkJYl\naBSpt6Up1SKGgdQHslrQrF1b3La5ttVWW/l1AfrB6y+sMTr22GPtT3/6k8koUiVryormamrB\nNG79I1CuzegDnZqKMGvWLL+IVR1VfVtLO0zpbUolTnO61bl1Wzf7TlBok1oki0sfAX3IU/eQ\nU045xb/BDfWpo+q5v21G9y69ofvxj3/s15/ogaSNHCQ3vOFNH7X1W+Nybebggw/2gDSXXy9f\nXn31Vb8IWtMqK3Fad/TFL34x+tM9S85t49zjh84rkUuawSNQ7rnR3zajb9c8+eSTft2RDHN9\n00ZrkPRsU4cXlz4C5dpMf/u/2sFXo0V6Nqnf/PLLL9u9997rN/eQYV1rxxqkPhDXQ0hTofSF\nX/3FndYj6S2bOhjuW0l+UaLm8OoNnaYhjB49Op4c/3pCoJI2ow8L68Non/nMZ/x0l49+9KP+\nI56VINL28mFaVFjEH/LtvffefnpWCHNMBwG9eZXTFLti98ADD/ipTv1pM5KpKcLazEMbeajD\nrBEBlacpV7j0EaikzWiESfcZGUnaGdN9iy/6wHn6rhiN+0OgkueGpsj1p81oBoQ2Z9CUb81s\n0MudQw89tNtOwP25DvLWjkAlbaa//V8Zzj/84Q99u9M9TYMM2jzo7LPPrt2FxkrKOAWYUxED\nMlBeWb+a481akIEiOvTlqL1oow829Bj6dT1QV9jfNqOt5DXFk2lSA1UjyZejEUONTmtBNQ4C\nlRDoT5vR6JHuM+oL6fMVuKFPYCD6v1rrJIOpkjWS1SKKgVQtssiFAAQgAAEIQAACEIAABFJH\ngFdIqasyFIYABCAAAQhAAAIQgAAEqkUAA6laZJELAQhAAAIQgAAEIAABCKSOAAZS6qoMhSEA\nAQhAAAIQgAAEIACBahHAQKoWWeRCAAIQgAAEIAABCEAAAqkjgIGUuipDYQhAAAIQgAAEIAAB\nCECgWgQwkKpFFrkQgAAEIAABCEAAAhCAQOoIYCClrspQGAIQgAAEIAABCEAAAhCoFgEMpGqR\nRS4EIAABCEAAAhCAAAQgkDoCGEipqzIUhgAEIAABCEAAAhCAAASqRQADqVpkkQsBCEAAAhCA\nAAQgAAEIpI4ABlLqqgyFIQABCEAAAhCAAAQgAIFqEcBAqhZZ5EIAAhCAAAQgAAEIQAACqSOA\ngZS6KkNhCEAAAhCAAAQgAAEIQKBaBDCQqkUWuRCAAAQgAAEIQAACEIBA6ghgIKWuylAYAhCA\nAAQgAAEIQAACEKgWAQykapFFLgQgAAEIQAACEIAABCCQOgIYSKmrMhSGAAQgAAEIQAACEIAA\nBKpFAAOpWmSRCwEIQAACEIAABCAAAQikjgAGUuqqDIUhAAEIQAACEIAABCAAgWoRwECqFlnk\nQgACEIAABCAAAQhAAAKpI4CBlLoqQ2EIQAACEIAABCAAAQhAoFoEMJCqRRa5EIAABCAAAQhA\nAAIQgEDqCGAgpa7KUBgCEIAABCAAAQhAAAIQqBYBDKRqkUUuBCAAAQhAAAIQgAAEIJA6AhhI\nqasyFIYABCAAAQhAAAIQgAAEqkUAA6laZJELAQhAAAIQgAAEIAABCKSOAAZS6qoMhSEAAQhA\nAAIQgAAEIACBahHAQKoWWeRCAAIQgAAEIAABCEAAAqkjgIGUuipDYQhAAAIQgAAEIAABCECg\nWgT+H4RzWM23l5mPAAAAAElFTkSuQmCC",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ggplot(arrests.race.yr %>% filter(homeless == 1 & arrest_year != 2017), aes(x = arrest_year, \n",
    "                                                                            y = arrests_percent, color = race.grp)) +\n",
    "  geom_line() +\n",
    "  geom_text(data = arrests.race.yr %>% filter(homeless == 1 & \n",
    "                                              arrest_year == 2016), aes(label = race.grp), hjust = 0.7,\n",
    "            vjust = 1) +\n",
    "  scale_y_continuous(limits = c(0, 50)) +\n",
    "  labs(x = \"\", y = \"% of arrests\", title = \"Racial Breakdown of homeless arrests\") +\n",
    "  theme(legend.position = 'none')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding: In 2011, one in 10 people arrests citywide were of homeless people; in 2016, it was 1 in 6"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use the grouped dataframe `arrest.totals` to calculate percentage of homeless arrests by year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>arrest_year</th><th scope=col>homeless</th><th scope=col>arrests_number</th><th scope=col>arrests_percent</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>2011    </td><td>1       </td><td>10496   </td><td>10.85408</td></tr>\n",
       "\t<tr><td>2012    </td><td>1       </td><td>11837   </td><td>11.69271</td></tr>\n",
       "\t<tr><td>2013    </td><td>1       </td><td>12237   </td><td>12.47070</td></tr>\n",
       "\t<tr><td>2014    </td><td>1       </td><td>12622   </td><td>13.41667</td></tr>\n",
       "\t<tr><td>2015    </td><td>1       </td><td>13418   </td><td>15.41112</td></tr>\n",
       "\t<tr><td>2016    </td><td>1       </td><td>14011   </td><td>16.75797</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|llll}\n",
       " arrest\\_year & homeless & arrests\\_number & arrests\\_percent\\\\\n",
       "\\hline\n",
       "\t 2011     & 1        & 10496    & 10.85408\\\\\n",
       "\t 2012     & 1        & 11837    & 11.69271\\\\\n",
       "\t 2013     & 1        & 12237    & 12.47070\\\\\n",
       "\t 2014     & 1        & 12622    & 13.41667\\\\\n",
       "\t 2015     & 1        & 13418    & 15.41112\\\\\n",
       "\t 2016     & 1        & 14011    & 16.75797\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "arrest_year | homeless | arrests_number | arrests_percent | \n",
       "|---|---|---|---|---|---|\n",
       "| 2011     | 1        | 10496    | 10.85408 | \n",
       "| 2012     | 1        | 11837    | 11.69271 | \n",
       "| 2013     | 1        | 12237    | 12.47070 | \n",
       "| 2014     | 1        | 12622    | 13.41667 | \n",
       "| 2015     | 1        | 13418    | 15.41112 | \n",
       "| 2016     | 1        | 14011    | 16.75797 | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  arrest_year homeless arrests_number arrests_percent\n",
       "1 2011        1        10496          10.85408       \n",
       "2 2012        1        11837          11.69271       \n",
       "3 2013        1        12237          12.47070       \n",
       "4 2014        1        12622          13.41667       \n",
       "5 2015        1        13418          15.41112       \n",
       "6 2016        1        14011          16.75797       "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "arrest.totals %>% \n",
    "    mutate(arrests_percent = arrests_number / sum(arrests_number) * 100) %>%\n",
    "    filter(homeless == 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding: The 14,000 arrests of homeless people in 2016 included more than 500 unique charges"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Filter the data to include homeless arrests in 2016 and calculate the number and percent of times each charge was cited. Note that this is done by analyzing each charge separately, so the `times_cited` column will not sum to the total number of arrests per year (arrestees can have multiple charges)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "arrest.reasons <- data %>% filter(homeless == 1 & arrest_year == 2016) %>% \n",
    "                           group_by(charge_code, charge_desc) %>% \n",
    "                           summarize(times_cited = n()) %>% \n",
    "                           ungroup() %>% \n",
    "                           mutate(percent_cited = times_cited/sum(times_cited) * 100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the number of unique charges"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "523"
      ],
      "text/latex": [
       "523"
      ],
      "text/markdown": [
       "523"
      ],
      "text/plain": [
       "[1] 523"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "length(unique(arrest.reasons$charge_code))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding: The most common offense was failure to appear in court for unpaid petty or minor citations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sort by percent of the time the charge was cited to get the top charges"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>charge_code</th><th scope=col>charge_desc</th><th scope=col>times_cited</th><th scope=col>percent_cited</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>853.7PC                        </td><td>fta-after-written-promise      </td><td>4447                           </td><td>21.145982                      </td></tr>\n",
       "\t<tr><td>11377(A)HS                     </td><td>possession-controlled-substance</td><td> 924                           </td><td> 4.393723                      </td></tr>\n",
       "\t<tr><td>459.5PC                        </td><td>                               </td><td> 674                           </td><td> 3.204945                      </td></tr>\n",
       "\t<tr><td>3000.08CPC                     </td><td>                               </td><td> 613                           </td><td> 2.914883                      </td></tr>\n",
       "\t<tr><td>3454(C)PC                      </td><td>                               </td><td> 579                           </td><td> 2.753210                      </td></tr>\n",
       "\t<tr><td>245(A)(1)PC                    </td><td>adw-wo-firearmgbi              </td><td> 530                           </td><td> 2.520209                      </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|llll}\n",
       " charge\\_code & charge\\_desc & times\\_cited & percent\\_cited\\\\\n",
       "\\hline\n",
       "\t 853.7PC                         & fta-after-written-promise       & 4447                            & 21.145982                      \\\\\n",
       "\t 11377(A)HS                      & possession-controlled-substance &  924                            &  4.393723                      \\\\\n",
       "\t 459.5PC                         &                                 &  674                            &  3.204945                      \\\\\n",
       "\t 3000.08CPC                      &                                 &  613                            &  2.914883                      \\\\\n",
       "\t 3454(C)PC                       &                                 &  579                            &  2.753210                      \\\\\n",
       "\t 245(A)(1)PC                     & adw-wo-firearmgbi               &  530                            &  2.520209                      \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "charge_code | charge_desc | times_cited | percent_cited | \n",
       "|---|---|---|---|---|---|\n",
       "| 853.7PC                         | fta-after-written-promise       | 4447                            | 21.145982                       | \n",
       "| 11377(A)HS                      | possession-controlled-substance |  924                            |  4.393723                       | \n",
       "| 459.5PC                         |                                 |  674                            |  3.204945                       | \n",
       "| 3000.08CPC                      |                                 |  613                            |  2.914883                       | \n",
       "| 3454(C)PC                       |                                 |  579                            |  2.753210                       | \n",
       "| 245(A)(1)PC                     | adw-wo-firearmgbi               |  530                            |  2.520209                       | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  charge_code charge_desc                     times_cited percent_cited\n",
       "1 853.7PC     fta-after-written-promise       4447        21.145982    \n",
       "2 11377(A)HS  possession-controlled-substance  924         4.393723    \n",
       "3 459.5PC                                      674         3.204945    \n",
       "4 3000.08CPC                                   613         2.914883    \n",
       "5 3454(C)PC                                    579         2.753210    \n",
       "6 245(A)(1)PC adw-wo-firearmgbi                530         2.520209    "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "head(arrest.reasons %>% arrange(desc(percent_cited)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Many codes did not come with charge descriptions in the data. Those that appear in the above table are described as follows:\n",
    "\n",
    "* 459.5PC: shoplifting\n",
    "* 3000.08CPC: parole warrant\n",
    "* 3454(C)PC: flash incarceration"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding: The top five charges were for non-violent or minor offenses"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some charge codes are grouped, largely according to [this](http://milliondollarhoods.org/wp-content/uploads/2017/10/Policing-the-House-2.0.FINAL_.pdf) UCLA report. For example. charge codes 40508(A)VC, 853.7PC, and 853.8PC all cover failure to appear."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "arrest.reasons$failure <- ifelse(arrest.reasons$charge_code == '40508(A)VC'|\n",
    "                                 arrest.reasons$charge_code == '853.7PC'|\n",
    "                                 arrest.reasons$charge_code == '853.8PC', 1, 0)\n",
    "\n",
    "arrest.reasons$trespass <- ifelse(arrest.reasons$charge_code == '419PC'|\n",
    "                                  arrest.reasons$charge_code == '602(K)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(O)(2)PC'|\n",
    "                                  arrest.reasons$charge_code == '602.5(A)PC'|\n",
    "                                  arrest.reasons$charge_code == '555PC'|\n",
    "                                  arrest.reasons$charge_code == '484F(A)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(L)(1)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(P)PC'|\n",
    "                                  arrest.reasons$charge_code == '602.5(B)PC'|\n",
    "                                  arrest.reasons$charge_code == '602PC'|\n",
    "                                  arrest.reasons$charge_code == '602(M)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(Q)PC'|\n",
    "                                  arrest.reasons$charge_code == '602.8PC'|\n",
    "                                  arrest.reasons$charge_code == '602(A)PC'|\n",
    "                                  arrest.reasons$charge_code == 'A602(N)1PC'|\n",
    "                                  arrest.reasons$charge_code == '602(S)PC'|\n",
    "                                  arrest.reasons$charge_code == '626.8(A)1PC'|\n",
    "                                  arrest.reasons$charge_code == '602(D)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(N)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(U)(1)PC'|\n",
    "                                  arrest.reasons$charge_code == '647(E)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(F)PC'|\n",
    "                                  arrest.reasons$charge_code == '602(O)PC'|\n",
    "                                  arrest.reasons$charge_code == '602.1(A)PC'| \n",
    "                                  arrest.reasons$charge_code == '647(H)PCLPP'| \n",
    "                                  arrest.reasons$charge_code == '602(J)PC'| \n",
    "                                  arrest.reasons$charge_code == '602(O)(1)PC'| \n",
    "                                  arrest.reasons$charge_code == '602.1(B)PC'| \n",
    "                                  arrest.reasons$charge_code == '369I(A)PC', 1, 0)\n",
    "\n",
    "arrest.reasons$shoplift <- ifelse(arrest.reasons$charge_code == '18 1708'|\n",
    "                                  arrest.reasons$charge_code == '484PCTFMV'|\n",
    "                                  arrest.reasons$charge_code == '485PC'|\n",
    "                                  arrest.reasons$charge_code == '488PC'|\n",
    "                                  arrest.reasons$charge_code == '459.5PC'|\n",
    "                                  arrest.reasons$charge_code == '484F(A)PC'|\n",
    "                                  arrest.reasons$charge_code == 'A488PC'|\n",
    "                                  arrest.reasons$charge_code == '490PC'|\n",
    "                                  arrest.reasons$charge_code == 'A484PC'|\n",
    "                                  arrest.reasons$charge_code == '484E(D)PC'|\n",
    "                                  arrest.reasons$charge_code == '666PC'|\n",
    "                                  arrest.reasons$charge_code == '484PC'|\n",
    "                                  arrest.reasons$charge_code == '490.2PC'|\n",
    "                                  arrest.reasons$charge_code == '666(A)PC'|\n",
    "                                  arrest.reasons$charge_code == '484(A)PC'|\n",
    "                                  arrest.reasons$charge_code == '490.5(A)PC'|\n",
    "                                  arrest.reasons$charge_code == '537(A)(1)PC'|\n",
    "                                  arrest.reasons$charge_code == '666.5PC'|\n",
    "                                  arrest.reasons$charge_code == '484E(A)PC'|\n",
    "                                  arrest.reasons$charge_code == '587CPC'|\n",
    "                                  arrest.reasons$charge_code == '666.5(A)PC'|\n",
    "                                  arrest.reasons$charge_code == '484E(B)PC', 1, 0)\n",
    "\n",
    "arrest.reasons$supervision_viol <- ifelse(arrest.reasons$charge_code == '1203.2PC'|\n",
    "                                          arrest.reasons$charge_code == '3000.08CPC'|\n",
    "                                          arrest.reasons$charge_code == '3454(C)PC'| \n",
    "                                          arrest.reasons$charge_code == '3455(B)1PC'|\n",
    "                                          arrest.reasons$charge_code == '1203.2(A)PC'|\n",
    "                                          arrest.reasons$charge_code == '3056PC'|\n",
    "                                          arrest.reasons$charge_code == '3455(A)4PC'|\n",
    "                                          arrest.reasons$charge_code == '3455(C)PC'|\n",
    "                                          arrest.reasons$charge_code == '3000.08FPC'|\n",
    "                                          arrest.reasons$charge_code == '3454PC'|\n",
    "                                          arrest.reasons$charge_code == '3455(A)PC'|\n",
    "                                          arrest.reasons$charge_code == '18 3606US', 1, 0)\n",
    "\n",
    "arrest.reasons$drug_poss <- ifelse(arrest.reasons$charge_code == '11377(A)HS'|\n",
    "                                   arrest.reasons$charge_code == '11377(A)1HS'|\n",
    "                                   arrest.reasons$charge_code == '11377HS'| \n",
    "                                   arrest.reasons$charge_code == '11350(A)HS'|\n",
    "                                   arrest.reasons$charge_code == '11350HS'|\n",
    "                                   arrest.reasons$charge_code == '11357HS'|\n",
    "                                   arrest.reasons$charge_code == '11357(A)HS'|\n",
    "                                   arrest.reasons$charge_code == '11357(B)HS'|\n",
    "                                   arrest.reasons$charge_code == '11357(C)HS'|\n",
    "                                   arrest.reasons$charge_code == '4573.6PC'|\n",
    "                                   arrest.reasons$charge_code == '11550(A)HS'|\n",
    "                                   arrest.reasons$charge_code == '11375(B)2HS'|\n",
    "                                   arrest.reasons$charge_code == '11351HS'|\n",
    "                                   arrest.reasons$charge_code == '4060BP', 1, 0)\n",
    "\n",
    "arrest.reasons$charge_desc_grouped <- ifelse(arrest.reasons$drug_poss == 1, 'drug_poss',\n",
    "                                             ifelse(arrest.reasons$trespass == 1, 'trespass',\n",
    "                                                    ifelse(arrest.reasons$shoplift == 1, 'shoplift',\n",
    "                                                           ifelse(arrest.reasons$supervision_viol == 1, 'supervision violation',\n",
    "                                                                  ifelse(arrest.reasons$failure == 1, 'failure to appear',\n",
    "                                                                         arrest.reasons$charge_code)))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get top five offenses using `charge_desc_grouped` as the charge identifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>charge_desc_grouped</th><th scope=col>times_cited</th><th scope=col>perc_cited</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>failure to appear    </td><td>4576                 </td><td>21.759391            </td></tr>\n",
       "\t<tr><td>drug_poss            </td><td>2147                 </td><td>10.209225            </td></tr>\n",
       "\t<tr><td>supervision violation</td><td>2085                 </td><td> 9.914408            </td></tr>\n",
       "\t<tr><td>shoplift             </td><td>1650                 </td><td> 7.845934            </td></tr>\n",
       "\t<tr><td>trespass             </td><td>1056                 </td><td> 5.021398            </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|lll}\n",
       " charge\\_desc\\_grouped & times\\_cited & perc\\_cited\\\\\n",
       "\\hline\n",
       "\t failure to appear     & 4576                  & 21.759391            \\\\\n",
       "\t drug\\_poss             & 2147                    & 10.209225              \\\\\n",
       "\t supervision violation & 2085                  &  9.914408            \\\\\n",
       "\t shoplift              & 1650                  &  7.845934            \\\\\n",
       "\t trespass              & 1056                  &  5.021398            \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "charge_desc_grouped | times_cited | perc_cited | \n",
       "|---|---|---|---|---|\n",
       "| failure to appear     | 4576                  | 21.759391             | \n",
       "| drug_poss             | 2147                  | 10.209225             | \n",
       "| supervision violation | 2085                  |  9.914408             | \n",
       "| shoplift              | 1650                  |  7.845934             | \n",
       "| trespass              | 1056                  |  5.021398             | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  charge_desc_grouped   times_cited perc_cited\n",
       "1 failure to appear     4576        21.759391 \n",
       "2 drug_poss             2147        10.209225 \n",
       "3 supervision violation 2085         9.914408 \n",
       "4 shoplift              1650         7.845934 \n",
       "5 trespass              1056         5.021398 "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "arrest.reasons %>% group_by(charge_desc_grouped) %>% \n",
    "    summarise(times_cited = sum(times_cited)) %>% \n",
    "    mutate(perc_cited = times_cited/sum(times_cited) * 100) %>% arrange(desc(times_cited)) %>% head(5)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.4.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}