{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Quick Demo: EDA & Resampling to answer basic comparison questions\n",
    "\n",
    "This is as a quick illustration of how we can combine our previous lessons about **resampling** and our skills with Tidy Data to get an answer to a *specific question*.\n",
    "\n",
    "This example should serve as a guide for how you might approach some of the questions you'll be asked to do in HW 2.\n",
    "\n",
    "We'll use the `starwars.csv` file that you'll explore in `03_challenge.ipynb` to answer the following question:\n",
    "\n",
    "### For non-human species on average, who is taller: males or females?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import polars as pl\n",
    "from polars import col\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's load the file and print the first few rows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><style>\n",
       ".dataframe > thead > tr,\n",
       ".dataframe > tbody > tr {\n",
       "  text-align: right;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       "</style>\n",
       "<small>shape: (5, 11)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>name</th><th>height</th><th>mass</th><th>hair_color</th><th>skin_color</th><th>eye_color</th><th>birth_year</th><th>sex</th><th>gender</th><th>homeworld</th><th>species</th></tr><tr><td>str</td><td>f64</td><td>f64</td><td>str</td><td>str</td><td>str</td><td>f64</td><td>str</td><td>str</td><td>str</td><td>str</td></tr></thead><tbody><tr><td>&quot;Luke Skywalker&quot;</td><td>172.0</td><td>77.0</td><td>&quot;blond&quot;</td><td>&quot;fair&quot;</td><td>&quot;blue&quot;</td><td>19.0</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Tatooine&quot;</td><td>&quot;Human&quot;</td></tr><tr><td>&quot;C-3PO&quot;</td><td>167.0</td><td>75.0</td><td>null</td><td>&quot;gold&quot;</td><td>&quot;yellow&quot;</td><td>112.0</td><td>&quot;none&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Tatooine&quot;</td><td>&quot;Droid&quot;</td></tr><tr><td>&quot;R2-D2&quot;</td><td>96.0</td><td>32.0</td><td>null</td><td>&quot;white, blue&quot;</td><td>&quot;red&quot;</td><td>33.0</td><td>&quot;none&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Naboo&quot;</td><td>&quot;Droid&quot;</td></tr><tr><td>&quot;Darth Vader&quot;</td><td>202.0</td><td>136.0</td><td>&quot;none&quot;</td><td>&quot;white&quot;</td><td>&quot;yellow&quot;</td><td>41.9</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Tatooine&quot;</td><td>&quot;Human&quot;</td></tr><tr><td>&quot;Leia Organa&quot;</td><td>150.0</td><td>49.0</td><td>&quot;brown&quot;</td><td>&quot;light&quot;</td><td>&quot;brown&quot;</td><td>19.0</td><td>&quot;female&quot;</td><td>&quot;feminine&quot;</td><td>&quot;Alderaan&quot;</td><td>&quot;Human&quot;</td></tr></tbody></table></div>"
      ],
      "text/plain": [
       "shape: (5, 11)\n",
       "┌────────────────┬────────┬───────┬────────────┬───┬────────┬───────────┬───────────┬─────────┐\n",
       "│ name           ┆ height ┆ mass  ┆ hair_color ┆ … ┆ sex    ┆ gender    ┆ homeworld ┆ species │\n",
       "│ ---            ┆ ---    ┆ ---   ┆ ---        ┆   ┆ ---    ┆ ---       ┆ ---       ┆ ---     │\n",
       "│ str            ┆ f64    ┆ f64   ┆ str        ┆   ┆ str    ┆ str       ┆ str       ┆ str     │\n",
       "╞════════════════╪════════╪═══════╪════════════╪═══╪════════╪═══════════╪═══════════╪═════════╡\n",
       "│ Luke Skywalker ┆ 172.0  ┆ 77.0  ┆ blond      ┆ … ┆ male   ┆ masculine ┆ Tatooine  ┆ Human   │\n",
       "│ C-3PO          ┆ 167.0  ┆ 75.0  ┆ null       ┆ … ┆ none   ┆ masculine ┆ Tatooine  ┆ Droid   │\n",
       "│ R2-D2          ┆ 96.0   ┆ 32.0  ┆ null       ┆ … ┆ none   ┆ masculine ┆ Naboo     ┆ Droid   │\n",
       "│ Darth Vader    ┆ 202.0  ┆ 136.0 ┆ none       ┆ … ┆ male   ┆ masculine ┆ Tatooine  ┆ Human   │\n",
       "│ Leia Organa    ┆ 150.0  ┆ 49.0  ┆ brown      ┆ … ┆ female ┆ feminine  ┆ Alderaan  ┆ Human   │\n",
       "└────────────────┴────────┴───────┴────────────┴───┴────────┴───────────┴───────────┴─────────┘"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Solution\n",
    "sw = pl.read_csv('starwars.csv')\n",
    "sw.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then let's filter the data down to non-human males and females only:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><style>\n",
       ".dataframe > thead > tr,\n",
       ".dataframe > tbody > tr {\n",
       "  text-align: right;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       "</style>\n",
       "<small>shape: (41, 11)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>name</th><th>height</th><th>mass</th><th>hair_color</th><th>skin_color</th><th>eye_color</th><th>birth_year</th><th>sex</th><th>gender</th><th>homeworld</th><th>species</th></tr><tr><td>str</td><td>f64</td><td>f64</td><td>str</td><td>str</td><td>str</td><td>f64</td><td>str</td><td>str</td><td>str</td><td>str</td></tr></thead><tbody><tr><td>&quot;Chewbacca&quot;</td><td>228.0</td><td>112.0</td><td>&quot;brown&quot;</td><td>&quot;unknown&quot;</td><td>&quot;blue&quot;</td><td>200.0</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Kashyyyk&quot;</td><td>&quot;Wookiee&quot;</td></tr><tr><td>&quot;Greedo&quot;</td><td>173.0</td><td>74.0</td><td>null</td><td>&quot;green&quot;</td><td>&quot;black&quot;</td><td>44.0</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Rodia&quot;</td><td>&quot;Rodian&quot;</td></tr><tr><td>&quot;Yoda&quot;</td><td>66.0</td><td>17.0</td><td>&quot;white&quot;</td><td>&quot;green&quot;</td><td>&quot;brown&quot;</td><td>896.0</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>null</td><td>&quot;Yoda&#x27;s species&quot;</td></tr><tr><td>&quot;Bossk&quot;</td><td>190.0</td><td>113.0</td><td>&quot;none&quot;</td><td>&quot;green&quot;</td><td>&quot;red&quot;</td><td>53.0</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Trandosha&quot;</td><td>&quot;Trandoshan&quot;</td></tr><tr><td>&quot;Ackbar&quot;</td><td>180.0</td><td>83.0</td><td>&quot;none&quot;</td><td>&quot;brown mottle&quot;</td><td>&quot;orange&quot;</td><td>41.0</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Mon Cala&quot;</td><td>&quot;Mon Calamari&quot;</td></tr><tr><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td><td>&hellip;</td></tr><tr><td>&quot;San Hill&quot;</td><td>191.0</td><td>null</td><td>&quot;none&quot;</td><td>&quot;grey&quot;</td><td>&quot;gold&quot;</td><td>null</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Muunilinst&quot;</td><td>&quot;Muun&quot;</td></tr><tr><td>&quot;Shaak Ti&quot;</td><td>178.0</td><td>57.0</td><td>&quot;none&quot;</td><td>&quot;red, blue, white&quot;</td><td>&quot;black&quot;</td><td>null</td><td>&quot;female&quot;</td><td>&quot;feminine&quot;</td><td>&quot;Shili&quot;</td><td>&quot;Togruta&quot;</td></tr><tr><td>&quot;Grievous&quot;</td><td>216.0</td><td>159.0</td><td>&quot;none&quot;</td><td>&quot;brown, white&quot;</td><td>&quot;green, yellow&quot;</td><td>null</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Kalee&quot;</td><td>&quot;Kaleesh&quot;</td></tr><tr><td>&quot;Tarfful&quot;</td><td>234.0</td><td>136.0</td><td>&quot;brown&quot;</td><td>&quot;brown&quot;</td><td>&quot;blue&quot;</td><td>null</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Kashyyyk&quot;</td><td>&quot;Wookiee&quot;</td></tr><tr><td>&quot;Tion Medon&quot;</td><td>206.0</td><td>80.0</td><td>&quot;none&quot;</td><td>&quot;grey&quot;</td><td>&quot;black&quot;</td><td>null</td><td>&quot;male&quot;</td><td>&quot;masculine&quot;</td><td>&quot;Utapau&quot;</td><td>&quot;Pau&#x27;an&quot;</td></tr></tbody></table></div>"
      ],
      "text/plain": [
       "shape: (41, 11)\n",
       "┌────────────┬────────┬───────┬────────────┬───┬────────┬───────────┬────────────┬────────────────┐\n",
       "│ name       ┆ height ┆ mass  ┆ hair_color ┆ … ┆ sex    ┆ gender    ┆ homeworld  ┆ species        │\n",
       "│ ---        ┆ ---    ┆ ---   ┆ ---        ┆   ┆ ---    ┆ ---       ┆ ---        ┆ ---            │\n",
       "│ str        ┆ f64    ┆ f64   ┆ str        ┆   ┆ str    ┆ str       ┆ str        ┆ str            │\n",
       "╞════════════╪════════╪═══════╪════════════╪═══╪════════╪═══════════╪════════════╪════════════════╡\n",
       "│ Chewbacca  ┆ 228.0  ┆ 112.0 ┆ brown      ┆ … ┆ male   ┆ masculine ┆ Kashyyyk   ┆ Wookiee        │\n",
       "│ Greedo     ┆ 173.0  ┆ 74.0  ┆ null       ┆ … ┆ male   ┆ masculine ┆ Rodia      ┆ Rodian         │\n",
       "│ Yoda       ┆ 66.0   ┆ 17.0  ┆ white      ┆ … ┆ male   ┆ masculine ┆ null       ┆ Yoda's species │\n",
       "│ Bossk      ┆ 190.0  ┆ 113.0 ┆ none       ┆ … ┆ male   ┆ masculine ┆ Trandosha  ┆ Trandoshan     │\n",
       "│ Ackbar     ┆ 180.0  ┆ 83.0  ┆ none       ┆ … ┆ male   ┆ masculine ┆ Mon Cala   ┆ Mon Calamari   │\n",
       "│ …          ┆ …      ┆ …     ┆ …          ┆ … ┆ …      ┆ …         ┆ …          ┆ …              │\n",
       "│ San Hill   ┆ 191.0  ┆ null  ┆ none       ┆ … ┆ male   ┆ masculine ┆ Muunilinst ┆ Muun           │\n",
       "│ Shaak Ti   ┆ 178.0  ┆ 57.0  ┆ none       ┆ … ┆ female ┆ feminine  ┆ Shili      ┆ Togruta        │\n",
       "│ Grievous   ┆ 216.0  ┆ 159.0 ┆ none       ┆ … ┆ male   ┆ masculine ┆ Kalee      ┆ Kaleesh        │\n",
       "│ Tarfful    ┆ 234.0  ┆ 136.0 ┆ brown      ┆ … ┆ male   ┆ masculine ┆ Kashyyyk   ┆ Wookiee        │\n",
       "│ Tion Medon ┆ 206.0  ┆ 80.0  ┆ none       ┆ … ┆ male   ┆ masculine ┆ Utapau     ┆ Pau'an         │\n",
       "└────────────┴────────┴───────┴────────────┴───┴────────┴───────────┴────────────┴────────────────┘"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered = sw.filter(\n",
    "    (~col('species').eq('Human')) & (col('sex').is_in(['female','male'])) \n",
    "    )\n",
    "\n",
    "filtered"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's **summarize** the data by calculating the *mean* and *median* of the height column separately for males and females.\n",
    "\n",
    "We an use a `.group_by` to do this"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><style>\n",
       ".dataframe > thead > tr,\n",
       ".dataframe > tbody > tr {\n",
       "  text-align: right;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       "</style>\n",
       "<small>shape: (2, 3)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>sex</th><th>height_mean</th><th>height_median</th></tr><tr><td>str</td><td>f64</td><td>f64</td></tr></thead><tbody><tr><td>&quot;female&quot;</td><td>179.571429</td><td>178.0</td></tr><tr><td>&quot;male&quot;</td><td>176.911765</td><td>189.0</td></tr></tbody></table></div>"
      ],
      "text/plain": [
       "shape: (2, 3)\n",
       "┌────────┬─────────────┬───────────────┐\n",
       "│ sex    ┆ height_mean ┆ height_median │\n",
       "│ ---    ┆ ---         ┆ ---           │\n",
       "│ str    ┆ f64         ┆ f64           │\n",
       "╞════════╪═════════════╪═══════════════╡\n",
       "│ female ┆ 179.571429  ┆ 178.0         │\n",
       "│ male   ┆ 176.911765  ┆ 189.0         │\n",
       "└────────┴─────────────┴───────────────┘"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered.group_by('sex').agg(\n",
    "    height_mean = col('height').mean(),\n",
    "    height_median = col('height').median(),\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That's interesting - it looks like male non-humans have a lower *mean* height than females, but not a lower *median* height.\n",
    "\n",
    "But these summary statistics don't tell us about the *distrubution* of heights by sex. So let's visualize those using `sns.catplot`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1460e0d50>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9MAAAPTCAYAAACzDVpFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAB7CAAAewgFu0HU+AACTaElEQVR4nOzdd3iUVf738c8kk54QSgokGVpoAaVIQi9BUJciLKDYIYq4llVxbVtFd13Lqqv+XDsi2LAgNuxKlwDSREokhZKEQBJqCimTzPNHHiIxbe4hMxMy79d1cT0zc59z7u+wP/Pkwzn3OSabzWYTAAAAAACwm5e7CwAAAAAA4FxDmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIPM7i4A7me1WnXo0CFJUvv27WU2838WAAAAANAQZqahQ4cOyWKxyGKxVIdqAAAAAED9CNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYJDZ3QUAwLnsSGGpikorFOxvVtsgX3eXAwAAABchTAOAQcVlVn289aDeWr9fu3JOVn/eNyZU1w7ppMn9ouTv4+3GCgEAAOBsJpvNZnN3EXCvrKwsWSwWSVJmZqZiYmLcXBHQfO0/UqSk13/U3vyiett0jwjWwhsGKbp1gAsrAwAAgCvxzDQA2Cm3oERXv7qhwSAtSam5hbp2/gYdKypzUWUAAABwNcI0ANjpf8vTlH38lF1t9+YX6aXV6U6uCAAAAO5CmAYAOxSVWvXRlmxDfT7YlKVSa4WTKgIAAIA7EaYBwA7f7T6sglKroT5Hi8q0ek++kyoCAACAOxGmAcAOeQWlLu0HAACA5o0wDQB28DU79uPS0X4AAABo3vgtDwDs0LtDK5f2AwAAQPNGmAYAO8R3bqte7UMM9RnQsbV6RxGmAQAAWiLCNADY6YYRXQy1n22wPQAAAM4dhGkAsNOMeIuuHtzRrrazR3TRpL5RTq4IAAAA7mJ2dwEAcC55ZOr5igr118urM1RQUvuorNaBProtsZvmjOrqhuoAAADgKiabzWZzdxFwr6ysLFksFklSZmamYmJi3FwR0PydKqvQJ9uytTo1T4WlFQrxNyuxR7gu7Rclfx9vd5cHAAAAJyNMgzANAAAAAAbxzDQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDPDZMb9myRY888ojGjx8vi8UiPz8/BQcHq0ePHkpKStKaNWsaHWPhwoUymUx2/Vm4cGGj4xUXF+uJJ57QoEGD1LZtWwUHBysuLk733HOPDhw40ATfGgAAAADQFMzuLsAdRo8erdWrV9f6vKysTKmpqUpNTdWiRYt03XXXaf78+fL19XV6Tenp6Zo4caJ++eWXGp+npKQoJSVF8+fP1zvvvKMJEyY4vRYAAAAAQMM8MkxnZ2dLkqKionT55Zdr5MiR6tixoyoqKpScnKynnnpK2dnZevPNN2W1WvXOO+80OubXX3+tqKioeq/HxMTUe62wsFCTJk2qDtJz5szRlVdeqYCAAK1YsUKPPvqoTpw4ocsvv1zJycnq27evwW8MAAAAAGhKHhmme/XqpUceeUTTp0+Xt7d3jWtDhgzRddddp+HDh2vPnj1avHixbrnlFo0cObLBMXv06KHOnTs7VM+TTz6plJQUSdJ//vMf3XvvvdXXhg4dqjFjxmjUqFEqLi7W3LlztXz5cofuAwAAAABoGh75zPSyZcs0Y8aMWkH6tLCwMD311FPV75csWeK0WsrLy/Xss89KkuLi4nT33XfXajN06FDNnj1bkrRixQpt3rzZafUAAAAAABrnkWHaHomJidWv09PTnXaflStX6vjx45KkWbNmycur7v9JkpKSql8vXbrUafUAAAAAABpHmK5HWVlZ9ev6Am5TOHPX8NGjR9fbLj4+XkFBQZKktWvXOq0eAAAAAEDjPPKZaXusWrWq+nWvXr0abZ+UlKTdu3fr2LFjatWqlbp166Zx48bplltuUXR0dL39du/ebdd9zGazYmNjtX379hp97JGVldXg9ZycHEPjAQAAAICnI0zXobKyUo899lj1+xkzZjTa58zwfeTIER05ckQbNmzQU089pWeeeUZ/+MMf6uyXmZkpSQoKClLr1q0bvIfFYtH27duVl5en0tJS+fn52fFtqvoBAAAAAJoOYboOTz/9tDZu3ChJmjp1quLj4+tt27VrV02bNk1Dhw6tDq0ZGRn68MMPtWTJEpWUlOjmm2+WyWTSTTfdVKt/QUGBJCk4OLjRuk4v85aqjtOyN0wDAAAAAJqWyWaz2dxdRHOyatUqjRs3TlarVREREdq+fbsiIyPrbHvixAm1atVKJpOpzuvLli3TtGnTVF5ersDAQKWnp6t9+/Y12sTGxiojI0MWi0UHDhxosLaZM2fqzTfflFQ1o93Q2dVnsmeZ96BBgwyPCwAAAACeig3IzrBz505NnTpVVqtVfn5+ev/99+sN0pIUGhpab5CWpEmTJmnevHmSpOLiYr322mu12vj7+0uqueFZfUpLS6tfBwQENNr+tJiYmAb/dOjQwe6xAAAAAACE6Wp79+7VxRdfrGPHjsnb21uLFy9ucHdte82ZM6c6cJ/5XPVpISEhkqqWbTemqKio+rU9y8IBAAAAAM5BmJZ08OBBjRs3TgcPHpTJZNKCBQs0derUJhk7IiJCYWFhkqTs7Oxa108vqS4qKqo+b7o+pzcrCw8P53lpAAAAAHAjjw/T+fn5uuiii5SRkSFJeu655zRz5swmvUdDj6X37t27+nVKSkq97axWq9LT0yVJcXFxTVccAAAAAMAwjw7TJ06c0CWXXKJdu3ZJkh577DHddtttTXqP3NxcHTlyRJIUFRVV6/qIESOqX9e1DPy0TZs2VS/zHj58eJPWCAAAAAAwxmPDdHFxsSZOnKgtW7ZIkv72t7/p/vvvb/L7vPLKK9Uz03U9g52YmKjQ0FBJ0qJFi+qdxV64cGH166Zagg4AAAAAcIxHhumysjJNnTpVP/zwgyTpzjvv1MMPP2xojH379mnr1q0Ntlm2bJn+9a9/Saratfv666+v1cbX11d33HGHJGn37t168skna7VJTk6u3gl89OjRSkhIMFQrAAAAAKBpmd1dgDtcddVV+uabbyRJF154oWbPnq0dO3bU297X11c9evSo8dm+ffs0ZswYDR06VJdeeqn69++viIgI2Ww2ZWRkaMmSJVqyZEn1TPOTTz6p6OjoOse/99579d5772nPnj267777lJaWpiuvvFIBAQFasWKFHnnkEVmtVgUEBOiZZ55pmr8EAAAAAIDDTLaGdsdqoRo6G7ounTp10r59+2p8tnLlSo0ZM6bRvoGBgXr66ad10003NdguLS1NEyZMUGpqap3XW7VqpbfffluTJk2yu257ZWVlyWKxSKraMfz0DuMAAAAAgLp55Mx0Uxg4cKDeeustJScna9OmTcrJyVF+fr6sVqvatGmjPn36aOzYsbrxxhsVERHR6HjdunXT1q1b9fzzz+uDDz5QWlqaysrKZLFYNGHCBN15553q1KmTC74ZAAAAAKAxHjkzjZqYmQYAAAAAYzxyAzIAAAAAAM4GYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAg8zuLgAAzlXr0vO1JjVfRaVWBfuZldgzQoO6tHV3WQAAAHABwjQAGPTpTwf1f9+nKi23sMbnL6xMV6/2IZo7rod+d157N1UHAAAAV2CZNwAY8NKqdN2xeGutIH1ayqEC3fL2Zi1at8+1hQEAAMClCNMAYKfvdh3WY1+mNNrOZpMe/GynfkjLd0FVAAAAcAfCNADY6aVV6Xa3tdmkl1dnOLEaAAAAuBNhGgDssDvnpDbtP2aoz5rUPO3LL3JSRQAAAHAnwjQA2MFokJaqZqe3HDDeDwAAAM0fYRoA7FBaXuFQv1MO9gMAAEDzRpgGADuEBvg41K91gG8TVwIAAIDmgDANAHYYGxcpX7OxH5lBvt4a1SPMSRUBAADAnQjTAGCHtkG+mnh+B0N9pgyIVoi/YzPaAAAAaN4I0wBgp9sv7Gb3cu92Qb66ZXSskysCAACAuxCmAcBOXcOD9fr1CWoT2HCgDgv206IbBsnSNtBFlQEAAMDVCNMAYMAFHdvo8ztG6g+ju6ptUM3NxcKC/fTHMd30xR0jdF50qJsqBAAAgCuYbDabzd1FwL2ysrJksVgkSZmZmYqJiXFzRcC5odRaoR3ZJ1VUalWIv1nnRYfKx5t/owQAAPAEZncXAADnKj+ztwZ2auPuMgAAAOAGTKEAAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIDYgA+DR1mcc0Zvr92vNnjwVlVUoyNdbiT0jdN3QTkro3Nbd5QEAAKCZIkwD8Egnist1y9ubtS79SI3PT5ZY9elPB/XpTwc1snuYXrjmAoX4+7ipSgAAADRXLPMG4HFOlVVo5oINtYL0b61JzdesBRtVUl7hosoAAABwriBMA/A4L65M009ZJ+xqu+XAcc1fk1Hv9fKKSh0vLlN5RWVTlQcAAIBzAMu8AXiU8opKLf4x01CfdzYc0C2J3eTtZZIkWSsq9fXOw3pz/T6tzzgqSTKZpBHdwnTtkE4aFxdZ3RYAAAAtE2EagEdZ+Uue8gpKDfU5eKJEa9PyNbpHuA6fLFHS6z9qd87JGm1stqpl4WtS89XP0lqvzYpXWLBfU5YOAACAZoRl3gA8SvaxYgf7ndLx4jJd9er6WkH6t37KPK5r529QQUm5Q/cCAABA80eYBuBRvBxcfu1lkp5bnqaMvCK72qccKtCrq+t/1hoAAADnNsI0AI/SJSzIoX7RbQK0ZHOWoT7v/pjJxmQAAAAtFGEagEcZHhumjm0DDfWJDQ/SyVNWnThlbNl2bkGp1qbmG+oDAACAcwNhGoBH8fIy6bohnQz1uW5IJx06WeLQ/RztBwAAgOaNMA3A41w/vLPG9Ay3q+1FvSM1c2hnmR181trRfgAAAGjeCNMAPI7Z20svXTdQlw+MUX1Z18skXTXIoheuuUBeXiZ1jwx26F49IkPOolIAAAA0V5wzDcAj+Zm99cTl/XTH2O56a8N+rd6Tr6JSq4L9zErsGa6rB3dUTJtfn60e2rWduoYH2b2btySdF91K/SytnVA9AAAA3M1jw/SWLVv01Vdfac2aNdqxY4dyc3Pl4+OjqKgoDRs2TLNnz9bIkSMbHKOkpERff/21vvvuO23cuFGpqakqKChQSEiIevbsqUsuuUR/+MMf1KFDhwbHSUxM1KpVq+yq22az2f0dATTO0jZQfxkfp7+Mb7idyWRS0rDOeuCTnXaPPWto57MrDgAAAM2WyeaB6Wz06NFavXp1o+2uu+46zZ8/X76+vrWubd++XSNGjFBBQUGDY4SEhGj+/PmaMWNGvW3cHaazsrJksVgkSZmZmYqJiWnyewAtgc1m0x3vbtNnPx1stO2M+Bj957J+LqgKAAAA7uCRM9PZ2dmSpKioKF1++eUaOXKkOnbsqIqKCiUnJ+upp55Sdna23nzzTVmtVr3zzju1xjh58mR1kB4+fLgmTZqk+Ph4tWvXTnl5eVq6dKnmz5+vgoICXX311QoJCdH48Q1PfcXHx+v1119v+i8MoEmYTCY9c0V/tW/lp0XJ+1VmrX2GtL+Pl24c0VV3X9zDDRUCAADAVTxyZnrSpEmaOXOmpk+fLm9v71rX8/PzNXz4cO3Zs0eStHr16lpLvtetW6dnn31W8+bNU+/eveu8zyeffKKpU6fKZrMpNjZWqampMplq73Z0emZ69OjRWrly5dl/QYOYmQaMO1pUpg82ZWpNar4KS60K8TdrdI9wXT7QotBAH3eXBwAAACfzyDBtj2XLlunSSy+VJN1xxx169tlnHRrnsssu04cffiip6jntAQMG1GpDmAYAAACAcwtHY9UjMTGx+nV6errD44wZM6ZJxgHQ/OQVlOrTnw5q8cYD+uyngzpaVObukgAAAOAiHvnMtD3Kyn79pdjLy/F/cygtLW2ScQA0HzuyT+jFVen6ZuchlVf8urjH1+ylCee11y2J3dSzPedLAwAAtGSE6Xqcubt2r169XDJOSkqKEhISlJKSovLycoWFhWngwIGaPn26rrrqKvn4OPYcZlZWVoPXc3JyHBoX8ETf7jqsP76zRaV1bD5WZq3Ux9sO6ptdh/XStQM1qke4GyoEAACAK/DMdB0qKys1dOhQbdy4UZL0448/Kj4+3vA4P/30kwYOHKiKigr16dNHO3bsqLOdPUdj9e7dW0uWLFFcXJzhOura9Kw+PDMN1G971nFd/lJynUH6twJ9vbX01mHq1b6VCyoDAACAq7HuuA5PP/10dZCeOnWqQ0G6tLRUN954oyoqKiRJjzzySL1tvby8NHbsWD311FP67rvvtHXrVq1evVrPPPNMdXjetWuXxowZowMHDjjwjQA0heeWp9kVpCWpuKxCL65knwQAAICWipnp31i1apXGjRsnq9WqiIgIbd++XZGRkYbHmTNnjubPny9JmjVrlhYuXFhv2+PHj6t169Z1XisvL9ecOXO0aNEiSVXhfunSpYZqsWeZ96BBgyQxMw3U5+DxUxr5nxWqqLT/R6avt5eS/3Kh2gX7ObEyAAAAuAPPTJ9h586dmjp1qqxWq/z8/PT+++87FKQfffTR6iA9cOBAPf/88w22ry9IS5KPj4/mz5+vDRs2KCUlRR999JGys7MVHR1tdz2EY+Dsrd6TZyhIS1JZRaXWpuVrSn/7/3sFAADAuYFl3v/f3r17dfHFF+vYsWPy9vbW4sWLNXr0aMPjvPzyy/rrX/8qSerZs6e+/PJLBQUFnVVtZrNZs2fPrn7f2PPVAJpeQYnVoX6FpY71AwAAQPNGmJZ08OBBjRs3TgcPHpTJZNKCBQs0depUw+MsXrxYt956qySpU6dO+u677xQe3jS7+fbu3bv6dXZ2dpOMCcB+QX6OLeQJ8mUBEAAAQEvk8WE6Pz9fF110kTIyMiRJzz33nGbOnGl4nE8//VQzZ85UZWWlOnTooO+//75Jl1fzaDvgXiO6hcnAxviSJLOXScNi2zmnIAAAALiVR4fpEydO6JJLLtGuXbskSY899phuu+02w+N8//33mjFjhqxWq9q1a6dvv/1WsbGxTVrr6RolKSoqqknHBtC4ju0CNaq7sZUmF/eJVEQrfydVBAAAAHfy2DBdXFysiRMnasuWLZKkv/3tb7r//vsNj7Nu3TpNmTJFpaWlatWqlb7++mv16dOnSWu1Wq1asGBB9ftRo0Y16fgA7PPHC7vJ7GXf9LSv2Us3j27af1QDAABA8+GRYbqsrExTp07VDz/8IEm688479fDDDxseZ9u2bZo4caKKiooUFBSkL774QgMHDjQ0xooVK3T8+PF6r5eXl+vGG29USkqKJOnSSy+VxWIxXCuAs5fQua2evLxfo4Ha19tL/3dlf/WNae2awgAAAOByHrkzzlVXXaVvvvlGknThhRdq9uzZ2rFjR73tfX191aNHjxqfpaen65JLLqkOwg8//LBCQ0MbHCciIkIRERE1Plu0aJEmT56syZMnKzExUT179lSrVq1UWFiozZs36+WXX9bu3bur+z/77LOOfGUATeT3A6IV1TpA/1uRpjWpeTpzOwMvkzSmZ4T+eGE3DejYxn1FAgAAwOlMNg/c2cpkcBehTp06ad++fTU+W7hwoa6//npD48ybN08PPvhgjc+SkpK0aNGiRvuef/75evfdd2vs6t1UsrKyqme7MzMzOZcasNO+/CKtTctXUalVwf5mjeoeLkvbQHeXBQAAABfwyJnp5uT+++9X//79lZycrF27dikvL09Hjx6Vn5+fIiMjFR8fr8suu0xTp06Vt7e3u8sFcIbOYUHqHHZ258gDAADg3OSRM9OoiZlpAAAAADDGIzcgAwAAAADgbBCmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wBwFiorbSoqtcpms7m7FAAAALiQ2d0FAMC5pqLSpu92H9Zb6/frh7R8Vdokby+TRnUP03VDOymxR4S8vEzuLhMAAABORJgGAANyC0o0e+Em/Zx9osbnFZU2rfglTyt+ydMFHVtr/qwEtQ3ydVOVAAAAcDaWeQOAnU6cKtc1r26oFaR/a8uB47rutQ0qKrW6qDIAAAC4GmEaAOz0woo0peYW2tV258GTemV1hpMrAgAAgLsQpgHADiXlFXp/U6ahPu/+eEDWikonVQQAAAB3IkwDgB2Wp+TqWHG5oT6HT5ZqTVq+kyoCAACAOxGmAcAOOSdKHOp3yMF+AAAAaN4I0wBgB28HT7ry5ogsAACAFokwDQB26B4Z4li/iOAmrgQAAADNAWEaAOwwLLaduoQFGerTu0MrDejYxkkVAQAAwJ0I0wBgB5PJpFlDOxnqkzSss3OKAQAAgNsRpgHATjOHdtaE89vb1XbaBdGakWBxckUAAI9SUS4dWC/t+Ubat1YqLXR3RYBHM7u7AAA4V3h5mfR/Vw5QRMhuvb1hv8orbLXa+Jq9dP3wzrr/kl5uqBAA0CIV5kobX5E2L5KKcn/93K+V1PcKacgtUrtY99UHeCiTzWar/dsgPEpWVpYslqoZtMzMTMXExLi5IqD5yyso1Xs/HtDq1HwVlVoV7GfWmF4RmhFvUdsgX3eXBwBoKQ7vkt6aLhUcrL+Nb4h0xRtS7IWuqwsAYRqEaQAAgGbpZI70SqJUeKjxtj6B0g1fSR36Ob0sAFV4ZhoAAABojpL/Z1+QlqTyYmnFo86tB0ANhGkAAACguSk/JW1721if1K+l4wecUw+AWgjTAAAAQHOz7wfp1DFjfWyVUsoXzqkHQC2EaQAAAKC5KTnuWD+jARyAwwjTAAAAQHNj9nesn09A09YBoF6EaQAAAKC5iUmQvHyM9+s0rOlrAVAnwjQAAADQ3IRESnGTjPVp31eyDHJOPQBqIUwDAAAAzdGw243NTo+Y67RSANRGmAYAAACao+iB0pT/SSbvxtuO/rN03nTn1wSgmtndBQAAAACoR78rpeAIafm/pexNta+H9ZBG3iP1u8L1tQEejjANAAAANGexF1b9ObhN2vOVVHJC8g2SOo+Uuo52d3WAxyJMAwAAAOeCqP5VfwA0CzwzDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDzO4uAADcpdRaoWU/5ejDLVnKOVEiL5PULSJYVw/upFHdw2Qymertm3m0WG+t36/VqfkqKrUq2M+sMb3CdfXgTopuHeDCbwEAAAB3MNlsNpu7i4B7ZWVlyWKxSJIyMzMVExPj5ooA5/tm5yH9eenPOlpUVuf1ruFBevGagerZPqTG56XWCv116Q59tDVLlXX89PT2MumKBIsemtxHPt4s/gEAAGip+E0PgMf58ucc3fzW5nqDtCRl5BXp8pfWac/hgurPrBWVuumNzfpwS91BWpIqKm16Z8MB/fGdLaqsrxEAAADOeYRpAB7laFGZ7np/W71h+EwnS6z64ztbqt+/tnavVu3Js+s+X+88rLc37He0TAAAADRzhGkAHuW9HzNVUl5pd/s9hwu1Lj1flZU2vbneWDh+I5kwDQAA0FKxARkAj/LujwcM91m8MVPlFTZlHTtlqF9qbqF+3HdUCZ3bGr4nAAAAmjdmpgF4jIpKm/YfKTbcb29+oTLyCh265968Iof6AQAAoHkjTAPwGJUOHl5QWSk5eu6Bo/cEAABA80aYBuAxfLy9FBHiZ7hfdJsARTl4dnR0G86cBgAAaIkI0wA8yuXxxs9Rv3xgjC7sFaGwYF9D/aJbB2h4bJjh+wEAAKD5I0wD8ChXD+4kby+T3e2jWwdobFykfM1euiLBYvBeHeVl4F4AAAA4dxCmAXiU6NYB+vvEOLva+np76akZ/arD921juun86FC7+g7s1EazR3RxuE4AAAA0b4RpAB7n+uFd9OClveXjXf+scWiAj16/PkFDurar/izQ16w3Zw/S4C4NH3U1snuYFl6fIH8f7yarGQAAAM2LyWZjq1lPl5WVJYulavlqZmamYmKMP1MKnItyT5Zo8cZMLdmSqUMnSuRlMik2PFhXD+6oqQOiFeRnrrdvcvoRvbV+v1an5qmo1KpgP7PG9IrQdUM6KZ5zpQEAAFo8wjQI0wAAAABgEMu8AQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgjw3TW7Zs0SOPPKLx48fLYrHIz89PwcHB6tGjh5KSkrRmzRpD43311VeaNm2aYmJi5Ofnp5iYGE2bNk1fffWV3WMUFxfriSee0KBBg9S2bVsFBwcrLi5O99xzjw4cOGD0KwIAAAAAnMRks9ls7i7C1UaPHq3Vq1c32u66667T/Pnz5evrW28bm82mm2++Wa+88kq9bW666Sa99NJLMplM9bZJT0/XxIkT9csvv9R5PTQ0VO+8844mTJjQaN1GZWVlyWKxSJIyMzMVExPT5PcAAAAAgJbEI2ems7OzJUlRUVG68847tWTJEm3cuFHJycn673//q+joaEnSm2++qaSkpAbH+vvf/14dpAcMGKDFixdr48aNWrx4sQYMGCBJeuWVV/SPf/yj3jEKCws1adKk6iA9Z84cff/991q3bp3+/e9/Kzg4WCdOnNDll1+u7du3n+3XBwAAAACcJY+cmZ40aZJmzpyp6dOny9vbu9b1/Px8DR8+XHv27JEkrV69WiNHjqzVLi0tTXFxcbJarYqPj9fq1asVEBBQfb24uFijR4/Wpk2bZDablZKSotjY2FrjPPjgg3rooYckSf/5z39077331rienJysUaNGyWq1asyYMVq+fPlZff/fYmYaAAAAAIzxyJnpZcuWacaMGXUGaUkKCwvTU089Vf1+yZIldbZ7+umnZbVaJUnPPfdcjSAtSYGBgXruueckSVarVc8880ytMcrLy/Xss89KkuLi4nT33XfXajN06FDNnj1bkrRixQpt3ry5kW8IAAAAAHAmjwzT9khMTKx+nZ6eXuu6zWbTJ598Iknq1auXhgwZUuc4Q4YMUc+ePSVJH3/8sX67EGDlypU6fvy4JGnWrFny8qr7f5Izl5svXbrU3q8BAAAAAHACwnQ9ysrKql/XFXD37t1b/ez16NGjGxzr9PWsrCzt27evxrUzdw1vaJz4+HgFBQVJktauXdtw8QAAAAAApzK7u4DmatWqVdWve/XqVev67t27G7x+pjOv7969W126dDE8jtlsVmxsrLZv316jjz2ysrIavJ6Tk2NoPAAAAADwdITpOlRWVuqxxx6rfj9jxoxabTIzM6tfN7Zh1+nNvX7b78z3QUFBat26daPjbN++XXl5eSotLZWfn1+D7eu6PwAAAADg7BGm6/D0009r48aNkqSpU6cqPj6+VpuCgoLq18HBwQ2Od3p5tlR1DFZd4zQ2Rl3j2BumgZakzFqpb3YdUnpukSpsNkW39teE8zsoxN/H3aWdFZvNprVp+fop87hKrZVqG+Sr8ed1UPtQf3eXBgAAgDoQpn9j1apV+vOf/yxJioiI0Isvvlhnu5KSkurXvr6+DY55Zug9depUneM0NkZj4zTkt7Phv5WTk6NBgwbZPR7gDmXWSv1veare2XhA+YVlNa499Nku/X5AtO65uKfaBjX+31Jz886GA5q/JkMZ+UU1Pv/357s1Li5S91zSQ90iQtxUHQAAAOpCmD7Dzp07NXXqVFmtVvn5+en9999XZGRknW39/X+dLTpzs7K6lJaWVr/+7fFZp8dpbIzGxmkI50bjXFdSXqGk1zdqfcbROq8Xl1XonQ0HlJx+RIvnDDmnZnP//vHPemv9gTqvWStt+mrnIa1Lz9fCGwbpgo5tXFwdAAAA6sNu3v/f3r17dfHFF+vYsWPy9vbW4sWLG9xdOyTk11mi3y7d/q2iol9nm367nPv0OI2N0dg4QEt2/4fb6w3SZ9qbX6TrF/6oykpbo22bg1dXZ9QbpM90ssSq2Qt/VG5BSaNtAQAA4BqEaUkHDx7UuHHjdPDgQZlMJi1YsEBTp05tsM+Zs72N7ZZ95jLr324GdnqcoqKi6vOmGxsnPDyc56XhMQ4cKdanPx20u/3unJP6PiXXiRU1jTJrpV5enWF3+2PF5XrbjuANAAAA1/D4MJ2fn6+LLrpIGRlVv9Q+99xzmjlzZqP9evfuXf06JSWlwbZnXo+Li3NoHKvVqvT09DrHAFqytzfsl83gRPNb6/c7p5gm9NXOQ8ovLG284Rne/fGArBWVTqoIAAAARnh0mD5x4oQuueQS7dq1S5L02GOP6bbbbrOrb5cuXRQVFSWp5pnUdVm9erUkKTo6Wp07d65xbcSIEdWvGxpn06ZN1cu8hw8fbleNQEuQnHHEJX1cLTndeI2HT5Zq7282KQMAAIB7eGyYLi4u1sSJE7VlyxZJ0t/+9jfdf//9dvc3mUyaMmWKpKoZ5fXr19fZbv369dUzzlOmTJHJZKpxPTExUaGhoZKkRYsWyVbPFNzChQurXze2BB1oSYrLKgz3KbNWNvsZ3FNlVof6FTnw9wEAAICm55G7eZeVlWnq1Kn64YcfJEl33nmnHn74YcPjzJ07V6+++qqsVqtuv/12rV69usYu26dOndLtt98uSTKbzZo7d26tMXx9fXXHHXfoX//6l3bv3q0nn3xS9957b402ycnJeu211yRJo0ePVkJCguFagXNVK3/jP6aCfL1l9q76t8LDJ0v03o+Z2nO4QJU2myJb+euygTHqExXa6Djr0vK17OccHS0sk5+Pl/pbWmv6wBi1suNM6+PFZVqyOUvbs06ovKJS7YJ9NblftAZ1aVv1vQIcOxc71MF+AAAAaFomW31ToS3Y9OnTtXTpUknShRdeqGeeeabWjPGZfH191aNHjzqv/eUvf9Fjjz0mSRowYIDuv/9+xcbGKj09XY8//ri2bt1a3e6RRx6pc4yCggLFx8drz549kqSbbrpJV155pQICArRixQo98sgjKiwsVEBAgNatW6f+/fs7+tXrlJWVVb0xWmZmJkdpoVl5YWWa/vPVL4b6TOkfpX9OPk9//2SHvvw5R9Y6dvce0LG1/v3789U7qlWtayt/ydW/lu1Sel7tJdUBPt66cpBFfxkfJ19z7cU9JeUVevjzXVqyOUsl5bVnx3u1D9EDl/aWtcKmmQs2Gvpe3SKC9d2f6j9lAAAAAK7jkWG6oeBcl06dOmnfvn11XqusrNScOXO0YMGCevvPnj1br7zyiry86l9Vn5aWpgkTJig1NbXO661atdLbb7+tSZMmGardHoRpNGdHCks19LHlKrPav2z79aQEPfLFbqXmNnzkXLCfWW/OHqQBZ5zf/NlPBzX3vW2qaOR4rZHdw7QgKUE+3r/+d11SXqGZCzZq496Gj/Hy9fbS81cP0CNfphh6BvrBS3sraXgXu9sDAADAeTz2memm4uXlpddee02ff/65pkyZoqioKPn6+ioqKkpTpkzRF198ofnz5zcYpCWpW7du2rp1qx5//HHFx8erdevWCgwMVM+ePXXXXXdp+/btTgnSQHPXLthPc0baHyDH9orQG8n7Gg3SklRYatWcNzapqLTq+eWMvELd/cFPjQZpSVqTmq8nvq45Y/6vZbsaDdKSVFZRqTve3aYbhneWvf+2FxsepMviLY03BAAAgEt45Mw0amJmGs2dzWbTXz/aocUbGz5neXi3dvrz+Dhd+txaQ+M//PvzdO2QTnrw051auG6f3f1C/Mxa/9exCvIz63hxmQY/8r1KDcyg35oYq/ah/nrw051qKL93CQvSGzcMkqVtoN1jAwAAwLmYmQbQ7JlMJj067Xw9f/UFGtS5ba3rPSKD9c8pfbTw+kFauiXL8Phvrd+vU2UV+tBg34JSqz7eli1JWrI5y1CQlqT3N2XqyoSOeu8PQ/W7Pu1l9qo5TR3Zyk9zx3XXR7cOI0gDAAA0Mx65mzeAc9PEvh00sW8HpeUWKD2vSJWVNkW1DlA/S+vqNj9nnTA8bsqhAqUeLlBBifHjqn7OOiENlrY7cN/8wjLlnDilhM5tldC5rQ6dKNGO7BMqtVaqTZCPBnVuW70rOQAAAJoXwjSAc063iBB1iwip81qZg+dLFzl47vPpjdGMbJB2pjNns9uH+qt9qL9D4wAAAMC1mPIA0KKEBfsZ7hPg4+3wMup2wb5V9w3xNdzXZJLaBRnvBwAAAPcjTANoUS7t18GhPjFtAjWgY2vDfaf0j64ao2+U4b7DY8PUzoHwDwAAAPcjTANoUSac30FtDc72Xjek8///fzsZ6tfP0lrnRYdKkgZ3bacekcGG+l87pKOh9gAAAGg+CNMAWhQ/s7cemNTb7vZXxFt0fkxVIL60X5SGdK29W3jd9/HSA5Pianz24KV95ONt38HRo3uE66Le7e2uEwAAAM0L50yDc6bRIr29Yb8e+GSnKho4wHn6BTF6fPr5NXbMPllSrpve2KT1GUfr7Rfk660/j4/TrpyTWpOap6JSq4L9zUrsEaHY8CA99lWKSsrr35BsdI9wvXjtBQr0ZQ9IAACAcxVhGoRptFiphwv05vr9+mhLtgpKq3br9vYyaWyvCF07pJNG9Qivs5+1olJf7jikN9fv18a9v4bqsGA/Tb8gWrsPndTqPfn13vfCXhHqGhakj7Zm60hRmaSqzcaGxbbTdUM66aLe7eXtZd8MNgAAAJonwjQI02jxSsordPD4KVVU2hQe4qfWgfY/U51XUKqjRWXy9/FSu2Bf3bBwU42AXZ+R3cP08rUDlVtQqlJrpdoF+zq00zgAAACaJ9YYAmjx/H281TXc2OZgp4WH+Ck8pCoEP/d9ql1BWpLWpObrnY0HdOPIrg7dFwAAAM0bG5ABgB2sFZV6e8MBQ33eWr9fLP4BAABomQjTAGCHVXvydOhkiaE++44UN7iRGQAAAM5dhGkAsMP+I8UO9TtwtKiJKwEAAEBzQJgG0OLZbDadOFWuY0VlDR6V1ZyVlFcov7BUJeUV7i4FAAAAYgMyAC1YbkGJFm/I1OKNB6qXaAf6emtK/2jNHNpJcR1a2T1Wp3aBDtVgaetYP0kqr6jU1zsP6c3k/dpwxsZngzq31bVDO2n8ee3l482/iQIAALgDR2OBo7HQIn21I0dz39umkvLKetv8YVRX/Xl8L5lMjZ/5bK2o1IjHVxh6brpzu0CtuCfRrvF/K/dkia5f+KN2HjxZb5te7UP0+vUJ6hAaYHh8AAAAnB2mNAC0OMtTDuu2d7Y2GKQl6eXVGXrsqxS7xjR7e+nqwR0N1XHN4E4OBekTp8p1zfwNDQZpSUo5VKBr5m/Q8eIyw/cAAADA2SFMA2hRrBWV+uvSHXY/G/3yqgylHi6wq+0fRnfVoM5t7Wo7snuYkoZ3rvX5weOntOdwgbKPn6q370ur0pWaW2jXfTLyivTCynS72gIAAKDp8Mw0gBbl212HDR9h9eb6/frnlPMabedn9tbr1yfozne36bvdh+ttN7FvBz11eb/q55mLSq1aujVbbyXv1y9nBPduEcG6dnBHTR8YoxB/H0lSqbVC7/+Yaaj+DzZl6k8X9ZC/j7ehfgAAAHAcYRpAi7J0a7bhPh9tybYrTEtSkJ9Z82fFa0f2Cb21fr/WpOarsNSqYD+zxvQK13VDOqtn+5Dq9geOFGvW6xu1N7/2EVlpuYV68LNdeu2HvVp4/SDFhgdr9Z58HSkytmz7WHG5VqTkavz5HQz1AwAAgOMI0wBalFyDs9KSVFBqVVGpVUF+9v9IPC86VI9N79tgm/zCUl3z2nplHq1/SbckZR49pWvnb9BHtw7XYQfql+RwPwAAADiGZ6YBtCjeXsY3/JIks7dj/Rry4sr0RoP0aTknSvTc8lSZHa6fH+cAAACuxG9fAFqUXgbOjj6tS1iQ/MxN+7xxSXmFlmzOMtTn463Z6ujgedZnLi0HAACA8xGmAbQoVw8ydnyVJF1j8Mgre6zak6cTp8oN9Skqq1BeQal6G/wHgZ6RIUqwc5dxAAAANA3CNIAW5bzoULuPr5KkYD+zLhsY0+R15BeWOtQvr6BU19dxpFZD6jqCCwAAAM5FmAbQ4jx7VX9Fhfo32s7X20vPX3OBDhwt1lc7cvTNzkPKyLPvfGdJKiy1atWePH3xc45W78lTcZm1xtiO8PPx1uXxFs2Ity/gT7sgWlc5MBsPAACAs8Nu3gBanA6hAVp663Dd9d42JWccqbNNp3aBGtsrQv/+fJf2HK4ZoId0baukYV30u/Pa19l3b36RXl2ToU+2ZquorKL68xA/s6ZdEK0bR3bV+TGhDtV+XlTVEu/Hp/dV+1b+mr92r4rPuMdpAT7eumFEZ919UU+H7gMAAICzY7LZbDZ3FwH3ysrKksVikSRlZmYqJqbpl7wC7vLLoQK9s2G/9hwuVEWlTe1D/TWxbwct/GGvkjOONtg3aVhnPTi5T43P1qXl66Y3N6uw1FpPLyk0wEevzYrXo1+maPP+Y3bXel50Ky27fWSNz06WlOvDzVla+Ute9fFdo3qE67KBMQoN8LF7bAAAADQtwjQI0/A417++USt+ybOr7dxx3TV3XA9JVcF82gs/1JiNrk8rf7PuvriH5n26y+66nr6in6YO4L8/AACAcwHPTAPwKOszjtgdpCXppVXpOlFctSv3/y1PtStIS9LJEqt2ZJ/UrYmxdrW/fnhngjQAAMA5hDANwKO8uX6/ofYl5ZX6YHOmcgtK9M3OQ4b6frb9oP4wKlb/nNJH4SF+dbZpF+Srv0+M07xL+9R5HQAAAM0TG5AB8Cjf7TpsuM83uw6rlb+PyiuMPRVTUl6pNWl5mjm0s64a1FFf7jiklb/kqrDEqmA/s0b2CNOE8zvIz+xtuCYAAAC4F2EagMcotVao1FppuN/JU+U6WVLu0D1PnKrq5+Ptpcn9ojS5X5RD4wAAAKB5YZk3AI/h6+0lby+T4X4Bvt7y93Fs9jjQl1lnAACAlogwDcBjmEwmDe7S1nC/IV3baUjXdob7eZmkQV2M9wMAAEDzR5gG4FGuG9LJUHsvk3TN4I7qFhGsoQYD9YW9IhTdOsBQHwAAAJwbCNMAPMpFvSPVIzLY7va/7x+tmDaBkqRbx8TK3lXiZi+T/jDavmOxAAAAcO4hTAPwKGZvLy1ISpClbeMzxkO7ttMj086vfj+ye7gemnKeTI0Eam8vkx6ddr4SOhtfUg4AAIBzA2EagMeJaROopbcM1xXxFvn71P4xGBbsq9sv7KZFNwyqtfHYdUM66bVZ8epnaV3n2PGd2mjh9Qm6PN7ijNIBAADQTJhsNpuxg1PR4mRlZcliqfrFPzMzUzExMW6uCHCdE6fK9fn2HOWcOCUvk0ndIoJ1SZ/28jU3/m+NO7JPaNWePBWVWhXi76PEnuGK69DKBVUDAADA3QjTIEwDAAAAgEFmdxcAAAAAoBGVlVLat9IvX0qlJyWfQKnLKKn37yWzr7urAzwSYRoAAABozrZ/IC3/l3R8f83Pt74pff1Xadgd0vA73FMb4MHYgAwAAABorpKfl5beWDtIn1aUJ337D2nZXa6tCwBhGgAAAGiW9q6Wvv6bfW03LZA2L3RqOQBqYpk3ADhBzolTWrwxUz9nHVd5hU3tgn01uV+UxvSMkJdXIwdVAwAgVc1Ky8Bewev+Jw1MclY1AH6DMA0ATaiw1Kq/ffSzlm3PUUVlzV+APtl2UDFtAvTgpX00rnekmyoEAJwTjh+QUr8x1udIqpSxSuo62jk1AaiBZd4A0ESKSq266pX1+mTbwVpB+rSsY6d005ub9Mm2bBdXBwA4p+T8JNkqjfc7uKXpawFQJ8I0ADSRBz7ZqZ+zTzTartIm3fvBdu3LL3JBVQCAc5K11LX9ABhGmAaAJpBfWKrPfjpod/uyikq9tb6enVkBAAgKc6xfYLumrQNAvQjTANAE3t+UqbIKY8vxPticpVJrhZMqAgCc0zoNl4IN7q/h5SPFTXZOPQBqIUwDQBPYnVNguM+JU+XKOV7ihGoAAOc8bx/pglnG+sRdKoWwwSXgKoRpAGgC5VYHNomRVG5wNhsA4EGG3CK1jbWvbUAbaYydZ1IDaBKEaQBoApGt/Az38TJJYcHG+wEAPERgW2nmx1JYj4bbBYVL1y6Vwrq5pCwAVQjTANAEpgyINtxndI9wtQnydUI1AIAWo3VHac4KacKTUnhczWutYqQL/y7dul6KvsA99QEezOzuAgCgJbigYxv1iWqlnQdP2t3nuqGdnFgRAKDF8AuWBs2p+nN0r1RyQvINltp2kby83V0d4LGYmQaAJvLPKefJz2zfj9Xx57XXmJ4RTq4IANDitO0iRfWvWtJNkAbciplpAB4rt6BE723M1IdbsnTweIm8vKTuESG6alBH/X5AlAJ96/8RuT7jiN5av19rUvNVVGpVsL9ZiT3C9efxvfTfb/aooNRab9+J53fQf6/oJ5PJ5IyvBQAAABcw2Ww2m7uLgHtlZWXJYrFIkjIzMxUTE+PmigDnezN5n/65bJfKK+r+Edg60EcvXHOBhsWG1fj8RHG5bn5rs5IzjtQ79tCu7TS4a1t9sClL2cdPSZLMXiaNi4vUzKGdNKxbWL19AQAAcG5gZhqAx1m0bp/mfbqzwTbHi8uV9PqPemv2YA3q0laSVFxm1XULNmh71okG+yZnHFF5RaW+vWuUTpVXqKyiUq0DfBXgy3I8AACAloJnpgF4lIPHT+mfy3bZ1bbMWqm73tumysqq2esXV6Y3GqRP27T/mBYm71O7YD91CA0gSAMAALQwhGkAHuWdDQdUUWn/0y3Zx09peUquyisqtXhjpqF7vb3+QHUQBwAAQMvCMm8AHuWDzcYCsSS9vylTFTab8gtLDfXLPn5K69KPaER3npEGAABoaZiZBuAxyisqdfiksUAsSVnHTin72CmH7pl1rNihfgAAAGjemJkG4DEcPYjKy0ty9BQrL46/AgD8VvkpqfBw1evgSMknwP6+RflSyQnJN0gKae+c+gDYhTANwGOYvb3UsW2gDhw1NlvcJSxYXcKCHLpnl3DH+gEAWqCcn6SNr0o7PpTK////X+QTJJ1/mTRojtT+/Lr7WUulHUulH+dL2Zt+/TyspxR/g9T/Ksk/1Pn1A6iBZd4APMqVgyyG+1yVYNGo7uGKbm1g5kBSt4hgJXRua/h+AIAWaOVj0sujpK1v/hqkJam8SNqySHpphLT6idr9Cg5J88dKH99cM0hLUv4v0lf3Sy8MlQ7bd1IFgKZDmAbgUa6It8jfx/4ffd0jgjWsW5i8vEy6bmgnQ/eaabA9AKCFWvu0tPLRxtstf1ha979f35eclN6cKh36ueF+J7OlN6ZIx/afXZ0ADCFMA/Ao7YL99NTl/eVlx6PMIf5m/e/qC6rf3ziii0b1CLfrPhf3jtS1gwnTAODxCnOl5f+2v/33/6x6LlqSNrws5do541yUK614xHh9ABxGmAbgcSb27aAXrx2o1oE+9bbpEhakD24eqp7tQ6o/M3t76dWZAzVtQHS9G5J5maSrBln0/DUXyMuexA4AaNk2L5Iqy+1vX1EqbXlDqqyQNr9u7F47P5KKjhjrA8BhbEAGwCNd0qe9RvcI17LtOVqyOVM5J0rkbTIpNiJYVw/uqMQe4TLVkZj9zN767xX9dee47np7wwGt3pOnwlKrQvx9lNgzXNcM7qiYNoFu+EYAgGbp5w8c6LNEiomvWr5tREWptPuTqk3JADgdYRqAx/L38dZlA2N02cAYw307tQvSXyfE6a8T4pxQGQCgxTh9BJbRPoW5Dt7PwX4ADGOZNwAAAOAs3r7G+5j9HOvn6P0AOIQwDQAAADhLh37G+7Tv+//PnHZg740OfY33AeAQwjQAAADgLI48v5wwW2rbRYq90Fi/Nl2k2LHG7wfAIR4bpnNzc7Vs2TI98MADGj9+vMLCwmQymWQymZSUlNRo/3379lW3t/dP586d6xwrMTHR7jEAAABwDunxO6ldN/vbh/WUuo2rej30VmP3GnKL6j1uAkCT89gNyCIjI11+z549e7r8ngCcJ6+gVOszjqio1Kpgf7OGxYapbRDPqgEAzuDlJV25WHp9vFSc33DboAjpqsW/BuJu46Qxf5NW2HFOdd8rpUE3nX29AOzmsWH6TBaLRXFxcfrmm2/s7hMdHa2ff/650XaPPvqo3nnnHUnSrFmzGmwbHx+v1183eJ4gAJfbefCEXlqVoa93HFJZRWX1575mL006v4NuToxVj8iQBkYAAHiU8B7Sjd9Kn9wu7V9bd5vOI6Up/5PadK75+ej7pKBwaeWjde8M7teqakY68S/MSgMu5rFh+oEHHlBCQoISEhIUGRmpffv2qUuXLnb39/Hx0Xnnnddgm4qKCq1cuVKSFBISot///vcNtg8KCmp0TADu9d2uw/rj4i0qKa+sda3MWqmlW7P11c5Devm6gRrZPdwNFQIAmqW2XaXrP5cO75K2vikd21f1eZsu0gUzpYhe9feNv14acK206xNpz1dSyQnJN6gqgPe9QvILdslXAFCTx4bphx56yOn3+O6773Tw4EFJ0mWXXabAwECn3xOA82zPOq7b3tmiUmvtIH2m4rIK/eHNzfro1uHq2Z4ZagDAGSJ7S7971Hg/bx/p/Muq/gBoFjx2AzJXeOONN6pfN7bEG0Dz99zytEaD9GnFZRV6cWWakysCAACAuxCmnaSgoEAff/yxJKlTp04aNWqUewsCcFYOHj+l5Sm5hvp8seOQjhSWOqkiAAAAuJPHLvN2tiVLlqi4uFiSNHPmTLuOtUpJSVFCQoJSUlJUXl6usLAwDRw4UNOnT9dVV10lHx8fh2rJyspq8HpOTo5D4wKeZPWePFVU2gz1KbNW6of0I5rcL8pJVQEAAMBdCNNOcuYS75kzZ9rV5/Dhwzp8+NddGrOzs5Wdna1PP/1Ujz/+uJYsWaK4uDjDtVgsFsN9ANRUWGp1qF9BSXkTVwIAAIDmwOVh2svLS15eXtq+fbt69+5tV5/09HR1795dXl5eslod+4XWlQ4cOKBVq1ZJkoYNG6Zu3bo12N7Ly0tjx47VhAkT1K9fP7Vr104FBQXasmWLXn75Ze3evVu7du3SmDFjtHHjRnXs2NEVXwPwOOszjmhdWr6KyyoU4u+jsXEROi86VJIU6OvYj8sgB/sBAACgeXPLb3k2m7Glkmfbz9Xeeuut6lrtmZVeunSpWrduXevzkSNH6tZbb9WcOXO0aNEiHT58WHPnztXSpUsN1ZOZmdng9ZycHA0aNMjQmEBL8ulPB/W/5anac7iwxudPf7dH/S2tdffFPTS8WzuZTJKRH0NmL5OGxrZr4moBAADQHJwTUyang6k9zx03B2+++aYkyc/PT1dccUWj7esK0qf5+Pho/vz52rBhg1JSUvTRRx8pOztb0dHRdtcTExNjd1vA0zy/Ik1PfP1Lvde3ZR5X0us/6onL+mpU93Ct2pNn99gX9Y5UZCv/pigTAAAAzcw5sZv3kSNHJElBQUFurqRxGzduVEpKiiRp8uTJDQZle5nNZs2ePbv6/ekl5ADOzlc7choM0qdVVNp035Lturh3pMxe9v2jnq+3l24eHXu2JQIAAKCZcluYtneWuaioSM8995wkKTa2+f9i6sjGY/Y48/ny7OzsJhsX8GQvrEy3u6210qYf0vP1xOV9Gw3Uvt5eeubK/upnaX2WFQIAAKC5cvoy765du9b5+cUXX9zoUU+lpaXKzc1VZWWlTCaTLr30UmeU2GTKy8v13nvvSZIiIiL0u9/9rsnGPleeFwfOFT9lHtf2rBOG+nyz87AevLSP3r5xsP63Ik1r0/JrPEPtZZISe0bojxd20wUd2zRxxQAAAGhOnB6m9+3bV+szm81meHZ1yJAhuu+++5qoKuf4/PPPlZ+fL0m6+uqrZTY33V/vrl27ql9HRXFmLXC2ftx31HAfa6VNWw4c0+/O66DBXdtpX36R1qTlq6jUqmA/s0b3CJelbaATqgUAAEBz4/QwPWvWrBrvFy1aJJPJ1OjzxCaTSf7+/urQoYOGDRumCy+8sNlvQHbmEu/ffu+zYbVatWDBgur3o0aNarKxAU91qqzCoX7FZ/TrHBakzmHNfy8HAAAAND2nh+nXX3+9xvtFixZJkv7973/bfc70ueDo0aP6/PPPJUnnn3+++vfvb1e/FStWaMCAAfX+w0J5ebnmzJlTvanZpZdeKovF0hQlAx4tNLDhx0zq7RdQ1e9UWYU+2ZatNan5Kiy1KtjfrMQe4bq0X5T8fbybslQAAAA0Qy4/GmvevHmSqp4pdqe1a9cqLS2t+v3p5dmSlJaWpoULF9Zon5SU1OB47777rsrKyiQZm5VetGiRJk+erMmTJysxMVE9e/ZUq1atVFhYqM2bN+vll1/W7t27JVX9nT377LN2jw2gfuPiIvXQZ7tUUWn/fgSt/M0aFhum/y1P1SurM3SyxFrj+ufbc/TvL3brtsRumjOq7v0iAAAA0DKYbB66s1VSUlL1LLk9GvtrGjJkiDZs2CBvb29lZWWpffv2TVrH+eefr3fffdcps/lZWVnVs92ZmZmcSw2PMeeNTfp212G7218/vLPKrJV6e8OBRtveOKKL/j6p5ay+AQAAQE0un5luiVJTU7VhwwZJ0kUXXWR3kJak+++/X/3791dycrJ27dqlvLw8HT16VH5+foqMjFR8fLwuu+wyTZ06Vd7eLB0FmtLccd31Q1p+jeeg6xMW7Kfo1gF6+PPddo09f+1e9e/YWpP6smEgAABAS+TWmenKykrt2rVLGRkZKigoUEVF47/QNuXZzajCzDQ82drUfN381mYVllrrbRMe4qeF1yfo7vd/UsqhArvHHtCxtT66dXhTlAkA8HQlJ6Rt70i/fFn12jdY6jJSGpgkhdg/kQOg6bglTJ86dUoPP/ywXn31VR05csTufiaTSVZr/b/wwjGEaXi6fflFWvDDXn20JVsFZ4TqdkG+mpFg0fXDOivzWLGmv5hseOwv7hip3lGtmrJcAICnWfm49MMzUnlx7WteZmnAtdL4JySzr8tLAzyZy5d5nzp1ShdeeKE2btzY6HPIAOAKncOC9M8p5+n+3/XST1nHVVxaoVYBPupvaS1fs5ck6audhxwae1fOScI0AMBxX9wrbXyl/uuVVmnzQunkQenKxZI3T3ECruLy/9qefvrp6ueLzzvvPP3xj3/UwIED1bZtW3l5ebm6HACoFuRXtVt3XcqslQ6N6Wg/AAC065OGg/SZUr+pmr0edY9TSwLwK5eH6ffee0+SNGzYMC1fvly+vixHAdD8hYf4ubQfAABa/5Kx9j++Jg2fy+w04CIunwpOT0+XyWTSfffdR5AGcM4YFxepYD9jv5y0DfLVqB51z3QDANCgvF+kA+uM9Sk4KKV+7Zx6ANTi8jB9OkB37NjR1bcGAIcF+Zk1dUC0oT6Xx8fIz8yRdgAAB+TadxRjk/UDYJjLw3SvXr0kSYcOObaZDwC4y+1juym6dYBdbbuEBenmUbFOrggA0GLZGj8ytk6VDvYDYJjLw3RSUpJsNps++OADV98aAM5KRIi/3pkzWJ3bBTbYrntEsN66cbDaBPEoCwCgHge3SZ/cJj3ZQ3q4vfREN+nDG6UDVRv1KiTKsXFbOdgPgGEuP2faZrNp3LhxWr16td544w1dddVVrrw96sA504AxxWVWfbz1oN5av1+7ck5Wf943JlTXDumkyf2i5O/D8m4AQB3KiqWlc6SUZfW36ZooXb5QevVC6WiG/WP7BEl3p0j+HMkIuILTwvSBAwfqvVZYWKibbrpJycnJmj59uq6++mr16tVLgYENz/ZIPGvtDIRpwHH5haUqKrUqxN9HbZmJBgA0xFomvT1d2ru68bZRF0hxk6Tv/2n/+AOTpEufdbg8AMY4LUx7ezc+K2Oz2WQymewe02QyyWq1nk1ZqANhGgAAwAXWPSd983f72w+fKx1IljI3NN62dSfpxu+l4HCHywNgjNOembbZbI3+sbfdb/sAAAAA5xSbreocaCO2vSNd8ZbUeWTD7cJ7SUnLCNKAizntRPfXX3/dWUMDAAAA55Z9a6Vje431KcqVDqyXZn0mpX9fFcb3fP3rTt+dhksJs6W4yZK3T9PXDKBBTgvTs2bNctbQAAAAwLnleP37CTXaz2SSuo2r+lNZKZUVSr5BkhebXQLu5PKjsQAAAACPY2CfoAb7lZ6USo5Lp46fbUUAzpLTZqYBAAAA/H9tuzrer6Jc2vWJ9OP8qg3JTovoIyXcIPW9UvILbpo6AdiNmWkAAADA2ToOkcJ6GusTEiV16C+9dpH04eyaQVqScndKn98tvTBUyvulyUoFYB+Xz0zfcMMNhvuYTCb5+/srNDRU3bt315AhQxQXF+eE6gAAAAAnGTRH+uIe+9sPuEZ6+3Lp8M8NtztxQFo0WZrzvRTKEaeAq7g8TC9cuNDQ2dL1iY+P13//+18NHz68CaoCAAAAnCz+Bil9ufTLF4237TxS8vJpPEifVnhIWv5vaeqLZ1cjALu5fJl3x44d1bFjR4WFhdU4P9rX11eRkZGKjIyUr69vjXOlw8LCFBMTo1atWlV//uOPP2r06NF6++23Xf0VAAAAAOO8vKXLF0kDrpXUwORS799LV70rbX3T2Pg7l0rFR8+mQgAGuDxM79u3Tx999JFCQkLk6+uru+66S1u3blVRUZEOHjyogwcPqqioSFu3btXcuXPl4+Oj4OBgffTRRzp27JgyMzP1+OOPKyQkRJWVlbrxxhuVmZnp6q8BAAAAGGf2laY8L92xRRp2e9Vz1CFRUrvu0qCbpNs2SjMWSTnbpBMGf8e1llRtVAbAJUy209O/LnL48GENGDBAJ0+e1LfffquhQ4c22D45OVnjxo1TSEiItm7dqg4dOkiStm7dqhEjRqikpET33HOPHn/8cVeU3yJlZWXJYrFIkjIzMxUTw7M2AAAAbvXzkqpNx4xK/KuUeH/T1wOgFpfPTD/11FM6dOiQ/vSnPzUapCVp6NCh+tOf/qTc3Fw98cQT1Z8PGDBAN9xwg2w2m7799ltnlgwAAAC4lrevg/18mrYOAPVyeZj+5JNPZDKZdMkll9jd53e/+50k6fPPP6/x+fjx4yVVLR0HAAAAWozIPmrwuer6tD+/yUsBUDeXh+msrCxJkp+fn919Trc93fe0qKgoSVJxcXETVQcAAAA0A+1ipa6jjfVp3UmKHeucegDU4vIwHRgYKEnatGmT3X1+/PHHGn1PKy0tlSS1adOmiaoDAAAAmokhtxlrP/hmycvlv94DHsvl/7UNHDhQNptNjz76qI4cOdJo+/z8fD322GMymUyKj4+vce2XX36RJEVERDilVgCerdRaoRW/5OrDzVn6fHuOMo+yCgYA4EI9LpZG3Wtf2z7TpCG3OLceADWYXX3DW2+9Vd98842ysrI0ZMgQPfPMM5owYYJMpprPhNhsNn3++ee66667lJmZKZPJpNtuq/mvc1999VWdIRsAzkZ+YaleXZOhDzZl6WhRWfXnXiZpVI9wzRnZVcO7hbmxQgCAx7jw71JQhLTqMam4jokonyBp0Bxp7DzJ5MAz1gAc5vKjsSTp5ptv1iuvvFIdoNu1a6f+/ftXzzDn5uZq27Zt1TPXNptNN998s1544YXqMQ4dOqSuXbuqpKRE77//vi677DJXf40Wg6OxgF/tyy/Sta9tUNaxU/W2MZmkv02I040ju7qwMgCAR7OWSjs/kn75Uio5LvkGS11GSf2ukvxbubs6wCO5JUxL0qOPPqp//etfKikpqSqkjplpqWrzsXnz5unPf/5zjevFxcXKy8uTJEVHR8tsdvkke4tBmAaqnCwp16XPrdX+I/Yt5/6/qwZocr8oJ1cFAACA5shtYVqqml1etGiRvv/+e+3YsUPHjh2TVLWhWJ8+fTR27FjNmjVLHTp0cFeJHoEwDVSZvyZDD3++2+72ndsFasU9ibX+MRAAAAAtn1vDNJoHwjRQtRrmwqdWaW9+kaF+i24YpNE9wp1UFQAAAJor9s4HAEkZ+UWGg7Qkfb/7sBOqAQAAQHNHmAYASQUlVpf2AwAAwLmNMA0AkgJ9vV3aDwAAAOc2p22B3bVr1ZExJpNJ6enptT53xG/HAoCmEhserPat/HXoZImhfpw3DQAA4JmcFqb37dsnqfaRV6c/dwQ75gJwFm8vk64a1FFPf7fH7j6Rrfx0ce9IJ1YFAACA5sppYXrWrFmGPgcAd7tmSEe9uX6f8gvL7Gp/y+hYmb15WgYAAMATcTQWOBoLOMP2rOOauWCjjheXN9juhuGdNbJ7uFan5qmo1KpgPx+N6RWuEd3CWEUDAADgAQjTIEwDv7E3v0hPf7tHX+04pLKKyhrXekaGqL8lVMkZR3XgaHGtvl3DgjT3oh6a3C/KVeUCAADADQjTIEwD9cgvLNXXOw/paGGZ/H281c/SWsnpR+x6rvqvE3rpplGxLqgSAAAA7uC0Z6aNyM7O1qFDh1RcXKz4+HgFBAS4uyQAUFiwn64Z3Kn6/Vc7Dtm9QdmjX6aoR2SIEntGOKs8AAAAuJHbds4pKCjQvHnzZLFY1LFjRw0aNEiJiYnau3dvjXbvvvuuZsyYoTlz5ripUgCo8vJq+4/ms9mkV9dkOLEaAAAAuJNbZqbT0tI0fvx4ZWRk6MxV5nVt2jN06FBdd911qqys1KxZszRixAhXlgoAkqQd2Se09cBxQ33WpR9Rel6hYsODnVMUAAAA3MblM9OlpaWaOHGi0tPTFRgYqPvuu0/Lli2rt32nTp00ZswYSdKnn37qqjIBoIatmccN97HZpJ8c6AcAAIDmz+Uz0y+99JJSU1MVFBSkNWvWqH///o32GT9+vL777jslJyc7v0AAqENpeYVD/UrKKxtvBAAAgHOOy2emly5dKpPJpDvvvNOuIC1Jffv2lSSlpqY6sTIAqF+bQF8H+/k0cSUAAABoDlwepnft2iVJuvjii+3u065dO0nS8ePHnVESADRqbFyE/MzGfmQG+5k1qke4kyoCAACAO7k8TBcUFEiSQkND7e5TUlIiSfLxYYYHgHu0DvTVpL5RhvpMHRCtIL9mcQIhAAAAmpjLw/TpWebDhw/b3efnn3+WJEVGRjqlJgCwx51ju6ttkH3LvSNC/HRLYqyTKwIAAIC7uDxMn35O+vvvv7e7z4IFC2QymTR48GAnVQUAjevYLlCLrh+ksGC/Btu1b+WvN2YPUlTrABdVBgAAAFdzeZieNm2abDabXn75Ze3fv7/R9g899JA2bNggSbriiiucXR4ANOj8mFB9eedI3Tm2uyJb1QzVUaH+uufiHvrizpHq1b6VmyoEAACAK5hsNpvNlTesqKhQ3759lZKSIovFoueff14TJkyQt7e3TCaTduzYoZ49e+qHH37Qf/7zH33xxReSpISEBK1fv96VpXqMrKwsWSwWSVJmZqZiYmLcXBFwbrBWVOqXwwUqKq1QsJ9ZPduHyNvL5O6yAAAA4AIuD9OSlJ6eruHDhys3N1cmk0mBgYEqKiqSyWRSly5ddPjwYRUXF0uSbDaboqKilJycXB340LQI0wAAAABgjMuXeUtSbGystm3bpokTJ8pms6moqEhSVXDOyMhQUVGRbDabbDabLr74Yv34448EaQAAAABAs+G2M1vat2+vzz77TDt37tQnn3yiTZs2KTc3VxUVFWrXrp0GDBigKVOmKD4+3l0lAgAAAABQJ7cfgNqnTx/16dPH3WUAAAAAAGA3tyzzBgAAAADgXOb2memCggLt3btXBQUFqqioaLT9qFGjXFAVAAAAAAD1c1uYfvXVV/XCCy9o+/btdvcxmUyyWq1OrAoAAAAAgMa5PExXVFRo+vTp+uyzzyRV7eANAAAAAMC5xOVh+qWXXtKnn34qSYqMjNT111+vgQMHqm3btvLy4hFuAAAAAEDz5/Iw/cYbb0iSevfurTVr1qhNmzauLgEAAAAAgLPi8qng3bt3y2Qy6R//+AdBGgAAAABwTnLbuuqePXu669YAAAAAAJwVl4fp7t27S5KOHj3q6lsDAAAAANAkXB6mr7zyStlsNi1btszVtwYAAAAAoEm4PEzfcccd6tu3r1588UWtWbPG1bcHAAAAAOCsuTxM+/n56ZtvvtHAgQN10UUX6b777tO2bdtUUlLi6lIAAAAAAHCIyWaz2ZwxsLe3d6NtbDabTCaT3WOaTCZZrdazKQt1yMrKksVikSRlZmYqJibGzRUBAAAAQPPmtHOm7c3oTsryAAAAAAA4jdPC9Lx585w1NAAAAAAAbuW0Zd44d7DMGwAAAACMcfkGZAAAAAAAnOsI0wAAAAAAGOSxYTo3N1fLli3TAw88oPHjxyssLEwmk0kmk0lJSUl2jbFw4cLqPo39WbhwYaPjFRcX64knntCgQYPUtm1bBQcHKy4uTvfcc48OHDhwdl8YAAAAANBknLYBWXMXGRnp7hJqSE9P18SJE/XLL7/U+DwlJUUpKSmaP3++3nnnHU2YMMFNFQIAAAAATvPYMH0mi8WiuLg4ffPNNw6P8fXXXysqKqre6w1t6lVYWKhJkyZVB+k5c+boyiuvVEBAgFasWKFHH31UJ06c0OWXX67k5GT17dvX4ToBAAAAAGfPY8P0Aw88oISEBCUkJCgyMlL79u1Tly5dHB6vR48e6ty5s0N9n3zySaWkpEiS/vOf/+jee++tvjZ06FCNGTNGo0aNUnFxsebOnavly5c7XCcAAAAA4Ox57DPTDz30kCZNmuT25d7l5eV69tlnJUlxcXG6++67a7UZOnSoZs+eLUlasWKFNm/e7NIaAQAAAAA1eWyYbi5Wrlyp48ePS5JmzZolL6+6/yc5c1O0pUuXuqAyAAAAAEB9CNNutmbNmurXo0ePrrddfHy8goKCJElr1651el0AAAAAgPp57DPTTS0pKUm7d+/WsWPH1KpVK3Xr1k3jxo3TLbfcoujo6Hr77d69u/p1r1696m1nNpsVGxur7du31+hjj6ysrAav5+TkGBoPAAAAADwdYbqJrFq1qvr1kSNHdOTIEW3YsEFPPfWUnnnmGf3hD3+os19mZqYkKSgoSK1bt27wHhaLRdu3b1deXp5KS0vl5+dnV20Wi8W+LwEAAAAAsAth+ix17dpV06ZN09ChQ6tDa0ZGhj788EMtWbJEJSUluvnmm2UymXTTTTfV6l9QUCBJCg4ObvRep5d5S1XHadkbpgEAAAAATYswfRamTp2qWbNmyWQy1fg8ISFBV1xxhZYtW6Zp06apvLxcd911lyZPnqz27dvXaFtSUiJJ8vX1bfR+Z4bnU6dO2V3n6dnv+uTk5GjQoEF2jwcAAAAAno4NyM5CaGhorSB9pkmTJmnevHmSpOLiYr322mu12vj7+0uSysrKGr1faWlp9euAgAC764yJiWnwT4cOHeweCwAAAABAmHa6OXPmVAfuM5+rPi0kJERS1bLtxhQVFVW/tmdZOAAAAADAOQjTThYREaGwsDBJUnZ2dq3rMTExkqqC8unzputzerl2eHg4z0sDAAAAgBsRpl3AZrPVe613797Vr1NSUuptZ7ValZ6eLkmKi4truuIAAAAAAIYRpp0sNzdXR44ckSRFRUXVuj5ixIjq13UtAz9t06ZN1cu8hw8f3sRVAgAAAACMIEw72SuvvFI9Mz169Oha1xMTExUaGipJWrRoUb2z2AsXLqx+PXXq1KYvFAAAAABgN8K0g/bt26etW7c22GbZsmX617/+Jalq1+7rr7++VhtfX1/dcccdkqTdu3frySefrNUmOTm5eifw0aNHKyEh4WzLBwAAAACcBY89Z3rt2rVKS0urfp+fn1/9Oi0trcZMsCQlJSXVeL9v3z6NGTNGQ4cO1aWXXqr+/fsrIiJCNptNGRkZWrJkiZYsWVI90/zkk08qOjq6zlruvfdevffee9qzZ4/uu+8+paWl6corr1RAQIBWrFihRx55RFarVQEBAXrmmWea5PsDAAAAABxnsjW0O1YLlpSUpEWLFtnd/rd/TStXrtSYMWMa7RcYGKinn35aN910U4Pt0tLSNGHCBKWmptZ5vVWrVnr77bc1adIku2u2V1ZWliwWi6SqHcNP7zAOAAAAAKibx85Mn62BAwfqrbfeUnJysjZt2qScnBzl5+fLarWqTZs26tOnj8aOHasbb7xRERERjY7XrVs3bd26Vc8//7w++OADpaWlqaysTBaLRRMmTNCdd96pTp06ueCbAQAAAAAa47Ez0/gVM9MAAAAAYAwbkAEAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQWZ3FwAA56rUwwVanZqvolKrgv3MSuwZrq7hwe4uCwAAAC5AmAYAg35Iy9f/fZ+qDXuP1vj8X59Lw2PDdMfY7hrUpa2bqgMAAIArsMwbAAx4f1OmZi7YWCtIS5LNJq1Ny9c189frk23ZbqgOAAAArkKYBgA7Jacf0V+W/qyKSluD7corbLrng5+05cAxF1UGAAAAVyNMA4CdXliZ1miQPq28wqaXVqY7uSIAAAC4C2EaAOywN79Ia9PyDfX5PiVXOSdOOakiAAAAuBNhGgDssC49Xzb7JqWrVVTatCGj9rPVAAAAOPcRpgHADqfKKhzqV1RmbeJKAAAA0BwQpgHADiH+jp0kGOLv08SVAAAAoDkgTAOAHUb3iJDZy2Soj5/ZSyO7hTmpIgAAALgTYRoA7NA+1F/j4iIN9Zl4fge1CfJ1UkUAAABwJ8I0ANjpjxd2U4CPt11tg/3MuiUx1skVAQAAwF0I0wBgp/OiQ/XCtRc0GqhD/Mx6ZeZAdY8McVFlAAAAcDXCNAAYMKZnhD7543BNvyBGfuaaP0IDfLx1ZYJFH/9xuIbF8qw0AABAS2ay2YyenIqWJisrSxaLRZKUmZmpmJgYN1cEnBuOF5dpw96jKiq1KtjPrMFd2yk0gN27AQAAPIFjZ70AANQ60FeX9Gnv7jIAAADgBizzBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADPLYMJ2bm6tly5bpgQce0Pjx4xUWFiaTySSTyaSkpCS7xigpKdEnn3yi22+/XYMHD1bbtm3l4+Ojtm3baujQoXrwwQeVk5PT6DiJiYnV927sDwAAAADA/czuLsBdIiMjz6r/9u3bNWLECBUUFNS6duzYMa1fv17r16/Xf//7X82fP18zZsw4q/sBAAAAAJoPjw3TZ7JYLIqLi9M333xjd5+TJ09WB+nhw4dr0qRJio+PV7t27ZSXl6elS5dq/vz5Kigo0NVXX62QkBCNHz++wTHj4+P1+uuvn9V3AQAAAAA4n8eG6QceeEAJCQlKSEhQZGSk9u3bpy5dutjd38vLSzNmzNC8efPUu3fvWtcvvvhijR8/XlOnTlVFRYVuv/12paamNrhUOygoSOedd55D3wcAAAAA4DoeG6Yfeuihs+o/bNgwDRs2rME2U6ZM0bRp0/Thhx8qPT1d27Zt04ABA87qvgAAAAAA9/PYDchcZcyYMdWv09PT3VgJAAAAAKCpEKadrLS0tPq1lxd/3QAAAADQEnjsMm9XWbVqVfXrXr16Ndg2JSVFCQkJSklJUXl5ucLCwjRw4EBNnz5dV111lXx8fByqISsrq8Hr9hzfBQAAAAD4FWHaiX766Sd9/vnnkqQ+ffrUuVHZmQ4fPqzDhw9Xv8/OzlZ2drY+/fRTPf7441qyZIni4uIM12GxWAz3AQAAAADUjzDtJKWlpbrxxhtVUVEhSXrkkUfqbevl5aWxY8dqwoQJ6tevn9q1a6eCggJt2bJFL7/8snbv3q1du3ZpzJgx2rhxozp27OiqrwEAAAAAqANh2kn++Mc/atOmTZKkWbNmafLkyfW2Xbp0qVq3bl3r85EjR+rWW2/VnDlztGjRIh0+fFhz587V0qVLDdWSmZnZ4PWcnBwNGjTI0JgAAAAA4MkI007w6KOPav78+ZKkgQMH6vnnn2+wfV1B+jQfHx/Nnz9fGzZsUEpKij766CNlZ2crOjra7npiYmLsbgsAAAAAaBzbSzexl19+WX/9618lST179tSXX36poKCgsxrTbDZr9uzZ1e/P3NQMAAAAAOB6hOkmtHjxYt16662SpE6dOum7775TeHh4k4x95uZl2dnZTTImAAAAAMAxLPNuIp9++qlmzpypyspKdejQQd9//32TLq+22WxNNhaAs1dSXqFPtmVrdWq+ikqtCvYza0zPCE3q10F+Zm93lwcAAAAnI0w3ge+//14zZsyQ1WpVu3bt9O233yo2NrZJ77Fr167q11FRUU06NgBjnl+RpldWZ+jEqfIany/bnqN/f7FbtybG6saRXd1UHQAAAFyBMH2W1q1bpylTpqi0tFStWrXS119/rT59+jTpPaxWqxYsWFD9ftSoUU06PgD7/f3jn/XW+gP1Xj9aVKaHP9+twydL9LeJDZ8tDwAAgHMXz0yfhW3btmnixIkqKipSUFCQvvjiCw0cONDQGCtWrNDx48frvV5eXq4bb7xRKSkpkqRLL71UFovlbMoG4KAPNmU2GKTP9Oqavfp8e46TKwIAAIC7eOzM9Nq1a5WWllb9Pj8/v/p1WlqaFi5cWKN9UlJSjffp6em65JJLqoPwww8/rNDQUO3YsaPee0ZERCgiIqLGZ4sWLdLkyZM1efJkJSYmqmfPnmrVqpUKCwu1efNmvfzyy9q9e3d1/2effdaBbwugKby2dq/B9hma2LeDk6oBAACAO5lsHrqzVVJSkhYtWmR3+9/+NS1cuFDXX3+9oXvOmzdPDz74oEN1nH/++Xr33Xdr7OrdVLKysqpnuzMzMzmXGqjD5v1HNf3FZMP9vrhjpHpHtXJCRQAAAHAnj52Zbi7uv/9+9e/fX8nJydq1a5fy8vJ09OhR+fn5KTIyUvHx8brssss0depUeXuzQzDgLjsPnnSo366ck4RpAACAFshjZ6bxK2amgcbNX5Ohhz/fbbjfI1PP19WDOzqhIgAAALgTG5ABgB3CQ/xc2g8AAADNG2EaAOwwNi5SwX7GnoxpE+ijUT3CnFQRAAAA3IkwDQB2CPYza+qAaEN9ZsRb5GdmrwMAAICWiDANAHa6fWw3RbcOsKtt53aB+sPoWCdXBAAAAHchTAOAnSJC/PX2jYPVqV1gg+26RQTrrRsHq22Qr4sqAwAAgKtxNBYAGNA5LEhf3jlSH23N1pvJ+5VyqKD62vnRobpuSCdN7h8lfx+WdwMAALRkHI0FjsYCzkJeQamKSq0K9jcrLJiduwEAADwFM9MAcBbCQ/w4/goAAMAD8cw0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGGR2dwEAcK76OeuEVqfmqbDUqmA/s8b0jFDvqFbuLgsAAAAuQJgGAIOWpxzWs9+l6qesEzU+f+LrXzSwUxvNHdddI7uHu6k6AAAAuALLvAHAgDfX79fsRZtqBenTNu8/pqTXf9QHmzJdXBkAAABciTANAHZak5qneZ/skM3WcLuKSpv+svRn/bjvqGsKAwAAgMsRpgHATi+sSFdlI0H6NGulTS+vSnduQQAAAHAbwjQA2CEtt1DJGUcM9VmekqusY8VOqggAAADuRJgGADusNxikJanSJpZ6AwAAtFCEaQCwQ0l5hUP9issc6wcAAIDmjTANAHZo5e/jUL/QAMf6AQAAoHkjTAOAHRJ7hcvH22Soj7+Pl0Z247xpAACAlogwDQB2iAjx18V92hvqc2nfKIUGMjMNAADQEhGmAcBOd1zYXUG+3na1beVv1q1jujm5IgAAALgLYRoA7NSzfYhenRmvED9zg+1CA3y0IClBXcKCXFQZAAAAXI0wDQAGDOsWpk9vH6FrBnesNUsd4m9W0rDOWnb7CMV3buumCgEAAOAKJpvNZnN3EXCvrKwsWSwWSVJmZqZiYmLcXBFwbigstWrz/mMqKrUqxN+sgZ3aKNC34VlrAAAAtAz81gcADgr2M2t0D3brBgAA8EQs8wYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAwiTAMAAAAAYBBhGgAAAAAAgwjTAAAAAAAYRJgGAAAAAMAgwjQAAAAAAAYRpgEAAAAAMIgwDQAAAACAQYRpAAAAAAAMIkwDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwiDANAAAAAIBBhGkAAAAAAAzy2DCdm5urZcuW6YEHHtD48eMVFhYmk8kkk8mkpKQkw+N99dVXmjZtmmJiYuTn56eYmBhNmzZNX331ld1jFBcX64knntCgQYPUtm1bBQcHKy4uTvfcc48OHDhguCYAAAAAgHOYbDabzd1FuIPJZKr32qxZs7Rw4UK7xrHZbLr55pv1yiuv1Nvmpptu0ksvvdTgPdPT0zVx4kT98ssvdV4PDQ3VO++8owkTJthVlxFZWVmyWCySpMzMTMXExDT5PQAAAACgJfHYmekzWSwWXXzxxQ71/fvf/14dpAcMGKDFixdr48aNWrx4sQYMGCBJeuWVV/SPf/yj3jEKCws1adKk6iA9Z84cff/991q3bp3+/e9/Kzg4WCdOnNDll1+u7du3O1QnAAAAAKDpeOzM9Lx585SQkKCEhARFRkZq37596tKliyT7Z6bT0tIUFxcnq9Wq+Ph4rV69WgEBAdXXi4uLNXr0aG3atElms1kpKSmKjY2tNc6DDz6ohx56SJL0n//8R/fee2+N68nJyRo1apSsVqvGjBmj5cuXn8U3r42ZaQAAAAAwxmNnph966CFNmjRJkZGRDo/x9NNPy2q1SpKee+65GkFakgIDA/Xcc89JkqxWq5555plaY5SXl+vZZ5+VJMXFxenuu++u1Wbo0KGaPXu2JGnFihXavHmzwzUDAAAAAM6ex4bps2Wz2fTJJ59Iknr16qUhQ4bU2W7IkCHq2bOnJOnjjz/WbxcCrFy5UsePH5dUNSPu5VX3/yRnboq2dOnSs6weAAAAAHA2CNMO2rt3r7KzsyVJo0ePbrDt6etZWVnat29fjWtr1qyp1a4u8fHxCgoKkiStXbvWkZIBAAAAAE3E7O4CzlW7d++uft2rV68G2555fffu3dXPZhsZx2w2KzY2Vtu3b6/Rxx5ZWVkNXs/JyTE0HgAAAAB4OsK0gzIzM6tfN7Zh1+nNvX7b78z3QUFBat26daPjbN++XXl5eSotLZWfn59dtZ55fwAAAADA2WOZt4MKCgqqXwcHBzfY9vTybKnqGKy6xmlsjMbGAQAAAAC4DjPTDiopKal+7evr22DbM2eQT506Vec4jY3R2DgN+e1s+G/l5ORo0KBBdo8HAAAAAJ6OMO0gf3//6tdlZWUNti0tLa1+/dvjs06P09gYjY3TEM6NBgAAAICmxTJvB4WEhFS/bmzJdVFRUfXr3y7nPj2OPcu2GxoHAAAAAOA6hGkHnTnb29hu2Wcus/7tZmCnxykqKqo+b7qxccLDw+3efAwAAAAA0PQI0w7q3bt39euUlJQG2555PS4uzqFxrFar0tPT6xwDAAAAAOBahGkHdenSRVFRUZKkVatWNdh29erVkqTo6Gh17ty5xrURI0ZUv25onE2bNlUv8x4+fLgjJQMAAAAAmghh2kEmk0lTpkyRVDWjvH79+jrbrV+/vnrGecqUKTKZTDWuJyYmKjQ0VJK0aNEi2Wy2OsdZuHBh9eupU6eebfkAAAAAgLNAmD4Lc+fOldlctSH67bffXuu4qlOnTun222+XJJnNZs2dO7fWGL6+vrrjjjskSbt379aTTz5Zq01ycrJee+01SdLo0aOVkJDQlF8DAAAAAGCQxx6NtXbtWqWlpVW/z8/Pr36dlpZWYyZYkpKSkmqN0aNHD91zzz167LHHtGnTJg0fPlz333+/YmNjlZ6erscff1xbt26VJN17773q3r17nbXce++9eu+997Rnzx7dd999SktL05VXXqmAgACtWLFCjzzyiKxWqwICAvTMM8+c9XcHAAAAAJwdk62+dcUtXFJSkhYtWmR3+/r+miorKzVnzhwtWLCg3r6zZ8/WK6+8Ii+v+hcCpKWlacKECUpNTa3zeqtWrfT2229r0qRJdtdsr6ysrOpdxjMzMzmXGgAAAAAawTLvs+Tl5aXXXntNn3/+uaZMmaKoqCj5+voqKipKU6ZM0RdffKH58+c3GKQlqVu3btq6dasef/xxxcfHq3Xr1goMDFTPnj111113afv27U4J0gAAAAAA4zx2Zhq/YmYaAAAAAIxhZhoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGeew50wBwNtal5eutDfu1Zk++CsusCvYzK7FnhK4b0kmDurR1d3kAAABwMsI0ABhwvLhMN725WRv3Hq3xeUGJVZ/9dFCf/XRQI7uH6YVrLlCIv4+bqgQAAICzscwbAOxUXGbVta9tqBWkf2tNar5mLdiokvIKF1UGAAAAVyNMA4Cdnl+Rph3ZJ+1qu+XAcb22dq+TKwIAAIC7EKYBwA5l1kq992OmoT7vbDigykqbkyoCAACAOxGmAcAOy1NylV9YZqhP9vFT+iE930kVAQAAwJ0I0wBgh4PHTznUL/uYY/0AAADQvBGmAcAOJpNj/bwc7QgAAIBmjTANAHboGh7sUL8u4UFNXAkAAACaA8I0ANhhZLcwxbQJMNSne0SwEjq3dVJFAAAAcCfCNADYwcvLpJlDOxnqY7Q9AAAAzh2EaQCw0w3DuyixZ7hdbX/Xp72uGUyYBgAAaKkI0wBgJ7O3l16+bqAuGxgjr3r2FfP2MumawR31v6sHyKu+RgAAADjnmWw2m83dRcC9srKyZLFYJEmZmZmKiYlxc0VA85d5tFhvbdivNXvyVVRmVbCfWWN6RujqwR0V1drYs9UAAAA49xCmQZgGAAAAAINY5g0AAAAAgEGEaQAAAAAADCJMAwAAAABgEGEaAAAAAACDCNMAAAAAABhEmAYAAAAAwCDCNAAAAAAABhGmAQAAAAAwyOzuAgDgXJWWW6g1qXkqKrUq2M+s0T0j1CUsyN1lAQAAwAUI0wBg0Lq0fP3f8lStzzha43PTsl0a0S1Md4ztroTObd1UHQAAAFyBZd4AYMAHmzJ13YKNtYK0JNls0prUfF396np99tNBN1QHAAAAVyFMA4Cd1mcc0Z+X/qyKSluD7corbPrT+9u09cAxF1UGAAAAVyNMA4Cdnl+R1miQPq28wqaXVqU7uSIAAAC4C2EaAOywN79Ia9PyDfX5bneuDp0ocVJFAAAAcCfCNADYYV16vmz2TUpXq6i0aX3GEecUBAAAALciTAOAHU6VVTjUr6jM2sSVAAAAoDkgTAOAHUL8HTtJMMTfp4krAQAAQHNAmAYAO4zuESGzl8lQHz+zl0Z2C3NSRQAAAHAnwjQA2KF9qL/GxkUY6jPx/A5qE+TrpIoAAADgToRpALDT7Rd2l7+PfT82g/3MuiUx1skVAQAAwF0I0wBgp/OiQ/XiNQMV4OPdYLtgP7NeuW6gukeGuKgyAAAAuBphGgAMGNMrQh/dNkzTBkTL11zzR6i/j5dmxMfo49uGaxjPSgMAALRoJpvN6MmpaGmysrJksVgkSZmZmYqJiXFzRcC54VhRmTbsPaLC0gqF+Js1pGs7hQawezcAAIAncOysFwCA2gT56nfndXB3GQAAAHADlnkDAAAAAGAQYRoAAAAAAIMI0wAAAAAAGESYBgAAAADAIMI0AAAAAAAGEaYBAAAAADCIMA0AAAAAgEGEaQAAAAAADCJMAwAAAABg0P9r796jqirzP45/jtwRQeSiCIhXNC9oeRvHC6Xl3RkxJ0xDJRsvM65yzVSrWs2oqa1ZMzp2mdKayvtUjpblfYyUdKTQEhV1vKaAYIBCCCiC7N8f/jxLksvZcI5H8v1ai7WOez/Pfr57/3G2n7P3fjZhGgAAAAAAkwjTAAAAAACYRJgGAAAAAMAkwjQAAAAAACYRpgEAAAAAMIkwDQAAAACASYRpAAAAAABMIkwDAAAAAGASYRoAAAAAAJMI0wAAAAAAmOTq7AIAoL4qLzd0OqdQhSVlauTpqtaBPmrQwOLssgAAAHAHEKYBwKRLRdf0r2/O6cPkdJ3Pv2Jd3qKJt8b3bqHHe7aQn7ebEysEAACAo1kMwzCcXQScKyMjQ+Hh4ZKk9PR0hYWFObki4O51NLNAk5clK/tySZVtQht7acWTPdU2uNEdrAwAAAB3Es9MA4CN0i8Va+IH31QbpCXpfP4VPfFesn4ouHqHKgMAAMCdRpgGABu9+eVJ5RZes6nthYKrWrLrtIMrAgAAgLMQpgHABj9eKdXnBzNN9Vn/XYaKr5U5qCIAAAA4E2EaAGyQcOwHXS0tN9Xn8tUyfXUix0EVAQAAwJkI0wBgg0tFtt3e/VMXa9kPAAAAdzfCNADYwMO1dl+Xnq4udq4EAAAAdwPCNADYoFu4f636dQ1vbN9CAAAAcFcgTAOADbqE+alrmJ+pPn1aB6htsI+DKgIAAIAzEaYBwEZTB7Qx1f63A1o5qBIAAAA4G2EaAGw0IipEMx9qa1Pb54a018AOTR1cEQAAAJzF1dkFAEB98uyQ9mre2Ev/+PKkMn+8etv68CZeemZQpMZ2D3NCdQAAALhTCNN18OCDDyoxMdFUn507d+rBBx+0/nv58uWKj4+3qe+yZcs0efJkU+MBsL/xvVsotme4Eo79oK9O5qio5LoaebrqwfZBejAyWA0aWJxdIgAAAByMMH0HNWjQQO3atXN2GQDswKWBRYM7NdPgTs2cXQoAAACcgDBdB8uWLVNRUVG1bY4eParY2FhJ0qBBgxQaGlpl2+3bt6t58+ZVrg8L47ZRAAAAALgbEKbroFWrmmfqXbVqlfXzxIkTq20bGRmpli1b1rUsAAAAAICDMZu3A5WXl2vNmjWSJB8fH40ZM8bJFQEAAAAA7IEw7UAJCQk6f/68JGns2LHy9vZ2ckUAAAAAAHsgTDvQypUrrZ9rusUbAAAAAFB/8My0gxQWFurTTz+VJLVo0aLC67CqMnnyZB07dkx5eXny9fVV27Zt9fDDD2vGjBnVTlxWk4yMjGrXZ2Vl1XrbAAAAAHAvIkw7yPr1660zfcfFxcliqfm9s7e+s/rixYu6ePGivvnmGy1atEivvfaapk2bVqtawsPDa9UPAAAAAFA5wrSDmLnFu3Xr1hozZoz69OljDb5nzpzR+vXrtW7dOl29elXTp0+XxWLR1KlTHVo3AAAAAKBmFsMwDGcX8XOTkZGhiIgIlZeX6xe/+IWSkpKqbPvjjz/K19e3yivXmzZt0pgxY1RaWipvb2+dPn1azZo1M11PdbKystSrVy9JUnp6Ou+zBgAAAIAaMAGZA6xevVrl5eWSpEmTJlXb1s/Pr9pbwEeOHKnZs2dLkoqLi/X++++bricsLKzav5CQENPbBAAAAIB7GWHaAVatWiVJ8vDwUGxsbJ2399vf/tYauG99rhoAAAAA4ByEaTvbv3+/jh49KunGVWV/f/86bzM4OFiBgYGSZH1vNQAAAADAeQjTdnbrxGM13eJtBo+2AwAAAMDdgzBtR6Wlpfroo48kSUFBQRo2bJhdtpudna2LFy9Kkpo3b26XbQIAAAAAao8wbUdbt25VTk6OJGn8+PFydbXPm8feffdd65Xp6Ohou2wTAAAAAFB7hGk7MvNuaUk6e/asDhw4UG2bTZs2ad68eZIkT09PxcfH161IAAAAAECd2efSKZSXl6dNmzZJkjp37qwHHnigxj5nz57VQw89pD59+mjUqFHq1q2bgoODZRiGzpw5o3Xr1mndunXWq9ILFy5UaGioQ/cDAAAAAFAzwrSdfPzxxyopKZFk21XpWyUlJSkpKanK9d7e3lq8eLGmTp1apxoBAAAAAPZBmLaTm++WdnFx0YQJE2zq0717d61evVpJSUnav3+/srKylJubq7KyMvn7+6tTp04aNGiQnnrqKQUHBzuyfAAAAACACRaDdy7d8zIyMhQeHi5JSk9PV1hYmJMrAgAAAIC7GxOQAQAAAABgEmEaAAAAAACTCNMAAAAAAJhEmAYAAAAAwCTCNAAAAAAAJhGmAQAAAAAwiTANAAAAAIBJhGkAAAAAAEwiTAMAAAAAYBJhGgAAAAAAkwjTAAAAAACYRJgGAAAAAMAkwjQAAAAAACYRpgEAAAAAMIkwDQAAAACASa7OLgAA6quSsutKPf+jCkuuy8fDVV1C/eTuym+UAAAA9wLCNACYlJl/RSv2ntXa/enKKy61Lg/08VBszzBN6tNSwb6eTqwQAAAAjmYxDMNwdhFwroyMDIWHh0uS0tPTFRYW5uSKgLvXd2l5mrJ8X4UQ/VOBPh5aHt9TnUP97mBlAAAAuJO4HxEAbHQmp1Dxy6oP0pKUW1iiSR8kK/1S8R2qDAAAAHcaYRoAbPTml6f045Xqg/RNF4uuaUniaQdXBAAAAGchTAOADS4Wlmjz4SxTfT47cF6Xr9oWvgEAAFC/EKYBwAZf/i9b18rKTfUpunZdX53IdVBFAAAAcCbCNADYwNbbu38q/8o1O1cCAACAuwFhGgBs4OHmUqt+XrXsBwAAgLsbYRoAbNCzpb/pPhaL1D3CfD8AAADc/QjTAGCDDs181cNkMO7fLkgRAQ0dVBEAAACciTANADaaHt3G5rYNLNK0Aa0dWA0AAACciTANADZ6uGNTvTCsQ43tLBZp9qhO6ts28A5UBQAAAGdwdXYBAFCfTI9uo+aNvfRGwkmdyi68bX2HZo006+FIDe3czAnVAQAA4E4hTAOASb/q2ly/6tpcSacv6quTOSoqKVMjT1c92D5YPVs2cXZ5AAAAuAMI0wBQS33aBKhPmwBnlwEAAAAn4JlpAAAAAABMIkwDAAAAAGASYRoAAAAAAJMI0wAAAAAAmESYBgAAAADAJMI0AAAAAAAmEaYBAAAAADCJMA0AAAAAgEmEaQAAAAAATCJMAwAAAABgEmEaAAAAAACTCNMAAAAAAJhEmAYAAAAAwCTCNAAAAAAAJhGmAQAAAAAwiTANAAAAAIBJhGkAAAAAAEwiTAMAAAAAYBJhGgAAAAAAkwjTAAAAAACYRJgGAAAAAMAkwjQAAAAAACYRpgEAAAAAMIkwDQAAAACASYRpAAAAAABMIkwDAAAAAGASYRoAAAAAAJMI0wAAAAAAmOTq7ALgfGVlZdbPWVlZTqwEAAAAAByrWbNmcnWtexQmTEM5OTnWz7169XJiJQAAAADgWOnp6QoLC6vzdrjNGwAAAAAAkyyGYRjOLgLOdfXqVR0+fFiSFBQUZJdbHoB7QVZWlvVujuTkZIWEhDi5IgDAzx3nHqDuuM0bduPp6amePXs6uwygXgsJCbHL7UIAANiKcw/gXNzmDQAAAACASYRpAAAAAABMIkwDAAAAAGASYRoAAAAAAJMI0wAAAAAAmESYBgAAAADAJMI0AAAAAAAmWQzDMJxdBAAAAAAA9QlXpgEAAAAAMIkwDQAAAACASYRpAAAAAABMIkwDAAAAAGASYRoAAAAAAJMI0wAAAAAAmESYBgAAAADAJMI0AAAAAAAmEaYBAAAAADCJMA0AAAAAgEmEaQC4g3bt2iWLxSKLxaJdu3Y5uxwAwP9LS0vTtGnT1KZNG3l6elq/qzds2ODs0pxi+fLl1mNw9uxZZ5cD3JVcnV0AAAAA4ExpaWnq3r27cnNznV0KgHqEMA0AAIB72vz585WbmytXV1ctWLBAAwYMkI+PjyQpIiLCydUBuFsRpgEAAHBP++KLLyRJo0eP1vPPP+/kagDUFzwzDQAAgHva+fPnJUmRkZFOrgRAfUKYBgAAwD3t2rVrkiQ3NzcnVwKgPiFMA6j35syZY51xVJIKCgo0Z84cdenSRT4+PmratKmGDx+uvXv3VuiXnZ2tl19+WZ06dVLDhg0VEBCgX//61zpw4ECVY505c0aLFi3SqFGj1LJlS3l5ecnLy0sRERGKjY3Vtm3b7LZfO3bs0BNPPKFWrVrJy8tLvr6+6tq1q55//nllZWXZbRwAuBfdOlv1TXPnzrUus1gsmjx5coU+hmFo3bp1evTRRxUeHi5PT0/5+/urV69emjdvnvLz86scb/LkybJYLGrZsqUk6cKFC3r22WcVGRkpb29vhYaG6rHHHtORI0cq9Dt79qyefvppRUZGysvLS02bNtWECRN0+vTpavcvNTVV8+fP15AhQxQWFiYPDw/5+PioXbt2mjRpkr7++mtTx6sqdTkmQL1nAEA9N3v2bEOSIclIS0szIiMjrf++9c/FxcVYu3atYRiGcfDgQSM0NLTSdh4eHkZCQsJt45w5c6bS9j/9e+KJJ4zS0tJKa925c6e13c6dOyttU1hYaMTExFQ7ho+Pj7Fx40a7HUMAuNcsW7asxu/zSZMmWdtnZ2cbffv2rbZ906ZNja+//rrS8SZNmmRIMiIiIoyUlBSjWbNmlW7D29vb2L17t2EYhpGQkGD4+flV2s7f399ITU2tdKxbzzXV/b3wwgs2HZ/vv/++0jZ1PSZAfccEZAB+Vn7zm98oIyNDL774ooYOHSpvb2/t2bNHs2fPVkFBgaZMmaIePXpo5MiRunLlihYsWKDo6Gi5ublp27ZtWrBggUpKShQfH6+TJ0/K3d3duu3r16/L3d1dQ4YM0SOPPKKOHTuqSZMmunTpkk6cOKG33npLR44c0erVq9W6dWvNnTvXdP3Xr1/XqFGjtHPnTlksFo0bN05jxoxRq1atVFpaquTkZC1atEhpaWl69NFHtXfvXnXv3t2ehxAA7gmjR49Wjx49JEldunSRJM2YMUO/+93vrG38/f0lSUVFRYqOjtaxY8fk7u6u+Ph4DR8+XOHh4SoqKtJXX32lv//97/rhhx80bNgwHThwoMpZwIuLixUTE6Nr167p1VdfVXR0tFxcXLRt2za9+uqrKi4uVlxcnHbs2KGYmBj5+vrqlVdeUe/evVVWVqb169frtddeU15enqZMmVLpFeaysjI1bNhQI0aM0MCBA9WhQwf5+voqOztbR44c0RtvvKFz587pL3/5iyIjIxUfH2/6+NnzmAD1lrPTPADU1a1Xpj08PCr9BXzz5s3WNkFBQUZgYKBx6tSp29q99dZb1naffPJJhXWFhYVGZmZmlXWUl5cbkydPNiQZDRs2NPLz829rU9OV6YULFxqSDDc3N2PLli2VjnPp0iWjU6dOhiSjX79+VdYDALDNze/l2bNnV7p+5syZhiTDz8/P2LdvX6Vtzp49a4SEhFjvUPqpm1emJdl0DgoKCjLatWtnZGdn39buueees7b77rvvblufk5Nj5OXlVbm/JSUlxiOPPGK9Ul5WVnZbm5quTNvjmAD1Hc9MA/hZmTVrlnr37n3b8uHDh1t/Ec/JydH8+fPVpk2b29rFx8fL09NTkrR79+4K6xo2bKiQkJAqx7ZYLFq0aJFcXFxUVFRkfdWKrUpLS7Vo0SJJ0syZMzVs2LBK2/n7++tvf/ubJGnPnj06deqUqXEAALbLzc3Ve++9J0l65ZVXrFezfyoiIkJ/+tOfJEkff/yxiouLq9zmvHnzajwH5eTk6M0331RQUNBt7WbMmGH9/NNzlSQFBgaqcePGVY7v7u5uPY+cO3dOKSkpVbatjCOOCVAfEaYB/KyMGzeuynVRUVGSboTexx57rNI2Xl5eateunaQbk41Vp7S0VBkZGTp27JhSU1OVmpqqzMxMBQQESJIOHjxoqvbk5GTrxGJV1XfTgAEDrJ+TkpJMjQMAsN327dt19epVSbZ/N5eWlurbb7+ttI2t5yB/f38NHjy40natWrVSo0aNJNV8rpKkkpISpaWl6ejRo9bzlWEY1vVmz1f2PiZAfcUz0wB+Vqp7R+jNX+kDAwOtz8FV1+7y5cu3rSstLdW7776rVatW6cCBA9bXqVQmNzfXtqL/3/79+62f+/TpY3O/CxcumBoHAGC7W7+bq7s76aeq+m4ODAxUkyZNqux38xzUtm3bCjONV9bu8uXLlZ6rpBvPNL/xxhv66KOPdOTIEV2/fr3KbdXlfGWPYwLUV4RpAD8r3t7eVa5r0KBBjW1ubffT/3hcunRJgwcPtvmX9StXrtjU7qbs7GxT7W/itjkAcBx7fzfbeg6q7blKuvE6rYEDB+r777+3pVTOV0AtEaYBwEbPPPOMNUiPHj1aTz75pKKiohQcHCxPT0/rFYQWLVooPT29wi10trj1P0S7du2y3i5ek+DgYFPjAABsd/O72d3d3dRtymFhYY4qqUZxcXH6/vvvZbFYFB8fr3Hjxum+++5TUFCQPDw8JEnl5eVycXGRpFqfr+rTMQEcgTANADYoKCjQxx9/LEkaP3681qxZU2XbvLy8Wo1xa3h2d3dX586da7UdAID93PxuvnbtmgICAkzd1uwM//vf/7Rnzx5J0osvvqgFCxZU2q625yqp/h0TwFGYgAwAbHDy5EmVlpZKqn6Ss+PHj6uwsLBWY9x///3Wz//5z39qtQ0AgH3Vt+/mI0eOWD9Xd7669blns+rbMQEchTANADYoKyuzfq7uma+lS5fWeox+/fpZJ6VZunSpCgoKar0tAIB9DBs2TG5ubpKkxYsXVzgf3I3uxPmqvh0TwFEI0wBgg1tnVV25cmWlbTZt2qQ333yz1mN4enrq2WeflXRjxtNx48apqKioyvaXL1/WP/7xj1qPBwCoWWhoqOLj4yXdeIXUtGnTqg2P2dnZ1ncwO8PNV2tJ0ooVKypts2TJEm3YsKHWY9S3YwI4Cs9MA4ANAgICNHz4cG3evFlbtmzR0KFDNW3aNLVo0ULZ2dlav369li9frtatWys/P185OTm1Guf5559XQkKCEhIStHXrVnXs2FHTp09Xnz59rK9BOX78uHbt2qUNGzbI09NTM2fOtPPeAgButWjRIu3du1epqan64IMP9PXXX2vq1Knq3r27fHx8lJ+fryNHjuiLL77Qli1b1KVLFz311FNOqfX+++9X586dlZqaqiVLlig/P18TJkxQSEiI0tPTtXr1aq1bt059+/bVf//731qPU5+OCeAohGkAsNGSJUvUr18/paWlafv27dq+fXuF9S1atNCGDRs0fPjwWo/h4uKijRs3avr06Vq5cqXS0tL00ksvVdmembwBwPF8fHyUmJioCRMmaNu2bTp69KhmzZpVZXtfX987V9xPWCwWrVq1SgMHDlReXp4+/PBDffjhhxXadOnSRf/+97/VvHnzWo9Tn44J4Cjc5g0ANgoPD9d3332n5557TpGRkfLw8JCfn5+6du2q2bNnKyUlRR07dqzzOF5eXlqxYoX279+vGTNmqFOnTvLz85Orq6saN26sbt26acqUKVq3bp2OHTtmhz0DANSkSZMm2rp1qxISEhQfH6927drJx8dHrq6uatKkiXr27Knf//732rJli3bs2OHUWrt166aUlBRNnz5dERERcnNzU5MmTdSrVy8tXLhQycnJdpmBuz4dE8ARLIbZF8sBAAAAAHCP48o0AAAAAAAmEaYBAAAAADCJMA0AAAAAgEmEaQAAAAAATCJMAwAAAABgEmEaAAAAAACTCNMAAAAAAJhEmAYAAAAAwCTCNAAAAAAAJhGmAQAAAAAwiTANAAAAAIBJhGkAAAAAAEwiTAMAAAAAYBJhGgAAAAAAkwjTAAAAAACYRJgGAAAAAMAkwjQAAAAAACYRpgEAAAAAMIkwDQAAAACASYRpAAAAAABMIkwDAAAAAGASYRoAAAAAAJMI0wAAwCaZmZl64YUX9MADD8jPz0/u7u5q1qyZunTposcff1zLly9XQUFBpX2Li4v12muv6aGHHlLTpk3l7u6u4OBgDR48WMuWLdP169dv67Nnzx65urrKYrFo6NChMgyj0m0XFBSoVatWslgsCgwMVFZWll33GwCAyliMqs5MAAAA/2/37t0aOXJklWH5po0bN2rkyJEVlu3bt08xMTE6f/58lf169eqlzz//XE2bNq2w/OWXX9aCBQskSa+//rqefvrp2/rGxcVp9erVkqRPPvlEMTExNu0TAAB1QZgGAADVKikpUevWrZWZmalGjRppxowZeuihhxQcHKzS0lKdO3dOSUlJWr9+vd5+++0KYfrw4cPq06ePioqKFBwcrBkzZqh///4KCAhQdna2Pv/8c73zzjsqKytT7969tXv3brm5uVn7l5WVqW/fvkpOTpanp6f27dunzp07W9evXbtWsbGxkqSnnnpK//znP+/cgQEA3NMI0wAAoFpffvmlBg0aJKnyK883lZWVqbi4WL6+vpIkwzDUrVs3HTp0SF27dtUXX3yhwMDA2/pt27ZNI0aMUHl5ud577z1NmTKlwvpTp06pW7duKioqUlRUlJKTk+Xh4aGMjAxFRUUpLy9Pbdu2VUpKiho2bGjnvQcAoHI8Mw0AAKp14cIF6+cBAwZU2c7V1dUapCVp8+bNOnTokCRp5cqVlQZpSRo6dKjGjh0rSVq2bNlt69u2bavXX39dknTo0CG99NJLMgxDkyZNUl5enlxdXbVmzRqCNADgjiJMAwCAaoWEhFg/VxZ2q/LZZ59Jktq3b6+oqKhq294M6fv27at0MrIpU6ZozJgxkqTFixfr8ccf15dffilJmj17tnr16mVzXQAA2AO3eQMAgGqVlpaqQ4cOOnPmjCSpZ8+eiomJUXR0tHr06CF3d/dK+91///1KSUkxPV52draCgoJuW37p0iV16dJFmZmZ1mV9+/ZVYmKiXFxcTI8DAEBdEKYBAECNjh49qrFjx+rYsWMVlnt5eSk6OlpxcXGKjY2tEGpDQ0MrBF9bnT17VhEREZWuW716teLi4iRJbm5uOn78uFq1amV6DAAA6oowDQAAbHL9+nVt3LhRGzduVGJiok6fPl1hfffu3bVlyxYFBwdLkpo1a6YffvhBffv21dKlS20ep3379hVm9L51/P79+yspKcm67IMPPlB8fHwt9wgAgNojTAMAgFrJysrS1q1b9fbbb+vbb7+VJI0ePVqffvqpJKlTp046evSoIiMjdfz48TqPN3fuXM2ZM0eS5Ovrq4KCAjVq1EgpKSlq3bp1nbcPAIAZTEAGAABqJSQkRE8++aSSkpL0wAMPSJI2bdqkK1euSLrxzLQknThxQufOnavTWN98843mz58vSRo4cKASExPl4eGhy5cvKy4urtJJywAAcCTCNAAAqBM3NzdFR0dLuvGu6fz8fEnSr371K2ubv/71r7XeflFRkZ544gmVlZXJ399fK1asULdu3bRgwQJJ0t69e/Xqq6/WfgcAAKgFwjQAAKjW7t27derUqSrXX7t2TYmJiZIkHx8f60zcjz76qO677z5J0pIlS/T+++9XO05qaqo2btx42/JnnnnGOv7SpUsVFhYmSfrDH/6ggQMHSpJeeeUV7du3z+SeAQBQezwzDQAAqjVnzhzNmzdP/fv314gRIxQVFaWgoCBduXJFJ06c0NKlS5WcnCxJmjVrlhYvXmzte/jwYf3yl79UYWGhJGnIkCEaP368dZKx7OxsHThwQJs2bdLevXv1xz/+UQsXLrT237Bhg2JiYiRJcXFxWrlyZYXaMjIyFBUVpby8PLVr104HDhxQw4YNHX1IAAAgTAMAgOrNmTNHc+fOrbHdmDFjtGbNGnl6elZYfujQIY0dO1YnT56scRtz587Vn//8Z0k3JjiLiopSbm6uWrZsqYMHD8rX1/e2PmvXrlVsbKwkaerUqXrnnXds2S0AAOqEMA0AAKpVXFysxMRE7dixQ0lJScrMzFR2drakG6+/6t27tyZOnKjhw4dXuY2ysjL961//0qeffqpvv/1WOTk5Ki8vV0BAgNq3b69+/fopJibGOpGZYRgaNmyYtm/frgYNGigxMVH9+vWrcvsTJ07UqlWrJEmfffZZhee1AQBwBMI0AAAAAAAmMQEZAAAAAAAmEaYBAAAAADCJMA0AAAAAgEmEaQAAAAAATCJMAwAAAABgEmEaAAAAAACTCNMAAAAAAJhEmAYAAAAAwCTCNAAAAAAAJhGmAQAAAAAwiTANAAAAAIBJhGkAAAAAAEwiTAMAAAAAYBJhGgAAAAAAkwjTAAAAAACYRJgGAAAAAMAkwjQAAAAAACYRpgEAAAAAMIkwDQAAAACASYRpAAAAAABMIkwDAAAAAGASYRoAAAAAAJMI0wAAAAAAmESYBgAAAADApP8DMm/mNEzj0sAAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 489,
       "width": 489
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(\n",
    "    data=filtered,\n",
    "    kind='swarm',\n",
    "    x='sex',\n",
    "    y='height',\n",
    "    hue='sex',\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Interesting the male character look like their heights are more *spread out*. We also have many more observations of male heights than female heights.\n",
    "\n",
    "Let's make a nice summary figure that combines the distributions of heights with summary statistics. \n",
    "\n",
    "We'll use *layers* to combine:\n",
    "\n",
    "- A stripplot with `color = 'black', alpha = 0.3`  \n",
    "- A boxplot with `color = 'skyblue'`  \n",
    "- A pointplot with `color = 'black'`\n",
    "\n",
    "*Hint: Look at the \"layering plots\" section of the tutorial notebook and use `grid.map()`*\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1654cbb90>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1654cbb90>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1654cbb90>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1654cbb90>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1654cbb90>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.02, 'Non-human height by Sex')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 520,
       "width": 494
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Setup grid with data\n",
    "grid = sns.FacetGrid(\n",
    "    data= filtered,\n",
    "    height=5,\n",
    "    aspect=1\n",
    ")\n",
    "\n",
    "# Strip - each individual character \n",
    "grid.map(sns.stripplot,'sex','height', order=['male','female'], color='black', alpha=0.3)\n",
    "\n",
    "# Box - medians\n",
    "grid.map(sns.boxplot,'sex','height', order=['male','female'], color='skyblue')\n",
    "\n",
    "# Points - averages\n",
    "grid.map(sns.pointplot,'sex','height', order=['male','female'], color='black')\n",
    "\n",
    "# Labels\n",
    "grid.set_axis_labels('Sex', 'Height (cm)', fontsize=14)\n",
    "grid.set_xticklabels(['Male', 'Female'])\n",
    "grid.figure.suptitle('Non-human height by Sex', fontsize=16, y=1.02)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's try to use resampling to compare these groups.  \n",
    "One way we can do this is by asking: *how likely are we to observe this average difference if characters were randomly assigned to one of the two sexes*?\n",
    "\n",
    "We can simulate this by **resampling without replacement**, i.e. **permuting** the sex label across characters, re-calculating the mean difference, and then repeating this process many times.\n",
    "\n",
    "We can use the [`permutation_test`](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.permutation_test.html) function, which needs us to give it:\n",
    "\n",
    "- a *tuple* of the two groups we want to compare\n",
    "- a function that takes each group and returns the statistic we want to make an inference about\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import permutation_test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First lets get all the *male* and *female* heights, and store them in two new variables that we'll give to `permutation_test`\n",
    "\n",
    "We can do this by *filtering* rows by sex and *selecting* the height column, and converting the result to a numpy array:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(34, 1)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "males = filtered.filter(col('sex') == 'male').select('height').to_numpy()\n",
    "males.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the array is 2 dimensional (it was originally a column), we can use `.squeeze()` to remove the extra dimension we don't need:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(34,)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "males = males.squeeze()\n",
    "males.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And repeat for females"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7,)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "females = filtered.filter(col('sex') == 'female').select('height').to_numpy().squeeze()\n",
    "females.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we'll create a function that takes the mean difference between both groups:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "def mean_diff(x, y):\n",
    "    return np.mean(x) - np.mean(y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have all the pieces that `permutation_test` needs so lets use it.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = permutation_test(\n",
    "    (males, females),             # <- this is a tuple! \n",
    "    statistic=mean_diff, \n",
    "    permutation_type='independent',\n",
    "    n_resamples=1000\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's create a simple `matplotlib` histogram to check out the result and plot the true mean difference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([  1.,   8.,  76., 180., 227., 243., 151.,  72.,  37.,   5.]),\n",
       " array([-59.33613445, -47.58739496, -35.83865546, -24.08991597,\n",
       "        -12.34117647,  -0.59243697,  11.15630252,  22.90504202,\n",
       "         34.65378151,  46.40252101,  58.1512605 ]),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x165404590>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 413,
       "width": 552
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(result.null_distribution);\n",
    "plt.axvline(result.statistic, color='black', linestyle='--')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that our observed mean difference (black line) is centered inside the **null distribution** we created by shuffling sex labels around. \n",
    "\n",
    "The `result.pvalue` captures the proportion of times we'd observe a mean difference as large as ours if the null distribution were true, i.e. if sex labels were randomly assigned.\n",
    "\n",
    "This suggests that we'd be pretty likely to observe this mean difference *even if* sex labels were shuffled - so we shouldn't make much of it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9010989010989011"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.pvalue"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This quick example is meant to give you a sense of how to test a specific comparison, simply using some of the resampling approaches we learned about before.\n",
    "\n",
    "Remember we also learned about the [`bootstrap`](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.bootstrap.html) function from `scipy.stats` that resamples **with replacement** to help us estimate our *uncertainty* about a statistical estimate and capture this in **confidence intervals**.\n",
    "\n",
    "Keep these in mind, along with the `.sample()` method that Polars DataFrames have to help you with future data analysis tasks!\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "201b",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}