collisions-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>body {\n",
       "    margin: 0;\n",
       "    font-family: Helvetica;\n",
       "}\n",
       "table.dataframe {\n",
       "    border-collapse: collapse;\n",
       "    border: none;\n",
       "}\n",
       "table.dataframe tr {\n",
       "    border: none;\n",
       "}\n",
       "table.dataframe td, table.dataframe th {\n",
       "    margin: 0;\n",
       "    border: 1px solid white;\n",
       "    padding-left: 0.25em;\n",
       "    padding-right: 0.25em;\n",
       "}\n",
       "table.dataframe th:not(:empty) {\n",
       "    background-color: #fec;\n",
       "    text-align: left;\n",
       "    font-weight: normal;\n",
       "}\n",
       "table.dataframe tr:nth-child(2) th:empty {\n",
       "    border-left: none;\n",
       "    border-right: 1px dashed #888;\n",
       "}\n",
       "table.dataframe td {\n",
       "    border: 2px solid #ccf;\n",
       "    background-color: #f4f4ff;\n",
       "}\n",
       "h3 {\n",
       "    color: white;\n",
       "    background-color: black;\n",
       "    padding: 0.5em;\n",
       "}\n",
       "</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.core.display import HTML\n",
    "css = open('style-table.css').read() + open('style-notebook.css').read()\n",
    "HTML('<style>{}</style>'.format(css))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import geocoder\n",
    "import csv\n",
    "location = [43.65501,-79.42021]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from numpy import random\n",
    "from scipy.spatial import distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "campus = [[43.657946,-79.39993],\n",
    "          [43.663502,-79.40005],\n",
    "          [43.663051,-79.402196],\n",
    "          [43.665429,-79.398975]\n",
    "         ]\n",
    "\n",
    "campus_dict = {(43.657946,-79.39993):4495.8,\n",
    "               (43.663502,-79.400050):2653,\n",
    "               (43.663051,-79.402196):3574,\n",
    "               (43.665429,-79.398975):2304\n",
    "              }\n",
    "\n",
    "waterfront = [[43.648208,-79.370923],\n",
    "              [43.642711,-79.394043],\n",
    "              [43.639944,-79.387032],\n",
    "              [43.640625,-79.3932],\n",
    "              [43.640093,-79.380152]\n",
    "             ]\n",
    "\n",
    "waterfront_dict = {(43.648208,-79.370923):330,\n",
    "              (43.642711,-79.394043):745,\n",
    "              (43.639944,-79.387032):128,\n",
    "              (43.640625,-79.3932):154,\n",
    "              (43.640093,-79.380152):235\n",
    "             }          \n",
    "castleFrank = [[43.673792,-79.368187]]\n",
    "\n",
    "castleFrank_dict = {(43.673792,-79.368187):3413}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>injury</th>\n",
       "      <th>safety_equip</th>\n",
       "      <th>road_class</th>\n",
       "      <th>cyclist_crash_type</th>\n",
       "      <th>age_of_cyclist</th>\n",
       "      <th>stname1</th>\n",
       "      <th>stet_1_type</th>\n",
       "      <th>stname_2</th>\n",
       "      <th>stet_2_type</th>\n",
       "      <th>long</th>\n",
       "      <th>...</th>\n",
       "      <th>acc_time2</th>\n",
       "      <th>traffic_control</th>\n",
       "      <th>road_surface2</th>\n",
       "      <th>date</th>\n",
       "      <th>date2</th>\n",
       "      <th>driver_action</th>\n",
       "      <th>driver_condition</th>\n",
       "      <th>year</th>\n",
       "      <th>intersection</th>\n",
       "      <th>coords</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Local</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>34.0</td>\n",
       "      <td>EUCLID</td>\n",
       "      <td>AV</td>\n",
       "      <td>ULSTER</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.413300</td>\n",
       "      <td>...</td>\n",
       "      <td>08:00:00</td>\n",
       "      <td>no control</td>\n",
       "      <td>loose snow</td>\n",
       "      <td>1986-01-14 00:00:00</td>\n",
       "      <td>1986-01-14 00:00:00</td>\n",
       "      <td>Lost control</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Euclid Ave &amp; Ulster St, Toronto, Ontario, M6G</td>\n",
       "      <td>[43.65906013500046, -79.41272077799965]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>34.0</td>\n",
       "      <td>SALEM</td>\n",
       "      <td>AV</td>\n",
       "      <td>FERNBANK</td>\n",
       "      <td>AV</td>\n",
       "      <td>-79.433801</td>\n",
       "      <td>...</td>\n",
       "      <td>10:00:00</td>\n",
       "      <td>no control</td>\n",
       "      <td>slush</td>\n",
       "      <td>1986-01-28 00:00:00</td>\n",
       "      <td>1986-01-28 00:00:00</td>\n",
       "      <td>Driving properly</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Fernbank Ave &amp; Salem Ave, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.66582101600045, -79.43379108299968]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8.0</th>\n",
       "      <td>Minor</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>24.0</td>\n",
       "      <td>DUNDAS</td>\n",
       "      <td>ST</td>\n",
       "      <td>BEVERLEY</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.394060</td>\n",
       "      <td>...</td>\n",
       "      <td>08:00:00</td>\n",
       "      <td>traffic signal</td>\n",
       "      <td>wet</td>\n",
       "      <td>1986-01-16 00:00:00</td>\n",
       "      <td>1986-01-16 00:00:00</td>\n",
       "      <td>Failed to yield right-of-way</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Beverley St &amp; Dundas St W, Toronto, Ontario, M5T</td>\n",
       "      <td>[43.65382045300049, -79.39385107199968]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>24.0</td>\n",
       "      <td>BLOOR</td>\n",
       "      <td>ST</td>\n",
       "      <td>MARGUERETTA</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.440590</td>\n",
       "      <td>...</td>\n",
       "      <td>19:00:00</td>\n",
       "      <td>no control</td>\n",
       "      <td>wet</td>\n",
       "      <td>1986-02-19 00:00:00</td>\n",
       "      <td>1986-02-19 00:00:00</td>\n",
       "      <td>Failed to yield right-of-way</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Bloor St W &amp; Margueretta, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.658975,-79.439718]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>Minor</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Minor Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>54.0</td>\n",
       "      <td>AVENUE</td>\n",
       "      <td>RD</td>\n",
       "      <td>SHANLY</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.430720</td>\n",
       "      <td>...</td>\n",
       "      <td>15:00:00</td>\n",
       "      <td>stop sign</td>\n",
       "      <td>dry</td>\n",
       "      <td>1986-03-01 00:00:00</td>\n",
       "      <td>1986-03-01 00:00:00</td>\n",
       "      <td>Driving properly</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Dovercourt Rd &amp; Shanly St, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.664316,-79.430499]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       injury safety_equip      road_class cyclist_crash_type  age_of_cyclist  \\\n",
       "id                                                                              \n",
       "4.0   Minimal   Unrecorded           Local         Unrecorded            34.0   \n",
       "7.0   Minimal   Unrecorded  Major Arterial         Unrecorded            34.0   \n",
       "8.0     Minor   Unrecorded  Major Arterial         Unrecorded            24.0   \n",
       "9.0   Minimal   Unrecorded  Major Arterial         Unrecorded            24.0   \n",
       "10.0    Minor   Unrecorded  Minor Arterial         Unrecorded            54.0   \n",
       "\n",
       "     stname1 stet_1_type     stname_2 stet_2_type       long  \\\n",
       "id                                                             \n",
       "4.0   EUCLID          AV       ULSTER          ST -79.413300   \n",
       "7.0    SALEM          AV     FERNBANK          AV -79.433801   \n",
       "8.0   DUNDAS          ST     BEVERLEY          ST -79.394060   \n",
       "9.0    BLOOR          ST  MARGUERETTA          ST -79.440590   \n",
       "10.0  AVENUE          RD       SHANLY          ST -79.430720   \n",
       "\n",
       "                       ...                     acc_time2  traffic_control  \\\n",
       "id                     ...                                                  \n",
       "4.0                    ...                      08:00:00       no control   \n",
       "7.0                    ...                      10:00:00       no control   \n",
       "8.0                    ...                      08:00:00   traffic signal   \n",
       "9.0                    ...                      19:00:00       no control   \n",
       "10.0                   ...                      15:00:00        stop sign   \n",
       "\n",
       "     road_surface2                 date                date2  \\\n",
       "id                                                             \n",
       "4.0     loose snow  1986-01-14 00:00:00  1986-01-14 00:00:00   \n",
       "7.0          slush  1986-01-28 00:00:00  1986-01-28 00:00:00   \n",
       "8.0            wet  1986-01-16 00:00:00  1986-01-16 00:00:00   \n",
       "9.0            wet  1986-02-19 00:00:00  1986-02-19 00:00:00   \n",
       "10.0           dry  1986-03-01 00:00:00  1986-03-01 00:00:00   \n",
       "\n",
       "                     driver_action driver_condition    year  \\\n",
       "id                                                            \n",
       "4.0                   Lost control           Normal  1986.0   \n",
       "7.0               Driving properly           Normal  1986.0   \n",
       "8.0   Failed to yield right-of-way           Normal  1986.0   \n",
       "9.0   Failed to yield right-of-way           Normal  1986.0   \n",
       "10.0              Driving properly           Normal  1986.0   \n",
       "\n",
       "                                          intersection  \\\n",
       "id                                                       \n",
       "4.0      Euclid Ave & Ulster St, Toronto, Ontario, M6G   \n",
       "7.0    Fernbank Ave & Salem Ave, Toronto, Ontario, M6H   \n",
       "8.0   Beverley St & Dundas St W, Toronto, Ontario, M5T   \n",
       "9.0    Bloor St W & Margueretta, Toronto, Ontario, M6H   \n",
       "10.0  Dovercourt Rd & Shanly St, Toronto, Ontario, M6H   \n",
       "\n",
       "                                       coords  \n",
       "id                                             \n",
       "4.0   [43.65906013500046, -79.41272077799965]  \n",
       "7.0   [43.66582101600045, -79.43379108299968]  \n",
       "8.0   [43.65382045300049, -79.39385107199968]  \n",
       "9.0                    [43.658975,-79.439718]  \n",
       "10.0                   [43.664316,-79.430499]  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "col_df = pd.DataFrame.from_csv('toronto_cycling_central_mode.csv', index_col='id')\n",
    "col_df = col_df.sort_index()\n",
    "col_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "coords = central_col_df[[\"stname1\",\"stname_2\",\"street_1_type\",\"street_2_type\"]]\n",
    "coords= coords.values.tolist()\n",
    "coordinate_list=[]\n",
    "for pair in coords:\n",
    "    if isinstance(pair[1],str): \n",
    "        if pair[1] <= pair[0]:\n",
    "            temp = pair[0]\n",
    "            pair[0] = pair[1]\n",
    "            pair[1] = temp\n",
    "            temp = pair[2]\n",
    "            pair[2] = pair[3]\n",
    "            pair[3] = temp\n",
    "coords2 = coords\n",
    "for pair2 in coords2:\n",
    "    pair2 = str(pair2[0]) + ' ' + str(pair2[2]) + ' & ' + str(pair2[1]) + ' ' + str(pair2[3]) + ', Toronto'\n",
    "    coordinate_list.append(pair2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "coords = col_df[[\"STNAME1\",\"STNAME 2\",\"STET 1 TYPE\",\"STET 2 TYPE\"]]\n",
    "coords= coords.values.tolist()\n",
    "coordinate_list=[]\n",
    "for pair in coords:\n",
    "    if isinstance(pair[1],str): \n",
    "        if pair[1] <= pair[0]:\n",
    "            temp = pair[0]\n",
    "            pair[0] = pair[1]\n",
    "            pair[1] = temp\n",
    "            temp = pair[2]\n",
    "            pair[2] = pair[3]\n",
    "            pair[3] = temp\n",
    "coords2 = coords\n",
    "for pair2 in coords2:\n",
    "    pair2 = str(pair2[0]) + ' ' + str(pair2[2]) + ' & ' + str(pair2[1]) + ' ' + str(pair2[3]) + ', Toronto'\n",
    "    coordinate_list.append(pair2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ped_counts_df = pd.DataFrame.from_csv('Vehicle and Pedestrian Counts/TrafficPedestrianVolumes_2011.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the Power fit for the Bike/Pedestrian ratio, we get a function that predicts the bike numbers at any one intersection."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ped_counts_df['bike_prediction'] = (500.2146799711*ped_counts_df['8HrPedVol']**(-0.8950759596))*ped_counts_df['8HrPedVol']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Main</th>\n",
       "      <th>Midblock Route</th>\n",
       "      <th>Side 1 Route</th>\n",
       "      <th>Side 2 Route</th>\n",
       "      <th>Activation Date</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>Count Date</th>\n",
       "      <th>8HrPedVol</th>\n",
       "      <th>8HrVehVol</th>\n",
       "      <th>24HrPedVol</th>\n",
       "      <th>24HrVehVol</th>\n",
       "      <th>bike_prediction</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PX2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>FRONT ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1948/11/15</td>\n",
       "      <td>43.649450</td>\n",
       "      <td>-79.371410</td>\n",
       "      <td>2011/09/08</td>\n",
       "      <td>17008</td>\n",
       "      <td>19335</td>\n",
       "      <td>34016</td>\n",
       "      <td>38670</td>\n",
       "      <td>1390.124396</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KING ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1950/08/23</td>\n",
       "      <td>43.650461</td>\n",
       "      <td>-79.371924</td>\n",
       "      <td>2011/09/07</td>\n",
       "      <td>37719</td>\n",
       "      <td>17665</td>\n",
       "      <td>75438</td>\n",
       "      <td>35330</td>\n",
       "      <td>1511.289334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>ADELAIDE ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1958/09/12</td>\n",
       "      <td>43.651534</td>\n",
       "      <td>-79.372360</td>\n",
       "      <td>2008/06/16</td>\n",
       "      <td>1991</td>\n",
       "      <td>19726</td>\n",
       "      <td>3982</td>\n",
       "      <td>39452</td>\n",
       "      <td>1109.962532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>RICHMOND ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1962/04/21</td>\n",
       "      <td>43.652718</td>\n",
       "      <td>-79.372824</td>\n",
       "      <td>2009/07/30</td>\n",
       "      <td>2696</td>\n",
       "      <td>24842</td>\n",
       "      <td>5392</td>\n",
       "      <td>49684</td>\n",
       "      <td>1145.833262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>QUEEN ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1928/08/24</td>\n",
       "      <td>43.653704</td>\n",
       "      <td>-79.373238</td>\n",
       "      <td>2011/05/18</td>\n",
       "      <td>3622</td>\n",
       "      <td>19772</td>\n",
       "      <td>7244</td>\n",
       "      <td>39544</td>\n",
       "      <td>1181.886259</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Main Midblock Route   Side 1 Route Side 2 Route Activation Date  \\\n",
       "PX2                                                                         \n",
       "2    JARVIS ST            NaN     FRONT ST E          NaN      1948/11/15   \n",
       "3    KING ST E            NaN      JARVIS ST          NaN      1950/08/23   \n",
       "4    JARVIS ST            NaN  ADELAIDE ST E          NaN      1958/09/12   \n",
       "5    JARVIS ST            NaN  RICHMOND ST E          NaN      1962/04/21   \n",
       "6    JARVIS ST            NaN     QUEEN ST E          NaN      1928/08/24   \n",
       "\n",
       "      Latitude  Longitude  Count Date  8HrPedVol  8HrVehVol  24HrPedVol  \\\n",
       "PX2                                                                       \n",
       "2    43.649450 -79.371410  2011/09/08      17008      19335       34016   \n",
       "3    43.650461 -79.371924  2011/09/07      37719      17665       75438   \n",
       "4    43.651534 -79.372360  2008/06/16       1991      19726        3982   \n",
       "5    43.652718 -79.372824  2009/07/30       2696      24842        5392   \n",
       "6    43.653704 -79.373238  2011/05/18       3622      19772        7244   \n",
       "\n",
       "     24HrVehVol  bike_prediction  \n",
       "PX2                               \n",
       "2         38670      1390.124396  \n",
       "3         35330      1511.289334  \n",
       "4         39452      1109.962532  \n",
       "5         49684      1145.833262  \n",
       "6         39544      1181.886259  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ped_counts_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "ped_coords = ped_counts_df[['Latitude','Longitude']]\n",
    "ped_coords = ped_coords.replace(np.nan,0)\n",
    "ped_coordinate_list = ped_coords.values.tolist()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Main</th>\n",
       "      <th>Midblock Route</th>\n",
       "      <th>Side 1 Route</th>\n",
       "      <th>Side 2 Route</th>\n",
       "      <th>Activation Date</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>Count Date</th>\n",
       "      <th>8HrPedVol</th>\n",
       "      <th>8HrVehVol</th>\n",
       "      <th>24HrPedVol</th>\n",
       "      <th>24HrVehVol</th>\n",
       "      <th>bike_prediction</th>\n",
       "      <th>coordinates</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PX2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>FRONT ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1948/11/15</td>\n",
       "      <td>43.649450</td>\n",
       "      <td>-79.371410</td>\n",
       "      <td>2011/09/08</td>\n",
       "      <td>17008</td>\n",
       "      <td>19335</td>\n",
       "      <td>34016</td>\n",
       "      <td>38670</td>\n",
       "      <td>1390.124396</td>\n",
       "      <td>(43.64945, -79.37141)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KING ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1950/08/23</td>\n",
       "      <td>43.650461</td>\n",
       "      <td>-79.371924</td>\n",
       "      <td>2011/09/07</td>\n",
       "      <td>37719</td>\n",
       "      <td>17665</td>\n",
       "      <td>75438</td>\n",
       "      <td>35330</td>\n",
       "      <td>1511.289334</td>\n",
       "      <td>(43.6504606, -79.3719239)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>ADELAIDE ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1958/09/12</td>\n",
       "      <td>43.651534</td>\n",
       "      <td>-79.372360</td>\n",
       "      <td>2008/06/16</td>\n",
       "      <td>1991</td>\n",
       "      <td>19726</td>\n",
       "      <td>3982</td>\n",
       "      <td>39452</td>\n",
       "      <td>1109.962532</td>\n",
       "      <td>(43.6515337, -79.37236)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>RICHMOND ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1962/04/21</td>\n",
       "      <td>43.652718</td>\n",
       "      <td>-79.372824</td>\n",
       "      <td>2009/07/30</td>\n",
       "      <td>2696</td>\n",
       "      <td>24842</td>\n",
       "      <td>5392</td>\n",
       "      <td>49684</td>\n",
       "      <td>1145.833262</td>\n",
       "      <td>(43.6527176, -79.372824)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>JARVIS ST</td>\n",
       "      <td>NaN</td>\n",
       "      <td>QUEEN ST E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1928/08/24</td>\n",
       "      <td>43.653704</td>\n",
       "      <td>-79.373238</td>\n",
       "      <td>2011/05/18</td>\n",
       "      <td>3622</td>\n",
       "      <td>19772</td>\n",
       "      <td>7244</td>\n",
       "      <td>39544</td>\n",
       "      <td>1181.886259</td>\n",
       "      <td>(43.653704, -79.373238)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Main Midblock Route   Side 1 Route Side 2 Route Activation Date  \\\n",
       "PX2                                                                         \n",
       "2    JARVIS ST            NaN     FRONT ST E          NaN      1948/11/15   \n",
       "3    KING ST E            NaN      JARVIS ST          NaN      1950/08/23   \n",
       "4    JARVIS ST            NaN  ADELAIDE ST E          NaN      1958/09/12   \n",
       "5    JARVIS ST            NaN  RICHMOND ST E          NaN      1962/04/21   \n",
       "6    JARVIS ST            NaN     QUEEN ST E          NaN      1928/08/24   \n",
       "\n",
       "      Latitude  Longitude  Count Date  8HrPedVol  8HrVehVol  24HrPedVol  \\\n",
       "PX2                                                                       \n",
       "2    43.649450 -79.371410  2011/09/08      17008      19335       34016   \n",
       "3    43.650461 -79.371924  2011/09/07      37719      17665       75438   \n",
       "4    43.651534 -79.372360  2008/06/16       1991      19726        3982   \n",
       "5    43.652718 -79.372824  2009/07/30       2696      24842        5392   \n",
       "6    43.653704 -79.373238  2011/05/18       3622      19772        7244   \n",
       "\n",
       "     24HrVehVol  bike_prediction                coordinates  \n",
       "PX2                                                          \n",
       "2         38670      1390.124396      (43.64945, -79.37141)  \n",
       "3         35330      1511.289334  (43.6504606, -79.3719239)  \n",
       "4         39452      1109.962532    (43.6515337, -79.37236)  \n",
       "5         49684      1145.833262   (43.6527176, -79.372824)  \n",
       "6         39544      1181.886259    (43.653704, -79.373238)  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ped_counts_df['coordinates'] = ped_counts_df[['Latitude','Longitude']].apply(tuple, axis=1)\n",
    "ped_counts_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ped_dict = ped_counts_df.set_index('coordinates').to_dict()['bike_prediction']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>injury</th>\n",
       "      <th>safety_equip</th>\n",
       "      <th>road_class</th>\n",
       "      <th>cyclist_crash_type</th>\n",
       "      <th>age_of_cyclist</th>\n",
       "      <th>stname1</th>\n",
       "      <th>stet_1_type</th>\n",
       "      <th>stname_2</th>\n",
       "      <th>stet_2_type</th>\n",
       "      <th>long</th>\n",
       "      <th>...</th>\n",
       "      <th>traffic_control</th>\n",
       "      <th>road_surface2</th>\n",
       "      <th>date</th>\n",
       "      <th>date2</th>\n",
       "      <th>driver_action</th>\n",
       "      <th>driver_condition</th>\n",
       "      <th>year</th>\n",
       "      <th>intersection</th>\n",
       "      <th>coords</th>\n",
       "      <th>coordinates</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Local</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>34.0</td>\n",
       "      <td>EUCLID</td>\n",
       "      <td>AV</td>\n",
       "      <td>ULSTER</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.413300</td>\n",
       "      <td>...</td>\n",
       "      <td>no control</td>\n",
       "      <td>loose snow</td>\n",
       "      <td>1986-01-14 00:00:00</td>\n",
       "      <td>1986-01-14 00:00:00</td>\n",
       "      <td>Lost control</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Euclid Ave &amp; Ulster St, Toronto, Ontario, M6G</td>\n",
       "      <td>[43.65906013500046, -79.41272077799965]</td>\n",
       "      <td>(43.6599, -79.4133)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>34.0</td>\n",
       "      <td>SALEM</td>\n",
       "      <td>AV</td>\n",
       "      <td>FERNBANK</td>\n",
       "      <td>AV</td>\n",
       "      <td>-79.433801</td>\n",
       "      <td>...</td>\n",
       "      <td>no control</td>\n",
       "      <td>slush</td>\n",
       "      <td>1986-01-28 00:00:00</td>\n",
       "      <td>1986-01-28 00:00:00</td>\n",
       "      <td>Driving properly</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Fernbank Ave &amp; Salem Ave, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.66582101600045, -79.43379108299968]</td>\n",
       "      <td>(43.665832, -79.433801)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8.0</th>\n",
       "      <td>Minor</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>24.0</td>\n",
       "      <td>DUNDAS</td>\n",
       "      <td>ST</td>\n",
       "      <td>BEVERLEY</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.394060</td>\n",
       "      <td>...</td>\n",
       "      <td>traffic signal</td>\n",
       "      <td>wet</td>\n",
       "      <td>1986-01-16 00:00:00</td>\n",
       "      <td>1986-01-16 00:00:00</td>\n",
       "      <td>Failed to yield right-of-way</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Beverley St &amp; Dundas St W, Toronto, Ontario, M5T</td>\n",
       "      <td>[43.65382045300049, -79.39385107199968]</td>\n",
       "      <td>(43.65377, -79.39406)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>24.0</td>\n",
       "      <td>BLOOR</td>\n",
       "      <td>ST</td>\n",
       "      <td>MARGUERETTA</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.440590</td>\n",
       "      <td>...</td>\n",
       "      <td>no control</td>\n",
       "      <td>wet</td>\n",
       "      <td>1986-02-19 00:00:00</td>\n",
       "      <td>1986-02-19 00:00:00</td>\n",
       "      <td>Failed to yield right-of-way</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Bloor St W &amp; Margueretta, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.658975,-79.439718]</td>\n",
       "      <td>(43.6588, -79.44059)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>Minor</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Minor Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>54.0</td>\n",
       "      <td>AVENUE</td>\n",
       "      <td>RD</td>\n",
       "      <td>SHANLY</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.430720</td>\n",
       "      <td>...</td>\n",
       "      <td>stop sign</td>\n",
       "      <td>dry</td>\n",
       "      <td>1986-03-01 00:00:00</td>\n",
       "      <td>1986-03-01 00:00:00</td>\n",
       "      <td>Driving properly</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Dovercourt Rd &amp; Shanly St, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.664316,-79.430499]</td>\n",
       "      <td>(43.66425, -79.43072)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       injury safety_equip      road_class cyclist_crash_type  age_of_cyclist  \\\n",
       "id                                                                              \n",
       "4.0   Minimal   Unrecorded           Local         Unrecorded            34.0   \n",
       "7.0   Minimal   Unrecorded  Major Arterial         Unrecorded            34.0   \n",
       "8.0     Minor   Unrecorded  Major Arterial         Unrecorded            24.0   \n",
       "9.0   Minimal   Unrecorded  Major Arterial         Unrecorded            24.0   \n",
       "10.0    Minor   Unrecorded  Minor Arterial         Unrecorded            54.0   \n",
       "\n",
       "     stname1 stet_1_type     stname_2 stet_2_type       long  \\\n",
       "id                                                             \n",
       "4.0   EUCLID          AV       ULSTER          ST -79.413300   \n",
       "7.0    SALEM          AV     FERNBANK          AV -79.433801   \n",
       "8.0   DUNDAS          ST     BEVERLEY          ST -79.394060   \n",
       "9.0    BLOOR          ST  MARGUERETTA          ST -79.440590   \n",
       "10.0  AVENUE          RD       SHANLY          ST -79.430720   \n",
       "\n",
       "               ...             traffic_control  road_surface2  \\\n",
       "id             ...                                              \n",
       "4.0            ...                  no control     loose snow   \n",
       "7.0            ...                  no control          slush   \n",
       "8.0            ...              traffic signal            wet   \n",
       "9.0            ...                  no control            wet   \n",
       "10.0           ...                   stop sign            dry   \n",
       "\n",
       "                     date                date2                 driver_action  \\\n",
       "id                                                                             \n",
       "4.0   1986-01-14 00:00:00  1986-01-14 00:00:00                  Lost control   \n",
       "7.0   1986-01-28 00:00:00  1986-01-28 00:00:00              Driving properly   \n",
       "8.0   1986-01-16 00:00:00  1986-01-16 00:00:00  Failed to yield right-of-way   \n",
       "9.0   1986-02-19 00:00:00  1986-02-19 00:00:00  Failed to yield right-of-way   \n",
       "10.0  1986-03-01 00:00:00  1986-03-01 00:00:00              Driving properly   \n",
       "\n",
       "     driver_condition    year  \\\n",
       "id                              \n",
       "4.0            Normal  1986.0   \n",
       "7.0            Normal  1986.0   \n",
       "8.0            Normal  1986.0   \n",
       "9.0            Normal  1986.0   \n",
       "10.0           Normal  1986.0   \n",
       "\n",
       "                                          intersection  \\\n",
       "id                                                       \n",
       "4.0      Euclid Ave & Ulster St, Toronto, Ontario, M6G   \n",
       "7.0    Fernbank Ave & Salem Ave, Toronto, Ontario, M6H   \n",
       "8.0   Beverley St & Dundas St W, Toronto, Ontario, M5T   \n",
       "9.0    Bloor St W & Margueretta, Toronto, Ontario, M6H   \n",
       "10.0  Dovercourt Rd & Shanly St, Toronto, Ontario, M6H   \n",
       "\n",
       "                                       coords              coordinates  \n",
       "id                                                                      \n",
       "4.0   [43.65906013500046, -79.41272077799965]      (43.6599, -79.4133)  \n",
       "7.0   [43.66582101600045, -79.43379108299968]  (43.665832, -79.433801)  \n",
       "8.0   [43.65382045300049, -79.39385107199968]    (43.65377, -79.39406)  \n",
       "9.0                    [43.658975,-79.439718]     (43.6588, -79.44059)  \n",
       "10.0                   [43.664316,-79.430499]    (43.66425, -79.43072)  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "col_df['coordinates'] = col_df[['lat','long']].apply(tuple, axis=1)\n",
    "col_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "central_col_df = col_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#only run after above cell already has produced output. Otherwise file doesn't exist. \n",
    "import csv\n",
    "with open('closest_traffic_count.csv', 'r') as f:\n",
    "    reader = csv.reader(f)\n",
    "    closest_traffic_count = list(reader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.path as mplPath\n",
    "import numpy as np\n",
    "\n",
    "waterfront_Path = mplPath.Path(np.array([[43.635497, -79.398156],\n",
    "[43.639000, -79.400725],\n",
    "[43.640822, -79.401427],\n",
    "[43.646984, -79.376977],\n",
    "[43.649889, -79.370343],\n",
    "[43.651614, -79.362725],\n",
    "[43.648090, -79.361191],\n",
    "[43.646451, -79.361937],\n",
    "[43.641209, -79.376739],\n",
    "[43.639969, -79.379965],\n",
    "[43.637698, -79.391847],\n",
    "[43.635666, -79.398368],\n",
    "[43.636489, -79.399603]]))\n",
    "\n",
    "waterfront_Path.contains_point((200, 100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "campus_Path = mplPath.Path(np.array([[43.659838, -79.399772],\n",
    "[43.661388, -79.401006],\n",
    "[43.665592, -79.402705],\n",
    "[43.666768, -79.401354],\n",
    "[43.668213, -79.393958],\n",
    "[43.663141, -79.392719],\n",
    "[43.659264, -79.394100],\n",
    "[43.658329, -79.398204]]\n",
    "))\n",
    "\n",
    "campus_Path.contains_point((43.661013, -79.397200))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "castleFrank_Path = mplPath.Path(np.array([[43.672105, -79.376696],\n",
    "[43.671562, -79.370962],\n",
    "[43.674418, -79.366821],\n",
    "[43.676086, -79.358731],\n",
    "[43.677056, -79.354021],\n",
    "[43.677040, -79.355126],\n",
    "[43.677622, -79.358516],\n",
    "[43.676194, -79.359503],\n",
    "[43.675170, -79.364760],\n",
    "[43.674580, -79.367539],\n",
    "[43.672019, -79.371112],\n",
    "[43.672710, -79.376927]]\n",
    "))\n",
    "\n",
    "castleFrank_Path.contains_point((43.676120, -79.359035))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import csv\n",
    "closest_traffic_point = []\n",
    "bike_traffic = []\n",
    "i = 0\n",
    "for i in range(0,len(central_col_df)):\n",
    "    point = central_col_df['coordinates'].iloc[i]\n",
    "    if waterfront_Path.contains_point(point):\n",
    "        closest = waterfront[distance.cdist([point], waterfront).argmin()]\n",
    "        closest_traffic_point.append(tuple(closest))\n",
    "        bike_traffic.append(waterfront_dict[tuple(closest)])\n",
    "    elif campus_Path.contains_point(point):\n",
    "        closest = campus[distance.cdist([point], campus).argmin()]\n",
    "        closest_traffic_point.append(tuple(closest))\n",
    "        bike_traffic.append(campus_dict[tuple(closest)])\n",
    "    elif castleFrank_Path.contains_point(point):\n",
    "        closest = castleFrank[distance.cdist([point], castleFrank).argmin()]\n",
    "        closest_traffic_point.append(tuple(closest))\n",
    "        bike_traffic.append(castleFrank_dict[tuple(closest)])\n",
    "    else:\n",
    "        closest = ped_coordinate_list[distance.cdist([point], ped_coordinate_list).argmin()]\n",
    "        closest_traffic_point.append(tuple(closest))\n",
    "        bike_traffic.append(ped_dict[tuple(closest)])\n",
    "        \n",
    "myfile3 = open('closest_intersection.csv', 'w')\n",
    "wr = csv.writer(myfile3)\n",
    "wr.writerow(closest_traffic_point)\n",
    "myfile3.close()\n",
    "\n",
    "myfile4 = open('closest_int_bike_predictions.csv', 'w')\n",
    "wr = csv.writer(myfile4)\n",
    "wr.writerow(bike_traffic)\n",
    "myfile4.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1145.9669774671361,\n",
       " 1043.663994458788,\n",
       " 1235.7081763712024,\n",
       " 1083.2871929951482,\n",
       " 1043.663994458788,\n",
       " 1409.2143012952026,\n",
       " 1120.2506065859702,\n",
       " 1432.4371794221831,\n",
       " 864.05610391248706,\n",
       " 1214.3963006224874]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bike_traffic[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>injury</th>\n",
       "      <th>safety_equip</th>\n",
       "      <th>road_class</th>\n",
       "      <th>cyclist_crash_type</th>\n",
       "      <th>age_of_cyclist</th>\n",
       "      <th>stname1</th>\n",
       "      <th>stet_1_type</th>\n",
       "      <th>stname_2</th>\n",
       "      <th>stet_2_type</th>\n",
       "      <th>long</th>\n",
       "      <th>...</th>\n",
       "      <th>date</th>\n",
       "      <th>date2</th>\n",
       "      <th>driver_action</th>\n",
       "      <th>driver_condition</th>\n",
       "      <th>year</th>\n",
       "      <th>intersection</th>\n",
       "      <th>coords</th>\n",
       "      <th>coordinates</th>\n",
       "      <th>closest_traffic</th>\n",
       "      <th>traffic_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Local</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>34.0</td>\n",
       "      <td>EUCLID</td>\n",
       "      <td>AV</td>\n",
       "      <td>ULSTER</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.413300</td>\n",
       "      <td>...</td>\n",
       "      <td>1986-01-14 00:00:00</td>\n",
       "      <td>1986-01-14 00:00:00</td>\n",
       "      <td>Lost control</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Euclid Ave &amp; Ulster St, Toronto, Ontario, M6G</td>\n",
       "      <td>[43.65906013500046, -79.41272077799965]</td>\n",
       "      <td>(43.6599, -79.4133)</td>\n",
       "      <td>(43.66055, -79.41466000000001)</td>\n",
       "      <td>1145.966977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>34.0</td>\n",
       "      <td>SALEM</td>\n",
       "      <td>AV</td>\n",
       "      <td>FERNBANK</td>\n",
       "      <td>AV</td>\n",
       "      <td>-79.433801</td>\n",
       "      <td>...</td>\n",
       "      <td>1986-01-28 00:00:00</td>\n",
       "      <td>1986-01-28 00:00:00</td>\n",
       "      <td>Driving properly</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Fernbank Ave &amp; Salem Ave, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.66582101600045, -79.43379108299968]</td>\n",
       "      <td>(43.665832, -79.433801)</td>\n",
       "      <td>(43.66717180000001, -79.43169820000001)</td>\n",
       "      <td>1043.663994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8.0</th>\n",
       "      <td>Minor</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>24.0</td>\n",
       "      <td>DUNDAS</td>\n",
       "      <td>ST</td>\n",
       "      <td>BEVERLEY</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.394060</td>\n",
       "      <td>...</td>\n",
       "      <td>1986-01-16 00:00:00</td>\n",
       "      <td>1986-01-16 00:00:00</td>\n",
       "      <td>Failed to yield right-of-way</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Beverley St &amp; Dundas St W, Toronto, Ontario, M5T</td>\n",
       "      <td>[43.65382045300049, -79.39385107199968]</td>\n",
       "      <td>(43.65377, -79.39406)</td>\n",
       "      <td>(43.65381040000001, -79.39383199999999)</td>\n",
       "      <td>1235.708176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9.0</th>\n",
       "      <td>Minimal</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Major Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>24.0</td>\n",
       "      <td>BLOOR</td>\n",
       "      <td>ST</td>\n",
       "      <td>MARGUERETTA</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.440590</td>\n",
       "      <td>...</td>\n",
       "      <td>1986-02-19 00:00:00</td>\n",
       "      <td>1986-02-19 00:00:00</td>\n",
       "      <td>Failed to yield right-of-way</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Bloor St W &amp; Margueretta, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.658975,-79.439718]</td>\n",
       "      <td>(43.6588, -79.44059)</td>\n",
       "      <td>(43.6591, -79.43918000000002)</td>\n",
       "      <td>1083.287193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>Minor</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>Minor Arterial</td>\n",
       "      <td>Unrecorded</td>\n",
       "      <td>54.0</td>\n",
       "      <td>AVENUE</td>\n",
       "      <td>RD</td>\n",
       "      <td>SHANLY</td>\n",
       "      <td>ST</td>\n",
       "      <td>-79.430720</td>\n",
       "      <td>...</td>\n",
       "      <td>1986-03-01 00:00:00</td>\n",
       "      <td>1986-03-01 00:00:00</td>\n",
       "      <td>Driving properly</td>\n",
       "      <td>Normal</td>\n",
       "      <td>1986.0</td>\n",
       "      <td>Dovercourt Rd &amp; Shanly St, Toronto, Ontario, M6H</td>\n",
       "      <td>[43.664316,-79.430499]</td>\n",
       "      <td>(43.66425, -79.43072)</td>\n",
       "      <td>(43.66717180000001, -79.43169820000001)</td>\n",
       "      <td>1043.663994</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       injury safety_equip      road_class cyclist_crash_type  age_of_cyclist  \\\n",
       "id                                                                              \n",
       "4.0   Minimal   Unrecorded           Local         Unrecorded            34.0   \n",
       "7.0   Minimal   Unrecorded  Major Arterial         Unrecorded            34.0   \n",
       "8.0     Minor   Unrecorded  Major Arterial         Unrecorded            24.0   \n",
       "9.0   Minimal   Unrecorded  Major Arterial         Unrecorded            24.0   \n",
       "10.0    Minor   Unrecorded  Minor Arterial         Unrecorded            54.0   \n",
       "\n",
       "     stname1 stet_1_type     stname_2 stet_2_type       long      ...        \\\n",
       "id                                                                ...         \n",
       "4.0   EUCLID          AV       ULSTER          ST -79.413300      ...         \n",
       "7.0    SALEM          AV     FERNBANK          AV -79.433801      ...         \n",
       "8.0   DUNDAS          ST     BEVERLEY          ST -79.394060      ...         \n",
       "9.0    BLOOR          ST  MARGUERETTA          ST -79.440590      ...         \n",
       "10.0  AVENUE          RD       SHANLY          ST -79.430720      ...         \n",
       "\n",
       "                     date                date2                 driver_action  \\\n",
       "id                                                                             \n",
       "4.0   1986-01-14 00:00:00  1986-01-14 00:00:00                  Lost control   \n",
       "7.0   1986-01-28 00:00:00  1986-01-28 00:00:00              Driving properly   \n",
       "8.0   1986-01-16 00:00:00  1986-01-16 00:00:00  Failed to yield right-of-way   \n",
       "9.0   1986-02-19 00:00:00  1986-02-19 00:00:00  Failed to yield right-of-way   \n",
       "10.0  1986-03-01 00:00:00  1986-03-01 00:00:00              Driving properly   \n",
       "\n",
       "     driver_condition    year  \\\n",
       "id                              \n",
       "4.0            Normal  1986.0   \n",
       "7.0            Normal  1986.0   \n",
       "8.0            Normal  1986.0   \n",
       "9.0            Normal  1986.0   \n",
       "10.0           Normal  1986.0   \n",
       "\n",
       "                                          intersection  \\\n",
       "id                                                       \n",
       "4.0      Euclid Ave & Ulster St, Toronto, Ontario, M6G   \n",
       "7.0    Fernbank Ave & Salem Ave, Toronto, Ontario, M6H   \n",
       "8.0   Beverley St & Dundas St W, Toronto, Ontario, M5T   \n",
       "9.0    Bloor St W & Margueretta, Toronto, Ontario, M6H   \n",
       "10.0  Dovercourt Rd & Shanly St, Toronto, Ontario, M6H   \n",
       "\n",
       "                                       coords              coordinates  \\\n",
       "id                                                                       \n",
       "4.0   [43.65906013500046, -79.41272077799965]      (43.6599, -79.4133)   \n",
       "7.0   [43.66582101600045, -79.43379108299968]  (43.665832, -79.433801)   \n",
       "8.0   [43.65382045300049, -79.39385107199968]    (43.65377, -79.39406)   \n",
       "9.0                    [43.658975,-79.439718]     (43.6588, -79.44059)   \n",
       "10.0                   [43.664316,-79.430499]    (43.66425, -79.43072)   \n",
       "\n",
       "                              closest_traffic  traffic_count  \n",
       "id                                                            \n",
       "4.0            (43.66055, -79.41466000000001)    1145.966977  \n",
       "7.0   (43.66717180000001, -79.43169820000001)    1043.663994  \n",
       "8.0   (43.65381040000001, -79.39383199999999)    1235.708176  \n",
       "9.0             (43.6591, -79.43918000000002)    1083.287193  \n",
       "10.0  (43.66717180000001, -79.43169820000001)    1043.663994  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "central_col_df['closest_traffic'] = tuple(closest_traffic_point)\n",
    "central_col_df['traffic_count'] = bike_traffic\n",
    "\n",
    "central_col_df.rename(columns={'closest_traffic': 'closest_ped_count', 'traffic_count':'predicted_bike_count'})\n",
    "central_col_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">traffic_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>size</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>intersection</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Galbraith Rd &amp; St George St, Toronto, Ontario, M5S</th>\n",
       "      <td>3.0</td>\n",
       "      <td>4495.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Huron St &amp; Russell St, Toronto, Ontario, M5S</th>\n",
       "      <td>3.0</td>\n",
       "      <td>4495.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Russell St &amp; St George St, Toronto, Ontario, M5S</th>\n",
       "      <td>11.0</td>\n",
       "      <td>3695.329056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Willcocks St &amp; Spadina Ave, Toronto, Ontario, M5S</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3574.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glen Morris St &amp; Spadina Ave, Toronto, Ontario, M5S</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3574.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Prince Edward Viad &amp; Bayview Ave, Toronto, Ontario, M4W</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3413.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Howard St &amp; Ontario St, Toronto, Ontario, M4X</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3413.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Howard St &amp; Rose Ave, Toronto, Ontario, M4X</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3413.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bloor St E &amp; Castle Frank Cres, Toronto, Ontario, M4W</th>\n",
       "      <td>20.0</td>\n",
       "      <td>3413.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bloor St E &amp; Parliament St, Toronto, Ontario, M4W</th>\n",
       "      <td>26.0</td>\n",
       "      <td>3413.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   traffic_count             \n",
       "                                                            size         mean\n",
       "intersection                                                                 \n",
       "Galbraith Rd & St George St, Toronto, Ontario, M5S           3.0  4495.800000\n",
       "Huron St & Russell St, Toronto, Ontario, M5S                 3.0  4495.800000\n",
       "Russell St & St George St, Toronto, Ontario, M5S            11.0  3695.329056\n",
       "Willcocks St & Spadina Ave, Toronto, Ontario, M5S            1.0  3574.000000\n",
       "Glen Morris St & Spadina Ave, Toronto, Ontario,...           1.0  3574.000000\n",
       "Prince Edward Viad & Bayview Ave, Toronto, Onta...           1.0  3413.000000\n",
       "Howard St & Ontario St, Toronto, Ontario, M4X                1.0  3413.000000\n",
       "Howard St & Rose Ave, Toronto, Ontario, M4X                  5.0  3413.000000\n",
       "Bloor St E & Castle Frank Cres, Toronto, Ontari...          20.0  3413.000000\n",
       "Bloor St E & Parliament St, Toronto, Ontario, M4W           26.0  3413.000000"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intersection_df = central_col_df.groupby('intersection').agg({'traffic_count':[np.size,np.mean]}).sort_values(by=[('traffic_count','mean')],ascending=False)\n",
    "intersection_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">traffic_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>size</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>intersection</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Queen St W &amp; Spadina Ave, Toronto, Ontario, M5V</th>\n",
       "      <td>74.0</td>\n",
       "      <td>1314.307288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bay St &amp; Dundas St W, Toronto, Ontario, M5G</th>\n",
       "      <td>71.0</td>\n",
       "      <td>1504.051740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queen St W &amp; University Ave, Toronto, Ontario, M5H</th>\n",
       "      <td>64.0</td>\n",
       "      <td>1337.865922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bloor St E &amp; Yonge St, Toronto, Ontario, M4W</th>\n",
       "      <td>63.0</td>\n",
       "      <td>1439.906704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bathurst St &amp; Queen St W, Toronto, Ontario, M5T</th>\n",
       "      <td>61.0</td>\n",
       "      <td>1270.732436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bathurst St &amp; College St, Toronto, Ontario, M5S</th>\n",
       "      <td>60.0</td>\n",
       "      <td>1248.605392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>King St W &amp; York St, Toronto, Ontario, M5H</th>\n",
       "      <td>58.0</td>\n",
       "      <td>1274.090650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bay St &amp; King St W, Toronto, Ontario, M5H</th>\n",
       "      <td>57.0</td>\n",
       "      <td>1436.424115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dundas St E &amp; Yonge St, Toronto, Ontario, M5B</th>\n",
       "      <td>57.0</td>\n",
       "      <td>1546.272908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>College St &amp; University Ave, Toronto, Ontario, M5G</th>\n",
       "      <td>55.0</td>\n",
       "      <td>1325.051617</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   traffic_count             \n",
       "                                                            size         mean\n",
       "intersection                                                                 \n",
       "Queen St W & Spadina Ave, Toronto, Ontario, M5V             74.0  1314.307288\n",
       "Bay St & Dundas St W, Toronto, Ontario, M5G                 71.0  1504.051740\n",
       "Queen St W & University Ave, Toronto, Ontario, M5H          64.0  1337.865922\n",
       "Bloor St E & Yonge St, Toronto, Ontario, M4W                63.0  1439.906704\n",
       "Bathurst St & Queen St W, Toronto, Ontario, M5T             61.0  1270.732436\n",
       "Bathurst St & College St, Toronto, Ontario, M5S             60.0  1248.605392\n",
       "King St W & York St, Toronto, Ontario, M5H                  58.0  1274.090650\n",
       "Bay St & King St W, Toronto, Ontario, M5H                   57.0  1436.424115\n",
       "Dundas St E & Yonge St, Toronto, Ontario, M5B               57.0  1546.272908\n",
       "College St & University Ave, Toronto, Ontario, M5G          55.0  1325.051617"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intersection_df = central_col_df.groupby('intersection').agg({'traffic_count':[np.size,np.mean]}).sort_values(by=[('traffic_count','size')],ascending=False)\n",
    "intersection_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">traffic_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>size</th>\n",
       "      <th>mean</th>\n",
       "      <th>normalized_accident_rate</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>intersection</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Front St W &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>29.0</td>\n",
       "      <td>276.607735</td>\n",
       "      <td>0.004368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Front St W &amp; John St, Toronto, Ontario, M5V</th>\n",
       "      <td>17.0</td>\n",
       "      <td>199.933112</td>\n",
       "      <td>0.003543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queens Quay W &amp; Rees St, Toronto, Ontario, M5J</th>\n",
       "      <td>9.0</td>\n",
       "      <td>139.888889</td>\n",
       "      <td>0.002681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bay St &amp; Queens Quay W, Toronto, Ontario, M5J</th>\n",
       "      <td>18.0</td>\n",
       "      <td>288.928665</td>\n",
       "      <td>0.002596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queens Quay W &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>18.0</td>\n",
       "      <td>289.316788</td>\n",
       "      <td>0.002592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yonge St &amp; The Esplanade, Toronto, Ontario, M5E</th>\n",
       "      <td>20.0</td>\n",
       "      <td>330.000000</td>\n",
       "      <td>0.002525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Station St &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>14.0</td>\n",
       "      <td>235.000000</td>\n",
       "      <td>0.002482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lake Shore Blvd W &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>14.0</td>\n",
       "      <td>235.000000</td>\n",
       "      <td>0.002482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queen St W &amp; Spadina Ave, Toronto, Ontario, M5V</th>\n",
       "      <td>74.0</td>\n",
       "      <td>1314.307288</td>\n",
       "      <td>0.002346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lake Shore Blvd E &amp; Yonge St, Toronto, Ontario, M5E</th>\n",
       "      <td>13.0</td>\n",
       "      <td>235.000000</td>\n",
       "      <td>0.002305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bathurst St &amp; College St, Toronto, Ontario, M5S</th>\n",
       "      <td>60.0</td>\n",
       "      <td>1248.605392</td>\n",
       "      <td>0.002002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bathurst St &amp; Queen St W, Toronto, Ontario, M5T</th>\n",
       "      <td>61.0</td>\n",
       "      <td>1270.732436</td>\n",
       "      <td>0.002000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queen St W &amp; University Ave, Toronto, Ontario, M5H</th>\n",
       "      <td>64.0</td>\n",
       "      <td>1337.865922</td>\n",
       "      <td>0.001993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bay St &amp; Dundas St W, Toronto, Ontario, M5G</th>\n",
       "      <td>71.0</td>\n",
       "      <td>1504.051740</td>\n",
       "      <td>0.001967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bay St &amp; Lake Shore Blvd W, Toronto, Ontario, M5J</th>\n",
       "      <td>11.0</td>\n",
       "      <td>235.000000</td>\n",
       "      <td>0.001950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>College St &amp; Shaw St, Toronto, Ontario, M6G</th>\n",
       "      <td>53.0</td>\n",
       "      <td>1153.924281</td>\n",
       "      <td>0.001914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>King St W &amp; York St, Toronto, Ontario, M5H</th>\n",
       "      <td>58.0</td>\n",
       "      <td>1274.090650</td>\n",
       "      <td>0.001897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lake Shore Blvd W &amp; Spadina Ave, Toronto, Ontario, M5V</th>\n",
       "      <td>7.0</td>\n",
       "      <td>154.000000</td>\n",
       "      <td>0.001894</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bloor St W &amp; Keele St, Toronto, Ontario, M6P</th>\n",
       "      <td>49.0</td>\n",
       "      <td>1113.959596</td>\n",
       "      <td>0.001833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bloor St E &amp; Yonge St, Toronto, Ontario, M4W</th>\n",
       "      <td>63.0</td>\n",
       "      <td>1439.906704</td>\n",
       "      <td>0.001823</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   traffic_count               \\\n",
       "                                                            size         mean   \n",
       "intersection                                                                    \n",
       "Front St W & York St, Toronto, Ontario, M5J                 29.0   276.607735   \n",
       "Front St W & John St, Toronto, Ontario, M5V                 17.0   199.933112   \n",
       "Queens Quay W & Rees St, Toronto, Ontario, M5J               9.0   139.888889   \n",
       "Bay St & Queens Quay W, Toronto, Ontario, M5J               18.0   288.928665   \n",
       "Queens Quay W & York St, Toronto, Ontario, M5J              18.0   289.316788   \n",
       "Yonge St & The Esplanade, Toronto, Ontario, M5E             20.0   330.000000   \n",
       "Station St & York St, Toronto, Ontario, M5J                 14.0   235.000000   \n",
       "Lake Shore Blvd W & York St, Toronto, Ontario, M5J          14.0   235.000000   \n",
       "Queen St W & Spadina Ave, Toronto, Ontario, M5V             74.0  1314.307288   \n",
       "Lake Shore Blvd E & Yonge St, Toronto, Ontario,...          13.0   235.000000   \n",
       "Bathurst St & College St, Toronto, Ontario, M5S             60.0  1248.605392   \n",
       "Bathurst St & Queen St W, Toronto, Ontario, M5T             61.0  1270.732436   \n",
       "Queen St W & University Ave, Toronto, Ontario, M5H          64.0  1337.865922   \n",
       "Bay St & Dundas St W, Toronto, Ontario, M5G                 71.0  1504.051740   \n",
       "Bay St & Lake Shore Blvd W, Toronto, Ontario, M5J           11.0   235.000000   \n",
       "College St & Shaw St, Toronto, Ontario, M6G                 53.0  1153.924281   \n",
       "King St W & York St, Toronto, Ontario, M5H                  58.0  1274.090650   \n",
       "Lake Shore Blvd W & Spadina Ave, Toronto, Ontar...           7.0   154.000000   \n",
       "Bloor St W & Keele St, Toronto, Ontario, M6P                49.0  1113.959596   \n",
       "Bloor St E & Yonge St, Toronto, Ontario, M4W                63.0  1439.906704   \n",
       "\n",
       "                                                                             \n",
       "                                                   normalized_accident_rate  \n",
       "intersection                                                                 \n",
       "Front St W & York St, Toronto, Ontario, M5J                        0.004368  \n",
       "Front St W & John St, Toronto, Ontario, M5V                        0.003543  \n",
       "Queens Quay W & Rees St, Toronto, Ontario, M5J                     0.002681  \n",
       "Bay St & Queens Quay W, Toronto, Ontario, M5J                      0.002596  \n",
       "Queens Quay W & York St, Toronto, Ontario, M5J                     0.002592  \n",
       "Yonge St & The Esplanade, Toronto, Ontario, M5E                    0.002525  \n",
       "Station St & York St, Toronto, Ontario, M5J                        0.002482  \n",
       "Lake Shore Blvd W & York St, Toronto, Ontario, M5J                 0.002482  \n",
       "Queen St W & Spadina Ave, Toronto, Ontario, M5V                    0.002346  \n",
       "Lake Shore Blvd E & Yonge St, Toronto, Ontario,...                 0.002305  \n",
       "Bathurst St & College St, Toronto, Ontario, M5S                    0.002002  \n",
       "Bathurst St & Queen St W, Toronto, Ontario, M5T                    0.002000  \n",
       "Queen St W & University Ave, Toronto, Ontario, M5H                 0.001993  \n",
       "Bay St & Dundas St W, Toronto, Ontario, M5G                        0.001967  \n",
       "Bay St & Lake Shore Blvd W, Toronto, Ontario, M5J                  0.001950  \n",
       "College St & Shaw St, Toronto, Ontario, M6G                        0.001914  \n",
       "King St W & York St, Toronto, Ontario, M5H                         0.001897  \n",
       "Lake Shore Blvd W & Spadina Ave, Toronto, Ontar...                 0.001894  \n",
       "Bloor St W & Keele St, Toronto, Ontario, M6P                       0.001833  \n",
       "Bloor St E & Yonge St, Toronto, Ontario, M4W                       0.001823  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intersection_df['traffic_count','normalized_accident_rate'] = intersection_df['traffic_count','size']/(24*intersection_df['traffic_count','mean'])\n",
    "intersection_df = intersection_df.sort_values(by=[('traffic_count','normalized_accident_rate')],ascending=False)\n",
    "intersection_df.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1449"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = intersection_df['traffic_count','normalized_accident_rate']\n",
    "df = df_test.to_frame(name='normalized yearly accident rate')\n",
    "df['total collisions'] = intersection_df['traffic_count','size']\n",
    "df['traffic estimate'] = intersection_df['traffic_count','mean']\n",
    "df = df.dropna()\n",
    "len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkIAAAJCCAYAAAAsp6gAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3Xl8FPXh//H35oSc5BJCGjCAEsCA3CCCCNoKAiKItIIa\nEJVyFAUURRBRLBSCoggF5SoKFa3cRcGqRUQBQcINX48QEAgCm2STQFiS7O8Pftm6JGAm2d1s2Nfz\n8ejjkczMzr4nsuk7n/nMjMlms9kEAADghXwqOwAAAEBloQgBAACvRRECAABeiyIEAAC8FkUIAAB4\nLYoQAADwWhQhAADgtShCAADAa/lVdoCqzmazyWzOU1FR1bgvpY+PSZGRwWR2MTK7B5ndg8zu4eNj\nUlRUSGXH8DoUoQoymUzy8TFVqQ/alZk3f71DhUWlDw5GhQepWVJjd0YsobTMno7M7kFm9yCze/j4\nmCo7gleiCEGW84UKjY4rdV1WToab0wAA4D7MEQIAAF6LIgQAALwWRQgAAHgtihAAAPBaFCEAAOC1\nKEIAAMBrUYQAAIDXoggBAACvRRECAABeiyIEAAC8FkUIAAB4LYoQAADwWhQhAADgtShCAADAa1GE\nAACA16IIAQAAr0URAgAAXosiBAAAvBZFCAAAeC2KEAAA8FoUIQAA4LUoQgAAwGtRhAAAgNeiCAEA\nAK9FEQIAAF6LIgQAALwWRQgAAHgtihAAAPBaFCEAAOC1KEIAAMBrUYQAAIDXoggBAACvRRECAABe\niyIEAAC8FkUIAAB4LYoQAADwWhQhAADgtShCAADAa1GEAACA16IIAQAAr0URAgAAXosiBAAAvBZF\nCAAAeC2KEAAA8FoUIQAA4LUoQgAAwGtRhAAAgNeiCAEAAK9FEQIAAF7Lr7IDwHMVFRUpNztLZ8+e\nveo2kZGR8vGhTwMAqiaKEK4q15KlA2lmqfqp0tfnZuv37RIVHR3t5mQAADgHRQjXVD0oRGE1Iis7\nBgAALsE5DQAA4LUoQgAAwGtRhAAAgNeiCAEAAK9FEQIAAF6LIgQAALwWRQgAAHgtihAAAPBa3FAR\nLlFUVCSz2XzV9TyaAwDgCShCcAmz2axN2w4rJCS8xDoezQEA8BQUIbhMSEg4j+cAAHg0ihA8Smmn\n1Pz8TCooOC+zOUcFBbarnlLjdBsAwCiKEMrtcmk5V+o6s/mcbEU2w/ss7ZSaj49J1asHKO3H7yUf\nf91wQ2yJ13G6DQBQHhQhlFtebra+TD2tG26wlliXcfKYQsKjFK6oEut+q0AFB4U5nFLz9TEpKChQ\nQcGnJR9/w6fbfmvitsRoEgB4K4oQKiQoOKzUYpJjybzqa8pboMrrWhO3pWuPJnH1GwBc3yhCqBTl\nKVDXYnSUqayuVaIslky1SaypyMiSpc3Pz6Tw8OqG3w8A4F4UIVwXKjLKVN4SlWPJ1Jep6aW+Z15u\nlu4uuCA/v+oqKHCcK1VUVCRJhid9V+QU37VeW5wnIMBXBQXnlZWV55CZUS8A1zOKEK7pwvlcWbJK\n/z/QvNwc+fj5l7remet8fEyyXgzQ+TyL5HPt113N+TzLVY/jl4yflX40X1FRJU+Nnf3llELCI2Xy\nMRl6z7xci1Z9+p2CQ2qo6IpJ42d/OSWTr3+p73fhfJ7uaFm/1FEms/mcNu/6UdWDgkt9z/K+tjhP\nTEyMqlULUH6+1Z75Wvv0BMVXFF5Z3jwZmd3DmZm5COP6ZrLZbFXjXzUAAICTMd4NAAC8FkUIAAB4\nLYoQAADwWhQhAADgtShCAADAa1GEAACA16IIAQAAr0URAgAAXosiBAAAvJbHPGIjPz9f06ZN05Yt\nW5Sdna0GDRpo+PDh6tixY6nbHzhwQCkpKTp48KACAwPVsmVLTZw4UZGRl58JVVRUpDfeeEMbN27U\nmTNnVKdOHQ0aNEi9evWSJO3YsUOPPPKIAgMD7fu02WyKjY3Vxo0bXX/AAACg0nlMEZo8ebL279+v\nBQsWKC4uTqtWrdKwYcO0Zs0a1atXz2Hb7OxsDRkyRL1799bs2bNltVo1btw4jRw5UsuWLZMkzZ07\nVytXrtS8efPUsGFDbd68WaNGjVJMTIzat28vSTKZTNqzZ4/bjxUAAHgGjzg1ZrFYtG7dOo0YMUIJ\nCQkKCAhQ//791aBBAy1fvrzE9uvWrVNhYaHGjBmjkJAQRUZGasyYMdq1a5cOHjwoSVq+fLkGDRqk\nJk2ayM/PT127dlXnzp21dOlSdx8eAADwUB5RhA4cOKDCwkI1a9bMYXlSUlKpIzZ79uxR48aN5ef3\nvwGtxMREBQYGKjU1VceOHZPZbFbTpk0dXte0aVOH/dlsNj3//PO644471L59ew0dOlRHjx517sEB\nAACP5RFFyGw2S5LCw8MdlkdEROjcuXOlbn/ltsWvN5vNMpvNMplMqlGjRon9ZWVlSZKCg4PVrFkz\nde7cWZ999plWrlypwsJCJScn6/z58846NAAA4ME8ogi5gs1mu+b6Jk2aaMWKFfrDH/4gPz8/xcbG\natq0acrIyNB///tfp70PAACuxv8XlZ9HTJaOioqSJGVlZal69er25ZmZmYqJiSmxfXR0tDIyMkos\nz87OVnR0tMP+fi0zM9N+VdnVcgQHB+v06dNlzm4ymWSxXFBhYVGZX1OZfH19FBZWncwuRmb3ILN7\nkNk9ijOXh9mcJx8fk5MTVX0REcG/uY1HFKFbbrlF/v7+Sk1NVWxsrH357t271aFDhxLbN2/eXK+9\n9pouXbokf39/SdLevXtltVrVokULxcfHKzo6WqmpqWrVqpX9dbt27VLz5s0lSR9//LEyMjI0aNAg\n+/oTJ04oLy9PderUMZS/sLBIBQVV44NWjMzuQWb3ILN7kNlzFRXZVFTEqFB5eMSpsZCQEPXt21dz\n5sxRWlqa8vPztXDhQh0/flwDBgzQ6dOn1a1bN3333XeSpB49eigwMFApKSnKzc1VRkaGZsyYodtu\nu00333yzJCk5OVlLlizR/v37ZbVatX79em3dutVefAIDAzVz5kytXbtWly5d0qlTp/Tiiy+qTp06\nV713EQAAuL54xIiQJI0fP14zZszQwIEDlZubq0aNGmnhwoWKj4/XiRMndPToUfsk5pCQEC1evFhT\npkzR7bffrsDAQHXp0kXjx4+372/IkCGyWq0aMWKEzGazbrzxRs2aNUstWrSQJHXp0kXTpk3TggUL\nNGnSJAUEBKhLly6aPn26AgICKuVnAAAA3MtkY4ZVhWVm5lWZoVc/Px9FRAST2cXI7B5kdg8yu0dx\n5vI4cybHyWmuDzExob+5jUecGgMAAKgMFCEAAOC1KEIAAMBrUYQAAIDXoggBAACvRRECAABeiyIE\nAAC8FkUIAAB4LYoQAADwWhQhAADgtTzmWWMAgKrl22+3a9Gi+UpLS1NoaKhat26rv/xljKpVq1bq\n9mfPntX8+W9p584dysvLU82aNdWjx33q339AqdsvWbJACxfO1+zZ83XrrS1ceSjl8umnn+if/3xP\nP/98XJGRUercuYuefHK4TCZTZUeDAYwIAQAMO378mMaNG60777xLa9Z8rNmz5+vIkcNKSZl61ddM\nmPCsMjJO6Z13/qFPPvlCo0aN1dtvz9X69atLbPv990f00UcfeGyp2L17l6ZMmaQBAx7Rv//9H/31\nrzP0n/9s1KJFb1d2NBjEiBAAr9exY2v95S9j9N13O7Vz5w75+/vrvvv66Mknh9u32bz5Cy1b9g+l\np6fJ399fbdveppEjR6tGjRqSLheDt956XXv27JbNZlN8fF09+eQwtW7dTpK0aNHb2rLlv7rjjju1\nYsVyjR//otq1u12zZ7+mLVs2Kzc3RxERkerZs7cefniQJOnixXy9/fZcffXVlzp37qxiYm5Q374P\n6oEH/mjf51dffanBgx/XvHlzlJFxUnXrJuiZZ8YrMbGR/dhGjhyttWtXKSAgQIsWvVfi+B96qK9O\nn85wWGazSSaT9Oijj2nw4CElXrNmzUrFxcXpwQcfkiTVqhWrQYOGaMKEcRox4mn7z+XXjhw5pFGj\nxio6OkaS1KpVG9Wvf5MOHjygHj1627e7dOmSpkyZpOTkx/TGGzMd9vGPfyzU+vVr9eGHa0r9b7lo\n0dvatOljDRnypBYsmKczZ86obt0bNWHCZNWr16DU13Tp0kFX9q3i4x83boJ+//tuJV7z0UcfqG3b\n9ura9feSpHr16qt//wFavPgdDRr0uHx8GGeoKihCACBp+fKlmjBhsl59dbp27tyh554bo7i436lH\nj/u0c+cOTZ48QRMmTNadd3bVuXNnNWXKJE2Y8KzeeuvyCMCECc+qTp0btWbNJ/L19dPChfM1fvyz\n+uij9QoLC5MknTnzi3Jzc/T1118rL++SlixZpL1792jRomWKiIjQkSOH9eyzo9SwYSO1adNOM2f+\nTQcP7teMGbP0u9/V0ddff6UJE55VaGiY/vCH7pKkU6dO6quvvtT8+YtkMvlo7Ni/aObMaXrnnX/Y\nj23t2lV6+eW/XrUILF/+keGf18GD+9WkSZLDskaNmqiwsFCHDh1Q+/YdSrymc+eu2rTpY7Vt2161\nasVq9+5dSk9P0+OPD3XY7p135io8vIb69u2vWbNSHNY9+uhjevTRx66Z7cyZX7Rt29davXq1srLy\nNHnyJI0bN1offri21O0//3xrWQ7ZwcGD+9Wr1/0Oyxo1aqLc3BwdO5auG29MMLxPVA4qKwBI6tCh\no1q2bC0fHx+1adNOrVu30ebNn0uSVq78ULfddru6dLlLJpNJ0dExGjp0hPbs2a1Tp05KkubPX6IX\nXnhJgYHV5Ofnp27d7lV+/gX99NMP9vfIycnRY489qYCAAElSbm6ufHxMCgy8/H3Dholas2aj2rRp\np/Pn87Rx4wYlJw9RnTo3ysfHR7ff3km33dZRGzast+/z/Pk8/eUvoxUcHKKgoCB17NhZaWk/Ohxb\n27btrlqCyisrK9Ne8IqFh9ewryvNc89NVLVq1fXgg/fpjjvaauzYv2jIkD/bR80kac+eVK1du1rj\nx08qd7ZLly5p5MinFBoaqpCQUCUnD9Hp0xk6dOhAufd5pcvHH+6wrEaNGrLZbMrMNDvtfeB6jAgB\ngFTiL/jatX+n7777VpJ07NhR/fzzcXXt+r9RDptN8vPz06lTJxUbW1v79+/VkiUL9OOPP+jixYuS\nbDKZTLJarfbXhISEKDQ01P59nz79tH3717rvvm669dbmat26re6+u5siIiJ04sQJ2Ww21atXv0TO\nTz/9xP59eHgNBQUF27+vXr26w3tePpa48v9gnGjixOd06ZJV77+/SjfcUFN79uzWpEnPKzg4WN27\n91R+fr6mTp2soUOHq1at2HK/T0hIqKKiou3f164dJ5vNpl9+Oa1GjZo441BwHaEIAYCkwsJCh+9t\nNptMpsuD5gEBgbrvvj56+ulnS33tiRM/69lnn1bv3n306qvTFR5eQydO/Kw//tHx1Imfn7/D9zfc\nUFOLFy/X4cOH9O232/XJJxu0aNE7euONv6ugoEA2m63Ee11e9r8JLWWZTHzl+16pPHOEIiOjlJ2d\n7bAsOzvLvu5Kx44d1TfffKW//32R4uJ+J+nyHKG77+6mf/1rhbp376nZs19TzZqx6t37gd88pmsp\nLCy44lgu/xyL/3teqTxzhCIiImWxOB5/VlaWTCaToqJKHj88F0UIAHR5svOvnTz5s2rWrClJqlOn\nro4cOeyw/uLFfOXm5ioqKlqHDx9UQcElPfLIY/bTQwcO7PvNkpKfny+TyaTExEZKTGykhx9O1ogR\nT+iTT/6t5OTHZDKZ9MMPPzic1vrppx9Up05dZxyyXXnmCN1yS1N99dWXDsv27EmVn5+fGjVqXGL7\nwsIiSVJRkWPhvFz4Lq9bu3aVgoOD1aPHXQ7bPPfcGN1zz7166qmxZcp24cIFnTt3ThERl0fKTpw4\nLkmqWbNWqduXZ47QLbc01YED+xyW7d27W2Fh4YqPd+5/H7gWc4QAQNLWrVv03Xc7VVBQoG3bvtbO\nnd/arwjq3/8hHTp0QCtWLFN+fr6ys7M0bdoUPfXUMElSXFy8JGn37suv3779G33xxWeSVGKk5dee\nf36Mpk59WZmZl+fUHD9+TL/8clp1696o8PAauvPOrvrHPxbo2LF0FRQU6Isv/qNt275W7959XPmj\nKJPevS+PIr3//nu6eDFfx44d1aJFb6tbtx72uTPz5r2lSZPGS7pcJuvVq6+FC+crIyNDhYWF2r17\nlz77bJPuuusPkqRVqzbovfc+1JIl/7T/T5Kef36ihgwZWnqQUvj5+Wvu3DdlsVhksVj0j38sVFxc\nvBo2THTa8T/44EPauXOH/vOfjbp06ZIOHz6oFSuWqX//hzz2kn+UjhEhANDl+Tr/+tcKPffcGPn7\n+2ngwEftp0QaNWqil1+eqn/8Y6Hefnuu/P391aZNe6WkvClJSkxspEGDHtdrr03X3/72qlq1aqPn\nn5+oWbNSNGvWjKu+5wsvTNasWdM1YMADunTJqqioaN1zz73q3buvJOm5517UW2+9rtGjRyg3N0dx\ncfF6+eWp6tixc5mPy2QyueT/mGvVilVKypuaM2eW3nnn7woJCdFdd92jP/95pH0bs/mcvQj6+vpq\nxow3NH/+Wxo8eIDy8y8oOjpGDz88SH/84+UbKhZfVn9l/vDwGgoJCZFUfPn8mqteASZJoaGhatu2\nvfr27avTp39RvXr1NX366848fDVpcosmT56qBQv+rr/+9WVFRESob9/+9lsfoOow2Uo7CQ1DMjPz\nVFBQVNkxysTPz0cREcFkdjEyu4ezMnfs2Fpjxz6v++5z/UiLN/+cnWXQoIe0ePHyUtctWvS21qxZ\nqX//e5NHZS6L4p9zeZw5k+PkNNeHmJjQ39yGU2MAgCpj69YtXPkFp+LUGACv56rTR3C+Dh06qkOH\njpUdA9cRihAAr/fllzsqOwKcZPDgJzR48BOVHQNVCKfGAACA16IIAQAAr0URAgAAXosiBAAAvBZF\nCAAAeC2KEAAA8FoUIQAA4LUoQgAAwGtRhAAAgNeiCAEAAK9FEQIAAF6LIgQAALwWRQgAAHgtihAA\nAPBaFCEAAOC1KEIAAMBrUYQAAIDXoggBAACvRRECAABeiyIEAAC8FkUIAAB4Lb/KDgDAvdJOWbRh\nW7oseVaFBQeoe7u6SogNq+xYAFApKEKAF0k7ZdGclftkzrn4v2UnLRreJ0k3xdeoxGQAUDk4NQZ4\nkQ3b0h1KkCSZcy5qw7b0SkoEAJWLIgR4EUue1dByAFXD2bNnKztClUURArxIWHCAoeUAcL2jCAFe\npHu7uooMDXRYFhkaqO7t6lZSIgCoXEyWBrxIQmyYhvdJ0sfb0pXNVWMAQBECvE1CbJiG3Z9U2TEA\nwCNwagwAAHgtihAAAPBaFCEAAOC1KEIAAMBrUYQAAIDXoggBAACvRRECAABeiyIEAAC8FkUIAAB4\nLYoQAADwWhQhAADgtShCAADAa1GEAACA16IIAQAAr0URAgAAXosiBAAAvBZFCAAAeC2KEAAA8FoU\nIQAA4LUoQgAAwGtRhAAAgNfymCKUn5+vl156SV27dlWrVq30xz/+UVu2bLnq9gcOHNCgQYPUtm1b\nderUSU8//bTMZrN9fVFRkV5//XXdc889atmype6//36tXbv2qvubO3euEhMT9e233zr1uAAAgOfy\nmCI0efJk7dq1SwsWLNDXX3+t+++/X8OGDdNPP/1UYtvs7GwNGTJEiYmJ+uyzz7R69Wrl5uZq5MiR\n9m3mzp2rlStXaubMmdq+fbtGjBih8ePH65tvvimxv0OHDum9996TyWRy6TECAADP4hFFyGKxaN26\ndRoxYoQSEhIUEBCg/v37q0GDBlq+fHmJ7detW6fCwkKNGTNGISEhioyM1JgxY7Rr1y4dPHhQkrR8\n+XINGjRITZo0kZ+fn7p27arOnTtr6dKlDvuyWq169tlnNWzYMLccKwAA8BweUYQOHDigwsJCNWvW\nzGF5UlKS9uzZU2L7PXv2qHHjxvLz87MvS0xMVGBgoFJTU3Xs2DGZzWY1bdrU4XVNmzYtsb9Zs2Yp\nIiJCAwcOlM1mc+JRAQAAT+f325u4XvHcnvDwcIflEREROnfuXKnbX7lt8evNZrPMZrNMJpNq1KhR\nYn9ZWVn273fu3KkPP/xQq1evrlB+X1+P6JNlUpyVzK5FZvcgs3uQ2T0qktXHR/LzqzrH6kk8ogi5\nwm+N7pw/f17jx4/X6NGjFRcXV6H3CgurXqHXVwYyuweZ3YPM7kFmzxUREayIiODKjlEleUQRioqK\nkiRlZWWpevX//aPNzMxUTExMie2jo6OVkZFRYnl2draio6Md9vdrmZmZioyMlCRNmzZNtWvX1p/+\n9KcK57dYLqiwsKjC+3EHX18fhYVVJ7OLkdk9yOweZHaP4szlkZmZJ3//PCcnqvrKUg49ogjdcsst\n8vf3V2pqqmJjY+3Ld+/erQ4dOpTYvnnz5nrttdd06dIl+fv7S5L27t0rq9WqFi1aKD4+XtHR0UpN\nTVWrVq3sr9u1a5datGghSfrggw8UEhKidu3aOex72LBhuu+++zRhwoQy5y8sLFJBQdX4oBUjs3uQ\n2T3I7B5k9lxFRfKK43QFjzihGBISor59+2rOnDlKS0tTfn6+Fi5cqOPHj2vAgAE6ffq0unXrpu++\n+06S1KNHDwUGBiolJUW5ubnKyMjQjBkzdNttt+nmm2+WJCUnJ2vJkiXav3+/rFar1q9fr61btyo5\nOVmStHnzZv373//WmjVr7P+TpFdffVWjRo2qlJ8DAABwL48YEZKk8ePHa8aMGRo4cKByc3PVqFEj\nLVy4UPHx8Tpx4oSOHj2q8+fPS7pcnBYvXqwpU6bo9ttvV2BgoLp06aLx48fb9zdkyBBZrVaNGDFC\nZrNZN954o2bNmmUfEapZs2aJDCaTSREREQoNDXXPQQMAgEplsnHNeIVlZuZVmSFJPz8fRUQEk9nF\nyOweZHYPMrtHcebyOHQoTdHR0U5OVPXFxPz2wIZHnBoDAACoDBQhAADgtShCAADAa1GEAACA16II\nAQAAr0URAgAAXosiBAAAvBZFCAAAeK1yFaHDhw9r7Nix6tatm1q2bKnjx48rPz9fc+fOdXY+AAAA\nlzFchLZv365+/fopNTVVzZo1k9VqlSSdOXNGixYt0tKlS50eEgAAwBUMF6HXX39dPXv21MaNGzVt\n2jT5+V1+XFl8fLxeeOEFrVixwukhAQAAXMFwETpy5IgGDx4sX1/fEutat26tY8eOOSUYAACAqxku\nQsHBwfbTYVfKzMxUYGBghUMBAAC4g+Ei1LhxY02bNk25ubkOy61Wq95++221bNnSaeEAAABcyc/o\nC0aOHKlHHnlEnTp1UlJSki5duqTnn39eR48e1YULF7Rs2TJX5AQAAHA6wyNCSUlJWrNmjXr37q2c\nnBzVrl1b+fn5uueee7R69WolJia6IicAAIDTGR4ROnnypOLj4/Xiiy+WWJeTk6M9e/aoWbNmTgkH\nAADgSoZHhLp27arMzMxS1504cUJDhgypcCgAAAB3KPOI0PTp0yVJNptNb731lqpVq+aw3mazad++\nfSoqKnJuQgAAABcpcxHKzc3V7t27ZTKZtHz58lK3CQsL08iRI50WDgAAwJXKXIRefvllSVJiYqK2\nbt2qqKgol4UCAABwB8NzhA4fPnzVEmQ2mzV27NgKhwIAAHAHw1eNSdLp06e1a9euEpOm9+7dq08/\n/dQpwQAAAFzNcBHavn27hg4dqgsXLshkMslms0mSTCaTfH19NWDAAKeHBAAAcAXDRWjWrFnq1q2b\nHnvsMT3wwAN6++235efnp9WrV8tms+nZZ591RU4AAACnMzxH6Pvvv9cTTzyh+vXrS5Jq1aql5s2b\na/LkyQoPD1dKSorTQwIAALiC4SJktVoVEBAgSQoKClJWVpZ9Xb9+/bR27VrnpQMAAHAhw0WoQYMG\n+uijj1RYWKg6depo9erV9nXHjx/XxYsXnRoQAADAVQzPEUpOTta4cePUvXt39enTRxMnTtT+/ftV\no0YN7dixQ+3bt3dFTgAAAKczXIR69eql2rVrKy4uTvXr11deXp7Wrl2rkydPqmfPnnrqqadckRMA\nAMDpDBchs9msli1bymQySbo8QpScnOzsXEClSztl0YZt6bLkWRUWHKDu7eoqITassmMBAJzI8Byh\nzp076/Tp067IAniMtFMWzVm5T7uOnNH3P2dr15EzmrNyn9JOWSo7GgDAiQwXoVtuuUVbtmxxRRbA\nY2zYli5zjuPEf3PORW3Yll5JiQAArmD41Fjv3r21aNEiffnll2rSpIlCQ0NLbMPdpVHVWfKshpYD\nAKomw0XoxRdflCSlpaWV+lwxk8lEEUKVFxYcYGg5AKBqMlyEPvvsM1fkADxK93Z1lXbS4nB6LDI0\nUN3b1a3EVAAAZzNchOLi4lyRA/AoCbFhGt4nSR9vS1c2V40BwHXLcBECvEVCbJiG3Z9U2TEAAC5k\n+KoxAACA6wVFCAAAeC2nFyGbzebsXQIAALiE4SLUtWtXZWZmlrru0KFD6tixY4VDAQAAuEOZJ0tv\n3rxZknTixAlt3bq1xI0UbTabvv76a+Xk5Dg3IQAAgIuUuQhNnTpVR48elclk0tixY6+63T333OOU\nYAAAAK5W5iL0ySefKDMzU+3bt9fs2bMVHh5eYpuwsDAlJiY6NSAAAICrGLqPUEREhJYuXaoWLVrI\nz49bEAEAgKrNcJtp06aNzpw5o4MHD1510nTv3r0rHAwAAMDVDBehdevW6YUXXtClS5dKvVTeZDJR\nhAAAQJVguAi99dZbateunQYPHqzIyEiZTCZX5AIAAHA5w0UoIyND8+bNU0JCgivyAAAAuI3hGyrW\nqVNH58+fd0UWAAAAtzJchEaNGqXXXntNZrPZFXkAAADcxvCpsX/9619KT09X586d9bvf/U5BQUEO\n600mkz788EOnBQQAAHAVw0UoLy9PsbGxio2NdUUeAAAAtzFchN59911X5AAAAHA7w3OEip0/f157\n9uzRpk34Dx2eAAAgAElEQVSb7JOnCwsLnRYMAADA1QwXoaKiIqWkpKh9+/bq37+/Ro0apXPnzikj\nI0M9e/bUL7/84oqcAAAATme4CM2bN0/Lli3T4MGDtWTJElWrVk2SFBwcrIiICM2aNcvpIQEAAFzB\n8Byhjz76SJMmTSrxGI3Q0FA9/fTT+stf/uK0cAAAAK5keETozJkzatmyZanrYmNjZbFYKhwKAADA\nHQwXodjYWO3bt6/UdYcOHVJMTEyFQwEAALiD4VNjHTt21OTJk5WXl6fbb79dJpNJmZmZOnLkiF59\n9VXdc889rsgJAADgdIaL0OjRo5WWlqaJEyfKZDLJZrOpf//+stls6tSpk5566ilX5AQAAHA6w0Uo\nKChICxcu1L59+5Samqrc3FyFhYWpefPmaty4sSsyAgAAuIThIlQsKSlJSUlJzswCAADgVmUqQqNG\njSrzDk0mE/cSAgAAVUKZitCVV4lZLBbl5uYqNDRUwcHBysnJUV5enmrUqMHDWAEAQJVRpiL0+eef\n27/eunWr5syZo1deeUX169e3Lz906JBeeuklbqgIAACqDMP3EUpJSdHo0aMdSpAkNWrUSGPGjNH0\n6dOdFg4AAMCVDBehH3/8UVFRUaWui4mJUVpaWoVDAQAAuIPhIhQTE6N3331XNpvNYXlRUZGWLVum\n6Ohop4UDAABwJcOXzw8dOlQTJ07Upk2b1LBhQwUHB+vChQs6ePCgzGazxo8f74qcAAAATme4CPXr\n10/x8fH66KOP9MMPP+jYsWOqXr262rRpo/vuu0+dO3d2QUwAAADnK9cNFdu1a6d27do5OwsAAIBb\nlakIbd68Wbfddpv8/f21efPm39z+jjvuMBwkPz9f06ZN05YtW5Sdna0GDRpo+PDh6tixY6nbHzhw\nQCkpKTp48KACAwPVsmVLTZw4UZGRkZIuz1l64403tHHjRp05c0Z16tTRoEGD1KtXL/s+Vq9eraVL\nl+ro0aMKCgpSkyZNNHr0aDVs2NBwfgAAUPWUqQg9+eST2rp1q6KiovTkk0/aH7ZaGpPJpEOHDhkO\nMnnyZO3fv18LFixQXFycVq1apWHDhmnNmjWqV6+ew7bZ2dkaMmSIevfurdmzZ8tqtWrcuHEaOXKk\nli1bJkmaO3euVq5cqXnz5qlhw4bavHmzRo0apZiYGLVv317/+c9/NGnSJL3++uvq3LmzsrOzNWHC\nBD3++OPavHmzTCaT4WMAAABVS5mK0NKlSxUeHm7/2tksFovWrVunmTNnKiEhQZLUv39/vf/++1q+\nfLkmTJjgsP26detUWFioMWPGyM/v8iGMGTNGvXv31sGDB9W4cWMtX75cQ4YMUZMmTSRJXbt2VefO\nnbV06VK1b99eN9xwg15//XV16dJFkhQREaH7779fn3/+uc6cOaMbbrjB6ccJAAA8S5mKUJs2bUp8\nfeHCBVWvXt2+3Gw2209LGXXgwAEVFhaqWbNmDsuTkpK0Z8+eEtvv2bNHjRs3tpcgSUpMTFRgYKBS\nU1MVEhIis9mspk2bOryuadOmWrJkif3rX0tPT9d7772nDh06UIIAAPAShidLWywWjR07VlFRUZo6\ndap9+ciRIxUUFKTXXntNoaGhhvZpNpslyT7qVCwiIkLnzp0rdfsrty1+vdlsltlslslkUo0aNUrs\nLysry2HZF198oZEjR6qoqEjdunXTlClTDGWXJF9fw7djqjTFWcnsWmR2DzK7B5ndoyJZfXwkP7+q\nc6yexHARmjlzpn744QcNGDDAYfkTTzyhqVOnaubMmXrppZecla/crjaH6Up33nmn9u/fr/T0dE2d\nOlUPPfSQPvjgA/n7+5f5vcLCqv/2Rh6GzO5BZvcgs3uQ2XNFRAQrIiK4smNUSYaL0H//+1+lpKSo\ndevWDsvvuOMOBQcHa8yYMYaLUPEjO7KyshxOt2VmZiomJqbE9tHR0crIyCixPDs7W9HR0Q77+7XM\nzMyrnr6rW7eu/va3v6lt27bavHmz7rrrrjLnt1guqLCwqMzbVyZfXx+FhVUns4uR2T3I7B5kdo/i\nzOWRmZknf/88Jyeq+spSDg0XoeKyUZqaNWuWKB9lccstt8jf31+pqamKjY21L9+9e7c6dOhQYvvm\nzZvrtdde06VLl+wjN3v37pXValWLFi0UHx+v6OhopaamqlWrVvbX7dq1Sy1atJAkvfDCCyoqKnI4\nvWe1WiVJvr6+hvIXFhapoKBqfNCKkdk9yOweZHYPMnuuoiJ5xXG6guETig0bNtTq1atLXbdo0SLd\ndNNNhkOEhISob9++mjNnjtLS0pSfn6+FCxfq+PHjGjBggE6fPq1u3brpu+++kyT16NFDgYGBSklJ\nUW5urjIyMjRjxgzddtttuvnmmyVJycnJWrJkifbv3y+r1ar169dr69atSk5OliR16NBB69at07//\n/W9ZrVaZzWZNnTpVMTExDuUJAABcvwyPCA0fPlx//vOf9fnnn6tZs2YKDg6WxWLRt99+q1OnTmnu\n3LnlCjJ+/HjNmDFDAwcOVG5urho1aqSFCxcqPj5eJ06c0NGjR3X+/HlJl4vT4sWLNWXKFN1+++0K\nDAxUly5dHJ5zNmTIEFmtVo0YMUJms1k33nijZs2aZR8R6t69u4qKijRv3jyNHz9eoaGhatq0qRYv\nXmx4sjcAAKiaTLayzir+lV27dumdd97Rvn37ZLFYFBUVpZtvvlmPPfaY2rZt64qcHi0zM6/KDEn6\n+fkoIiKYzC5GZvcgs3uQ2T2KM5fHoUNpV5224s1iYn57YKNczxpr2bKlWrZsWZ6XAgAAeIxy3XTg\n448/1nvvveewbOrUqdqwYYNTQgEAALiD4SK0atUqPf300zp16pTD8oKCAj3zzDNXnUgNAADgaQyf\nGlu8eLGee+45+9VXxSZOnKh69epp4cKF6t27t7PyAQAAuIzhEaH09HR17dq11HWdO3fW8ePHKxwK\nAADAHQwXoaioKP3f//1fqesOHjyoiIiICocCAABwB8Onxnr16qUJEybo+PHjSkpKst9HaMeOHVq0\naJEGDhzoipwAAABOV64bKprNZs2YMUNFRZfvzWCz2eTj46PevXtr5MiRTg8JAADgCoaLkL+/v15+\n+WWNGjVKBw4cUE5OjiIjI3XTTTdxMycAAFCllOuGitLluUKdOnVyWPbLL7/oqaee0vLlyyscDAAA\nwNXKVYQOHDigrVu3Ojxp3maz6dChQzp06JDTwgEAALiS4SK0ceNGPf300yoqKpLJZNKvH1VWu3Zt\njRo1yqkBAQAAXMXw5fPz58/X448/rj179qhatWr69NNPtWXLFo0ePVqNGzfWgw8+6IqcAAAATme4\nCKWlpemBBx5QYGCgfUQoJiZGTzzxhFq2bKlXXnnFFTkBAACcrlwPXS0WHByss2fP2r+/++679fnn\nn1c4FAAAgDsYLkKNGjXSO++8owsXLqh+/foOT6FPTU11ajgAAABXMjxZeujQoRo2bJgGDRqkP/3p\nTxo1apR27typsLAw/fTTT+rVq5crcgIAADid4SLUqVMnffzxx6pZs6YSEhI0e/ZsrVu3TlarVT17\n9tQjjzziipwAAABOV677CMXHx9u/vvvuu3X33Xc7LRC8V9opizZsS5clz6qw4AB1b1dXCbFhlR0L\nAHAdK/edpQFnSjtl0ZyV+2TOufi/ZSctGt4nSTfF16jEZACA61mFrhoDnGXDtnSHEiRJ5pyL2rAt\nvZISAQC8AUUIHsGSZzW0HAAAZ6AIwSOEBQcYWg4AgDMYLkKTJk3iwapwuu7t6ioyNNBhWWRooLq3\nq1tJiQAA3qBcD1394IMP1KhRI/Xv31/33nuvQkJCXJENXiQhNkzD+yTp423pyuaqMQCAmxguQlu3\nbtXWrVv18ccfKyUlRdOmTVO3bt304IMP6tZbb3VFRniJhNgwDbs/qbJjAAC8iOEi5Ovrq06dOqlT\np06aPHmyvvzyS23YsEHJycmKj49X//791bdvX1WvXt0VeYFKtXXfSX20+SflWwtVLcBXfe+opw5J\ntSs7FgCgnCo0WTogIEB33XWXnn76aQ0cOFBpaWmaMmWKOnfurGXLljkrI+ARtu47qUUbDisr16p8\na6Gycq1atOGwtu47WdnRAADlVO4bKp4/f14bNmzQypUrtXv3btWtW1djxoxR9+7dtWnTJqWkpCgn\nJ0dDhw51Zl6g0ny0+SfZbI7LbLbLyxkVAoCqyXAR+vbbb7Vy5Up98sknunTpkrp06aJFixapffv2\n9m0efvhhRUREaPr06RQhXDfyrYWGlgMAPJ/hIvTwww+rVq1aGjJkiPr166cbbrih1O1at24ts9lc\n4YCAp6gW4Ftq6akW4FsJaQAAzmC4CL311lvq0qWLfHyuPb2oZs2a2r9/f7mDAZ6m7x31tGjDYYfT\nYybT5eUAgKqpTEXoyonP//znP6+5/YABA8qfCPBQxfOAVm7+SReucdVY2imLPt6ervMXCxUU6Ktu\nbbkfEgB4qjIVoVdeeaXMOzSZTBQhXLc6JNW+5sTotFMWzVm5z+EBsj+dsGh4nyTKEAB4oDIVoc8+\n+8zVOYDrwoZt6Q4lSJLMORe1YVu6hnOzSADwOGUqQnFxcfavFy9erH79+vFYDVQp7roRoiXPamg5\nAKByGb6h4uzZs3X27FlXZAFcwp03QgwLDjC0HABQuQwXocGDB+uNN95Qdna2K/IATnetGyE6W/d2\ndRUZGuiwLDI0UN3b1XX6ewEAKs7w5fP79+/Xjz/+qA4dOiguLk6hoaEO600mkz788EOnBQQqyp03\nQkyIDdPwPkn6ZPsx5V0sUHCgr+7hqjEA8FiGi1BeXp5q1aqlWrVquSIP4HTuvhFiQmyYRj7QVBER\nwcrMzFNBQZFL3gcAUHGGi9C777571XWFhYWcMoPH4UaIAICrqdDT56907NgxdevWzZm7BCqsQ1Jt\nDe6eqIiQAFUL8FWNkAAN7p7Ig1IBAMZHhGw2m5YuXaotW7YoKyvLYfmJEyfk68tzl+B5futGiAAA\n72R4RGju3LmaMWOGcnJydPDgQfn4+Cg3N1cHDhxQixYt9Oabb7oiJwAAgNMZLkJr1qzR9OnTtWLF\nCgUGBmrmzJn65JNPtGLFCmVkZCgyMtIVOQEAAJzOcBE6deqUmjdvfvnFPj4qKCiQJDVr1kzDhg3T\nyy+/7NyEAAAALmK4CAUFBdmvDKtRo4aOHz9uX5eYmKi9e/c6Lx0AAIALGZ4s3aZNG02aNEl///vf\n1bRpU82aNUvx8fEKDw/Xe++9V+IGiwAAVEVppyzasC1dljyrwoID1L0dN0e9HhkuQqNHj9aTTz6p\n8+fP6/HHH9fAgQPVvXt3+/oxY8Y4NSAAAO6WdsqiOSv3yZxz8X/LTlo0vE8SZeg6Y7gIJSQkaNOm\nTfbvN2zYoE8//VQFBQW69dZb7fOHAACoqjZsS3coQZJkzrmoDdvSNfz+pEpKBVcwXISuVKtWLT38\n8MPOyAI4BcPZACrKkmc1tBxVV7nuLH348GGNHTtW3bp1U8uWLXX8+HHl5+dr7ty5zs4HGFI8nL3r\nyBl9/3O2dh05ozkr9yntlKWyowGoQsKCAwwtR9VluAht375d/fr1U2pqqpo1ayar9XI7PnPmjBYt\nWqSlS5c6PSRQVtcazgaAsurerq4iQwMdlkWGBqp7u7qVlAiuYrgIvf766+rZs6c2btyoadOmyc/v\n8tm1+Ph4vfDCC1qxYoXTQwJlxXA2AGdIiA3T8D5JatUwRjf9LlwtG8YwUfo6ZXiO0JEjRzRlypRS\nnynWunVrvfjii04JBpQHw9kAnCUhNkzDmBh93TM8IhQcHGw/HXalzMxMBQYGlroOcAeGswEARhgu\nQo0bN9a0adOUm5vrsNxqtertt99Wy5YtnRYOMIrhbACAEYZPjY0cOVKPPPKIOnXqpKSkJF26dEnP\nP/+8jh49qgsXLmjZsmWuyAmUGcPZAICyMjwilJSUpNWrV6t3797KyclR7dq1lZ+fr3vuuUerV69W\nYmKiK3ICAAA4neERodOnT6tu3bpMigYAAFWe4RGhLl26aNiwYfrvf/8rm83mikwAAABuYbgIPfPM\nMzp37pyGDh2qzp07680339SJEydckQ0AAMClDBeh5ORkrVixQp999pkGDBigzz//XHfffbcee+wx\nbdq0SQUFBa7ICQAA4HTletaYJMXFxemJJ57Q6tWrtX79et10000aPXq0OnfurPnz5+v8+fPOzAkA\nAOB05S5Cxb755hvNmzdPH3zwgUJDQ3XnnXdq+fLl6tmzp9LS0pyREQAAwCUMXzUmST///LNWrVql\nVatW6eTJk2rRooUmT56sP/zhDwoICNDFixc1atQovfjii3r33XednRkAAMApDBehRx55RDt37lT1\n6tXVq1cv/elPf9LNN9/ssE1gYKCeeeYZ9e7d22lBAQAAnM1wEcrJydGkSZPUs2dPBQUFXXW7mjVr\navjw4RUKBwAA4EqGi9CqVavKtF1ISIiGDh1qOBAAAIC7VHiyNAAAQFVFEQIAAF6LIgQAALwWRQgA\nAHitMk2W3rx5s6Gd3nHHHeUKAwAA4E5lKkJPPvmkTCaT/WnzJpPJvs5mszl8L0mHDh0yHCQ/P1/T\npk3Tli1blJ2drQYNGmj48OHq2LFjqdsfOHBAKSkpOnjwoAIDA9WyZUtNnDhRkZGRkqSioiK98cYb\n2rhxo86cOaM6depo0KBB6tWrl30fO3bs0OzZs3XkyBH5+fnp1ltv1XPPPac6deoYzg8AAKqeMhWh\npUuX2r/OzMxUSkqKfv/736t58+YKDg5WTk6Ovv32W3355Zd66aWXyhVk8uTJ2r9/vxYsWKC4uDit\nWrVKw4YN05o1a1SvXj2HbbOzszVkyBD17t1bs2fPltVq1bhx4zRy5EgtW7ZMkjR37lytXLlS8+bN\nU8OGDbV582aNGjVKMTExat++vb7//ns9/vjjGjVqlBYtWqScnBw988wzGj58uNatW1euYwAAAFVL\nmYpQmzZt7F8//fTTeuKJJ9SvXz+HbX7/+9+rXr16ev/999W+fXtDISwWi9atW6eZM2cqISFBktS/\nf3+9//77Wr58uSZMmOCw/bp161RYWKgxY8bIz+/yIYwZM0a9e/fWwYMH1bhxYy1fvlxDhgxRkyZN\nJEldu3ZV586dtXTpUrVv316//PKL+vXrp8GDB0uSIiMjNXDgQA0bNkxnz55VdHS0oWMAgOtR2imL\nPt6ervMXCxUU6Ktm9aOU+sM5WfKsCgsOUPd2dZUQG1bZMYFyM3xDxS1btmj06NGlruvQoYNSUlIM\nhzhw4IAKCwvVrFkzh+VJSUnas2dPie337Nmjxo0b20uQJCUmJiowMFCpqakKCQmR2WxW06ZNHV7X\ntGlTLVmyxJ61Q4cODuvT09NVrVo11ahRw/AxAMD1Ju2URXNW7pM556J92a4jZ/T/Z0lc3uakRcP7\nJP1mGUo7ZdGGbekUKHiccj10df/+/YqPjy+xvDxzgyTJbDZLksLDwx2WR0RE6Ny5c6Vuf+W2xa83\nm80ym80ymUwlCk1ERISysrJKzXD48GG9+eabGjVqlEPBKgtf36pz8V1xVjK7Fpndg8yu9fH2dIcS\nJMmhBEmSOeeiPtmerpEPOP4h+2s/nbRo7qp9Omf5376OnrJo5APNVK+2a8pQVfo5F6tIVh8fyc+v\n6hyrJzFchLp06aIJEyZo//79aty4sYKCgnThwgXt3btX//rXv9SpUydX5DTMduWn9Rq++OILPfPM\nM3r44YeVnJxs+L3Cwqobfk1lI7N7kNk9yOwa5y8Wlmm7vIuFiogIvur6TWsPOJQgSTpnuahNO4/r\n+UfbXOVVzhEWVl3fH8/Uvz7/Xlk5F1UjNFAPdLlJN8VHuPR93S0iIvia/w1wdYaL0KRJk2QymfTu\nu+/KarX+b0d+furSpUu5JktHRUVJkrKyslS9+v9+OWRmZiomJqbE9tHR0crIyCixPDs7W9HR0Q77\n+7XMzEz7VWXFFi5cqDlz5mjSpEm67777DGeXJIvlggoLi8r1Wnfz9fVRWFj1Kpv5++NZWv91mn14\nvcdtCS77i7IiqvrPmcyuU5UyBwX6lmm74EBfZWbmXXX92czzV1l+4Zqvq4jin3Pq4QzNWpHqUMSO\nHDW7dDSqvIozl0dmZp78/V3zs6zKylIODReh4OBg/e1vf9PkyZN19OhR5ebmKigoSHXr1lVwcPna\n6C233CJ/f3+lpqYqNjbWvnz37t0l5vFIUvPmzfXaa6/p0qVL8vf3lyTt3btXVqtVLVq0UHx8vKKj\no5WamqpWrVrZX7dr1y41b97c/v2CBQu0ePFivffee2rcuHG5sktSYWGRCgo8+xfalapi5u+PZ+nN\nD/c4DNX/dKJs8xMqS1X8OZPZPapC5m5t6+qnExaHz5zJ5Hh6LDI0UPe0rXvNYwkNCrjKcn+X/wzW\nbPmp1NGodVvTNPz+JJe+tzsVFcnj/z15qnKfUKxWrZqio6NVvXp1NWjQoNwlSLr8pPq+fftqzpw5\nSktLU35+vhYuXKjjx49rwIABOn36tLp166bvvvtOktSjRw8FBgYqJSVFubm5ysjI0IwZM3Tbbbfp\n5ptvliQlJydryZIl2r9/v6xWq9avX6+tW7dq0KBBki5P0H7zzTc1f/78CpUguM/6r9NKzFcw51zU\nhm3plZQIuL4lxIZpeJ8ktU68QY0TItU6MUaDuyeqVcMY3fS7cLVsGFOmP0S6t6uryNBAh2WRoYHq\n3q6uK+NLkix5VkPL4X3KNVn6n//8p+bPn6/Tp0/LZDJp48aNCg4O1ujRo/XWW28pJCTE8D7Hjx+v\nGTNmaODAgcrNzVWjRo20cOFCxcfH68SJEzp69KjOn788vBoSEqLFixdrypQpuv322xUYGKguXbpo\n/Pjx9v0NGTJEVqtVI0aMkNls1o033qhZs2apRYsW9mMoKCjQgAED7K8pvjnkK6+84nDjRXgGfqEB\n7pcQG6aRDzRVRESwMjPzVFBQpA5JtQ3vY3ifJH28LV3Zbr5qLCy49NGoqy2H9zHZjMwqlrR8+XK9\n+uqruvfee9W2bVu9/PLLWr9+vapXr65HH31UHTp0cCgk3qD4l0NV4Ofn4/ALrSoozjz5na+18/CZ\nEutbNozxuCHuqvxzJrNrGc3sCZedV+Wf864Dp0qcUo8MDfTIU+rFmcvj0KE07n9XipiY0N/cxvCI\n0HvvvadnnnnGfnXVlClTJF2ewDx+/Hg9//zzXleE8Nuc8cu8x20JJeYruGt4HagMpd3Hp6z37Snr\n/iu7ZLlavdqVNxqFqsFwEfr555/VpUuXUtclJCTY7wkEFHPWL3N+ocHbbNhW8j4+xfPiKjoK6uqS\n5UkSYsM0zMNGjeE5DBehG264QUePHi31waRpaWnclRklOPOXOb/Q4E1cOS/OlSULqEoMXzXWqlUr\nvfTSS/rmm29UVPS/c8WHDh3StGnT1LlzZ2fmw3WASc5A+bhyoi+fS+Ayw0XomWeeUXBwsAYPHqym\nTZvqwoULuvfee9WnTx/5+vpq7NixrsiJKoyrNoDyceVl53wugcsMnxqLiorS6tWr9emnn2rPnj3K\nyclRWFiYmjdvrjvvvNPwc7pw/everq7STjLJGd7FGRORXXnZOZ9L4DLDl8+jpKp4Sam7M6edspT7\nl3lVvnSXzK7lqZlLm4hcfMn2TfE1PCZzWT+XXPLvHlw+73wuuXz+kUce0b333qv+/fuXWHf27Fl1\n7Nix3E+hx/WLSc7wJteaiDwq3nMuKHHF59KbrkbD9cHwHKEdO3Zo8uTJmjBhgsNDV4sxwATA23nz\nRORrlUDAE5XrWWMpKSn6/PPPNXDgQJ0+fdphnclkckowAKiqvHkisjeXQFRN5SpCbdu21UcffaSC\nggL16dNHO3fudHYuAKiyKvMho5XNm0sgqqZyP30+NjZW77//vjp06KDk5GQtXbrUmbkAoMoqvtrL\n6FParwfeXAJRNRmeLP3rU18BAQGaPn26mjRpounTp+urr75yajgAqKq89QKBynzSPFAehotQaZOh\nH330USUmJuqpp55ySigAQNXlrSUQVZPhInT48OFSl7dt21arVq3SN998U+FQAICy84T79gBVVZmK\n0A8//KB69erJx8dHP/zwwzW3TUrirwAAcBfu2wNUTJmKUM+ePfXVV18pKipKPXr0uOol8jabTSaT\niRsqAoCb8BR5oGLKVIT++te/KjQ01P419woC4M086VQU9+0BKqZMRej++++3f92nTx+XhQEAT3e1\nU1H3d0pQ6g/n3F6OuG8PUDFlKkLLli0r8w5NJpMeeuihcgcCAE92tVNR7278P1l/9XBPd83T4Sny\nQMWUqQi98sorZd4hRQjA9exqp5ysVzzh3F3zdLhvD1AxZSpCn332matzAOXmSfM1cP2zXios87ZH\n0jM19b1dLv93WZb79vA5AUpXpiIUFxdXpp2ZzWb99a9/VUpKSoVCAWXFpcNwt3OWi7+90f+Xm1+g\n73/OlvS/f5c3xddwVbSr4nNy/TObzykyMlI+PuV+cpbXMnxDRUk6ffq0du3apczMTIfle/fu1aef\nfuqUYKg6KvMvTS4dhrsVFBb99kalKP53OaoSihCfk+vf5l0/KjIyStHR0ZUdpcoxXIS2b9+uoUOH\n6sKFCzKZTPZHbphMJvn6+mrAgAFODwnPVdl/aXLpMNytWoCv8q0lT4+FVvdXwzo1lJ1n1fHTOcq/\nVLIw/WI+7/C9u/6I4HNy/aseFFzZEaosw0Vo1qxZ6tatmx577DE98MADevvtt+Xn56fVq1fLZrPp\n2WefdUVOeKjK/kuTS4fhbn3vqKdFGw7r149dNJmkB7vUV4ek2pKk0W99pfxLJUtGzoVL9q9/Oum+\nPyL4nABXZ/hk4vfff68nnnhC9evXlyTVqlVLzZs31+TJkxUeHs78IC9T2X9pdm9XV5GhgQ7LuHQY\nrpr2ii0AACAASURBVNQhqbYGd09UREiAqgX4qkZIgAZ3T7SXIEkKDSq9YPx6+fqv0676R4Sz8TkB\nrs7wiJDValVAwOUPc1BQkLKyshQfHy9J6tevn/70pz9p3Lhxzk0Jj1XZf2kmxIbp/k4JWrn5J12w\nFqpagK/u75TABFC4VIek2g7F50o3RFTX8V9yS11ezJ1/RHCJPXB1hkeEGjRooI8++kiFhYWqU6eO\nVq9ebV93/PhxXbxY9isqUPVV9l+aaacsWvVlmjJzrcq3Fior16pVX6Yp7ZTFLe8PlKYsnwt3/xFR\nfIn98wNbavj9XC0GFDM8IpScnKxx48ape/fu6tOnjyZOnKj9+/erRo0a2rFjh9q3b++KnPBQlf2X\nZmXPUQJKU5bPRY/bEvTTifLdEZp7AuFKF87nVXaEKstwEerVq5dq166tuLg41a9fX3l5eVq7dq1O\nnjypnj176qmnnnJFTniwstzMzVUqe44ScDW/9bmoV7v0siRJc1btu2rJqewrNeGZ7mhZX5GRkZUd\no0oq132EWrVqZf86OTlZycnJzsoDGFLZc5TgHVw1AnNlWSpLySnvKCijSNe3yMgobqZYTuUqQhcv\nXlR6erosFov9PkK/1rp16woHA8qCB07CiPKUAXeOwJSl5JRnFJRRJODqDBehTZs2acKECcrJyZEk\nhxsq2mw2mUwmHTp0yLkpgauo7DlKqDrKWwbcOQ+tLCWnPKOgzKUDrs5wEUpJSVGjRo00YMAAhYeH\ny2QyuSIXUGaVOUcJVcfVysD/a+/Ow5sq0/6Bf9Om6V66AbIUKGsLdANEhCrrO4OICILKiICI46gj\ngiAOoo6KOjCCAvObAir7CDqCgEKBVxEHEbS8FAotFVlalkILpWmbplua9Pz+YBqa5KQ9SdPkpPl+\nrotLenLOyZ1joTfPc9/Ps+2HCwj097E6SuTMOjQpSY49o6CspSOyzuZEqLCwEJ988gm6dOnSDOEQ\nETUPaz/0L1wrhd5wZ4r//JVitG8dDENtLUICVfD2Ev/HXnPUoUlJcuwZBWUtHZF1NidCvXv3RkFB\nARMhInIr1n7o10+CAEBTqYfmyp0NpUMCVAjxV0JTqTceExuBcUQxstQkx9ZRUNbSEVlncyL01ltv\n4d1338WtW7cQGxsLf39/i3Pat7e+4ioRUWOao8NJLBlQeimgr7Vs+KhPU6FDbOcw9PRTWk1OHFmM\n3BxTvaylI7LO5kTo1q1buHr1KubPn2/1HBZLE5G9mqvDSSwZKK/U42y90R9r9IbaBpMTdyhGZi0d\nkTibE6FFixahVatWmD59OoulicjhmjOpkLJuj5jGamlYjEzkvmxOhPLz87F9+3b06NGjOeIhIg/n\nys1Ivb0UuH6rApqKO+8lpZaGxchE7svmRKhnz54oL+eeJkTUPFy1GWmd3HyNzbU0LEYmcl82J0Lv\nvPMOli9fjunTp6N3796ixdJix4iIpHB1UmFPLQ2LkYncl82J0JQpU2AwGPDTTz9ZPYfF0kRkL7Gk\nIrF7hEO6yJpzvy0WIxO5J5sToaeffpoF0kTUrOonFQ11kfWICpV8T+63RURi7BoRCg0N5S63ROQU\nDW2NERTgg4pqAwJ8vfHAPQ2P7rhDizsROZ/NidCwYcPw7bff4q677mqOeMgNNed0A5HUrTFyrjU8\nusMWdyISY/OwTt++fXH48OHmiIXcUN10Q/pvhTifV4r03wqRsiMTufkaV4dGLYTUrTHqRndsvQ9b\n3Ik8m80jQuPHj8f69evx448/ok+fPggODrY4Z8qUKQ4JjuSP0w3U3GzZGqOh0R1r3WiJ3SOQsjOT\nI5pEHsrmROivf/0rACA3NxffffedxesKhYKJkAfhdAM1N7EuslslVbh8o8ziXGs7xdfdZ8L90dhx\nKAeVOgP8VN5Ijr8LO3/MZQE1kQezORH6/vvvmyMOclOcbiBnMG9N/2DrSdHzGupozc3XYOePuSjW\n3k7Sq3QG7E+7Cp2+1uQ8jmgSeRabE6EOHTqYfF1TUwMfHx+HBUTuxdWL35FnMtTWih7XG8SPA+LT\nuOZJUB3zEU02BJDcqdVFCA8PZ0e3HWxOhAAgNTUV//rXv5CdnY2amhr4+voiLi4Of/rTn5CcnOzo\nGEnGuKIuNZVYkgGgwcTDnpFIW6Zr69+H6w+ROziUfhHh4RGIjIx0dShux+ZEaNeuXViwYAESExMx\nefJkBAYGoqysDCdOnMCzzz6L1atXY+jQoc0RK8kUV9Qle+Xma7DyywxoKvXGY79eKobSCybHzBMP\ne0YibZmu7RAZYPw9GwLIHfgHBLo6BLdlcyK0ceNGzJw5E/Pnz7d4bdGiRVi1ahUTISKSZNsPF0wS\nHgCoqNZbnGeeeNSNRO5Pu4Lyaj0Cfb0xupEFFcWSJwUAy94z4MdT+Rh/XzcAbAggaulsToRycnKw\nfPly0deeeOIJTJo0qclBEZFnyLuplXyueeIR3S4EsybFIywsEMXF5dBbqfepf775NG5mThF0NZbX\nVekMxt+zIYCoZbM5EVIqlaiqqhJ9Ta/Xcx8yIpJMbDTGGkckHubTuHP/+RN0NZYjO34qb+Pv2RBA\n1LLZnAjFx8cjJSUFy5cvN+kWq66uxj/+8Q8kJCQ4NEAicl+NdVtFtQnG2SvFFteZL5hob+LR2PtP\nHNoV6/eehVAvI1Mobh+vw4YAopbN5kRo9uzZeOqpp5CcnIw+ffogKCgIZWVlOHPmDGpqarBp06bm\niJOI3IyUbqtHh3fDym2noam4MyoTEqDCo8O74tSFImPikdg9wub2dSnvPySuPQCYLLI4cWhX4/E6\nbAggarlsToSSkpKwY8cObNq0CVlZWcjLy0NwcDDGjh2LqVOnIjo6ujniJCI3I6XbKrpdCGY/Gi86\n2lKXjDSU0PSICm3S+wO3kyHzxIfI3RQV3UKtlfW1qGF2rSPUrVs3LFq0yNGxEFELIrXbqrHRloYS\nmtkNJELs9iJPIhhqXB2C27IrEdLpdLhy5QpKSkogCJbljnfffXeTAyMi9+aobit7Exp2e5EniWzT\njqtK28nmROjIkSOYP38+iostCxwFQYBCocCvv/7qkOCISD6kbDNR/5wavcHiHgoFkNg9wqb3tTeh\nYbcXEUlhcyL097//He3atcO8efMQHh7OdnnyKJ6655RYnc75K8Vo3zoYhtpaY0Gz+U7u5gQByLhQ\nZFNNjr0JDbu9iEgKmxOhK1eu4N///jd69erVHPEQyZYn7zklVqejqdRDU6/1/dT5QjSypuHt62ys\n0WlKQsNuLyJqjM2JUNu2bTkPSR7Jk/eckpK8SEmCAPtqdJjQEFFzsTkRevrpp7F69WosWbIEKhWL\nDslztJQuJHum9xxVYCw2peWp041EjsT2efvZnAidO3cO2dnZGDFiBGJjYxEQEGDyukKhwIoVKxwW\nIJFctIQuJKlr8pgnJ4ndIyzqdMSYb2IaEqBC+8gAGGoF0STHk6cbiRyJ7fP2szkR+uGHHwAAKpUK\nFy9etHidxdPUUrWELiRr03vbfriAoAAfVFQbINQKuHazzGRX+NzrGky4P9q42nN+UTm0lZa7xHe+\nKxiRrfwk1/J48nQjkSOxfd5+NidCBw8ebI44iGSvJXQh3VRXiB4/n1cKQ631LVDVZdXIuFBkTE5y\n8zWiW2NM/X0vm56HtXisHScicjS7FlQk8lRyL9ptrN6mrFJ8+LyhJKhO/VqohrbGcEQ81o4TETka\nEyGiFkJKvU1TZq69vUwvbiwplBJPcIAKJVrLYvPgAPepuyIi98ZEiNyeJ3cd1f/shSWVFkmFeb1N\nWYX9Iy221v9Jqf9pE+aPqze1Fte2CfO3O04iIlswESK35sldR2KfXcxNdQVSdmZCU66TNAVmjd5g\nW2uulOUGWkIBOpEcFBXdglpdhPDwcBZN20g2T6uqqgpvv/02Ro4ciQEDBmDy5Mk4fPiw1fPPnDmD\nGTNm4J577sH999+Pl19+GWq12vh6bW0tli9fjtGjR6N///6YMGECvvnmG4v77N+/H4MHD8bkyZOb\n5XNR82po1KGlE/vsYm4UVyL9t0KczyuFyB7JkplPjTVGynID0e1CMOH+aIQFqeCn8kZokAoT7o9u\n8UkskaP5+/nh2NkbJj8HSRrZJELvvPMO0tPTsXbtWhw9ehQTJkzACy+8gJycHItzS0tL8cwzzyAm\nJgbff/89du3aBa1Wi1mzZhnPWbVqFXbs2IEPP/wQaWlpePHFF7Fw4UL8/PPPxnNeeeUVrF69Gl27\ndnXKZyTHaymLHNpDymf08faCTuqSz42wdWqsQ2RAo8dz8zXY+WMuirU6VOkMKNHqsPPHXOTma5oU\nK5Gnuat9J4SEhLk6DLcki0RIo9Fg9+7dePHFFxEdHQ2VSoXHH38c3bt3x9atWy3O3717NwwGA+bN\nm4egoCCEh4dj3rx5SE9PR3Z2NgBg69atmDFjBvr06QOlUomRI0di2LBh2Lx5s/E+nTp1wvbt2xEV\nFeW0z0qO1RIWObSXtc8YFqRCj46t0L9Xa9wVIZ6MKL0UxhGYAF9vSe9n69TYj6fyGz3uySN6RCQP\nskiEzpw5A4PBgISEBJPjcXFxOHXqlMX5p06dQu/evaFU3ilxiomJga+vLzIyMnDlyhWo1WrEx8eb\nXBcfH29yv5deegk+Pj4O/jTkTGMGdUZ4sK/JMU+pMUnsHmHRBaZQAI8M7YrXnuyPP0+Is1p0HBTg\ng6g2QejWoRWCJHZo2ZpcVukMjR735BE9IpIHWRRL181ptmrVyuR4WFgYioqKRM83P7fuerVaDbVa\nDYVCgdDQUJPXw8LCUFJS4sDIb/P2lkU+KUldrFJjzrmuwZ6jucaOrLGDo9G1vXPrNxqKuUdUKF56\nNAGpRy+htFyHVoE+eNAFMZqz9Tnb49TFIouaH0EATl8swtCkjgCAh4ZE4/zVYmgqTFeBLtHqjB1m\nSu/Gp7wiQnzx0JBoKJXSP4+/yls0GfL39Tbep1WQeHIVGqSS9F7OeM6Oxpidw51jtodCoYBCASiV\nCpv+nJJMEqHmIDSlKtRGISHu1+orJebzV4uRsiMThSWVxmOXC7R47am70SPK+XPR1mLuHxaI/n3a\nOTkaaRz5vXH+ajG2HzyPkrJqhAb7oqhUvFD67JUSLNlyAqHBvrin911orFFMbxA/ISRQhY5tghAa\n7ItJI3pI+n9eP8aIUH8Um7XzeymApx7sg7CwQADAE6NjcblAa/I91jrUH38YHWs8R4qW+mdQbhiz\nfAUE+ABQIDQ00KY/OySTRCgiIgIAUFJSAn//O9+0xcXFaN26tcX5kZGRKCgosDheWlqKyMhIk/vV\nV1xcjPDwcEeGDgDQaCphsLF+wlW8vb0QEuIvKeat+381+QEFAIUllfh8/6+YNSnBylWOZ0vMcuHo\nmHOua/D/tp9CkeZO8mPtH31lFTXIzr09ynosKx/21krX1tZCrzdAp9NDo6lCcXG5zTEG+SuhAFBj\nEODv641Jw7sjqXuE8V6RQSr8+ZE4ixG9yCBVo+8H8HvDWRizc9TFbI+KihooFEBJSTmUSvHaQE8k\nJSmURSLUt29f+Pj4ICMjA+3a3fmX/cmTJzFkyBCL85OSkvDRRx+hpqbGWONz+vRp6HQ69OvXD1FR\nUYiMjERGRgYGDBhgvC49PR1JSUkOj99gqIXeQZ05ziIl5lKRFX+B29Mqrvi8LfU5A40vCrn7SK5J\nggFAUoLTlMelrdTj3NVSAEDONcu1mcxjLq+ssYhRW6lH/16tTTZQNX8endoE4fnxfU3jtjHwlvy9\nISeMWb5u3LgOANDFtPGIz+tIsphIDAoKwsSJE5GSkoLc3FxUVVVh3bp1uHr1KqZMmYIbN27ggQce\nwIkTJwAAY8eOha+vL5YtWwatVouCggIsXboUgwcPRs+ePQEATz31FDZu3IisrCzodDrs2bMHR44c\nwYwZMyze35nTaO7EkzuynKluYcS6tX7SfytEyo5MkxZyVxcPm3dyicV84Vqp6LWujp3IE9Tqa1Cr\n5x599pDFiBAALFy4EEuXLsWTTz4JrVaL2NhYrFu3DlFRUbh27RouXbqEiorbO1IHBQVhw4YNeO+9\n95CcnAxfX1+MGDECCxcuNN7vmWeegU6nw4svvgi1Wo0uXbpgxYoV6NevHwDg+PHjePrpp6FQKKDX\n6yEIAuLj46FQKPDuu+9i3LhxLnkOcsJVf53DWgv55v1nUVp+e30dW1vXm0P9hEYs5oZqjYioed3V\nvhMAcFVpOygEDoc0WXFxudsMRSqVXggLC5Qcc26+ptEdxpubrTHLgS0xL/4sHefzxEdTGhKg8kKF\nruF7B6i8oPDyQnmVvsHzpKg/xWUtZqWXAvp61dnhwb4O2e5EbOoQAPalXUZFtQEBvt544B732GOu\npX8/y4U7x2yPdduOAgAG922HyMhIR4bl1lq3Dm70HNmMCJE8NbbDODWdLSMmCgC+Km/4/fdXhbqy\nwfOVSiUeHd4Vpy4UobRch7ybZagUSZ68FDDpLlMoYNKabz4SaC3mHlGhCPRTOjRxFttT7fzVUkCo\nhabyToInVsdERNQYJkLUZEcyr+OrQzmo0hngp/LGxKFdMSSuvavDchuJ3SNw4lyhpH3ABNxekLBK\nZ4CUDS80FTocySxAoP/tpoLqGvF/GSu9FUjo3hrl1XoE+nojvluEMXkSS2isTZtOGtbN4YmI2DSc\npsKy7sh8Z3siIimYCFGTHMm8jvV7zxp/iFfpDFi/9ywAMBmSKOOC5cKIUki95MK1Uqv1O3UMtcCs\nSfEmUwkN/f+LbheCPz8S55RpU1uKrVmYTUS2YiJETfLVoRzR1Y2/OpTjtESosdZzOaofc/6txtfL\naYrGkiDg9s7y/9h+qsF6G7Hn7IxpU1umDlmYTZ7q5s3be/jV1rZ1cSTuh4kQNYmU/aSak1j9SO51\nedeKiMXsarWCgONnC41fm9fbuPI5i03DhQSoLGqE2NFInoyt8/ZjIkRN4mdlPyk/lbQdzRuTm69p\nsDOood3L5VorIhazOQWkTX35+nghrmsESst1yLmugaGx/TREKL0UFqNG5s/Qlc/Z2jQcAOxPu2Ks\naxrtJl1jRM2B7fP2YyJETTJxaFeTGiHgdsfRxKFdm3xvsVEI85EKd9y93Fps3l4K+Ci94KfyNm6I\n2hgvReMJk3lSFRKgQvvIABhqBYQEqnCzuBJXb2objNPVz9la96J5XRMRka2YCFGDGqu/qasD2nEo\nB5UO7hqTMgrhjqtfW4vNUCvA8N+OMKlqDALSfyts8JzOdwUjspWf1aLmlJ2ZoolQ/Tjd8TkTEUnB\nRIiskloXMiSufbMURksZhXDH1a9taZdvTGOF0CEBKkz9fa8Gp4ykPEN3fM5ERFIwESKrXF1/I2UU\nwplt3I5ib7u8VEH+SrSLCJT8LOqeYUP1Nu74nImIpGAiRFa5ui5E6iiEu61+3dzPT+lte7FkdLuQ\nRutt3O05ExFJwUSIrHJ1XYiUkQp35O0lZU1oifdSAOazYyVanbHYWu5LCRCRY3AdIfsxESKr5FAX\nImWkQu52Hb6Ivb9cgaFWgLeXAgG+jllaIMRfCYOABjdUlftSAkTkGFxHyH5MhMgq1oXYp/7aR8Wa\nKuQXVRhf0xsEaCrs3wk+LEiFyFB/hASqUF6px9krxY1eI+elBIjIMbiOkP2YCFGDWBdiG0euGu2r\nBKrr5Uzhwb4m01yLP0uXdB+2uBMRWcdEiMgKe/Ywk7JqtFQqHx/EdQu1OhonJcFhizsRUcOYCJHN\n3HGTU1vZu7fWTXWF1ddsVVWjN1kR+votrclzT+weIboHV/1Vo1vi/xsiIkdiIkQ2ccdNTu1h7xpK\nxVrHbaSqN1s12nwRxtzrGky4PxqnLhSxhouIyE5MhMgmrl5k0Vmsjew0NuKjq5G+PUZjzBddNP9a\nXVaNjAtFLeq5E5F96trn1WoVwsPDWTRtAyZCZBNXL7LYnOpP+V0vEk94GhvxqdE3bcloP5U3/FTe\n8PXxxo3iykbPbwnPnYiarlZfg8CgYBw7ewPh4RGIjIx0dUhug4kQ2cTViyw2F6ndXlXVeqTszDSp\n08m4UGT8uqk7Z6yaOxTA7Y1QpSRC7v7cicgx7mrfCSGh4dCUqF0ditthIkQmGiuElsMii81BareX\nvhYN1u04ithzVihMp8dawnMnInI1JkJkJKUQ2l0XWWwswbN3iqm5Nk8Ve86J3SNYGE1E5GBMhMhI\naiG0+SKLRzKv4/99dRpVOgP8VN6YOLQrhsS1d1rcjZGS4MlhiikixNfka7HFLOX0XImIWgKWlZOR\nPYXQRzKvY/3esyjR6lClM6BEq8P6vWdxJPN6c4Vps4YSvDpjBnVGeLCv+aXNRmG276qvjxdX8CYi\ncgEmQmRkTyH0V4dyRNu8vzqU48jQmkRKghfdLgQT7o9GWJAKfir7N0UND/bFuCGdERbsC39fJfxV\nln/EFAAeGtwZA3q1Ro+OrdC/V2u8+kQ/TnMRkd1u3szH9WtXcfNmPtTqItTWut8G1a7CqTEysqcQ\nurxKfMdja8ddwVoiV1hSicWfpRvrb7YdvABNpf0boioA9IkOQ0L31rheVIGKagPyC7Wo1JkmYgKA\na7cquP4PETlMrb4GgkEHfz8/ttDbiIkQGdlTCG2tWLi5iojtkdg9wqTTq06JVocS7e0k5dSFW9Ab\nmha0AODw6QL8nFUAfSP/GOP6P0TkSHXt8wDYQm8jJkJkwtbd5iNC/ETXu4kI8XNkWE1yJLOg0XOa\nmgSZ3EvCiLQri7M9Ya84IiKpmAhRk3RsEySaCHVsE9Totc76gZx3U+vwezaFK9f/aaiDrkdUqEti\nIiJyJSZCZOJI5nV8dShHciu8vQssNufmreYJVmW1/XU/jqL0UiC6fYjLR2Aa6qCbzUSIiDwQEyEy\nqmuFr6vvqdIZsC71LL5Pz4OP0lv0h7i9Cyw6cvPW+omPt5cXrheWNanouTn0iArF/D8kuTqMFr1X\nHBGRPZgIkZFYKzwAXCq4M7UkNmpja10R4LgfyFL3CHMmLwVQW+85hgSoMGlYN9cFVE9L3SuOiMhe\nTITIqEpnaPQcdVk1Nu8/i9JyXZNWknbUD2Spe4Q5U4fWQbgrPADl1XoE+npj9D3yKUZuqXvFEXm6\nmzfzoS0vBwBUlGugVqsQHh4OLy8uF9gYJkJk5KfylpQMXb5xZ4SoSmfA+r1nAdi2/YOjfiDLcUqn\nTZg/Zk2KR1hYIIqLy6GX0kbmJO66VxwRNaxuHSEAXEvIRkyEyGji0K4mNUJS1a0kbUsi5KgfyHKb\n0nGH0RV7pjKJSN7qryMEcC0hWzARIqO6RGbHoRxU6gxQeisgCEB5VeOFx1JGksw54gdyYvcInDhX\n6LQFHBW4vXBinZAAFdpHBsBQK3B0hYjIDTERIhND4tqbjOzk5mtMRm1+u1ICbaXl9hlN2Z+rKTIu\nFDl1FeuwYBW6tm/FaSUiohaCiRA1yHzUxrzFHri9k/rEoV2dEo/5GkE31RVOed86tQI4rURE1IIw\nESKbmE+f2ds1JlVuvgb70i6jotoAoVbA+avFkrawaC7BAfKqSSIioqZhIkQ2M58+E+OI7TOcvUaQ\nt5cCST0iUVquQ2FJpXFD1vrahPk7JRYiIlvUb58H7rTQ18d2enFMhMjh7N0+wzx5Kq+sceoaQZGt\n/IzTXmKfwR06wojIM9Vvnwdut9Bn51XB63o+AECrLcXvBsWwnV4EEyFyOHu2zxBLPBTNGqWl+hvF\ncr0dInIn5u3zJB0TIXI4e7bPEEuenNgMJjraw/V2iIhaPiZC1KANe7Px0+kCCLg9QpMcfxdmjOlt\nco75lJauRnxNIW8v62M8zu7+qtOjYyuroz2OqHMiIiJ5YyJEVm3Ym43DpwuMXwuA8eu6ZEhsSkvp\nLZ7wKBTWE6FirfP3C/P2UuC1J/uLvmZvnRMREbkXlo+Tidx8DVJ2ZmLxZ+kmSVB9P9U7LjalpTeI\nT2rpDdb73rWVja9e7WhR9WqCzDVU50RERC0HR4TISGq7ev00x5ZNT125L5ivjxeqa+4kYoF+Skz9\nfS+r59tT50RERO6HiRAZiY2CiKk/wWUtufHx9kJNvREgsWLk+vVHzcFP5W1c8LF9ZJBNHWDWPpeU\nZK7+IpABvt544B7WFhFR8zJfR8icYNADaOe8gNwIEyEP9o/tp5Bxocj4tb9KWsN6cvxdxt+PGdQZ\nudc1FuvtJMffhcOn8o2rT0+4P9okGTCvP3I0BYBVc4eaHLOlA8za52psHSGxUbWca81XW8SCbiIC\nLNcRqq9cW4ZhA7ojPJzt9WKYCHmo5V+eNEmCAKBS1/DYjFjXmNh6O4ndI7Dzx1wU/3dl5iqdAdsO\nXsCRzBsw1NYiJFCF9N8KHf6Z6vNqoENNCnvXEbJnDSV7saCbiOo0tI6QpkSN8PAIriptBRMhD3Xy\nXFHjJ+H2KIiUH6z1U6gjmQUWyYCmUg/NlWJbw7RbQ4XQUtmzjpAza4ucmXQREbVUTITIwoBerSWP\ngtjSPu8sjRVCN6em1BbZigXdRERNx0SILNgyCmJL+3xzsiV5a0721hbZw5lJFxFRS8VEyEO1DfPF\njWLLDrE2ob423UcOow/3iax27Sp1tUX7066gvFqPQF9vjG6mrjFnJl1ERC0VEyEPpdaIJzA3S6rx\n9JKDxq/DgpT48MX7rd7HlaMP1rb8EHMk8zq+OpSDqv92sU0c2hVD4tqbnOPIDizhv1VTzTk2xo1h\niahOQ+3zFeUaqNWmf1eHh4ezePq/mAh5KH2ttB/RxVo9/rT0B3RpFyL6g3bMoM44f6UYGieuDB3b\nOQzz/5Bk9XXzhKZDZAB2H70M4b8fuUpnwPq9ZwHAmAw5qgMrN1+DlV9mmDyPc5eLMfuxxGZJULgx\nLBEBDbfP+/v5ITuvCl7X8wEAWm0pfjcoBpGRkc4MUbaYCHkopZcCNRJreWoMAs7nlQKwkhwoK0Mr\nHgAAGYxJREFUnPevipAAFSYN62b1dbGE5sQ5GJOgOoIAfHUox5gIOaoDa9sPFyySQk2lHtt+uIBX\nn+gn+T5ERLZoqH2eGsZEyEM9OLgLdh3Otfk68+Rg7y+XoalovjqhdhH+6Ng6uMF6m/ojQIUllSjR\nmsZjngTVqdIZjL93VAdW3k2tTceJiMi1mAh5qEeGdkNtbS32/XIF+loB3l4Kyd1e9ZODnGslzRUi\nAv2UeGZsH/SICkVYWCCKi8uh15tu3Cp1fzQxfipv4+8d1YFl9Qm6dkUBIiKygomQBxt/XzeMv+/O\nNNO8f/6IYm3jtT71k4NibY1DY/L18UKntsFNWslZCoUCmDi0q/FrR3VgRbUJxlmRhSOj2gTbHCMR\nETU/JkIebPbK/6Cs8s4IS7C/F4L9vUyOmVMogMTuEc0WU6e2wXjtyf6Sz5cydSW295l515ijOrAe\nHd4NK7edNpkubKyuiYiIXIeJkId68cODFglPQwlQHUEA1qWexbrUs80Sl61TUdbODwtSITLU3ySh\nqT/6JcYRHVjR7UIw+9F4p6wjRERETcdEyENpJCQ9zmbPaJO1KS1Xbjwa3S4EsybFW61rIiJytIbW\nETIntq6QGE9Za4iJEMmGIAAZF4osFjpsCBcVJCJqeB0hc+brConxpLWGmAiRU/mpvOGn8oavjzdu\nFFdavG7Plh1cVJCIPB3XEbIfEyEPFeLv5fTpsfBgFZb9ORkAkLIzUzQRaqxGyJHbYBARETER8lDV\nNc6vW6kVgMWfpSMkUIXE7hE2t6vnXHfMNhhERER1mAh5qGrnbQ1mVKLVGVd9zr2uwYT7o3HqQpHk\n2p49R3Mdsg0GERFRHSZC5BLqsmpkXCiyKYFx1DYYREREdZgIkcMoFIC3lwIhAT5QlzWenNiawDhq\nGwwiopbmWt5VFJc4bsujCm0Z8sJqUV5hWcspVwoFENWxIxQK2/Y0YiLkoXyVjp8e696hFUICVSiv\nrJGUCNmawIwdHI2ca03fBoOIqKVp1SoEXt6O+5HuGx6OPI0K17UVDrtncysrvoE2rVvDz8/PpuuY\nCHmoaWN649Nvsh16z/N5pQBubykR4q+EpvJOpqVQmO4Cb08C07U91wwiIhLTtn0UvB2YCLkjXWWp\nXdd59lPzYI5OgurTVOgQ2zkMPf2UxoQlsXuETYXR1nDNICIiciTZJEJVVVVYsmQJDh8+jNLSUnTv\n3h1//vOfcd9994mef+bMGSxbtgzZ2dnw9fVF//798eabbyI8/PaCUrW1tVi5ciX+93//F4WFhejU\nqRNmzJiBcePGSb4H2U9vqLVIWGxZMZqIiMgZZLOJyDvvvIP09HSsXbsWR48exYQJE/DCCy8gJyfH\n4tzS0lI888wziImJwffff49du3ZBq9Vi1qxZxnNWrVqFHTt24MMPP0RaWhpefPFFLFy4ED///LPk\ne5D9WMBMRETuQBaJkEajwe7du/Hiiy8iOjoaKpUKjz/+OLp3746tW7danL97924YDAbMmzcPQUFB\nCA8Px7x585Ceno7s7NtTPlu3bsWMGTPQp08fKJVKjBw5EsOGDcPmzZsl34OsG9CrNXp0bIWYTqEI\nCTBNeljATERE7kIWU2NnzpyBwWBAQkKCyfG4uDicOnXK4vxTp06hd+/eUCrvhB8TEwNfX19kZGQg\nKCgIarUa8fHxJtfFx8dj48aNku7Ru3dvB37ClsVHqTCZ9srN17CAmYiI3JIsEiG1Wg0AaNWqlcnx\nsLAwFBUViZ5vfm7d9Wq1Gmq1GgqFAqGhoRb3K/nvOguN3cMW3t6yGFiTxBGxBvn7QKm8c58eUaHo\nERXawBVNUxezOz5nxty8GLNzMGbnaEqsAbVFUAju81mbQ9tWSvj5qUx+Pkkhi0SoOQj1e7WbWUiI\nv9Pey9W8FMBTD/ZBWFig09/bHZ8zY3YOxuwcjFm+xo0Z6uoQ3JYsEqGIiAgAQElJCfz973zTFhcX\no3Xr1hbnR0ZGoqCgwOJ4aWkpIiMjTe5XX3FxsbEjrLF7tHS7P3zY1SEQERG5nCzG0fr27QsfHx9k\nZGSYHD958iSSkpIszk9KSsKvv/6Kmpoa47HTp09Dp9OhX79+iIqKQmRkpMX90tPTjfdr6B5i70lE\nREQtjywSoaCgIEycOBEpKSnIzc1FVVUV1q1bh6tXr2LKlCm4ceMGHnjgAZw4cQIAMHbsWPj6+mLZ\nsmXQarUoKCjA0qVLMXjwYPTs2RMA8NRTT2Hjxo3IysqCTqfDnj17cOTIEcyYMaPRe/Tq1ctlz4KI\niIicRyE4s5imATU1NVi6dClSU1Oh1WoRGxuL+fPno3///rh27RpGjRqFTz/9FMnJyQCACxcu4L33\n3kNGRgZ8fX0xYsQILFy4EMHBwcZ7rlq1Cl9++SXUajW6dOmCl156CaNGjTK+LuUeRERE1HLJJhEi\nIiIicjZZTI0RERERuQITISIiIvJYTISIiIjIYzERIiIiIo/FRIiIiIg8FhMhIiIi8lhMhOx09uxZ\njB8/HjExMdDpdBav79mzB4888gj69euH3//+9/jwww+duv+ZuaqqKrz99tsYOXIkBgwYgMmTJ+Pw\n4cMui0dMXl4epk2bhpiYGOTm5pq8dvToUUyePBl33303RowYgb/+9a+orKx0UaR3FBcX44033sCw\nYcPQr18/jB8/Hnv37jW+Lse4L168iFmzZuHee+9F//79MW7cOGzbts34uhxjri83NxdJSUl47bXX\njMfkGPOIESPQt29fJCQkICEhAfHx8UhISMCxY8dkGzMA7Nq1C2PGjEF8fDxGjRqF9evXG1+TW8zH\njx83Ptf6v3r37o2FCxfKMmbg9vfwrFmzkJycjAEDBuCRRx7B119/bXxdjjG3aALZbMuWLUJycrIw\nd+5cISYmRqiurjZ5PS0tTejdu7eQmpoqVFdXC+fOnROGDx8urFy50kURC8KCBQuEsWPHCjk5OUJ1\ndbXwxRdfCH379hUuXrzospjq++6774TBgwcLr732mhATEyPk5OQYX8vNzRXi4uKEDRs2CJWVlUJe\nXp4wYcIE4dVXX3VhxLf94Q9/EKZOnSoUFBQIer1e2LJlixAbGyucOnVKuHTpkuzirq6uFoYMGSK8\n+eabQllZmaDX64WvvvpK6NWrl3Do0CFZxlyfwWAQHnvsMeHuu+8WFixYIAiCfL8/hg8fLuzcuVP0\nNbnGvHfvXuGee+4Rjh49Kuh0OuHYsWPCmDFjZPv9LKa4uFgYMmSIkJaWJsuYDQaDMHToUOHll18W\nSkpKhJqaGmH37t1CTEyMcPToUVnG3NIxEbLDpk2bhMLCQmHHjh2iidCsWbOEZ5991uKagQMHCgaD\nwZmhCoIgCKWlpUKfPn2E/fv3mxwfP3688O677zo9HjHbt28XcnNzhbS0NItEaMmSJcKDDz5ocv6B\nAweEPn36CEVFRc4O1aisrEx44403hN9++83k+MCBA4WPP/5YlnEXFxcL27ZtE8rKykyO33333cKa\nNWtkGXN9q1atEiZNmiS8+uqrxkRIrjE3lAjJNeYxY8YIH3/8sehrco3Z3Ny5c4XXX39dEAR5xlxQ\nUCD06tVL+PHHH02ODx482C3+DLZEnBqzw7Rp0xrcof7UqVNISEgwORYfHw+NRmMx5eMMZ86cgcFg\nsIgpLi4Op06dcno8YiZOnIguXbqIvmbteer1emRmZjohOnFBQUF49913jfvbAYBarUZ5eTnatWsn\ny7hDQ0MxadIkBAUFAQC0Wi3WrVsHQRAwatQoWcZc59dff8W6deuwZMkSeHnd+atLzjHv27cPDz30\nEAYMGIDx48cbpz/kGHNhYSEuXryIoKAgTJ06Ff3798dDDz0k65jNpaWl4dChQ5g/fz4Aecbctm1b\nJCUl4csvv8StW7dQW1uLvXv3oqKiAsOHD5dlzC2d0tUByI3BYEBFRYXV14OCgqBQKBq8h1qtRqtW\nrUyOhYWFQRAEFBUVoVu3bg6JVSq1Wg0AojEVFRU5NRZ7qNVqJCYmmhwLCwszviYXOp0O8+bNQ8+e\nPTFmzBikpKTIOu4hQ4ZArVaja9eu+OSTT9CtWzfZPuuamhr85S9/wfPPP2/x50euMcfExKBTp05Y\nvHgx/P39sWPHDixYsACBgYGyjDk/Px8A8MUXX2DZsmXo3LkzvvzySyxYsABt2rSRZczmVqxYgenT\npxv/rpNrzCkpKfjjH/+I5ORkKBQKBAQE4G9/+xt69uwp25hbMiZCZtLT0zFt2jSLZEcQBCgUCuzd\nuxfR0dEuio7kqrCwELNmzYJOp8PatWvh7e3t6pAadeTIEWi1WuzYsQMzZ87EmjVrXB2SVStWrEBA\nQABmzpzp6lAkW7VqlcnXU6ZMwaFDh/D55583+o8pV3ryySeNo5xTp07F119/jR07drg4qsalpaXh\nzJkz+Pjjj10dSoNqamowc+ZMdO7cGatXr0ZISAj+85//4LXXXkNISIirw/NITITMDBw4EGfPnm3S\nPSIjI1FSUmJyrLi4GAqFosEpteYSEREBACgpKYG/v79JTK1bt3Z6PLay9jzrXnO1s2fP4rnnnkO/\nfv3w/vvvG5+x3OMGbo9wTps2DT///DM2bNiA1q1byy7m9PR0bN261eoPY3d4znU6deqEtLQ0Wcbc\npk0bADBOm9bp1KkTbt26JcvvjfpSU1MxaNAgk2RCjs/5l19+wdmzZ/HJJ58Y//4dPXo0vv76a3z+\n+eeyf84tEWuEmkFSUpJF7c3x48cRGhrqktGkvn37wsfHBxkZGSbHT548iaSkJKfHYytrz1OpVCIu\nLs5FUd128eJFzJgxA4899hg++ugjk0RTjnEfOnQIw4YNg1arNTleU1MDHx8fWca8fft26PV6PPHE\nExg0aBAGDRqE1NRUpKam4t5770W/fv0svrddHXNeXh4WLVoEjUZjcvzcuXPo3LmzLJ9zmzZtEBoa\niqysLJPjly9fRseOHWUZcx1BEPDtt99ixIgRJsflGHNtba3Jf+sYDAYIgiDLmFs8l5Zqu7mvvvpK\ntGssIyND6NOnj7Bnzx6hurpaOH36tDBkyBBhzZo1LopUEN5++23hwQcfFHJycoTKykph7dq1QkJC\ngnDlyhWXxSTml19+EXr16mXSNZaXlyckJiYK69evFyorK4WLFy8KY8aMEd58800XRnq7DXbChAnC\n+++/L/q6HOMuLS0V7rvvPmHu3LlCUVGRoNPphN27dwt9+vQRUlNTZRmzRqMRCgoKTH7Nnj1bmDNn\njnDjxg1ZxlxZWSkMHTpUmDNnjlBUVCRUVVUJmzZtEmJjY4XDhw/LMmZBEIQ1a9YI/fr1E3766Seh\nurpa+Oyzz4TY2FghOztbtjELwu3lCHr16iWcPHnS5LgcY9ZoNMblV27duiXodDrh22+/lfWfwZZO\nIQguXOXPTcXHx0OhUMBgMMBgMEClUgEAHn74YSxatAgAcODAAaxcuRKXL19GREQEHn/8cTz33HMu\ni7mmpgZLly5FamoqtFotYmNjMX/+fPTv399lMdU3evRo5Ofno7a2Fnq9Hj4+PlAoFBgwYADWrVuH\n48eP44MPPsBvv/2G4OBgjB07FvPmzYOPj4/LYk5PT8eTTz5pjKF+3Yec47506RKWLVuGX375BYIg\noFOnTpg2bRomTJgAALKM2VzdYoqLFy8GIM+YL126hKVLl+LkyZOoqKhAz549MXfuXAwaNEi2MQPA\nmjVr8O9//xtFRUXo3LkzFixYgCFDhsg65mPHjmH69OnYv38/OnfubPKaHGO+ePEiPvzwQ6SlpaGm\npgYdO3bEzJkzMXHiRNnG3JIxESIiIiKPxRohIiIi8lhMhIiIiMhjMREiIiIij8VEiIiIiDwWEyEi\nIiLyWEyEiIiIyGMxESIiIiKPxUSIiIiIPBYTISIiIvJYTISICAAwYsQIzJ49u9nfZ+rUqZg0aZLx\n65iYGHzwwQcAbm+VEBMTg0OHDkm617Vr1xATE4MtW7Y0S6xE1PIxESKSqdTUVOPeQ864Tg769euH\nI0eOYPDgwZLOb9++PY4cOWKSWBER2ULp6gCISNyJEydMNnJt7uvkQKlUIiIiQvL5CoXCpvOJiMxx\nRIhIhqZOnYotW7YgKysLsbGx2LVrFwDgxo0beOWVV3Dvvfeib9++GDVqFJYvXw69Xt/gdadPn8bM\nmTPRv39/JCQk4MEHH8Tnn39uc1ynT5/G9OnTkZSUhOTkZMybNw83btwwvq7VavHWW2/hvvvuQ9++\nfTF06FAsWrQIWq1W0v3Np8Z0Oh2WLFmCESNGID4+HsnJyViwYAFKSkoAiE+NXbx4Ec8//zwGDhyI\nuLg4PPDAA9iwYYPJ+8TExGDdunVYtWoVhg0bhqSkJEyePBlnzpwxnqPRaPD666/j/vvvR1xcHIYN\nG4b33nsP1dXVNj83IpIvJkJEMpSSkoLExETExsbiyJEjGDNmDHQ6HaZNm4Zff/0VH330Efbt24fn\nn38emzdvxqJFi6xeV15ejqeffhoqlQrbtm3Dvn378MQTT+Cdd97BgQMHJMd05coVPPXUU+jQoQO2\nbduG1atXIycnB3/84x8hCAIA4LnnnsN//vMfvP3229i/fz/eeOMN7N+/36bao/qjWatWrcK+ffuw\nZMkSfPfdd1i5ciWys7Px6quvil6rVqsxZcoUlJSU4JNPPkFqaioee+wxLFu2DJ988onJudu3b0dJ\nSQnWr1+PTZs2obCwEAsWLDC+/u677yIzMxP//Oc/ceDAAbz33ns4ePAglixZIvmzEJH8cWqMSIZC\nQkLg4+MDvV6P8PBwAMCePXtw5coVfPnll4iLiwMAREVF4fLly9iwYQP+8pe/iF7n7e2NHTt2ICws\nDMHBwQCAKVOmYPXq1Th8+DBGjRolKaZNmzbBx8cHixYtglJ5+6+ORYsW4bPPPkNBQQFu3LiB48eP\nY8WKFRg5ciQAoGPHjlCr1Xj77beRm5uL6OjoRt+nLqkCgOzsbPTq1QsDBw4EALRt2xaffvopNBqN\n6LXbtm1DWVkZ/vnPfxqnzGbMmIGsrCxs3rwZzz77rPFcPz8/LFy40Pj1ww8/jNWrV6OyshL+/v7I\nzs7GwIEDER8fb3zvTZs2mcRHRO6PI0JEbiIrKws+Pj7GJKhOfHw89Ho9zp8/L3qdt7c3rl+/jvnz\n52P48OHo168fkpKSoFarjVNMUmRmZqJ3797GJAgA4uLi8Pe//x3t2rVDZmYmFAoFkpKSTK5LSEiA\nIAjIzs624dPeNnLkSBw+fBizZ8/G3r17oVar0bZtW/To0UP0/KysLHTs2NGibig+Ph5FRUUoLCw0\niau+usSx7pmMHDkS27Ztw+uvv44DBw5Aq9UiKioKnTp1svlzEJF8MREichNarRb+/v4WxwMDAyEI\nAsrLy0Wvy8rKwsyZM1FdXY3Fixdjx44d+Oabb9C6dWub3l+j0SAwMLDB+AAgKCjIIj4AVuNryOOP\nP441a9agqqoKr7/+OpKTk/H000/j4sWLVmMQi1EshoCAAJNz6qbk6kZ85s6di7/97W+4evUqXn75\nZQwaNAgvvfSSSU0UEbk/JkJEbiI4OBgVFRUWUzNarRYKhcI47WUuNTUVCoUCKSkpGDRoELp06YKO\nHTuitLTUpvePiIho8JqQkBBjPPWVlZUZ47fH0KFD8fHHH+PYsWNYvXo1bt26ZTLFVV9wcLBoYXbd\nsboYpRo3bhw2b96MY8eOYdmyZcjKysIrr7xi+4cgItliIkQkY/WTnoSEBOj1emRkZJicc/z4cfj5\n+aFnz56i19XU1MDHx8dkBGTfvn2oqqqyqd6lZ8+eyMrKQlVVlfFYdnY2nnjiCZw7dw7x8fEQBAH/\n93//ZxGfQqEw1tpIJQgCvvvuOxQUFAAAfHx8MHToULz00ku4fv26aJ1QQkICrl27ZjFqc/z4cbRr\n1844/dWYqqoq7N2715jE+fv7Y/To0cZidSJqOZgIEclUq1atkJeXh9OnT+PGjRsYOXIkunbtitdf\nfx1paWm4evUqtmzZgs8//xzTpk2Dn5+f6HWJiYmoqqrChg0bcO3aNezcuRNbtmxBYmIiLly4gLy8\nPEnxTJ06FQDw6quvIjc3F5mZmcbW+G7duiEuLg6DBg3CBx98gIMHD+Lq1avYvXs3UlJS8OCDD6JD\nhw42fX6FQoG1a9dizpw5OH78OAoKCnDmzBl88cUX6Nmzp+joziOPPIKwsDDMmTMHp0+fxqVLl5CS\nkoKDBw/i+eefl/zeSqUSH3zwAV599VWcPn0aBQUFOHHiBL755htj4TYRtQxMhIhkatq0afD19cXU\nqVOxb98++Pj4YOPGjejTpw/mzJmDMWPGYPPmzZgzZw5efvllq9eNHTsW06dPx9q1a/Hwww/jwIED\nWLFiBaZPn46bN29iypQpxmsbWoixa9eu2LhxI4qLi/HII4/gueeeQ4cOHfDpp5/C29sbwO32/ZEj\nR+Ktt97CAw88gOXLl2Py5MlYvHixyb3qv49CobD4us6qVavQqVMnzJkzB//zP/+DF154AeHh4Viz\nZo3o+WFhYfjXv/6FsLAwzJw5E+PGjcO3336L999/H48++qjV9zSnVCqxadMmeHl54U9/+hN+97vf\n4ZVXXkFiYiLb54laGIXAXlAiIiLyUBwRIiIiIo/FRIiIiIg8FhMhIiIi8lhMhIiIiMhjMREiIiIi\nj8VEiIiIiDwWEyEiIiLyWEyEiIiIyGMxESIiIiKPxUSIiIiIPBYTISIiIvJY/x8rPpeRBsmwvQAA\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f92a65434a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_context(\"notebook\", font_scale=1.1)\n",
    "scatter = sns.jointplot(x='total collisions', y = 'normalized yearly accident rate', data = df, ylim= (0,0.005))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>normalized accident rate</th>\n",
       "      <th>total collisions</th>\n",
       "      <th>traffic estimate</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>intersection</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Front St W &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>0.104842</td>\n",
       "      <td>29.0</td>\n",
       "      <td>276.607735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Front St W &amp; John St, Toronto, Ontario, M5V</th>\n",
       "      <td>0.085028</td>\n",
       "      <td>17.0</td>\n",
       "      <td>199.933112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queens Quay W &amp; Rees St, Toronto, Ontario, M5J</th>\n",
       "      <td>0.064337</td>\n",
       "      <td>9.0</td>\n",
       "      <td>139.888889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bay St &amp; Queens Quay W, Toronto, Ontario, M5J</th>\n",
       "      <td>0.062299</td>\n",
       "      <td>18.0</td>\n",
       "      <td>288.928665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queens Quay W &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>0.062216</td>\n",
       "      <td>18.0</td>\n",
       "      <td>289.316788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yonge St &amp; The Esplanade, Toronto, Ontario, M5E</th>\n",
       "      <td>0.060606</td>\n",
       "      <td>20.0</td>\n",
       "      <td>330.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Station St &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>0.059574</td>\n",
       "      <td>14.0</td>\n",
       "      <td>235.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lake Shore Blvd W &amp; York St, Toronto, Ontario, M5J</th>\n",
       "      <td>0.059574</td>\n",
       "      <td>14.0</td>\n",
       "      <td>235.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queen St W &amp; Spadina Ave, Toronto, Ontario, M5V</th>\n",
       "      <td>0.056303</td>\n",
       "      <td>74.0</td>\n",
       "      <td>1314.307288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lake Shore Blvd E &amp; Yonge St, Toronto, Ontario, M5E</th>\n",
       "      <td>0.055319</td>\n",
       "      <td>13.0</td>\n",
       "      <td>235.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    normalized accident rate  \\\n",
       "intersection                                                                   \n",
       "Front St W & York St, Toronto, Ontario, M5J                         0.104842   \n",
       "Front St W & John St, Toronto, Ontario, M5V                         0.085028   \n",
       "Queens Quay W & Rees St, Toronto, Ontario, M5J                      0.064337   \n",
       "Bay St & Queens Quay W, Toronto, Ontario, M5J                       0.062299   \n",
       "Queens Quay W & York St, Toronto, Ontario, M5J                      0.062216   \n",
       "Yonge St & The Esplanade, Toronto, Ontario, M5E                     0.060606   \n",
       "Station St & York St, Toronto, Ontario, M5J                         0.059574   \n",
       "Lake Shore Blvd W & York St, Toronto, Ontario, M5J                  0.059574   \n",
       "Queen St W & Spadina Ave, Toronto, Ontario, M5V                     0.056303   \n",
       "Lake Shore Blvd E & Yonge St, Toronto, Ontario,...                  0.055319   \n",
       "\n",
       "                                                    total collisions  \\\n",
       "intersection                                                           \n",
       "Front St W & York St, Toronto, Ontario, M5J                     29.0   \n",
       "Front St W & John St, Toronto, Ontario, M5V                     17.0   \n",
       "Queens Quay W & Rees St, Toronto, Ontario, M5J                   9.0   \n",
       "Bay St & Queens Quay W, Toronto, Ontario, M5J                   18.0   \n",
       "Queens Quay W & York St, Toronto, Ontario, M5J                  18.0   \n",
       "Yonge St & The Esplanade, Toronto, Ontario, M5E                 20.0   \n",
       "Station St & York St, Toronto, Ontario, M5J                     14.0   \n",
       "Lake Shore Blvd W & York St, Toronto, Ontario, M5J              14.0   \n",
       "Queen St W & Spadina Ave, Toronto, Ontario, M5V                 74.0   \n",
       "Lake Shore Blvd E & Yonge St, Toronto, Ontario,...              13.0   \n",
       "\n",
       "                                                    traffic estimate  \n",
       "intersection                                                          \n",
       "Front St W & York St, Toronto, Ontario, M5J               276.607735  \n",
       "Front St W & John St, Toronto, Ontario, M5V               199.933112  \n",
       "Queens Quay W & Rees St, Toronto, Ontario, M5J            139.888889  \n",
       "Bay St & Queens Quay W, Toronto, Ontario, M5J             288.928665  \n",
       "Queens Quay W & York St, Toronto, Ontario, M5J            289.316788  \n",
       "Yonge St & The Esplanade, Toronto, Ontario, M5E           330.000000  \n",
       "Station St & York St, Toronto, Ontario, M5J               235.000000  \n",
       "Lake Shore Blvd W & York St, Toronto, Ontario, M5J        235.000000  \n",
       "Queen St W & Spadina Ave, Toronto, Ontario, M5V          1314.307288  \n",
       "Lake Shore Blvd E & Yonge St, Toronto, Ontario,...        235.000000  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intersection_df.to_csv('totals_test.csv')\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import scipy.stats as stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f92a73a6e80>"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAskAAAHyCAYAAAAZYFMeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nohkJkoUQQgghhGjm/wNMU+c1lySq6AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f92a73a2c88>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df[\"traffic estimate\"], bins = 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f92a78eb710>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAq8AAAHyCAYAAADFmScpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCNFoSPEqhBBCCCEaDSlehRBCCCFEo/H/E/QLM+weEhwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f92a78ebf60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df[\"total collisions\"], bins = 15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f92a6117160>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqcAAAHyCAYAAADWTmfdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f92a600a080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df[\"total collisions\"], fit=stats.gamma,  bins=range(0, 75, 1), kde=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A gamma distribution looks like a good choice. We know that we are dealing with sums of poission distributions where each intersection is basically a poisson process, with the mean accidents for a year being the parameter. So the above histogram is really summing up the number of accidents each year, over these poisson processes. While we could try and fit each intersection individually, this is a whole lot of work. And it's not clear how one would use these estimated parameters to compute the prior distribution for the ensemble. \n",
    "\n",
    "A gamma distribution looks nice, and intuitively makes sense since the prior for a poisson distribution is a gamma. Trying to read up on the intuition for how a ensemble of Poisson processes, each with their own paramater (not merely the same), hasn't lead me to any intuition for why I should stick with a gamma distribution for the entire ensemble, but it's a nice fit. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.3457292036111696, 0.99999999999999978, 13.068355505386879)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats.gamma.fit(df['total collisions'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}