Speed Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import datetime as dt\n",
    "import os\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import statsmodels.formula.api as sm\n",
    "import statsmodels.genmod\n",
    "from sklearn import linear_model\n",
    "from sklearn import ensemble\n",
    "from sklearn import model_selection\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import scipy.sparse as sps\n",
    "\n",
    "sns.set(color_codes=True)\n",
    "%matplotlib inline\n",
    "\n",
    "PLOT_DIR = \"Images\"\n",
    "MODEL_DIR = \"Model Output\"\n",
    "\n",
    "DATA_DIR = \"Data\"\n",
    "EMME_VOLUME_DATA = os.path.join(DATA_DIR, \"EMME 2011 VOLUME SUMMARY.OCT2015.v2.csv\")\n",
    "EMME_LINK_DATA = os.path.join(DATA_DIR, \"EMME_link_data.csv\")\n",
    "RAW_DATA_DIR = os.path.join(DATA_DIR, \"Geoprocessed Data\", \"Processed CSVs\")\n",
    "CLEANED_DATA_DIR = os.path.join(DATA_DIR, \"Cleaned Data\")\n",
    "\n",
    "CLEANED_DATA = os.path.join(CLEANED_DATA_DIR, \"cleaned_data.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Cleaning\n",
    "\n",
    "This is the data cleaning step, taking the raw CSVs produced by ArcGIS and turning them into a single large data table."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "csv_list = [f for f in os.listdir(RAW_DATA_DIR)\n",
    "            if os.path.splitext(f)[1] == \".csv\"]\n",
    "\n",
    "df_list = []\n",
    "for csv in csv_list:\n",
    "    csv_data = pd.read_csv(os.path.join(RAW_DATA_DIR, csv), parse_dates=[\"STARTED_AT\"])\n",
    "    \n",
    "    # clean up trip-specific stuff\n",
    "    csv_data = csv_data[csv_data[\"CUMUL_METE\"] != 0]\n",
    "    csv_data['trip_length'] = csv_data['CUMUL_METE'].max()\n",
    "    csv_data.loc[csv_data[\"SIG_DIST\"] < 0, \"SIG_DIST\"] = 1000\n",
    "    \n",
    "    csv_data[\"TRIP_ID\"] = int(os.path.splitext(csv)[0])\n",
    "    \n",
    "    df_list.append(csv_data)\n",
    "\n",
    "data = pd.concat(df_list)\n",
    "    \n",
    "data['bike_lanes'] = data['BIKE_CODE'] == 11\n",
    "data['sharrows'] = data['BIKE_CODE'].isin((3, 4))\n",
    "data['bike_path'] = data['BIKE_CODE'].isin((6, 7))\n",
    "    \n",
    "#data[data['agency_id'] == \"\"] = 0\n",
    "#data[\"streetcar\"] = pd.to_numeric(data[\"agency_id\"])\n",
    "#data.drop(\"agency_id\", axis=1, inplace=True)\n",
    "  \n",
    "data['SLOPE_TF'] = data['SLOPE_TF'] * data['LINK_DIR'] * (-1)\n",
    "\n",
    "dates = pd.DatetimeIndex(data.RECORDED_A)\n",
    "data[\"time_of_day\"] = \"Off Peak\"\n",
    "data.loc[(dates.hour >= 7) & (dates.hour < 10), \"time_of_day\"] = \"AM Peak\"\n",
    "data.loc[(dates.hour >= 16) & (dates.hour < 19), \"time_of_day\"] = \"PM Peak\"\n",
    "    \n",
    "data.loc[data[\"PURPOSE\"] == \"other\", \"PURPOSE\"] = \"Other\"\n",
    "data.loc[data[\"PURPOSE\"] == \"Work-related\", \"PURPOSE\"] = \"Work-Related\"\n",
    "data.loc[data[\"PURPOSE\"] == \"Domicile-travail\", \"PURPOSE\"] = \"Commute\"\n",
    "\n",
    "df_list = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "volume_df = pd.read_csv(EMME_VOLUME_DATA)\n",
    "link_df = pd.read_csv(EMME_LINK_DATA)\n",
    "\n",
    "same_dir = data[\"LINK_DIR\"] == 1\n",
    "data.loc[same_dir, \"EMME_ID\"] = data.loc[same_dir, \"EMME_MATCH\"]\n",
    "data.loc[~same_dir, \"EMME_ID\"] = data.loc[~same_dir, \"EMME_CONTR\"]\n",
    "    \n",
    "# Add time-variant volume info for each point\n",
    "data = pd.merge(data, volume_df, left_on=\"EMME_ID\", right_on=\"LINK_ID\", how=\"left\")\n",
    "\n",
    "time_categories = ((\"AM_VOL\", 6, 10),\n",
    "                    (\"MID_VOL\", 10, 16),\n",
    "                    (\"PM_VOL\", 16, 19),\n",
    "                    (\"EVE_VOL\", 19, 6))\n",
    "dt_index = pd.DatetimeIndex(data[\"RECORDED_A\"])\n",
    "for period, start, stop in time_categories:\n",
    "    data[period] = data[period].fillna(0)\n",
    "    if start < stop:\n",
    "        index = (start <= dt_index.hour) & (dt_index.hour < stop)\n",
    "    else:\n",
    "        index = (start <= dt_index.hour) | (dt_index.hour < stop)\n",
    "    data.loc[index, \"volume\"] = data.loc[index, period]\n",
    "           \n",
    "# Add time-invariant link-level data from EMME\n",
    "LINK_INFO_COLS = [\"ID\", \"DATA2\", \"LANES\", \"VDF\"]\n",
    "    \n",
    "data = pd.merge(data, link_df.loc[:, LINK_INFO_COLS], left_on=\"EMME_ID\", right_on=\"ID\", how=\"left\")\n",
    "data[\"speed_limit\"] = data[\"DATA2\"]\n",
    "data.drop(\"DATA2\", axis=1, inplace=True)\n",
    "\n",
    "data.loc[data.volume == 0, \"volume\"] = 0\n",
    "data.loc[data.LANES == 0, \"LANES\"] = 1\n",
    "data.loc[data.speed_limit == 0, \"speed_limit\"] = 40\n",
    "\n",
    "data[\"lane_cap\"] = 250 # guess at capacity for small roads\n",
    "data.loc[data.VDF.isin((11, 12, 14, 16)), \"lane_cap\"] = 1800 # Controlled Highways\n",
    "data.loc[data.VDF.isin((13, 15, 17)), \"lane_cap\"] = 1400 # Highway ramps\n",
    "data.loc[data.VDF == 20, \"lane_cap\"] = 1000 # Two lane rural highway\n",
    "data.loc[data.VDF == 21, \"lane_cap\"] = 1100 # Multi-lane rural highway\n",
    "data.loc[data.VDF == 22, \"lane_cap\"] = 500  # Two-lane rural collector\n",
    "data.loc[data.VDF == 30, \"lane_cap\"] = 900  # Low density major arterial, no access\n",
    "data.loc[data.VDF.isin((40, 41)), \"lane_cap\"] = 800 # Low density arterial, some access\n",
    "data.loc[data.VDF == 42, \"lane_cap\"] = 700 # Low density major arterial, direct access\n",
    "data.loc[data.VDF == 50, \"lane_cap\"] = 600 # High density minor arterial\n",
    "data.loc[data.VDF == 51, \"lane_cap\"] = 500 # Collector\n",
    "\n",
    "data.loc[np.isnan(data.speed_limit), \"speed_limit\"] = 40\n",
    "data.loc[np.isnan(data.LANES), \"LANES\"] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_list = None\n",
    "volume_df = None\n",
    "link_df = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "drop_columns = (\"WINTER\", \"RIDER_HIST\", \"WORKZIP\", \"INCOME\", \"CYCLINGFRE\",\n",
    "                \"AGE\", \"CYCLING_LE\", \"GENDER\", \"RIDER_TYPE\", \"SCHOOLZIP\",\n",
    "                \"HOMEZIP\", \"CYCLINGEXP\")\n",
    "for col in drop_columns:\n",
    "    data = data.drop(col, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data['SPEED_DEMEANED'] = data['SPEED']- data.groupby('APP_USER_I')['SPEED'].transform('mean')\n",
    "data['SPEED_TRIP_DEMEANED'] = data['SPEED']- data.groupby('TRIP_ID')['SPEED'].transform('mean')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data['CONGESTION'] = data['volume']/(data['LANES']*data['lane_cap'])\n",
    "data.loc[data['STOP_DIST'] < 0, 'STOP_DIST'] = data.loc[data['STOP_DIST'] < 0, 'trip_length']\n",
    "data['TRIP_COMPLETION'] = data['CUMUL_METE']/data['trip_length']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Save data to a CSV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'data' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-1-cb215d6e19ce>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mto_csv\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mos\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mCLEANED_DATA_DIR\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"cleaned_data_final.csv\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mencoding\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\"utf8\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'data' is not defined"
     ]
    }
   ],
   "source": [
    "data.to_csv(os.path.join(CLEANED_DATA_DIR, \"cleaned_data_final.csv\"), encoding=\"utf8\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Read in saved data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#data = pd.read_csv(CLEANED_DATA, parse_dates=[\"RECORDED_A\"])\n",
    "data = pd.read_csv(os.path.join(CLEANED_DATA_DIR, \"cleaned_data_final.csv\"), \n",
    "                   parse_dates=[\"RECORDED_A\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Data Summary\n",
    "\n",
    "This is where we summarize the data / take an initial look at it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of observations: 3511876\n",
      "Number of unique users: 518\n",
      "\n",
      "Variable\tMin\tMax\tMean\tStandard Deviation\n",
      "SIG_DIST  \t0.0\t17203.3\t463.75\t956.49\n",
      "STOP_DIST \t0.0\t20806.9\t1989.15\t2716.11\n",
      "CONGESTION\t0.0\t7.9\t0.18\t0.28\n",
      "speed_limit\t40.0\t100.0\t43.52\t6.28\n",
      "LANES     \t1.0\t4.0\t1.57\t0.66\n",
      "SLOPE_TF  \t-0.1\t0.1\t0.00\t0.02\n",
      "lane_cap  \t250.0\t1800.0\t509.96\t241.81\n",
      "\n",
      "Variable\tProportion\n",
      "Bike Lanes\t0.22\n",
      "Sharrows\t0.09\n",
      "Bike Path\t0.12\n",
      "Major Arterial\t0.32\n",
      "Minor Arterial\t0.26\n",
      "Local Road\t0.42\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of observations: %d\" % len(data))\n",
    "print(\"Number of unique users: %d\" % len(data[\"APP_USER_I\"].unique()))\n",
    "\n",
    "cont_variables = (\"SIG_DIST\", \"STOP_DIST\", \"CONGESTION\", \"speed_limit\", \"LANES\", \"SLOPE_TF\", \"lane_cap\")\n",
    "print(\"\\nVariable\\tMin\\tMax\\tMean\\tStandard Deviation\")\n",
    "for var in cont_variables:\n",
    "    print(\"%s\\t%0.1f\\t%0.1f\\t%0.2f\\t%0.2f\" % (var.ljust(10), np.min(data[var]), np.max(data[var]), np.mean(data[var]), np.std(data[var])))\n",
    "\n",
    "cat_variables = ((\"Bike Lanes\", \"bike_lanes\", True), (\"Sharrows\", \"sharrows\", True), (\"Bike Path\", \"bike_path\", True),\n",
    "                 (\"Major Arterial\", \"RDCLASS\", 1), (\"Minor Arterial\", \"RDCLASS\", 2), (\"Local Road\", \"RDCLASS\", 3))\n",
    "print(\"\\nVariable\\tProportion\")\n",
    "for var, col, value in cat_variables:\n",
    "    print(\"%s\\t%0.2f\" % (var, np.mean(data[col] == value)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With the loaded data, we make plots to get an idea of the relationship between speed and\n",
    "other variables in the model; in particular, slope and distance to intersection."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAElCAYAAAAFukKMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1dcfca8f7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hexbin(\"SLOPE_TF\", \"SPEED\", data=data, bins=\"log\", cmap=plt.cm.YlOrRd)\n",
    "plt.xlim(-0.05, 0.05)\n",
    "plt.ylim(0, 20)\n",
    "plt.xlabel(\"Slope (grads/100)\")\n",
    "plt.ylabel(\"Residual of Speed from other Model Variables (m/s)\")\n",
    "plt.title(\"Histogram of Slope-Speed\\nobservations\")\n",
    "cbar = plt.colorbar()\n",
    "cbar.set_label(\"log10(Counts)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAElCAYAAAAFukKMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDYF1Kq6XDwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1dcd86b4c88>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hexbin(data[\"SIG_DIST\"], data[\"SPEED\"], bins=\"log\", cmap=plt.cm.YlOrRd)\n",
    "plt.xlim(0,500)\n",
    "plt.ylim(0, 20)\n",
    "plt.xlabel(\"Distance to Nearest Intersection (m)\")\n",
    "plt.ylabel(\"Speed (m/s)\")\n",
    "plt.title(\"Histogram of Speed-Intersection Distance\\nobservations\")\n",
    "cbar = plt.colorbar()\n",
    "cbar.set_label(\"log10(Counts)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Linear Models\n",
    "\n",
    "Now, we read in the cleaned data, and run a linear model on the speed,\n",
    "regressed on the `explanatory_road` variables defined below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>          <td>SPEED</td>      <th>  R-squared:         </th>  <td>   0.137</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th>  <td>   0.137</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th>  <td>4.647e+04</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sat, 29 Jul 2017</td> <th>  Prob (F-statistic):</th>   <td>  0.00</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>22:25:15</td>     <th>  Log-Likelihood:    </th> <td>-7.0991e+06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>3511876</td>     <th>  AIC:               </th>  <td>1.420e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>3511863</td>     <th>  BIC:               </th>  <td>1.420e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    12</td>      <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "             <td></td>               <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>              <td>    4.1832</td> <td>    0.011</td> <td>  392.991</td> <td> 0.000</td> <td>    4.162     4.204</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_lanes[T.True]</th>     <td>    0.3715</td> <td>    0.003</td> <td>  131.858</td> <td> 0.000</td> <td>    0.366     0.377</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sharrows[T.True]</th>       <td>    0.2321</td> <td>    0.004</td> <td>   63.657</td> <td> 0.000</td> <td>    0.225     0.239</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_path[T.True]</th>      <td>    0.6642</td> <td>    0.003</td> <td>  190.220</td> <td> 0.000</td> <td>    0.657     0.671</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SIG_DIST <= 35[T.True]</th> <td>   -1.3618</td> <td>    0.003</td> <td> -481.549</td> <td> 0.000</td> <td>   -1.367    -1.356</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>STOP_DIST < 20[T.True]</th> <td>   -0.6032</td> <td>    0.006</td> <td> -109.295</td> <td> 0.000</td> <td>   -0.614    -0.592</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.2]</th>        <td>    0.2213</td> <td>    0.003</td> <td>   70.467</td> <td> 0.000</td> <td>    0.215     0.228</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.3]</th>        <td>   -0.0987</td> <td>    0.004</td> <td>  -23.869</td> <td> 0.000</td> <td>   -0.107    -0.091</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>CONGESTION</th>             <td>   -0.2745</td> <td>    0.004</td> <td>  -63.536</td> <td> 0.000</td> <td>   -0.283    -0.266</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>speed_limit</th>            <td>    0.0217</td> <td>    0.000</td> <td>   81.990</td> <td> 0.000</td> <td>    0.021     0.022</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>LANES</th>                  <td>    0.0588</td> <td>    0.002</td> <td>   23.753</td> <td> 0.000</td> <td>    0.054     0.064</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SLOPE_TF</th>               <td>  -26.4575</td> <td>    0.061</td> <td> -433.763</td> <td> 0.000</td> <td>  -26.577   -26.338</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>lane_cap</th>               <td>    0.0007</td> <td> 1.03e-05</td> <td>   66.070</td> <td> 0.000</td> <td>    0.001     0.001</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>100935.202</td> <th>  Durbin-Watson:     </th>  <td>   0.138</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>   <td> 0.000</td>   <th>  Jarque-Bera (JB):  </th> <td>193361.997</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>            <td>-0.218</td>   <th>  Prob(JB):          </th>  <td>    0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>        <td> 4.064</td>   <th>  Cond. No.          </th>  <td>3.54e+04</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                  SPEED   R-squared:                       0.137\n",
       "Model:                            OLS   Adj. R-squared:                  0.137\n",
       "Method:                 Least Squares   F-statistic:                 4.647e+04\n",
       "Date:                Sat, 29 Jul 2017   Prob (F-statistic):               0.00\n",
       "Time:                        22:25:15   Log-Likelihood:            -7.0991e+06\n",
       "No. Observations:             3511876   AIC:                         1.420e+07\n",
       "Df Residuals:                 3511863   BIC:                         1.420e+07\n",
       "Df Model:                          12                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==========================================================================================\n",
       "                             coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "------------------------------------------------------------------------------------------\n",
       "Intercept                  4.1832      0.011    392.991      0.000         4.162     4.204\n",
       "bike_lanes[T.True]         0.3715      0.003    131.858      0.000         0.366     0.377\n",
       "sharrows[T.True]           0.2321      0.004     63.657      0.000         0.225     0.239\n",
       "bike_path[T.True]          0.6642      0.003    190.220      0.000         0.657     0.671\n",
       "SIG_DIST <= 35[T.True]    -1.3618      0.003   -481.549      0.000        -1.367    -1.356\n",
       "STOP_DIST < 20[T.True]    -0.6032      0.006   -109.295      0.000        -0.614    -0.592\n",
       "C(RDCLASS)[T.2]            0.2213      0.003     70.467      0.000         0.215     0.228\n",
       "C(RDCLASS)[T.3]           -0.0987      0.004    -23.869      0.000        -0.107    -0.091\n",
       "CONGESTION                -0.2745      0.004    -63.536      0.000        -0.283    -0.266\n",
       "speed_limit                0.0217      0.000     81.990      0.000         0.021     0.022\n",
       "LANES                      0.0588      0.002     23.753      0.000         0.054     0.064\n",
       "SLOPE_TF                 -26.4575      0.061   -433.763      0.000       -26.577   -26.338\n",
       "lane_cap                   0.0007   1.03e-05     66.070      0.000         0.001     0.001\n",
       "==============================================================================\n",
       "Omnibus:                   100935.202   Durbin-Watson:                   0.138\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           193361.997\n",
       "Skew:                          -0.218   Prob(JB):                         0.00\n",
       "Kurtosis:                       4.064   Cond. No.                     3.54e+04\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 3.54e+04. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "explanatory_road = (\"bike_lanes\", \"sharrows\", \"bike_path\", \"SIG_DIST <= 35\", \"STOP_DIST < 20\",\n",
    "                    \"CONGESTION\", \"speed_limit\", \"LANES\", \"C(RDCLASS)\", \n",
    "                    \"SLOPE_TF\", \"lane_cap\")\n",
    "\n",
    "formula = \"SPEED ~ \" + \" + \".join(explanatory_road)\n",
    "model = sm.ols(formula=formula, data=data).fit()\n",
    "\n",
    "data[\"residual\"] = model.resid\n",
    "data[\"predicted\"] = model.fittedvalues\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To control for any user-specific characteristics that might be correlated with both speed and choice of route (e.g. cyclist experience, fitness, comfort riding in traffic, etc), an additional regression is run, with User ID as an additional dummy category.  Unfortunately, this adds 518 extra variables to the estimation, which eats up memory, hence the sampling.  \n",
    "\n",
    "The estimates are quite similar, with the exception of number of lanes having a different sign in this regression compared to the previous one."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Actual R-squared: 0.321214\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLSAR Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>SPEED_DEMEANED</td>  <th>  R-squared:         </th>  <td>   0.135</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                  <td>GLSAR</td>      <th>  Adj. R-squared:    </th>  <td>   0.135</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th>  <td>4.562e+04</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sat, 29 Jul 2017</td> <th>  Prob (F-statistic):</th>   <td>  0.00</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>22:54:14</td>     <th>  Log-Likelihood:    </th> <td>-6.6776e+06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>3511875</td>     <th>  AIC:               </th>  <td>1.336e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>3511862</td>     <th>  BIC:               </th>  <td>1.336e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    12</td>      <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "             <td></td>               <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>              <td>   -0.6322</td> <td>    0.009</td> <td>  -66.964</td> <td> 0.000</td> <td>   -0.651    -0.614</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_lanes[T.True]</th>     <td>    0.3206</td> <td>    0.002</td> <td>  128.306</td> <td> 0.000</td> <td>    0.316     0.325</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sharrows[T.True]</th>       <td>    0.2291</td> <td>    0.003</td> <td>   70.861</td> <td> 0.000</td> <td>    0.223     0.235</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_path[T.True]</th>      <td>    0.3620</td> <td>    0.003</td> <td>  116.890</td> <td> 0.000</td> <td>    0.356     0.368</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SIG_DIST <= 35[T.True]</th> <td>   -1.2814</td> <td>    0.003</td> <td> -510.894</td> <td> 0.000</td> <td>   -1.286    -1.276</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>STOP_DIST < 20[T.True]</th> <td>   -0.4655</td> <td>    0.005</td> <td>  -95.111</td> <td> 0.000</td> <td>   -0.475    -0.456</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.2]</th>        <td>    0.0956</td> <td>    0.003</td> <td>   34.317</td> <td> 0.000</td> <td>    0.090     0.101</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.3]</th>        <td>   -0.1872</td> <td>    0.004</td> <td>  -51.045</td> <td> 0.000</td> <td>   -0.194    -0.180</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>CONGESTION</th>             <td>    0.0948</td> <td>    0.004</td> <td>   24.748</td> <td> 0.000</td> <td>    0.087     0.102</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>speed_limit</th>            <td>    0.0141</td> <td>    0.000</td> <td>   60.160</td> <td> 0.000</td> <td>    0.014     0.015</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>LANES</th>                  <td>    0.0402</td> <td>    0.002</td> <td>   18.297</td> <td> 0.000</td> <td>    0.036     0.044</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SLOPE_TF</th>               <td>  -24.9213</td> <td>    0.054</td> <td> -460.685</td> <td> 0.000</td> <td>  -25.027   -24.815</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>lane_cap</th>               <td>    0.0002</td> <td> 9.17e-06</td> <td>   17.547</td> <td> 0.000</td> <td>    0.000     0.000</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>279295.791</td> <th>  Durbin-Watson:     </th>  <td>   0.171</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>   <td> 0.000</td>   <th>  Jarque-Bera (JB):  </th> <td>785119.216</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>            <td>-0.441</td>   <th>  Prob(JB):          </th>  <td>    0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>        <td> 5.142</td>   <th>  Cond. No.          </th>  <td>3.54e+04</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                           GLSAR Regression Results                           \n",
       "==============================================================================\n",
       "Dep. Variable:         SPEED_DEMEANED   R-squared:                       0.135\n",
       "Model:                          GLSAR   Adj. R-squared:                  0.135\n",
       "Method:                 Least Squares   F-statistic:                 4.562e+04\n",
       "Date:                Sat, 29 Jul 2017   Prob (F-statistic):               0.00\n",
       "Time:                        22:54:14   Log-Likelihood:            -6.6776e+06\n",
       "No. Observations:             3511875   AIC:                         1.336e+07\n",
       "Df Residuals:                 3511862   BIC:                         1.336e+07\n",
       "Df Model:                          12                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==========================================================================================\n",
       "                             coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "------------------------------------------------------------------------------------------\n",
       "Intercept                 -0.6322      0.009    -66.964      0.000        -0.651    -0.614\n",
       "bike_lanes[T.True]         0.3206      0.002    128.306      0.000         0.316     0.325\n",
       "sharrows[T.True]           0.2291      0.003     70.861      0.000         0.223     0.235\n",
       "bike_path[T.True]          0.3620      0.003    116.890      0.000         0.356     0.368\n",
       "SIG_DIST <= 35[T.True]    -1.2814      0.003   -510.894      0.000        -1.286    -1.276\n",
       "STOP_DIST < 20[T.True]    -0.4655      0.005    -95.111      0.000        -0.475    -0.456\n",
       "C(RDCLASS)[T.2]            0.0956      0.003     34.317      0.000         0.090     0.101\n",
       "C(RDCLASS)[T.3]           -0.1872      0.004    -51.045      0.000        -0.194    -0.180\n",
       "CONGESTION                 0.0948      0.004     24.748      0.000         0.087     0.102\n",
       "speed_limit                0.0141      0.000     60.160      0.000         0.014     0.015\n",
       "LANES                      0.0402      0.002     18.297      0.000         0.036     0.044\n",
       "SLOPE_TF                 -24.9213      0.054   -460.685      0.000       -25.027   -24.815\n",
       "lane_cap                   0.0002   9.17e-06     17.547      0.000         0.000     0.000\n",
       "==============================================================================\n",
       "Omnibus:                   279295.791   Durbin-Watson:                   0.171\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           785119.216\n",
       "Skew:                          -0.441   Prob(JB):                         0.00\n",
       "Kurtosis:                       5.142   Cond. No.                     3.54e+04\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 3.54e+04. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "formula = \"SPEED_DEMEANED ~ \" + \" + \".join(explanatory_road)\n",
    "model = sm.glsar(formula=formula, data=data, rho=1).fit()\n",
    "\n",
    "data[\"fixed_residual\"] = model.resid\n",
    "data[\"fixed_predicted\"] = model.fittedvalues + data.groupby('APP_USER_I')['SPEED'].transform('mean')\n",
    "print(\"Actual R-squared: %f\" % (1 - np.var(data[\"fixed_residual\"]) / np.var(data[\"SPEED\"])))\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Actual R-squared: 0.322826\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLSAR Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>SPEED_DEMEANED</td>  <th>  R-squared:         </th>  <td>   0.137</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                  <td>GLSAR</td>      <th>  Adj. R-squared:    </th>  <td>   0.137</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th>  <td>2.321e+04</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sat, 29 Jul 2017</td> <th>  Prob (F-statistic):</th>   <td>  0.00</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>22:55:12</td>     <th>  Log-Likelihood:    </th> <td>-6.6734e+06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>3511875</td>     <th>  AIC:               </th>  <td>1.335e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>3511850</td>     <th>  BIC:               </th>  <td>1.335e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    24</td>      <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "               <td></td>                 <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>                  <td>    0.0970</td> <td>    0.036</td> <td>    2.726</td> <td> 0.006</td> <td>    0.027     0.167</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_lanes[T.True]</th>         <td>    0.3197</td> <td>    0.002</td> <td>  127.978</td> <td> 0.000</td> <td>    0.315     0.325</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sharrows[T.True]</th>           <td>    0.2256</td> <td>    0.003</td> <td>   69.785</td> <td> 0.000</td> <td>    0.219     0.232</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_path[T.True]</th>          <td>    0.3790</td> <td>    0.003</td> <td>  121.360</td> <td> 0.000</td> <td>    0.373     0.385</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SIG_DIST <= 35[T.True]</th>     <td>   -1.2819</td> <td>    0.003</td> <td> -511.653</td> <td> 0.000</td> <td>   -1.287    -1.277</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>STOP_DIST <= 20[T.True]</th>    <td>   -0.4679</td> <td>    0.005</td> <td>  -95.696</td> <td> 0.000</td> <td>   -0.478    -0.458</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.2]</th>            <td>    0.0997</td> <td>    0.003</td> <td>   35.790</td> <td> 0.000</td> <td>    0.094     0.105</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.3]</th>            <td>   -0.1801</td> <td>    0.004</td> <td>  -49.104</td> <td> 0.000</td> <td>   -0.187    -0.173</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(time_of_day)[T.Off Peak]</th> <td>   -0.6865</td> <td>    0.034</td> <td>  -20.013</td> <td> 0.000</td> <td>   -0.754    -0.619</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(time_of_day)[T.PM Peak]</th>  <td>   -0.7413</td> <td>    0.035</td> <td>  -21.417</td> <td> 0.000</td> <td>   -0.809    -0.673</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Errand]</th>       <td>   -0.2501</td> <td>    0.006</td> <td>  -40.873</td> <td> 0.000</td> <td>   -0.262    -0.238</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Exercise]</th>     <td>   -0.1046</td> <td>    0.005</td> <td>  -20.668</td> <td> 0.000</td> <td>   -0.115    -0.095</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Home]</th>         <td>   -0.1238</td> <td>    0.005</td> <td>  -24.855</td> <td> 0.000</td> <td>   -0.134    -0.114</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Leisure]</th>      <td>   -0.2796</td> <td>    0.004</td> <td>  -63.070</td> <td> 0.000</td> <td>   -0.288    -0.271</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Other]</th>        <td>   -0.0943</td> <td>    0.004</td> <td>  -25.672</td> <td> 0.000</td> <td>   -0.101    -0.087</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.School]</th>       <td>    0.0289</td> <td>    0.007</td> <td>    3.924</td> <td> 0.000</td> <td>    0.014     0.043</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Shopping]</th>     <td>   -0.1982</td> <td>    0.006</td> <td>  -35.196</td> <td> 0.000</td> <td>   -0.209    -0.187</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Social]</th>       <td>   -0.0596</td> <td>    0.004</td> <td>  -14.899</td> <td> 0.000</td> <td>   -0.067    -0.052</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Sport]</th>        <td>   -0.2897</td> <td>    0.032</td> <td>   -9.049</td> <td> 0.000</td> <td>   -0.352    -0.227</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(PURPOSE)[T.Work-Related]</th> <td>   -0.0060</td> <td>    0.005</td> <td>   -1.144</td> <td> 0.253</td> <td>   -0.016     0.004</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>CONGESTION</th>                 <td>    0.0972</td> <td>    0.004</td> <td>   25.088</td> <td> 0.000</td> <td>    0.090     0.105</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>speed_limit</th>                <td>    0.0139</td> <td>    0.000</td> <td>   59.421</td> <td> 0.000</td> <td>    0.013     0.014</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>LANES</th>                      <td>    0.0372</td> <td>    0.002</td> <td>   16.955</td> <td> 0.000</td> <td>    0.033     0.042</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>lane_cap</th>                   <td>    0.0002</td> <td> 9.17e-06</td> <td>   18.049</td> <td> 0.000</td> <td>    0.000     0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SLOPE_TF</th>                   <td>  -24.9195</td> <td>    0.054</td> <td> -460.982</td> <td> 0.000</td> <td>  -25.025   -24.814</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>280624.442</td> <th>  Durbin-Watson:     </th>  <td>   0.172</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>   <td> 0.000</td>   <th>  Jarque-Bera (JB):  </th> <td>790276.660</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>            <td>-0.443</td>   <th>  Prob(JB):          </th>  <td>    0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>        <td> 5.148</td>   <th>  Cond. No.          </th>  <td>3.91e+04</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                           GLSAR Regression Results                           \n",
       "==============================================================================\n",
       "Dep. Variable:         SPEED_DEMEANED   R-squared:                       0.137\n",
       "Model:                          GLSAR   Adj. R-squared:                  0.137\n",
       "Method:                 Least Squares   F-statistic:                 2.321e+04\n",
       "Date:                Sat, 29 Jul 2017   Prob (F-statistic):               0.00\n",
       "Time:                        22:55:12   Log-Likelihood:            -6.6734e+06\n",
       "No. Observations:             3511875   AIC:                         1.335e+07\n",
       "Df Residuals:                 3511850   BIC:                         1.335e+07\n",
       "Df Model:                          24                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================================\n",
       "                                 coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "----------------------------------------------------------------------------------------------\n",
       "Intercept                      0.0970      0.036      2.726      0.006         0.027     0.167\n",
       "bike_lanes[T.True]             0.3197      0.002    127.978      0.000         0.315     0.325\n",
       "sharrows[T.True]               0.2256      0.003     69.785      0.000         0.219     0.232\n",
       "bike_path[T.True]              0.3790      0.003    121.360      0.000         0.373     0.385\n",
       "SIG_DIST <= 35[T.True]        -1.2819      0.003   -511.653      0.000        -1.287    -1.277\n",
       "STOP_DIST <= 20[T.True]       -0.4679      0.005    -95.696      0.000        -0.478    -0.458\n",
       "C(RDCLASS)[T.2]                0.0997      0.003     35.790      0.000         0.094     0.105\n",
       "C(RDCLASS)[T.3]               -0.1801      0.004    -49.104      0.000        -0.187    -0.173\n",
       "C(time_of_day)[T.Off Peak]    -0.6865      0.034    -20.013      0.000        -0.754    -0.619\n",
       "C(time_of_day)[T.PM Peak]     -0.7413      0.035    -21.417      0.000        -0.809    -0.673\n",
       "C(PURPOSE)[T.Errand]          -0.2501      0.006    -40.873      0.000        -0.262    -0.238\n",
       "C(PURPOSE)[T.Exercise]        -0.1046      0.005    -20.668      0.000        -0.115    -0.095\n",
       "C(PURPOSE)[T.Home]            -0.1238      0.005    -24.855      0.000        -0.134    -0.114\n",
       "C(PURPOSE)[T.Leisure]         -0.2796      0.004    -63.070      0.000        -0.288    -0.271\n",
       "C(PURPOSE)[T.Other]           -0.0943      0.004    -25.672      0.000        -0.101    -0.087\n",
       "C(PURPOSE)[T.School]           0.0289      0.007      3.924      0.000         0.014     0.043\n",
       "C(PURPOSE)[T.Shopping]        -0.1982      0.006    -35.196      0.000        -0.209    -0.187\n",
       "C(PURPOSE)[T.Social]          -0.0596      0.004    -14.899      0.000        -0.067    -0.052\n",
       "C(PURPOSE)[T.Sport]           -0.2897      0.032     -9.049      0.000        -0.352    -0.227\n",
       "C(PURPOSE)[T.Work-Related]    -0.0060      0.005     -1.144      0.253        -0.016     0.004\n",
       "CONGESTION                     0.0972      0.004     25.088      0.000         0.090     0.105\n",
       "speed_limit                    0.0139      0.000     59.421      0.000         0.013     0.014\n",
       "LANES                          0.0372      0.002     16.955      0.000         0.033     0.042\n",
       "lane_cap                       0.0002   9.17e-06     18.049      0.000         0.000     0.000\n",
       "SLOPE_TF                     -24.9195      0.054   -460.982      0.000       -25.025   -24.814\n",
       "==============================================================================\n",
       "Omnibus:                   280624.442   Durbin-Watson:                   0.172\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           790276.660\n",
       "Skew:                          -0.443   Prob(JB):                         0.00\n",
       "Kurtosis:                       5.148   Cond. No.                     3.91e+04\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 3.91e+04. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "explanatory_road_trip = (\"bike_lanes\", \"sharrows\", \"bike_path\", \"SIG_DIST <= 35\", \"STOP_DIST <= 20\",\n",
    "                    \"CONGESTION\", \"speed_limit\", \"LANES\", \"lane_cap\", \"C(RDCLASS)\", \"SLOPE_TF\",\n",
    "                    \"C(time_of_day)\", \"C(PURPOSE)\")\n",
    "\n",
    "formula = \"SPEED_DEMEANED ~ \" + \" + \".join(explanatory_road_trip)\n",
    "model = sm.glsar(formula=formula, data=data, rho=1).fit()\n",
    "\n",
    "#data[\"predicted\"] = model.predict()\n",
    "#data[\"residual\"] = data.SPEED - data.predicted\n",
    "print(\"Actual R-squared: %f\" % (1 - np.var(model.resid) / np.var(data[\"SPEED\"])))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Actual R-squared: 0.387114\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLSAR Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>SPEED_TRIP_DEMEANED</td> <th>  R-squared:         </th>  <td>   0.126</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLSAR</td>        <th>  Adj. R-squared:    </th>  <td>   0.126</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>              <td>Least Squares</td>    <th>  F-statistic:       </th>  <td>4.236e+04</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>              <td>Sat, 29 Jul 2017</td>   <th>  Prob (F-statistic):</th>   <td>  0.00</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                  <td>22:33:08</td>       <th>  Log-Likelihood:    </th> <td>-6.4982e+06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>       <td>3511875</td>       <th>  AIC:               </th>  <td>1.300e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>           <td>3511862</td>       <th>  BIC:               </th>  <td>1.300e+07</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>               <td>    12</td>        <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>       <td>nonrobust</td>      <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "             <td></td>               <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>              <td>   -0.5037</td> <td>    0.009</td> <td>  -56.149</td> <td> 0.000</td> <td>   -0.521    -0.486</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_lanes[T.True]</th>     <td>    0.3195</td> <td>    0.002</td> <td>  134.565</td> <td> 0.000</td> <td>    0.315     0.324</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sharrows[T.True]</th>       <td>    0.2001</td> <td>    0.003</td> <td>   65.126</td> <td> 0.000</td> <td>    0.194     0.206</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_path[T.True]</th>      <td>    0.2987</td> <td>    0.003</td> <td>  101.518</td> <td> 0.000</td> <td>    0.293     0.304</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SIG_DIST <= 35[T.True]</th> <td>   -1.2781</td> <td>    0.002</td> <td> -536.293</td> <td> 0.000</td> <td>   -1.283    -1.273</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>STOP_DIST < 20[T.True]</th> <td>   -0.4500</td> <td>    0.005</td> <td>  -96.745</td> <td> 0.000</td> <td>   -0.459    -0.441</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.2]</th>        <td>    0.0928</td> <td>    0.003</td> <td>   35.061</td> <td> 0.000</td> <td>    0.088     0.098</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.3]</th>        <td>   -0.1428</td> <td>    0.003</td> <td>  -40.975</td> <td> 0.000</td> <td>   -0.150    -0.136</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>CONGESTION</th>             <td>    0.0828</td> <td>    0.004</td> <td>   22.738</td> <td> 0.000</td> <td>    0.076     0.090</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>speed_limit</th>            <td>    0.0102</td> <td>    0.000</td> <td>   45.632</td> <td> 0.000</td> <td>    0.010     0.011</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>LANES</th>                  <td>    0.0519</td> <td>    0.002</td> <td>   24.873</td> <td> 0.000</td> <td>    0.048     0.056</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SLOPE_TF</th>               <td>  -20.1754</td> <td>    0.051</td> <td> -392.492</td> <td> 0.000</td> <td>  -20.276   -20.075</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>lane_cap</th>               <td>    0.0002</td> <td> 8.71e-06</td> <td>   21.961</td> <td> 0.000</td> <td>    0.000     0.000</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>379301.732</td> <th>  Durbin-Watson:     </th>  <td>   0.186</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>   <td> 0.000</td>   <th>  Jarque-Bera (JB):  </th> <td>1099689.326</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>            <td>-0.589</td>   <th>  Prob(JB):          </th>  <td>    0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>        <td> 5.475</td>   <th>  Cond. No.          </th>  <td>3.54e+04</td>  \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLSAR Regression Results                           \n",
       "===============================================================================\n",
       "Dep. Variable:     SPEED_TRIP_DEMEANED   R-squared:                       0.126\n",
       "Model:                           GLSAR   Adj. R-squared:                  0.126\n",
       "Method:                  Least Squares   F-statistic:                 4.236e+04\n",
       "Date:                 Sat, 29 Jul 2017   Prob (F-statistic):               0.00\n",
       "Time:                         22:33:08   Log-Likelihood:            -6.4982e+06\n",
       "No. Observations:              3511875   AIC:                         1.300e+07\n",
       "Df Residuals:                  3511862   BIC:                         1.300e+07\n",
       "Df Model:                           12                                         \n",
       "Covariance Type:             nonrobust                                         \n",
       "==========================================================================================\n",
       "                             coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "------------------------------------------------------------------------------------------\n",
       "Intercept                 -0.5037      0.009    -56.149      0.000        -0.521    -0.486\n",
       "bike_lanes[T.True]         0.3195      0.002    134.565      0.000         0.315     0.324\n",
       "sharrows[T.True]           0.2001      0.003     65.126      0.000         0.194     0.206\n",
       "bike_path[T.True]          0.2987      0.003    101.518      0.000         0.293     0.304\n",
       "SIG_DIST <= 35[T.True]    -1.2781      0.002   -536.293      0.000        -1.283    -1.273\n",
       "STOP_DIST < 20[T.True]    -0.4500      0.005    -96.745      0.000        -0.459    -0.441\n",
       "C(RDCLASS)[T.2]            0.0928      0.003     35.061      0.000         0.088     0.098\n",
       "C(RDCLASS)[T.3]           -0.1428      0.003    -40.975      0.000        -0.150    -0.136\n",
       "CONGESTION                 0.0828      0.004     22.738      0.000         0.076     0.090\n",
       "speed_limit                0.0102      0.000     45.632      0.000         0.010     0.011\n",
       "LANES                      0.0519      0.002     24.873      0.000         0.048     0.056\n",
       "SLOPE_TF                 -20.1754      0.051   -392.492      0.000       -20.276   -20.075\n",
       "lane_cap                   0.0002   8.71e-06     21.961      0.000         0.000     0.000\n",
       "==============================================================================\n",
       "Omnibus:                   379301.732   Durbin-Watson:                   0.186\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):          1099689.326\n",
       "Skew:                          -0.589   Prob(JB):                         0.00\n",
       "Kurtosis:                       5.475   Cond. No.                     3.54e+04\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 3.54e+04. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "formula = \"SPEED_TRIP_DEMEANED ~ \" + \" + \".join(explanatory_road)\n",
    "model = sm.glsar(formula=formula, data=data, rho=1).fit()\n",
    "\n",
    "#data[\"predicted\"] = model.predict()\n",
    "#data[\"residual\"] = data.SPEED - data.predicted\n",
    "\n",
    "print(\"Actual R-squared: %f\" % (1 - np.var(model.resid) / np.var(data[\"SPEED\"])))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Because we aren't making a more complex model of how cyclists behave around intersections, \n",
    "we're just assuming that signalized intersections have a constant effect on speed when cyclists \n",
    "are within a certain distance of them.\n",
    "This makes intersections a \"black box\" in this model.  It would be very interesting to see\n",
    "an analysis in the future that tries to do a better job of modelling how cyclists' speed\n",
    "varies around intersections (e.g. taking into account what kind of turn they do, how far away\n",
    "they are from the intersection on approach/departure from it, what kind of intersection it is,\n",
    "etc.)\n",
    "\n",
    "Below, we show how we could select a cutoff for the \"black box\" intersection\n",
    "dummy variable.  We simply select a variety of cutoffs to use in the model, \n",
    "and choose the one that gives the highest model R-squared.  It can be seen that\n",
    "the R-squared peaks at a cutoff of 35m on either side of the intersection, so this is\n",
    "the value that we use in our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x1c1bc996e10>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAElCAYAAAAcHW5vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzP4/DxjyTCkTZ5gAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c12b0455c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cutoffs = range(5, 101, 1)\n",
    "r_squareds = []\n",
    "for cutoff in cutoffs:\n",
    "    explanatory_road = (\"bike_lanes\", \"sharrows\", \"bike_path\", \"SIG_DIST <= %d\"%cutoff, \"volume\",\n",
    "                        \"I(volume/(LANES*lane_cap))\", \"speed_limit\", \"LANES\", \"C(RDCLASS)\", \"SLOPE_TF\")\n",
    "    formula = \"SPEED ~ \" + \" + \".join(explanatory_road)\n",
    "    test_model = sm.ols(formula=formula, data=data.sample(frac=0.1)).fit()\n",
    "    \n",
    "    r_squared = test_model.rsquared\n",
    "    r_squareds.append(r_squared)\n",
    "    \n",
    "plt.plot(cutoffs, r_squareds)\n",
    "plt.ylim(0,)\n",
    "plt.title('Plot of intersection \"Black Box\" Cutoff Distance\\nvs. Model R-Squared')\n",
    "plt.xlabel('Intersection \"Black Box\" Cutoff Distance (m)')\n",
    "plt.ylabel('Model R-Squared')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x1caf05eec50>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAElCAYAAAAcHW5vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREQSU/AQEZHEFDxERCSx/w88wo7Dswa/YAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1cadbf5f668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cutoffs = range(5, 101, 1)\n",
    "r_squareds = []\n",
    "for cutoff in cutoffs:\n",
    "    explanatory_road = (\"bike_lanes\", \"sharrows\", \"bike_path\", \"SIG_DIST < 35\", \"STOP_DIST <= %d\"%cutoff, \"volume\",\n",
    "                        \"I(volume/(LANES*lane_cap))\", \"speed_limit\", \"LANES\", \"C(RDCLASS)\", \"SLOPE_TF\")\n",
    "    formula = \"SPEED ~ \" + \" + \".join(explanatory_road)\n",
    "    test_model = sm.ols(formula=formula, data=data.sample(frac=0.1)).fit()\n",
    "    \n",
    "    r_squared = test_model.rsquared\n",
    "    r_squareds.append(r_squared)\n",
    "    \n",
    "plt.plot(cutoffs, r_squareds)\n",
    "plt.ylim(0,)\n",
    "plt.title('Plot of intersection \"Black Box\" Cutoff Distance\\nvs. Model R-Squared')\n",
    "plt.xlabel('Intersection \"Black Box\" Cutoff Distance (m)')\n",
    "plt.ylabel('Model R-Squared')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We test for heteroskedasticity in our data using the Breusch-Pagan test using the following code.\n",
    "\n",
    "While the model is statistically significant, the R-squared of the model is\n",
    "very small.  \n",
    "However, with such a small R-squared, we can be pretty confident that \n",
    "heteroskedasticity is not hugely affecting our estimates; \n",
    "we are working with several million data points, so even this very\n",
    "small amount of heteroskedasticity creates a statistically significant relationship."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared: 0.011030\tF-stat: 19583.580176\tp-val: 0.000000\n"
     ]
    }
   ],
   "source": [
    "hetero_model = sm.ols(formula=\"I(residual**2) ~ predicted + I(predicted**2)\", \n",
    "                      data=data).fit()\n",
    "\n",
    "#hetero_model.summary()\n",
    "print(\"R-squared: %f\\tF-stat: %f\\tp-val: %f\" \n",
    "      % (hetero_model.rsquared, hetero_model.fvalue, hetero_model.f_pvalue))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared: 0.025801\tF-stat: 7750.784244\tp-val: 0.000000\n"
     ]
    }
   ],
   "source": [
    "hetero_model = sm.ols(formula=\"I(residual**2) ~ \" + \" + \".join(explanatory_road), \n",
    "                      data=data).fit()\n",
    "\n",
    "#hetero_model.summary()\n",
    "print(\"R-squared: %f\\tF-stat: %f\\tp-val: %f\" \n",
    "      % (hetero_model.rsquared, hetero_model.fvalue, hetero_model.f_pvalue))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Durbin-Watson Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Durbin-Watson statistic: 1573901\n",
      "9234689.21283\n"
     ]
    }
   ],
   "source": [
    "1.0 / np.sum(data.fixed_residual**2)))\n",
    "print(np.sum(data.fixed_residual**2))print(\"Durbin-Watson statistic: %d\" % (np.sum( np.diff( data.fixed_residual )**2.0 ))) #* "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While the above indicates that the data is not particularly heteroskedastic with respect to the\n",
    "predicted value, we do expect to see heteroskedasticity with respect to one of the predictors:\n",
    "the intersection black box dummy variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAg4AAAGpCAYAAADlSr7yAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2a4bdb665c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(x=\"SIG_DIST\", y=\"residual\", data=data, bins=\"log\", kind=\"hex\", xlim=(0,400))\n",
    "sns.plt.xlabel(\"Distance to Nearest Intersection (m)\")\n",
    "sns.plt.ylabel(\"Model Residual (m/s)\")\n",
    "sns.plt.title(\"Histogram of Model Residual-Intersection Distance\\nobservations\")\n",
    "cbar = sns.plt.colorbar()\n",
    "cbar.set_label(\"log10(Counts)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logarithmic Model\n",
    "\n",
    "As an economics student, my instinct is to model the elasticities (% changes in outcome as a result of a change in predictors) by regressing on the logarithms of the continuous variables.\n",
    "\n",
    "**NOTE** The naive logarithmic model can't work, because we have observations of a speed of 0. The literature seems to point to the Box-Cox method of taking log(x + c), where c is some constant (here, half the smallest nonzero value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>np.log(SPEED + 0.5 * 0.000014)</td> <th>  R-squared:         </th>  <td>   0.041</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                          <td>OLS</td>              <th>  Adj. R-squared:    </th>  <td>   0.041</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>                    <td>Least Squares</td>         <th>  F-statistic:       </th>  <td>   749.8</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                    <td>Wed, 22 Feb 2017</td>        <th>  Prob (F-statistic):</th>   <td>  0.00</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                        <td>22:15:51</td>            <th>  Log-Likelihood:    </th> <td>-3.0734e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>             <td>175590</td>             <th>  AIC:               </th>  <td>6.147e+05</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>                 <td>175579</td>             <th>  BIC:               </th>  <td>6.148e+05</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>                     <td>    10</td>             <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>             <td>nonrobust</td>           <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                   <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>                      <td>    1.0712</td> <td>    0.034</td> <td>   31.236</td> <td> 0.000</td> <td>    1.004     1.138</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_lanes[T.True]</th>             <td>    0.1181</td> <td>    0.010</td> <td>   12.303</td> <td> 0.000</td> <td>    0.099     0.137</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sharrows[T.True]</th>               <td>    0.1339</td> <td>    0.012</td> <td>   10.793</td> <td> 0.000</td> <td>    0.110     0.158</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bike_path[T.True]</th>              <td>    0.2218</td> <td>    0.012</td> <td>   18.956</td> <td> 0.000</td> <td>    0.199     0.245</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SIG_DIST <= 35[T.True]</th>         <td>   -0.7206</td> <td>    0.010</td> <td>  -74.913</td> <td> 0.000</td> <td>   -0.739    -0.702</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.2]</th>                <td>    0.0780</td> <td>    0.011</td> <td>    7.289</td> <td> 0.000</td> <td>    0.057     0.099</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>C(RDCLASS)[T.3]</th>                <td>   -0.1020</td> <td>    0.013</td> <td>   -7.765</td> <td> 0.000</td> <td>   -0.128    -0.076</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>I(volume / (LANES * lane_cap))</th> <td>   -0.1243</td> <td>    0.014</td> <td>   -8.843</td> <td> 0.000</td> <td>   -0.152    -0.097</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>speed_limit</th>                    <td>    0.0124</td> <td>    0.001</td> <td>   18.047</td> <td> 0.000</td> <td>    0.011     0.014</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>LANES</th>                          <td>   -0.0161</td> <td>    0.008</td> <td>   -2.148</td> <td> 0.032</td> <td>   -0.031    -0.001</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>I(SLOPE_TF / 100)</th>              <td> -495.5674</td> <td>   20.740</td> <td>  -23.895</td> <td> 0.000</td> <td> -536.217  -454.918</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>241220.250</td> <th>  Durbin-Watson:     </th>   <td>   2.002</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>   <td> 0.000</td>   <th>  Jarque-Bera (JB):  </th> <td>42820229.834</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>            <td>-8.281</td>   <th>  Prob(JB):          </th>   <td>    0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>        <td>77.689</td>   <th>  Cond. No.          </th>   <td>2.74e+05</td>  \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                  OLS Regression Results                                  \n",
       "==========================================================================================\n",
       "Dep. Variable:     np.log(SPEED + 0.5 * 0.000014)   R-squared:                       0.041\n",
       "Model:                                        OLS   Adj. R-squared:                  0.041\n",
       "Method:                             Least Squares   F-statistic:                     749.8\n",
       "Date:                            Wed, 22 Feb 2017   Prob (F-statistic):               0.00\n",
       "Time:                                    22:15:51   Log-Likelihood:            -3.0734e+05\n",
       "No. Observations:                          175590   AIC:                         6.147e+05\n",
       "Df Residuals:                              175579   BIC:                         6.148e+05\n",
       "Df Model:                                      10                                         \n",
       "Covariance Type:                        nonrobust                                         \n",
       "==================================================================================================\n",
       "                                     coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "--------------------------------------------------------------------------------------------------\n",
       "Intercept                          1.0712      0.034     31.236      0.000         1.004     1.138\n",
       "bike_lanes[T.True]                 0.1181      0.010     12.303      0.000         0.099     0.137\n",
       "sharrows[T.True]                   0.1339      0.012     10.793      0.000         0.110     0.158\n",
       "bike_path[T.True]                  0.2218      0.012     18.956      0.000         0.199     0.245\n",
       "SIG_DIST <= 35[T.True]            -0.7206      0.010    -74.913      0.000        -0.739    -0.702\n",
       "C(RDCLASS)[T.2]                    0.0780      0.011      7.289      0.000         0.057     0.099\n",
       "C(RDCLASS)[T.3]                   -0.1020      0.013     -7.765      0.000        -0.128    -0.076\n",
       "I(volume / (LANES * lane_cap))    -0.1243      0.014     -8.843      0.000        -0.152    -0.097\n",
       "speed_limit                        0.0124      0.001     18.047      0.000         0.011     0.014\n",
       "LANES                             -0.0161      0.008     -2.148      0.032        -0.031    -0.001\n",
       "I(SLOPE_TF / 100)               -495.5674     20.740    -23.895      0.000      -536.217  -454.918\n",
       "==============================================================================\n",
       "Omnibus:                   241220.250   Durbin-Watson:                   2.002\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):         42820229.834\n",
       "Skew:                          -8.281   Prob(JB):                         0.00\n",
       "Kurtosis:                      77.689   Cond. No.                     2.74e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 2.74e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "explanatory_road = (\"bike_lanes\", \"sharrows\", \"bike_path\", \"SIG_DIST <= 35\", \n",
    "                    \"I(volume/(LANES*lane_cap))\", \"speed_limit\", \"LANES\", \"C(RDCLASS)\", \n",
    "                    \"I(SLOPE_TF / 100)\")\n",
    "\n",
    "nonzero_min = np.min(data.loc[data[\"SPEED\"] != 0, \"SPEED\"])\n",
    "formula = \"np.log(SPEED + 0.5 * %f) ~ \" % nonzero_min + \" + \".join(explanatory_road)\n",
    "model = sm.ols(formula=formula, data=data).fit()\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On second thought, it doesn't make a whole lot of sense for our variables to have a multiplicative relationship.  If anything, the observation of 0 speed (with a real physical interpretation rather than an indication of unobserved data) should tell us that the log model isn't correct."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest\n",
    "\n",
    "To work in a more complex nonparametric model, we can use random forest regression.  Random forests are quick easy ways to assess just how much information there is in the data; it should tell us if there is some aspect of the speed that we could model more accurately."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Naive R^2: 0.836176\n",
      "Cross-validated(ish) R^2: 0.226547\n",
      "\n",
      "Relative feature importances:\n",
      "bike_lanes, 0.008825\n",
      "sharrows, 0.005564\n",
      "bike_path, 0.005942\n",
      "SIG_DIST, 0.712953\n",
      "volume, 0.195939\n",
      "speed_limit, 0.009443\n",
      "LANES, 0.017817\n",
      "RDCLASS, 0.018783\n",
      "lane_cap, 0.024735\n"
     ]
    }
   ],
   "source": [
    "#explanatory_road = (\"bike_lanes\", \"sharrows\", \"bike_path\", \"SIG_DIST <= 35\", \n",
    "#                    \"I(volume/(LANES*lane_cap))\", \"speed_limit\", \"LANES\", \"C(RDCLASS)\", \n",
    "#                    \"I(SLOPE_TF / 100)\")\n",
    "explanatory_road = (\"bike_lanes\", \"sharrows\", \"bike_path\", \"SIG_DIST\",\n",
    "                    \"volume\", \"speed_limit\", \"LANES\", \"RDCLASS\", \"lane_cap\")\n",
    "\n",
    "train = data.sample(frac=0.3)\n",
    "test = data.sample(frac=0.6)\n",
    "rf_model = ensemble.RandomForestRegressor(n_estimators=20)\n",
    "rf_model = rf_model.fit(train.loc[:,explanatory_road].values.astype(float), train[\"SPEED\"])\n",
    "cval_score = rf_model.score(test.loc[:,explanatory_road].values.astype(float), test[\"SPEED\"])\n",
    "\n",
    "rf_model = rf_model.fit(data.loc[:,explanatory_road].values.astype(float), data[\"SPEED\"])\n",
    "naive_score = rf_model.score(data.loc[:,explanatory_road].values.astype(float), data[\"SPEED\"])\n",
    "\n",
    "print(\"Naive R^2: %f\\nCross-validated(ish) R^2: %f\" % (naive_score, cval_score))\n",
    "print(\"\\nRelative feature importances:\")\n",
    "for exp_var, val in zip(explanatory_road, rf_model.feature_importances_):\n",
    "    print(\"%s, %f\" % (exp_var, val))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Note**: This may seem like an absolutely huge R^2 given our max R^2 of about 0.3something for our linear model with basically all the variables included (including dummies), but it's worth noting that random forests are really good at overfitting data.  The cross-validated score is a much better analogue of the linear model R^2, and it's a lot more in line with what the linear model gave us.\n",
    "\n",
    "We can also see with the feature importances that the distance to intersection plays an outsize role in predicting the speed. This might give us a better place to start looking."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualizing the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/andrew/.anaconda3/envs/uoft2016/lib/python3.5/site-packages/statsmodels/nonparametric/kdetools.py:20: VisibleDeprecationWarning: using a non-integer number instead of an integer will result in an error in the future\n",
      "  y = X[:m/2+1] + np.r_[0,X[m/2+1:],0]*1j\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAGoCAYAAAATsnHAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xl8XHW9//HXmclkT5qlSZq06ZrybQsFCpSWsgtXQFAQ\nRRHFBfHq/Xnv9V6v272IistP5apXf/e6K7gi6nUXRKAIlB1algLtadO96ZZ9X2Y5vz8mCWnIMjOZ\nmXNm8n4+Hnk0y5k5n6TtvPP5nu/5fi3HcRAREfEan9sFiIiITEQBJSIinqSAEhERT1JAiYiIJymg\nRETEk3LcLiAWzc3drk81LC8vpL29z+0yJqTaEqPa4ufVuiAza6uqKrFcKCdjqIOKUU6O3+0SJqXa\nEqPa4ufVukC1ZSMFlIiIeJICSkREPEkBJSIinpQRkyREsk0k4hBxHCwL/D79nigyEQWUSBpFHIc7\n79/Jxi0HcRywgLNX1/LOSw05fgWVyFgKKJE0CUci3H73dh578QhlRQHmFAXo7AvyyNbDtHUP8ME3\nrqYgT/8lRUbof4NIGjiOw/f/9DJPbTtGXUU+bzxvKUUFeQRDEf7w6B5e3tvOrXds4RPvOJ28gKYk\ni4AmSYikxaYXDvPUtmMsmFvAm85fRlFBHgCBHB9Xn7uUVYvL2Xe0h9882OhypSLeoYASSbH27kF+\n+UAjeQEfr1u3kIL83OO+7vNZXLJ2IRUledy/uYlte9tcqlTEWxRQIin28/t20D8Y4uwTKykrLZzw\nmECOj8vPWoRlwQ/+/DK9/cE0VyniPQookRR6Zvsxtuxopr6qgNNXzJ/y2NrKIs46cR7tPUP8+K6X\n0lShiHcpoERSpKc/yM/u20GO3+Li0+djWdOvC3rWifOoKMnjnif2cbC5Jw1ViniXAkokRX71QCNd\nvUOsMxVUlZfE9Bi/z+LCNfNxHLjjXhvHcX0hfxHXKKBEUuClvW08svUwNWV5rDtpQVyPXVpXyvL6\nMrYf6OSFXa0pqlDE+xRQIknW0x/kR3dvx2fBRafVxb1ChGVZvG7DEiwL7rh/B+FIJEWVinibAkok\niULhCN/63VZauwZYt6KSBTVlCT1PTUUhJy+rpLljgMdfPJLkKkUygwJKJInu3LiT7fs7WD6/iHNO\nqZ/Rc521ah5+n8XvN+0mFFYXJbOPAkokCSKOw28f3s0DW5qoLsvjsjMXxzRrbyqlRbmcvKyStu4h\nHlMXJbOQAkpkhgaHwnz7dy/y58f2Ul4c4PUbFpI/brWIRK1fVYPfZ/EHdVEyCymgRBLkOA7PNbbw\nmR89zeYdzSysLuC6ixuonFOctHOUFOZyasNc2nuGeOSFw0l7XpFMoNXMReIUiTi8uKeNvz61n237\n2rEsOK2hjAtOW5SSPZ3Wrarh+V0t/PGR3Zy9upZAjn6vlNlBASUyTnffEJGIQ27AT47fIhhyGBgK\nsfdINzsOdLDZPkZr1yAAS2uLOPukGmrnzklZPcUFAU5tmMszdjObXjjEa06L774qkUylgBIBXtjV\nwoPPHmLvkS46eoamPDY3x8fJS+Zw0tIKFlQnNo08XutW1vBcYwt/fGQP555cSyBHe0ZJ9lNAyazW\n1TfEL+7fyZMvHwWgpCCHZXVF+C2LUMQhEnHI8Vv4fRYVpbnMn1tCfU1Z2ofZigoCnLa8iqe2H+Oh\n5w5x8Rkzm8IukgkUUDJr7TvSxad/+BSdvUPUVeZz0Zo6aqtSN1Q3U2eurObZxhb+9Ogezjuljlzt\nvCtZTldbZVY62NzDTd9+lM7eIc5eVcnbLjKeDieAwvwAp59QRXd/iL892+R2OSIpp4CSWedgcw+3\n3vEsnT1DXLymmrNPWYg/BbPvUmHtimpyc3z8+bG9DA6F3S5HJKUy43+lSJIcbe/jq3c+R09/kNdv\nWMBp02wi6DUFeTmcsaKa3oEQ928+4HY5IimlgJJZo61rgK/84jk6e4e44JQqzjplkdslJeQMU0Ve\nwM9fnthH/2DI7XJEUkYBJbNCW9cAX7nzOVq7BtiwqpIzV2XuvUT5uTmsXVFN32CYu5/Y63Y5Iimj\ngJKsd6Stjy/+bDNH2vpYe0I5Z5+c+VO0z1hRRXFBgL8+eYCWjn63yxFJCQWUZLXGg5186Webae0a\n5JwTK7ngtEUzXmXcC3Jz/Fxwah2hiMOdG3e4XY5ISug+KMlKkYjDnx/fyx8f2YuDw2tOreaMlZk1\nIWI6KxeV8+zOFrbsbGXbvnZWLip3uySRpFIHJVnFcRxe2tPGF376DL/ftIfiAj9vPndh1oUTRLeG\nv+j06LW0H9+zTdPOJeuog5Ks0DcQ5NmdLWx6/hA7DnYCsLK+hAtPW0BxYb7L1aXOvIpCTjdVbLab\n+fl9NjdcvsrtkkSSRgElGat/MMRzO1t4attRXtzTRjjiALCsroizVlZTl6aFXN12/il1HDjWwyNb\nj3DikkrWrapxuySRpFBASUbpHwzxfGMLT28/xtbdbaO7zNaU59FQW4xZWMHc8uRtGJgJcvw+3rBh\nMT++x+ZHf9nG/KoiFlTNrp+BZCcFlHheMBTmhV2tPLntGC80tjAUioZSdVkeDXXFnLCwnOryEper\ndFdFaT6XnFnPnx/fx613bOHjbz+d+XOL3C5LZEYUUOJJ4UiEbXvbefLlo2zZ2Uz/YHQCQGVpLsvr\nilmxqILqitkdSuOtWlzBYDDCfc8ciIbUdadRp5CSDKaAEk851NLL37Y08eS2o/T0BwEoLczhxOXl\nrFhUQe3ckqy4jylV1iyfi+M43L/5IF/46TN84MqTWL200u2yRBKigBLXRSIOz+9qYePmg7y8tx2A\nwjw/py6bwwn15SyaV6ZQisNpJ1SRG/Dx16cO8PVfPc/V5y/ldeuz4wZlmV0UUOKa3oEgm54/zANb\nDtLSOQBAfVUBJy8tZ+Xianw+vaAm6qQllVSWFvC7Tbv4zUO72XekmxsuX0l+rv7LS+bQv1ZJu4PN\nPWzcfJDHXzrCUDBCwG9x0uJSTjuhmnmVuq6ULLWVhbzrkhX8/pHdPGM3c7i1j39602qqywvdLk0k\nJgooSYuBoRDPN7by0HNNbN/fAcCcogBnmgpOXT6PwvyAyxVmp6KCANdedAIPbD7As42t3PKjp3Vd\nSjKGAkpSZmAoxKZnm9j49D627modnR6+uKaQ1UvKMIs0jJcOfp/F361dyLzKIu59+pXrUpetX4RP\n16XEwxRQklQDQyFe2NUavZF2TChVlubSUFfMSk0Pd83qpZXMnZPP7zbt5jcP7ealPW289/JVVM7J\n3qWgJLMpoGTGpgql1cvKqa8sZN7cOZpF5gG1lUW869IV3PPkPrbv7+DmHz7JVecs4YI188kN+N0u\nT+Q4CihJSO9AkOd2trBlRzMv7mkjOK5TMgsrqKkopry8iI6OPperlbGK8gNcfd4ytu5u44EtB7nz\ngUb+8uR+Xru2njNX1qijEs9QQElMIo7DwWM9bN/fwdbdrWzf1z66OGvVnDyW1haNhpI6Je+zLIuT\nl1WyfMEcnnr5CFt2tvDrB3fx6wd3sXheCSsWlrOkrpRFNcXMnVOga4XiCgWUjAqFI/QNhOgdCNLR\nPcixjn6OtPWx/2gP+4500zcYGj22tiKfpfOKOGFhBVWzbHHWbFKQl8P5axZw5qp5bN/fjr2/jX1H\nu9l7pHv0mECOj5ryQurmFtJQX05Rrp+K0jzKS/MpL84jkKNt5SQ1FFBZxHEcgqEI/YMh+gZDDAyF\n6RsM0T8cOr0DIXr6g/T2B6N/DoTo7Q/SOxCkbyA0eu1oIhUluSyeV8r8ygKWzK+grFjDQNmkIC+H\nNcurWLO8iqFgmCNtfRxs7qK5o4+O7iBH23o52NzDU9uOveqxpUW5lBQGKMzLib7l51CYFyAv109u\nwEdewE9uwE9ewEdujj/6+Rzf8J/+4a9Hjwvk+I7rwB3HwXGiHXwk4hCOOPh9Fjk5Ps1AnAUyIqB+\ncs92AJzhj52Rd4Y/M/KxM/ZBo59zxn08/jnGPM+rHjtyrEN+XoCBweArjxh/Tmfcc0zyPMd9+fgP\njztm/JccByKRCMGwQzgcIRiOEAo7hEIRIo7DwFCY/sHQ6LBbrPJzfeTn+qkozR1+AfGRF/BRmOun\ntDiPyjmFzKssJk8X0GeN3ICfhTUlLKx5Zbal4zh09wUZCEfYf6iD7t5BegaC9PaH6O4P0drZz6Fg\nZMJ/0/GwiHZsESe6BFZkiiccCaqA30dgOPBGHh/w+/D7LHw+C//I+5aF32/h91mvCsGJjP+0M9UX\niQ6bWtbwn8PfjEX0cwUFAQYHQmCBb/iLPgv+7fq1sf1gZilrsr8cL5k/f/4iYORX9tGCLcs/4a9Q\nlm/iz09i/A/AAnAiYWfc87xy3lc+/6rzOJGwM9H7kxl5rtHn9Pmske9r7Pkty4fl81uW5bOwfJZl\n+SyfP+CzfH7L8uX4LJ/fF/04x2/5c3w+fyDH8vl9Pl8gxx8owJ9baPkD+WFfID9kWT7v/6VLVnAi\nISsSDvmdcNAXCQd94dAA4WC/44SDEScSDkfCobATCYVxHBwnHHGc6G9YluWLvtRbPsvy+aOv51b0\nlyTHCeNEWyvHiYQdxwlHnEgk4kRCjhMJR5xIKOJEwk4kHIy+70ScSDg4cuyr/u1blt8a/r8V/Xjk\n/9nI9+BEXvV/2nHGjDZEIhEHx7KwnIgTiibr2GPHvQ44Tnjsh0NNTU17ZvZTzl4ZEVAiIjL76Oqm\niIh4kgJKREQ8SQElIiKepIASERFPUkCJiIgnKaBERMSTFFAiIuJJCigREfEkBZSIiHhSRqzF1949\nqOUuRCTrlJfkxbws2we+eK/T1NLPDz52YVZtf1JVVTLpN6MOSkQkAwT80ZfrwWB4miOzhwJKRCQD\nBHKijcaQAkpERLxEHZSIiHjSKwE1+cai2UYBJSKSAXKGh/jUQYmIiKeMdlBDCigREfGQQI6uQYmI\niAcF/BriExERDxrtoGbREF9GrCQhIplj584d3PrFz+PzWTQsP4GPfvym474eCoX43C03c+TwIfz+\nHD756c9SVzcfx3H41v98gz//8Xf85b6HADh8+BBvv/ZNrFy5CseB8ooKvvDF/0xp/T/7ye08sPE+\nfD4fN9z4fjacfe5xX5+ozsce2cTPfvojLAscB+zt2/jVb/5I5dy5SasrNxANqP7BUNKe0+sUUCKS\nVF//6pf5t499ghUrVvGpT36CJx5/lPVnnT369XvvuZuSklJu+dwXefKJx/nWf3+Dz3/xVn7yox9S\nW1v3qudbvGgJ3/zOD9NS+6FDTWy8/15+cPvP6O7u4v03vpuzNpyDZb2yGs9EdW4451w2nBMNsoMH\nD/Df3/haUsMJID/gB6B3QAElIml015//wBOPPUpvby/Nzcd469vewRWvv5Lnnt3Md7713+QEAtTU\nzOPfb/o0lmXx2c98kuZjxxjo7+fGv/8HNpxzLv/n/e9l6bIGLMvi9W+4iv/88hfIzc0lkJvLF/5v\ntOv47C0309PdRTgc5sMf+QQnmBW8+Y1XcP4FF/LC889RUlLK177xTX7wvW9zqOkghw8f4lvfvW30\nBfpHt32fp558YrRTsCz46Cc+yeLFSwAIhYIcPnyIFStWAXDuuefz9FNPHBdQzzz9JK+74g0AnLlu\nPV/43KcAeMu111FQUMj3v/vN4342Dq9eivOJxx/l8KFDvPFN14x+7vDhQ9z08Y9Qv2gRB/btY+WJ\nJx7XvUUiEf7xH943WjvAvNpaPvWZz48es+WZpzlrwzn4/X7Kysqpratjz+5dLF3WMHrMZHWO+MF3\nv82N7/vAVH/dCckb7qD61EGJSLrt2bObn97xa7o6O7n+7ddw+RVv4Gtf+TLf/M4PKCkp5X/+33/x\nwP33snbdetav38Bll7+eQ00H+Y9PfGT0t/eGhuVcdfWb+dpXvsSbrrmWSy+7nM3PPE1LSzMb77+X\n1atP5h3vfA/bt73Mf331Vr79vds41HSQy6+4kn/60L/xvhuuZ+fOHUB0KO7b37v9uBrffcP7ePcN\n75v0e+jo6KCkpHT04/KKClpaWo47prW1lbLycgAsy8Jn+QiFQhQUFE74nK2tLdz0iY/Q3NzMm655\nC5dcevlxgTdWY+MOvvzVr1NVVc0N77yOxsYdNDScAIDP5+Nb3526E2ttbaGsrPyV+svKaWlpOS6g\nJqsToKWlmba2VpafYKY8TyLyhjuovoFg0p/bqxRQIh6x5rQzsCyLOWVllJSU0t7exsED+/nER/8V\nx4GBgX7Ky8spKSnlpZde5He//V98Pouurs7R51h14kkAnHf+hdz6xc9zYP9eLrr4EhYtXsK2l1/i\nPe/9ewBWrFxFU9MBAIqKikZfgKuqqunt6T7uuWbCcabfiCDiTL4ywpw5Zbz/H/6RSy+7gp7ubm54\n13WcsXY9lZWVEx5fv3ARVVXVAKw6aTX79+4dDahExFD+ce7+8x+59LLLEz7fVEY7KA3xiUi6OZEx\nL9SOg8/no6qq+lXXX+7+8x/p7u7iez/8MZ0dHbznXW8b/VogEADgjLXruP2nd/LIpgf53C0384//\n/GF8Pt9xgREJR8/n9x//MjByTM7wc4013RBfWVk5nR0do8c3HzvG3LlVxz3H3KoqWltaaWiIdmkA\nOTkTvxQVFhZy+RVXAjCnrIwVq05k3949kwZUZNzPkDHXjsYO8Y18efwQ39yqKvbv2/tK/c1Hqao6\nvv6pPPrIw3w+RZM4cvwWPp9FrzooEUm3rVufx3EcOjs76O3riw41WRZ79uxmyZKl/PqXv2DN6afT\n0dFBXd18AP72wP0Eg69+wfrfX93JhnPO5ZJLL8dxYOeO7axcdSKbn3mKE09azYtbnz9u2CpW0w3x\n5eTksHjJEl54/jlOPuVUHvzbRt5y7XXHHbNu3Vk8sPFe1q0/i00PP8jpp6897utju5bNzzzNI5se\n4kP/+hH6+/to3GGzcNGiSc9/qOkgra2tlJeX89JLW3nTW64d/VosQ3xnnLGOO+/4Ge97/wdpb2+j\npbmZJUuXTXjsRN3Voaam0Q4u2SzLIi/gUwclIulXW1vHv3/swzQ1HeQfPvhPAPzHJz/D52+5mdzc\nXObOreKNb3ozRUXFfPTD/8yLW1/g9W+4iurqGm77wXfHNgssqK/npk98hKLiYvJy8/jkpz9LXl4+\nn7/lZv7xH27EcRw++olPAsc1GaOTIcbOWovXhz78Mb78fz+L4ziceNJqzli7DoCPf+Rf+PJXvs7F\nr72Up558gvff+C5yc/O4+TOfA+Cr//kldjXupLe3hw9+4L2ce94FvOXat3P3XX/kfTdcTyQS4V3v\nuZG5c6t44vFHOdTUxNVvfstx565fuIjvfPP/sXt3I6ecsoYlS5bGVXvNvHlcedWb+MD73o1l+fjY\nv98MHD8pY6I6r73uejo7OykuKUn45xaL/IB/VnVQVixjxG7TjrqS7e768x/YvWsX//ShD7tdSkbo\n7+/jzjt+NnpNDaKz+P7j4//G7T/5hYuVxSeeHXXv37TN+dXDBzjWMcD3PnrBjH6J8BLtqCsiWaWl\npYWL/u6SV33eIjtetCeTn5dDOOIwNEu23NAQn4gHjEwEkNjU1y981edqa+u47Sd3uFBN+uTnRl+y\neweC5OX6Xa4m9dRBiYhkiMK8aEB1982O61AKKBGRDFEwElD9Qy5Xkh4KKBGRDFGYrw5KREQ8SEN8\nIiLiSaNDfH0a4hMREQ9RB5Umxph8Y0yjMeadbtUgIpJJCvLVQaXLzUCri+cXEckoBbl+LBRQKWWM\nMcAK4C43zi8ikoksyyI/168hvhT7KvBhyPJ1SUREkqwgP2fWdFBpX+rIGHM98Jht2/uijZRCSkRk\nOsXFuRSVFFJWks/urk7mlBWSG8ju5Y7cWIvvcmCJMeb1wAJgwBhzwLbtB1yoRUQkI/T0DBEM91Ew\nHEo797ZSXVbgclUzV1U1+RYlaQ8o27ZHdxAzxnwa2KNwEhGJTWlRdKfj9q6BrAioqeg+KBGRDFJc\nkAtAW/egy5Wknqvbbdi2fYub5xcRyTQlhcMd1CwIKHVQIiIZpHQ4oFo7B1yuJPUUUCIiGaSsJA+A\nw629LleSegooEZEMkpvjp6QwRwElIiLeU1laQGdvkP7BkNulpJQCSkQkw1SW5gNwpK3P5UpSSwEl\nIpJhKkpnx3UoBZSISIapGr5Bd9+RHpcrSS0FlIhIhqkpL8SyYPehTrdLSSkFlIhIhgnk+KgszWf/\nsW7CkYjb5aSMAkpEJAPVVhYRDDk0NWfvdSgFlIhIBqqtLARg75FulytJHQWUiEgGGgmoxqbsvQ6l\ngBIRyUBVcwrIzfGxY3+726WkjAJKRCQD+XwW86uKONYxQGdPdq5sroASEclQC6qKAdh5MDuH+RRQ\nIiIZaiSgdhzscLmS1FBAiYhkqNrKQvw+ix0HFFAiIuIhOX4f8yoKOHCsJytXNldAiYhksAVVJTgO\n7MrCZY8UUCIiGWxBdREAOw5kX0DluF2AiLyiccy1hIb6MhcrkUwxf+5IQGXf/VAKKBEXNMZwUVth\nJbHIz82hak4+uw91EQxFCORkz8BY9nwnIhmg8UBHTOEkEo+FNSWEwk7Wbb+hgBJJg5kGk0JNprJo\nXgkAL+/NrmE+BZRIiiUrXBRSMpn66mIsC17e2+Z2KUmla1AiKaJAkXTJC/iprShkz+Eu+gZCFOZn\nx0u7OiiRJEvldSaFnkxmcW0pEQfsLJrNp4ASSRJNgBA3LR6+DvV8Y4vLlSRPdvSBIjM0UbDEMrXb\njUBqPNChaefyKnWVRRQX5PDM9mO8/e9MVkw3V0DJrDdZyKgbkkzi81msXFTB09uP8eKeVtYsr3K7\npBnL/IgVSZCG5CTbrFxUDsATLx11uZLkUECJZCAFq0ykpryA8pJcnmtszorVzRVQMivpBV6ykWVZ\nrFpcQTDk8Mz2Y26XM2MKKJEMpZCViaxeUonPgr88uY+I47hdzowooGTW0Qu7ZLPSolxOXFzBkbZ+\nttjNbpczI5rFJ7OKV8Kp8eDEdTQs0PRxmbl1q2rYuqeNPz22h9NNFZZluV1SQtRBiaRR48GOScMp\nlq+LxKKiNB9TX8aBY71s3Z256/MpoGTWcLt7iid4Yj3W7e9JvOusE2sA+NOje3Ay9FqUAkokDRLp\nitRJyUxUlxfSML+UXYe62Jyh16IUUDIruNlpzCRoFFIyExesmY/fZ/GLjTsYHAq7XU7cFFCS9TI1\nnGJ9Dg3zyWQqSvJZu6Ka9u4h7npin9vlxE0BJZIi6n7EC9avqqG4IId7ntzHsfY+t8uJiwJKsppb\n3UWyw0lhJ4nKDfi5cM0CQmGHX2zc6XY5cVFASdbycjjtauocfUvG82qYT6ayYmEZ9dVFPN/YmlET\nJhRQkpW8Gk4ThVK8QSUSL8uyeO3ahfh9Fj/563Z6+oNulxQTBZRkFTe30IglnGby9enOoS5KplJZ\nms85q2vp7gtyZ4YM9WmpI/G0yV50R3aU9cqL8kzDaexxy+bPSUZJIq+ydkU12/e389iLR1i7oppT\nGua6XdKU1EGJZ4x0P2PfpjvWbalYmmgmw31e+JmId/l8Fq9bvwifBT++Zzt9A97eM0oBJZ7g9gvr\nSNDEEzixHpdI4Ez1GM3ok5moKivgrJPm0dEzxK/+1uh2OVNSQInrvBBOE31uukVdYzGTbkgTJyRV\n1q+soWpOPg8/f4iX9np3MVkFlMxq067SMO7rXlltXJMlZCb8fh+XrV+EZcHtd29jYMibQ30KKHFV\nJryYxjv0NyIZHVCiz5EJP1dx17yKQtatrKGta5D/fXCX2+VMSAEls1YqO6HpgmXngc7Rt5k+l0ii\nNpw0j4qSPB7Y0oS9v93tcl5FASWumY2/5U8USrEG1XhaRFZmKsfv43XrF2EBt921jcGgt1Y8dy2g\njDG3GmMeM8Y8aYx5o1t1yOzkRvc0XQhN9XV1UZIqdXOLOGNFNc2dA/xh0x63yzmOKwFljLkAWGXb\n9gbgMuDrbtQhkmxeCxJ1URKLc1bXUlaUy1+f3s+ew11ulzPKrQ7qIeCa4fc7gEJjjOVSLSIpF+sQ\nXrxdlBdmFErmC+T4uGTdQhwHbrvrZULhiNslAS4FlG3bjm3b/cMf3gjcbdu240Yt4o5M30RwIsnq\nnhK5HjUVdVESi0U1JZyyrJKmlj7u9sjmhq6uxWeMuRJ4D/BaN+sQyWSNBztoWFDmdhmSYsXFuRSV\nFKb0HFee38Duw1386dG9/N36xSycV5rS803Hchx3GhdjzCXALcAltm1P+Stje/eguqss49UVxxM1\nVfeUaEe0vH7iRWMnWkw2loAaWWBXvKO8JC/mSxv3b9rm5BUUpbIcABqbOvntw7tZUlvCTdefgc+X\n2qsvVVUlk57ArUkSpcCtwBXThZPIbJXsoT6RWDTMn8OKhWXsOdzNxs0HXa3FrUkSbwUqgV8ZY/5m\njHnAGLPApVpklsik7ine88XyvelalMTqotMXkJ/r5zcP7aK5o3/6B6SIK9egbNv+PvB9N84tkkyp\nnla+80DnpEN9IqlSlB/gotMXcNfj+/jRX7bzkWtPxbLSP9FaK0nIrDAbpmPPhu9R0mfVonKW1pay\nbV87j2w97EoNCijJeoks8hpLZxTLenvJMNHzaBFZSTXLsnjt2noCOT7u3LiTzp7BtNeggJKsFO8K\n5COhNPaFf6qgSsbQ3uHDhzl82J3fTEViUVqUy/mn1NE/GObXLqx4roCSrJDIjrixGh9eyQqnse9P\nF1SxdmMa5pNkO7VhLtVl+Tz24hF2pvnflwJKMloyAimewIn12KkCZbIwireb8tq6f5KdfD6Lv1u7\nEICf/tUmHEnfMkgKKMlY2dgtJGPIT9twSLLNn1vESUsqONjcy4PPHkrbeRVQkpGSFU7p7kJiCaDJ\njknmZAmReJ1/Sh25AR+/fXgXXX1DaTmnAkokySYb3tOECMlkRQUBzl1dS/9gmN8/vDst51RAScbJ\n1O4pHgoz8aJTl1dRXpLLQ88foqmlN+XnU0CJZBDN5hM3+X0WF5y6AMeBXz+wM+XnU0BJRvF695TM\n4b1UdlHK7/rdAAAfUUlEQVSaKCGJaphfSn11ES/sbuPlvW0pPZcCSiTDeXmoUrKPZVlcuCa6tvcv\nN+4kksItmxRQMut4ceZePI/VNhzitnkVhaxaVM6B5l4ef/FIys6jgJKM4fXrKl4LDq//vCSznXdK\nHX6fxW8e2sVgMJyScyigZFbxynBY59Gdr3oTySSlRbmsXVFNR88Q9z59ICXnUECJjLPzQGfc3VA8\nkyMmC6PJPq8p5+JV61bVUJDn5+7H99LZm/ybdxVQMmvE0j2NDZpYgyqeMJuuU0q0k/JKZyizS17A\nzzmraxkMRvjDpuTfvKuAklkhlr2bJgua6b42mfGdT6zhE8txXrveJbPXycvmUlGSF715t7knqc+t\ngBKJ0UhQjX1LFw3ziVf5fRYXrJmP48Ad9+/ASeK0cwWUZL107Xw7VqLdU6LHJ0I360qyLKsrZUlt\nCdv2dfD09mNJe14FlEgW01RzSQfLsrj49Hr8Pos7N+5kYCiUlOdVQElWc6N7Gm+ibqjjSONxbyKZ\nrrwkjzNX1tDRM8SfHt2blOdUQEnaJTK0lEgn4NbMtumuF00USBN9TvdGSaZZv6qG0sIAf316P4eS\nsNq5AkpmLTeuPU3VLU3XSU0XfJpqLm4L5Pi46PR6IhH4+X0znzChgJKs5MUX62QP5WmquXhRw/yR\nCRPtM54woYCSWcmLM/cg+SEmkm7JnDChgJKs44WJEeMlGjzJuA6lmXySbsmaMKGAEkmCeG+k7Ty6\ni86juyb8mrooyQbJmDChgJJZJR1DexDPskaTB1Ws5xLxomRMmFBAiSTo8OHDMQXG+I5ookCKJ6RE\nMkXD/FIWz4tOmHh5b3vcj1dAyayRzO5pqmCaqnuKNYjGhpruh5JMZVkW559SB8D/PtgYdxelgJK0\nSvX6b6meXh5r1zTi+KCZOpzURUk2qqkoZMXCMvYd7WGz3RzXYxVQItMYCaVYginTuh0tGCvpcM7J\ntVgW/OahXYQjkZgfp4CSWSGe4b2xgeT1SQm6WVcyQUVJPicvreRoez+Pbj0S8+MUUCLDZhpIUy1r\nFOvw3djjNN1cssmGk+bh91nc+9T+mB+jgBJB07dFUq2kMJeldaUcau3j4LHYdt5VQMms51Y4dRyN\nvUMa250ls96G+rKkPZfIdFYtKgfgiZePxnS8AkpmtWS92Mc7OWIknOIJKZFMt7RuDrk5Pp546UhM\nU84VUJL1JptIkMpw0vUjkVcL5Pg4ob6Mtu5BdjV1TXu8AkokxcZPkBjfNY3/ONn3QzUs0DCeeMfK\n4WG+WLbiUEBJRvDqi2yq73tSJybZZn5VEQBNzdNPlMiZ6ovGmCkDzLbt2O+4EskyyQynjqONlNU0\nJO35RLwqN8dPUX4OR9v7pj12yoACQsDIlSxr+E9n+H0H8CdapMhspEkRItH9oppaegmFp+5xpgwo\n27Yn7aCMMcsTrE0kq2lYTmRqZcV5HGzupaVzgNp5cyY9broOCgBjjB+4BJg7/Kk84CZg8QzrFIlZ\nw4KyhHaHXV4/J+lLAmXamnsiXlJekgfAsWmG+WIKKOBnQDlwCvAIsB74dOLliaTGsvlzUr6ieTy0\nQrnIqxXkRaOnbyA05XGxzuJbYNv2pYBt2/Y1wDnA2pkUKJKpvNQ9La+ffHhExKuCoei1p7zcqacx\nxDvNPMcYk2/b9j7gxMRKE5HpqPOSbDYUCgOQF0hOQD1gjPkY8HtgizHmrjgeK5I0Xr0fasRUEyTc\nmMHn9Z+XzE7BYGwdVEzXoGzb/rQxxm/bdtgY8xhQA9w70yJFJHmWzddwn2SG4PD08vxpOqhYZ/Hd\nMPzn2E+/FbgtoepEZqGymgbdByUCDAxFh/jyc6eOoFiH6c4d83YR8B/A+YmXJ5I6E3USE00mqK2t\nTUc5IjJOS2c/gRxrdLr5ZGId4nvP2I+NMYXA7YmXJyIis1Ek4tDWNciCqiJ8PmvKYxOa6GDbdh+g\nhcMkbtogT2R2a+8ZJBxxWFBdPO2xsV6D2sQra/IBzAdeSKw8MMZ8jejNvhHgX2zbfibR55LZJ9EV\nJSZSW1sb975Qc2qWu3ovlIYmJZO1dPQDsKAqSQEFfHLM+w7QBTwfb2EAxpjzgAbbtjcYY1YQnWix\nIZHnEslWc2qWuV2CSEo0dw4Ar2y7MZUph/iMMecNB4oz5g2glOiEiURcRPR+Kmzb3g6UGWOmj1KR\nOMQ6UUJE0mtkH6hFNSXTHjtdB/WF4T/zgNXAdqJbbBjgSeC8BOqbB4wd0msZ/pzm384SDfVlNB5I\nzhCd15TNa9Bq5iKTCEccmlp6qasspKQwd9rjp9tu41wAY8yPgTfYtn1k+ON64HNJqBde2WdKxBXp\nvA412b1Q2qxQplNcnEtRSaHbZczI/iPdhMIOp5pqqqpm3kGNaBgJJwDbtg8YY5YkWOMhoh3TiDog\nvlcHmdWSNUEiHebULJt2XT2Fk8Sip2eIYHj6XWi9bNvuFgAWVhXR3NwNMGVQxRpQLcaYXxDdaiNC\ndFJDoj+pe4HPAN83xpwGNNm23Zvgc4nEJRV7Q8UrnhUlyuZNHl66piaZ5sCx6PUnE+PtJrHeB3Ut\n8ADRa0+rgMeAa+IvD2zbfhzYbIx5FPg68MFEnkcym+6HikpW96R1+MTrQuEIB5q7qSkvYE7x1CtI\njJiygzLG1Nq2fRioBTYOv42oBnoSKdS27f9I5HEi8Yhn88J03w81VTBpirlko4PHegiGHE5pmDv9\nwcOmG+L7KnAd0WAamWI+MqnBAZbGW6RIths/ky+W61Ai2W7XoS4ATl5WGfNjppvFd93wn4lOiBCR\nJJhTs9ztEkQS5jgOjU0d5AV8nBDH8H5M16CMMZcZY94x/P7PjTE7jTFXJ1irSNZIdnBoeE+yUVv3\nIJ29QU5aUkmOP/YlYGM98lPAPcaYy4jeqLsG+Of4yxR5hVsTJSab/ZbMNe7Gz75T8Mhs1ngwei34\n5IbYh/cg9oDqs227Bbgc+Klt2z1AOK4ziUhcpppiLpJJXtrbht9nsWZ5VVyPizWg8o0xHwUuAzYa\nY5YDmtcqnpeO6dfJGuZTlyXZ6Fh7Py2dA5y8rJLigkBcj401oP6e6BYb77JtewC4BPhEfGWKzG7J\nCCBttSGZ5qW9bQBsOGneNEe+WkwBZdv2S8APgZEJ7D+zbfv+uM8mkgQNC1J37SrRAJioi5poiG6y\nkFL3JNkoEnF4eW8bBXl+Tl4W+/1PI2KdxfevRAPqluFP3WyM+eQUDxGRSYwPI4WTZKv9x7rpHQhx\n5soaAjnxb+Ae6yPeRnQH3Lbhjz8KXBH32USSZKZdVCrWsYu1i4oeu2z0bSJjH6d7oCRTbdkRXRz2\n7NWJjUzEGlDdtm1HRj4Yfj8yxfEinhHvRAld5xGZuY6eQRqbOlk8r4RldaUJPUesAbXLGPNpoNwY\nc7Ux5pfAywmdUSQD1NbWJhRU8XRRItls845mAF67th7LSmzbv1gD6hmgF2gC3kF0N12tQi6uSuVk\niRFudFMKNMl0g8EwW3e1MqcowBkrqhN+nlj3g3o9cL1t219J+EwiGSrelc5nssq5SDbYuruVoVCE\nKzYsjmtpo/FiDagCYI8xxgaGRj5p2/Z5CZ9ZJAkaFpSlZYfdkU4q3i05Roxf4Xyq48YaO2Soa2OS\nCcLhCM9sP0aO3+L8U+tm9FyxBtTnZnQWEQ9KZHfdWLupVHdR2k1XvOrFPW109QW5+IwFlBTmzui5\nYgoo27YfmtFZRFIoli4qns0Lp5PI5oYwfReV7mtP2tVYki0cjvD4S0fI8Vtctm7RjJ8v8cFBkVks\n0Vl+8YSQ7n+STPPi3mj3dP6p8ykviW1b96kooCQrpGNG30SmCqnJAmaikNLMPcl04YjD4y8ewe+z\neN36mXdPoICSWS4Z13Jm0kmVzWuIKZw0QUK87sXdrXT1BbkgSd0TKKDERY0HUj/7Ll0mC5CphunU\nNUm2GAqFeWTrYQI5Pl53VnK6J1BASRZxa5gvFaYKNs3gE695ZnszvQMhLjmzPmndEyigZBaZbE2+\nZL3guzUMl45NGUUm0zsQ5MltRynKz0nKzL2xFFDiimwa3htropCKdzbe+OMTDb5s6ijFux578QjB\nUISrzl1KQV6st9bGRgElIiIJaesa4PnGFqrK8me8asREFFCSdl7snpJ5XWcmXdR0x+n6k3jJw88f\nIuLANRc0zGjNvckooCSrzGRYa3n9HM8FQKqua2kVCZmpppZedhzsZEltCaebqpScI7kDhiIeF8uS\nR9OFVCzr9yWyHFIiK0dogoS4wXEcHnz2IABvfc3yhPd7mo46KEkrLw7vxSvRTmuqAEr2skaaICGp\ntPNgJ00tfaxZPpcTUtiNK6BEEjRdSE12LWp8GE0WTuMf77XhR5mdwhGHh55rwmfBmy9YltJzKaAk\n60zXPSRzWCzR0BgJqpl0ThreEzc8t7OZ9p4hzj91PrWVRSk9lwJK0iYbhvcmkorOZiaTIzS8J6nS\n0x9k0wuHKczzc+U5S1J+PgWUSBJMFlLJmoWn4T3xgr8928RQKMKbL2igtGhmmxHGQgElWSmdw3zJ\nluruSVPMJRH7jnSzbV87i+cVc94pyb8pdyIKKJEkSUYXFeuxXg5YyT7hcIT7njmAZcE7L12Bz5ea\naeXjKaAka3mpi5pJV6ThPXHb09uP0dY9yIVr5rN4XmnazquAEkmiqcJkqpBKdAv5sTQ5QlKhs2eQ\nx146QnFBDleftzSt51ZASVZzo4uaLqTGBtF0wTTRc2l4T9Jp45YmQmGHt118AoX5gbSeWwEls54b\nL/jJ6JgSoQkSEo+dBztpbOpk+YJS1q+qSfv5FVAiKZCM60bxPIeG9yTZgqEIGzcfwGfBOy9dmbL1\n9qaigJKsF8uLd7qH+lL5WJFkePylI3T1BbnkzIXMn5vaFSMmo4CStPH68FImXNuZSY1e//mLd7R2\nDfDUtqOUFefyhrNTv2LEZBRQMivEOgSW7JBKZOXzeI/X8J4kk+M43Pf0ASIOvOO1hrxcv2u1KKBE\nxlk2f05Kgmqmx2VChyeZb9u+dvYf62H10grWLJ/rai3asFBmjYYFZTQejH3B2pFAmG6Dw1iNhM9E\nGx6m+pqThvckFoNDYf72bBM5fou3v9a4MjFiLAWUpFVDfVnGrWo+tnNJRlglEkaTdU8a3pNkemTr\nYXoHQrzx3CVUlxW4XY6G+GR2mekL+sjwXyYNt6l7klgcbe9jy45mqsvyuXTdIrfLARRQIglLV1Bl\nUhhKZnIch3ufPoADXH/JCgI53ogGb1QhkkbJHhZLZVBN9bzaWkOS5YXdrRxu7WPtimpOXFLhdjmj\nFFCSdtn6oplpQ38iAH2DIR567hB5AR/XXrTc7XKOo4CSWSmVkwuSFVTqniQdHnruEANDYd547lLK\nS/LcLuc4CiiZtVI9A24mIaVOTNKhqaWXrbtbmT+3kIvOWOB2Oa+iaeYiKRTvvVSxBJO6J0kGx3F4\nYMtBIDoxwu/zXr/ivYpE0ihd9xHFEjzqmiSddhzs5HBrH6ebKk7w6C80ae+gjDF+4IfAMsAPfMS2\n7cfSXYe4y0s37Ma7wkSixgfQrqbOuENJ3ZMkQzji8NBzTfgsePP5y9wuZ1JudFDXAz22bZ8L3Aj8\nlws1iLguFR2Twkli8XxjCx09Q1ywZj41FYVulzMpNwLqp8CHh99vBrwz6V5mrUxYMigTahTvGwyG\nefTFw+QFfK5upRGLtA/x2bYdBsLDH/4LcEe6axCZSLqG+hKhoT1Jli07mukfDHPVuUsoLcp1u5wp\npTSgjDHvJTqM5wDW8J+ftm37PmPMB4E1wOtTWYNIPEaCwKtBJbNXcXEuRSUzG44bCobZvKOZwvwc\n3nbpSgrzA0mqLjVSGlC2bf+Q6ISI4wwH1+XAlcMdlYineCmo1D0JQE/PEMFw34yeY8uOZvoGQlyx\nYRG93QP0dg8kqbrEVVWVTPq1tF+DMsYsBd4PXG3bdjDd5xfvyIQX1YYFZa5e+9F1J0mWSMThqW1H\nyfFbXHx6vdvlxMSNG3XfS3RixN3GmJFhv9fath1yoRaRmIwNinR1VbGGUyYEvbhv+/52uvqCvOa0\n+Z6/9jTCjUkSNwE3pfu84k1euh8qVl4a/hOJheM4PPHyUXwWXHLmQrfLiZmWOhLPG+kQvBZk4zuc\nZAaWuidJpkOtfbR0DnC6qaLKAzvlxkoBJa6brIsa/+Lr1aAa4cYwoEgsXtjVAsAFp853uZL4KKDE\nE+LpBDJhWHAm91RpYoQk02AwzLZ97VSW5rFycbnb5cRFASUZaWw3NVm4uR1iqb7xV8N7Eott+9oJ\nhR3OO6UOn2W5XU5ctJq5ZLSpXqS98AIebzek7kmS7fnGFiwLzjm5zu1S4qaAkqyWiSEV03N64PsS\n72vrGuBoez8nLanw3G65sVBASdbLlBdzdU+SbDsPRjfKXLuixuVKEqOAEkkDhY+4YceBDnwWnLp8\nrtulJEQBJbNCpnRRscim70VSp7svyOG2Pk6oL6O4wNuLwk5GASWSJlN1UeqwJNkam6IzSE87ocrl\nShKngBJxmcJJUmHk+tOa5QooEc/zwtCYwkjSIRxxaGruobaygMo5+W6XkzAFlEiajQ2puO+T8kDI\nivcdbesjGHZYsajC7VJmRCtJiLhAnZSk0v5j3QCYDP+FRh2UzCrqQGQ22H+0BwCzMLPW3htPASUi\nkkVGrj/NqyhgToZsTDgZBZRIhlD3J7Fo6egnGHY4oT6zuydQQImIZJXDbX0ALK0rdbmSmVNAiYhk\nkcOt0YBaUpv5AaVZfCLTmGpoze09p0TGO9zaQyDHR93cQrdLmTF1UDLrxHotp6G+bNpjYzkmGXT9\nSWIxFArT2jnI4nkl+H2Z//Ke+d+BSArEGwgKEPGCo239OGTH8B4ooGSWmixQ0tURiaTCkbbsuf4E\nugYls1iyg6ihviwl16QUmBKrw629ACypLXG5kuRQByUikiUOt/ZSmJ9DVVmB26UkhQJKJInU7Yhb\n+gZDdPYGWVpbimVZbpeTFAooEQ9T4EmsjmTR/U8jFFAiSaZQETe0dPYDUF9d7HIlyaOAEhHJAq2d\nAwDMrypyuZLkUUCJeJQ6MYlHS+cAfp+VNRMkQAElkhIKF0knx3Fo7RqgpryAHH/2vKxnz3cikkUU\ncBKPnv4gQ6EIdVXZc/0JFFAiIhmvrXsQgHkVmb9A7FgKKJEUSbQLUvck8eruHQJg7px8lytJLgWU\niEiG6+yLBlRlqQJKRFJE3ZMkoqs3CEBFaZ7LlSSXAkokheIJHIWTJKqrVx2UiCRAwSOp1tU7SFF+\nDrkBv9ulJJUCSsQDFGIyE32DIeYU57pdRtIpoETSYKoAUjjJTITDEQaDEUqLFFAikqCJdutVOMlM\n9Q2GASgtzL6A0o66ImmmUJJk6huMzuArycKAUgclIpLB+gZCAJQWBlyuJPkUUCIiGaxvMBpQJboG\nJSIiXvJKB6WAEhERDxkJqBIN8YmIiJeMTJJQByUiIp4yeg1KASUiIl7SPxDC54OCvOxa5ggUUCIi\nGW0wGCY/4MeyLLdLSToFlIhIBhsKhsnPzc41FxRQIiIZbCgUIT8Lh/dAASUikrEcxyEYilCQpR2U\na9+VMaYG2AZcZdv2w27VISKSqUJhh4gD+XnZGVBudlC3ArtcPL+ISEYbCkVXMs/P1RBf0hhjLgS6\ngK1unF9EJBsMBSOAAippjDEB4FPATUD2zYsUEUmTVzqo7BziS+l3ZYx5L3Aj4BANIwe4B/i+bdtd\nxhhQSImITKu4OJeiksLjPtfeF13mqLK8kKqqEjfKSinLcZy0ntAY8wjRzs0ClgHHgGts29422WPa\nuwfTW6SISBqUl+TF/Av6/Zu2OXkFRcd9btehTn7z0G7efMEyXrd+UdLrS4eqqpJJfwZp7wtt2z5n\n5H1jzO3A7VOFk4iITGxwKDrEV6BrUCmhzkhEJEGDwWhAFRVk31Yb4OJ9UAC2bd/g5vlFRDJZ/2A0\noArzs3OShNsdlIiIJGhgKLrVRlF+dnZQCigRkQw1MHwNqkgdlIiIeEn/SAeVpdegFFAiIhmqu3eI\n3BwfhVqLT0REvKSrL0hFaV5WblYICigRkYwUDEUYGApTUZrvdikpo4ASEclAXX1DAFQqoERExEu6\nexVQIiLiQe09gwDMLVNAiYiIh7R1RQOqtrJomiMzlwJKRCQDtXYNADCvonCaIzOXAkpEJAO1dQ1Q\nWhigIEvvgQIFlIhIxgmGInT1BZlXmb3dEyigREQyTkdP9l9/AgWUiEjGaRu+/lSbxdefQAElIpJx\n2rqjHZSG+ERExFPaZsEMPlBAiYhknNauAfw+i7lzCtwuJaUUUCIiGcRxHNq6B6kuL8Dny85VzEco\noEREMsjAUJihYISa8uwe3gMFlIhIRunoiS4Sm81r8I1QQImIZJDO3ugMvqosv/4ECigRkYzSqQ5K\nRES8qEMdlIiIeFHXyEaFc9RBiYiIh3T3BckL+LJ6FfMRCigRkQzS0z9EeUme22WkheU4jts1iIiI\nvIo6KBER8SQFlIiIeJICSkREPEkBJSIinqSAEhERT1JAiYiIJymgRETEk7L/VuQkMMbcCpwD+IEv\n2bb9O5dLwhhTAPwIqAHygM/btn2Xq0WNY4zJB14EPmvb9k/crgfAGHM+8GuidVnAC7Ztf8jdql5h\njHk78FEgCHzKtu2/uFwSAMaYG4DrAYfoz+1027ZL3a0qyhhTBPwEKAdyif57u9fdqqKMMRbwHeAk\nYBD4gG3bO9ytKnMooKZhjLkAWGXb9gZjTAXwLOB6QAGvB562bfsrxpiFwH2ApwIKuBlodbuICTxo\n2/Zb3C5ivOF/X58C1gAlwC2AJwLKtu3bgNsAjDHnAde4W9Fx3g1st237JmNMLfAAsNLdkkZdCZTa\ntn22MWYp8A2i/3clBgqo6T0EPDn8fgdQaIyxbNt2dQkO27Z/NebDhcABt2qZiDHGACvwXmhCtAPw\noouB+2zb7gP6gA+4XM9kPgVc53YRY7QAq4ffrwCaXaxlvOXAUwC2be82xizywutHplBATWP4H1L/\n8Ic3And76R+XMeZRYD5whdu1jPNV4INEf7v1mlXGmN8TfTH7rG3b97td0LDFQJEx5g9AGXCLbdsP\nuFvS8YwxZwD7bds+5nYtI2zb/qUx5t3GmJ1Ef26Xu13TGFuBfzHGfINoWC0B5uKtEPUsTZKIkTHm\nSuA9wD+6XctYtm2fTXQY4edu1zLCGHM98Jht2/uGP+WljmUn8Bnbtq8iGp4/NMZ45Rc1i2hoXkX0\n39rt7pYzoRuJXvv0jOHrdvts214OXAR80+WSRtm2fQ/RDuoh4J+BbXjr/4OnKaBiYIy5BPh34FLb\ntrvdrgfAGHOaMWYBgG3bzwM5xpi5Lpc14nLgSmPM40Rf0D5pjHmNyzUBYNv2Idu2fz38/m7gCNEO\n1AuOEg12Z7i2bg/9nY64AHjM7SLGORv4K4Bt2y8AdcOTEzzBtu1P2bZ9rm3bHwQqvNR9ep1XfnP0\nLGNMKXArcJFt251u1zPGecAi4F+NMTVAkW3bLS7XBIBt29eOvG+M+TSwxytDVcaY64Ba27a/aoyZ\nB1QDTS6XNeJe4PbhWaMVeOjvFGB4AkK3bdsht2sZpxFYD/zOGLOIaI2eGIY3xpwMfMi27fcaYy4F\nNrtdUyZRQE3vrUAl8Kvh38oc4J22bR90tyy+Q3R46mEgH/g/LteTKf4I3DE8ZBsgOu3XEy+4tm0f\nMsb8L/AE0X9nnhpOBmoBL/72/13gNmPMg0RvBXm/u+UcZytgGWOeJHot++0u15NRtB+UiIh4kq5B\niYiIJymgRETEkxRQIiLiSQooERHxJAWUiIh4kgJKREQ8SQElAhhjfmqMeacxpsYY88tpjn1bnM99\nkTHmbzOrUGT2UUCJjGHb9lHbtt86zWG3GGPi/b+jGw5F4qSVJCRjDW8++HlgH9FVotuJrpn4c+AF\n4EXbtr9kjPkCsAEoAB6ybfvjw6uC3AacOPz44uHnXAQ8Ytt2vTGmiuiCrXOAENGVHa4BGoCNxpg3\nEt276VPDJQWB99m2vc8Yc9VwbQeILsUjInFSByWZ7jTgI8OrurcSXcx0BdEVy79kjHkzUGfb9oW2\nba8HlhtjLie699IJtm2fCbwTOGXMc450O18E7rJt+1yiIfQO27Y/M/y11xDdIfXbwBtt274Q+B+i\n24wA/DdwtW3blwGRFHzfIllPHZRkupds2z4y/P5jRLeqaLNte6RruRA4yxjzANFtDkqJdlu5w8dj\n23b/8Fpp461jOHBs294EbBr+/Mi25ycRXZ/ut8MdmQ+IDO+Mmz9ma+8HeGVDPRGJkQJKMt3YUYCR\nxXyHxnxuEPiebdtfG/sgY8y/cXxn45/guR2mHmUYJLoP0XFbiRhjKmN4bhGZhob4JNOtGN5uBOAc\n4A/jvv4IcLUxxg9gjLnZGLMMeJnoFg0YY0qIdkvjPQZcOnzMucaYsRsIBoAdwFxjzInDx5xnjLnR\ntu1WIDx8HogOJ4pInNRBSaZ7GfiiMWY50WtQm8Z+0bbt3xpj1gGPGWNCwBZgN7AHuM4Y8wTRSRIT\nbcJ3M9H9md4w/PHIlib3AM8AbwDeQXTbk/7hr/398J//AvzBGLOL6C6+IhInbbchGWt4Ft/nbNs+\nz+1aRCT5NMQnIiKepA5KREQ8SR2UiIh4kgJKREQ8SQElIiKepIASERFPUkCJiIgn/X+h4Bn3o/By\newAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f60682b80f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(x=\"predicted\", y=\"residual\", data=data.sample(frac=0.05), \n",
    "              kind=\"kde\", ylim=(-5, 5))\n",
    "sns.plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/andrew/.anaconda3/envs/uoft2016/lib/python3.5/site-packages/statsmodels/nonparametric/kdetools.py:20: VisibleDeprecationWarning: using a non-integer number instead of an integer will result in an error in the future\n",
      "  y = X[:m/2+1] + np.r_[0,X[m/2+1:],0]*1j\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAGoCAYAAAATsnHAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niYiIKymgRETElf4/FeH3MeAGRzgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f6067f41c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(x=\"SLOPE_TF\", y=\"residual\", data=data.sample(frac=0.05), \n",
    "              kind=\"kde\", xlim=(-0.04, 0.04), ylim=(-5, 5))\n",
    "sns.plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Andrew\\Anaconda3\\envs\\uoft2016\\lib\\site-packages\\statsmodels\\nonparametric\\kdetools.py:20: VisibleDeprecationWarning: using a non-integer number instead of an integer will result in an error in the future\n",
      "  y = X[:m/2+1] + np.r_[0,X[m/2+1:],0]*1j\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.JointGrid at 0x2a4e646d748>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAakAAAGpCAYAAAA3LMlbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niIi4lkJKRERcSyElIiKu9f8Bnjzj238UsmoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2a4e646d320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(x=\"SIG_DIST\", y=\"residual\", data=data.sample(frac=0.05), \n",
    "              kind=\"kde\", xlim=(0, 500), ylim=(-5, 5))\n",
    "sns.plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "grouped = data.groupby(\"SOURCEOID\")\n",
    "link_aggregated = grouped.aggregate(\n",
    "    {\"residual\": {\"resid_mean\": np.mean,\n",
    "                  \"resid_std\": np.std,\n",
    "                  \"rmse\": lambda x: np.sqrt(np.mean(x**2))},\n",
    "     \"predicted\": {\"predicted_mean\": np.mean},\n",
    "     \"SPEED\": {\"speed_mean\": np.mean,\n",
    "               \"counts\": len,\n",
    "               \"speed_dev_from_network_mean\": lambda x: np.mean(x) - data.SPEED.mean()},\n",
    "     \"APP_USER_I\": {\"unique_users\": lambda x: len(np.unique(x))}})\n",
    "\n",
    "link_aggregated.to_csv(os.path.join(MODEL_DIR, \"link_with_model_data.csv\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbYAAAEZCAYAAAD7QwBLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1cae8cf5f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x=\"SPEED\", y=\"PURPOSE\", data=data)\n",
    "sns.plt.ylabel(\"Trip Purpose\")\n",
    "sns.plt.xlabel(\"Observed Speed\")\n",
    "sns.plt.title(\"Plot of distributions of speed based on trip purpose\")\n",
    "sns.plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x1ce79af1b00>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbYAAAEZCAYAAAD7QwBLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1cae8cf5d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x=\"residual\", y=\"PURPOSE\", data=data)\n",
    "sns.plt.ylabel(\"Trip Purpose\")\n",
    "sns.plt.xlabel(\"Residual from Predicted Speed\")\n",
    "sns.plt.title(\"Plot of distributions of model residuals based on trip purpose\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dates = pd.DatetimeIndex(data.RECORDED_A)\n",
    "data[\"time_of_day\"] = \"Off Peak\"\n",
    "data.loc[(dates.hour >= 8) & (dates.hour < 10), \"time_of_day\"] = \"AM Peak\"\n",
    "data.loc[(dates.hour >= 16) & (dates.hour < 19), \"time_of_day\"] = \"PM Peak\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x2a488146b38>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaEAAAEZCAYAAAA+MZraAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhEREJDFKQiIikpj/D/kAgUxjrQS/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2a482046278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x=\"SPEED\", y=\"time_of_day\", data=data)\n",
    "sns.plt.ylabel(\"Trip Time of Day\")\n",
    "sns.plt.xlabel(\"Observed Speed\")\n",
    "sns.plt.title(\"Plot of distributions of speed based on trip time\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x2a500d4bcc0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaEAAAEZCAYAAAA+MZraAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3X2cFWX9//HXngVpuTEXRPMmvO+zoKZfFdG+JoZKit+k\nNDVRMtHS0h9ZSd9CTTK1kjTt5utNijcIGqa1ZpSU91oqat4in1BRTLzhZgkU5GZ3f39cM4fZwzln\nz7J7ds6y7+fjwYMzM9dc85nrzM5nrpk5M1XNzc2IiIikIZN2ACIi0n0pCYmISGqUhEREJDVKQiIi\nkholIRERSY2SkIiIpEZJqIzMbAczW2dmz0T//hn9OzWafoqZ/bGEeq4zs/9q47L3MrNXzOwpMxtU\npNwAM2uKPn/OzK5spd5RZvbDAtOy85vZA2Z2TBtj3tzM7ksMP2Nmm7eljo5kZueZ2RtmdkMnLOsF\nMzu4lTIlbS8bufwfmtnJecZnt4921N1kZv3LVb4jdMR6Fqm7xXadZ3qbtvNK+ztprx5pB9ANrHT3\nfeIBM9sWeNHMZkejSvmh1uHANW1c7tHA/e7+tVbKVcUxuPsfgdZ2ckOB2nwTSpy/mP5R/XF9+xQp\n2xnGASe6+99TjiOpLD/sc/cLC0zKbh/t0Nb50/jxYkesZyEttutcG7GdV9rfSbsoCXUyd19oZvOA\nTyTHm9l2wNXAjtGom939cjO7GNgWmGZmX3b32TnzXQB8CVgL/Av4f8ChwDeAjJnVuPvYnHmOAS4G\nPgCeSow/Bfiiu38uKnMe0Bj9mwCsAc6M6v0P8ApwGtAHWAbcEs8fVXmMmX0fqAGmu/ulZrYD8KK7\n94uWmRyeAvQ2s2eA/YB1wJbuvjTPep7t7u+Z2QPAP4D/BgYBj7j7l82sGvhlNH4N8BpwqruvLLHd\nbwe2B24wsx+4+x057XRstF47AguAXwNnA7sBP3f3Kwp9P+7+rpkNjta3BnCgd6L+TwE/icY1AZPc\nfSYFRPFkvwd3P9TMTgO+Tti5LomW62Z2EHA54SxIM/Bjd/+9md0IvODuV5SyfeQOm9kngF9FMWwL\nPAuc4O5rohgws60J28iAqMqZ7v6DPKtUBVxqZkOjzxe4+5/MrHf0Xe1G2BGvAMa4+7x826u7Pxr1\nEK4C9gB6AvdF05oKrWee9v00cBnhu1oTxXNvtP5fiL6j3YDVwJfdfU5OFbnb9SrgD8AngZOB2cCW\nwOeAE6PvZjvg38Ap7v5OK/WtS8xf6nY5jrCPaLF9FGqDctLpuE5mZgcCuwBP5EyaBtzn7p8EDgLG\nmtnx7n4+sJDwx5abgE4FPgvs6+57Ay8BN7n7dELP6bd5EtBWwA3AF9x9KPBGThzx0eBlwNfdfX/g\nAuAQd38yUe8FUbkhwMHufmjO/AD9gP2BA4GTzeyzecokh08l6jm6e1M8vsB63pyYf2d3Hw7sCYww\ns+HRMg9x972i9XyN8Eefq1C7f4n17X5HnvkOIuwgdgO2JuxwRwBHEXZsheK+MbHca6PxVwE7RPNs\nQdjJnOzu+wGjgWvMbPs8MSRlv4fotN6XgYPcfV9gMnBXVG4ScHnUJqcBI5KVtGH7yB0+nbDt/Tdh\nh7dz1BbJMl8FXo3W62BgVzPrV2B9XoliHwvcbGYDgCOBBnf/lLvXERLH2VH5DbbXaPzPgaeiddkH\nGAh8u4T1jNujP3AHYSe9N/AV4Nbo4IloPc5y9z2BvxMO1nLlbtc9gXp3H+zuT9OyTT8VrcfuwDOE\nA6nW6kvOX8p2ORw4hfzbR6dTEiq/3onrQS8AlxB2bG/FBaIjvP8G/g/A3ZcDNxH+6GJVeeo+ArjR\n3T+Mhq8i7ISL9XAPAp5PHPVcW6DcbcAfzOw3hKPOywqUe97dPygw7Xp3b3b3FcDvCKcVN0Zr6/lH\nAHd/n9A76w+8AKwzsyfM7CLgLnd/PFlpO9odYLa7L4w+zwdmRZ9fBXpFdeeL+9BoB/hJYGq03L8D\nL0ZlDgS2IbT9P4GZhCP7fAk0Kfk9HEU40Pl7VMdlwBZRgpsB/NrMbgX2BSbm1FPq9pHrf4HFZjaB\n0FvZBugbTYvb8C/AsWb2J+AM4HvRtpHPNQDu/hIwBzjQ3e8kJKSzo2uPhySWUWh7/R/gjKgdniac\nxtqzDes5DJjn7k9F8cwBHmV9knva3d+OPj8TLbsUjxYYP8vdX40+/wYYWWJ9sVK2y1EU3j46nU7H\nlV+La0IF5DsYyBCOmNoyXzXhOy2044Rw1JSc3pivkLtfEF2QH0k4+vse4Ugy1/tFlpWsu4pwSqo5\nJ+7Niswfa209VyWmNQNV7v4fM9ubcGQ5AvitmV3l7lcVqTce11q7Qzj1krS2DXHH8n0P1cAcdz8w\nnmBm2wDvEU7dFJL8HqqBqe7+/UQd27n7MuA6M7ub8L0eCUwysz0T8+Z+P40505IxJ7+726P5ZgD3\nEE6NttgO3f0pM9sJOIzwncw2s9G5BwfRcpI3CVQBa83sTOBrhN7BNGAp0WnUfNurme0btcVxcbJJ\nXMA/lBL+Dsi/jVQTtpG15Nn2CtSTq9Dfzbqc5RSKq5BStsti20enU0+o/FrdKKMj+MeBswDM7KOE\n0ynxUcw68u8Y7wVOjY5uAMYDD7l7vg0v9giwe2LH85XcAmZWbWbzgb7ufh3h3HGdmfUsEks+X47q\nqwVOIBzVLwN6mlldVCZ5B906wh9ILG67Nq+nmR1FOP//D3e/iHAtYq9kmRLafWO1Fvd7hFNJp0fL\n3YdwdE4Uz27RdQiiRDqPcJ2lVLOAE83sY1Ed3wD+Fn1+DNjH3W8h9EY+SssbTR4BhhTYPhYBe5jZ\nZlEv9HOJaSOBi6JTl1WEHkTyu8TMfgz8wN3vdvdzCKcnW1wbjVTFy43aZlfC6evPEnqWNxLa5HNA\ndaHtlZDw7wW+HdXVi9BrPosS/g4ij4dZbb+ojt2BTwMPFiifT+52nW99Y4dGBx0Qvp98N/oU+jtp\nTVyu4PaRBiWh8iv1jpuTgcPM7HnChn9HtKOAcBHzt2Z2WM48NxA2nifN7CVgb4ofLePui4ExwHQz\ne4pwLaI5p0wj8M2ozNOEo9tTo53+fcDRZnYVxTUD/4nmfxS4yt0fiU55fRf4i5k9QcsjvbeBf5rZ\nnOhcfBxXsfUsdI1iJuEUV3wn4oGE6yG5TqJwu5f63RWKoVjcYwg7gucIF9TnQPb7ORaYbGbPEq59\nneTub5YYC+4+C/gp8Neoji8RLqBDuGZxUfS93Ee46WFBYt5820dsFvAQ4UaKh4DnE9MmEk6HPUk4\nvfkgIXkk2+NKYG8zez76Tl4jnEbL1QzsHF14v45wXWMZ8DPgzGj8Xwmn13ZtZXsdTzgl/gLhZonn\ngMtaWc9kWy4BjgN+FW0jtwJfcfdX8pUvoNB2nVzf2L+BqWY2J4rpnI2oL1+92eFWto9OV6VXOYiI\npM+iuy7d/ei0Y+lM6gmJiEhq1BMSEZHUqCckIiKpURISEZHU6HdCBaxb19jc0LCy9YIpq63tTaXH\n2RViBMXZ0RRnx+oqcQ4c2K/UW8YB9YQK6tGj2G39laMrxNkVYgTF2dEUZ8fqKnG2lZKQiIikRklI\nRERSoyQkIiKpURISEZHU6O44aZdLL51EQ8PSomVWrVpJU1Mzffr0KVqutrY/EydO6sDoRKTSKQlJ\nuzQ0LGXJkiVU9awpWKZ5bXja/erGwnduxmVEpHtREpJ2q+pZQ99dCz9z8f1X7gYoqYyIdC+6JiQi\nIqlREhIRkdQoCYmISGqUhEREJDVKQiIikholIRERSY2SkIiIpEZJSEREUqMkJCIiqVESEhGR1CgJ\niYhIapSEREQkNUpCm5gZM6YxY8a0tMOoeGonkcqgJLSJmT37CWbPfiLtMCqe2kmkMigJiYhIapSE\nREQkNUpCIiKSGiUhERFJjZKQiIikRklIRERSoyQkIiKpURISEZHUKAmJiEhqlIRERCQ1PTpjIWa2\nI/AzoD/QE3gO+J67v29mOwEzgceBq4FpwAx3Py8x/xrg0WiwJzAH+Lq7N7UhhlOAOnf/fvvXqKXT\nTz+ZpqYQSnV1Nc3NzdlhqUxLliwGYNy4MSlHUrqqqiqam5vzDvfq1QuApqYmampqGDhwawYN2oGG\nhqX8619z6dGjB3379uOdd94GYKeddqFfv34A1NUNAeDdd99h6NADAFiw4HXmzp2Tnf7uu+/w9NNP\nAjBq1NHZaePHn8vUqVNoaFhKXd0QBg3aMRvfwIHDmDp1CgBjx45r8XnWrJnMnTsnu+z486BBO2aX\nvWLFCgYN2oGhQw9oUS6Oae7cOcye/Thbb/0xRo4clR0G2HrrjzFo0I7Z+Yp54YUXWLZsZd6y8fJi\npdRXyNy5c1iw4PVW40quY3K+vn0/Qv/+H2vTvG2ZlmzLzlT2JGRmHwHuBsa5+1PRuC8DtwGfAz4N\n3OPuE8zsAuBKd/91TjWL3X1Eos7bgVHAPW0Mp7n1Im2XTDiNjY3lWIRIiwSUO7x69ers57Vr17J8\n+XLmz3+1xba5fPny7OdXX52X/ew+N6rjQxYufAuABQveYNWqldnpq1d/mK2rvv6u7DSAhx66n6am\nJtznMmjQDtnxn/70MB566H4gJJ7k57iOeNnx50GDdmix7PnzX2XhwreyO876+juBsCOtr7+TefOc\nXr0+wsiRo7LDAL16fYRBg3YoKWlMnz6dtWsb85aNlxdrTxKqr7+TBQveaDWu5Dom56uqgo9/vG3z\ntmVasi07U2f0hI4CHowTEIC732JmZ5rZwcD3gRozWwGcCqw2s3+7e32ijqr4g5n1BPoC75vZ5sAN\nhB4WwHh3f8nMzgKOAXoDi4EvJObfEvgDcIG7P9DelTv99JPbW0VZjBs3hilTpqcdRkXqSr2f9ii1\nN55MKO4vF52eOzxhwvjsckIiWT//xRdfnJ12/vkTsp8vueTCbB3JunLnj9fB/eXsUXw8fdasmdnP\nq1atZOrUKS3mjetK9rbymTt3Di+++GL2c7Ls3LlzNointfqKLSeuq1hcyXK569zWeQutS7Fpq1at\nZNasmZ2aiDojCe0MvJpn/OvR/z8mnCa7yMyqgLdzEhBAfzO7P/rcBMx09wfN7CfA39z9WjPbFbiR\n0LMa4O6HApjZX4Ch0bwfI/TKxieTYntU8mm3CRPGl30ZDQ1Lae6AS4vNjWtoaPiwU2KWjhOf1szn\niSfWP6U87mFBy15YqXJ7JPX1d7UYjntZ+eYrpeeQr2zuMkupr5TlFKsnN55CdZUyb6F1aW096+vv\n2uSS0FvA/nnG7wosAHYqoY4lydNxCXsCnzGzEwi9pdpo/Bozuw34ANiOcB0J4AhgIZ10LUxERIrr\njJ1xPTDRzPZLXBM6HVjk7q+b2fAS6qgqMP5l4Cl3v93MBgKnmdmewOfd/QAzqwGeTsx/EzAVuMPM\nhrr7qnasFwCZTKZie0OTJ/+i7MuYMGE8S5evbL1gK6qqN6N2896dEnN3OR3XGQYM2LJgb2jYsGHZ\n3tC2226X7Q3tsstube4NjR59LACXXXZxNHwMt99+a3b68OEjeOCBvxWcr1i96+s8tuC0UusrZTnF\n6skXT6kxlLoura3n6NHHtLo+Hanst2i7+weEGxAuMLNHzOwfhNNjJ7ahmkI3FFwKnGBmDwB/Bl4E\nXiFcL3oE+Cuh57NtIp6XCYno521dl3yuv/7W1gulQNeDCusubZPJlPbnXVPTm5qa3mQyGcwGYzaY\nmpreLaYn60pOmzz5F9lpNTW9s/ObDeb8888nk8mQyWS4+OLJ2c/nnffDbB3xspPzJ+uPY6qrG0Jd\n3ZBs3SNHjsJsMJlMhpqa3owdOy47HI+L5yumrm4Ie+yxR96yyeXF/zb2xoS4rtbiSi4zuc41Nb3p\n3btt87Z1Wtxum+KNCbj7fGB0gWk3Jz7/sECZbQuMX0ripoOEw1qJ56fFprdVsjekW7SlXCrxFm0I\nvZB8t2jH0/J9Hj36mDbdor1+vmNbfI5vK04Ow/pbtEsxZswYli3L35vf2J5PobriW7RbK5dvvvgW\n7bbM25ZpybbsTFW5t31KVvOiRSvSjqFVAwf2IxlnfGG/M05rxctbunwlfXc9umCZ91+5G6DVMv07\n6XQc5G+n3LasVIqzYynOjjVwYL9Cl0/y0hMTREQkNUpCIiKSGiUhERFJjZKQiIikRklIRERSoyQk\nIiKpURISEZHUKAmJiEhqlIRERCQ1SkIiIpIavdJgEzN06LC0Q+gS1E4ilUFJaBNz/PEnpR1Cl6B2\nEqkMOh0nIiKpURISEZHUKAmJiEhqlIRERCQ1SkIiIpIaJSEREUmNkpCIiKRGSUhERFKjJCQiIqlR\nEhIRkdQoCYmISGqUhEREJDV6gKm0W/PaVbz/yt1FpwMllOnd0aGJSIVTEpJ2qa3t32qZVaugqamZ\nPn2KJZneJdUlIpsWJSFpl4kTJ7VaZuDAfixatKL8wYhIl6NrQiIikholIRERSY2SkIiIpKbVJGRm\nx5lZz84IRkREupdSekJHAvPM7NdmNrTcAYmISPfRahJy93HAEOAfwA/N7GkzO9fMtip7dCIiskkr\n6ZqQu68E3gAWAJsDewH3mdnZZYxNREQ2ca3+TsjMLgFOBOYDU4Bz3P1DM9s8Gver8oYoIiKbqlJ+\nrNoIHOru85Mj3X25mR1RnrBERKQ7KCUJXQocaWYHAVVANbCTu//A3WeXNToREdmklZKE7iQ8WXJX\n4BHgYMJNCiIiIu1SShIyYDfgKsI1oXOB35UzKKksl146iYaGpa2W++CDDwDo06dPi/HV1RkaG5ta\njKut7V/Sc+dEZNNWShJ6192bzWwu8El3v8XMepU7MKkcDQ1LWbJ0MZma4ptL0+p1AKzJrC1ebtW6\nDotNRLq2UpLQS2b2S+BqYJqZbQvoCQrdTKamB7VHDCpapuEvCwBKLiciUsrvhL4OzHD3OcCFwDbA\nmLJGJSIi3UIpPaHBwNZmdhTwgrsXfj2miIhIGxRMQtFjeX4H7AHMA5rDaPs7cJK7L+ucEEVEZFNV\n7HTcL4FHga3dfZi7HwBsDTwPXNkZwYmIyKat2Om4T7r7CckR7r7GzCYCz5Y3LBER6Q6K9YQ+zDfS\n3ZuBpnzTRERE2qJYEmreyGkiIiIlKXY6bnczey3P+CrCbdoiIiLtUiwJfaLTohARkW6pYBJy9zc6\nMxAREel+SnqzqoiISDkUTEJmdnBnBiIiIt1PsZ7QrwHM7MlOikVKMGPGNGbMmJZ2GF2O2k2kMhW7\nMWGhmf0b2DLnLrkqoNnddy5vaJLP7NlPAHD88SelHEnXonYTqUzFktCRwPbAH4GjOyccERHpTord\nHdcELAD2MrM9gEOi8g+6ux7bIyIi7dbq3XFmdjJQD+wM7AD83szGlTswERHZ9JXyPqFzgf3dfQmA\nmV0CPAhMKWNcIiLSDZTyO6HqOAEBuPti9ABTERHpAKX0hJ4zsyuBG6Lh04DnyheSiIh0F6X0hL4K\nrCacfrsJWAN8o4wxiYhIN9FqT8jdVwH/2wmxiIhIN6Nnx4mISGqUhEREJDWl3JiAmfUBdgFeAHq7\n+wclzDMcmAG8FI36CDDN3X9tZjcBo4Gt3H1tVH4f4CngEHd/OFHPhcAY4K1o1GbAee7+UCmxJ+qZ\nD5i7r2nLfKUYN24MAAMGbMmyZQ1kMhnWrl2bnT5gwJasWrWSLbaoZauttubNNxcAsGTJ4o4ORQqI\n2zr+rjpKdXU1PXr0oG/ffqxatZJtttmOZcsaeP/9FfTt24+1a9ew8867MnLkKBYseJ1Bg3akrm4I\nc+fOYcGC13n33XdoaFhKbW1/hg49gLq6IQDZ6XH59pg7dw5Au+vpKrrb+nZ1rSYhMzsUuBaoBj4F\nPG9mJ7n7rBLqv8/dx0T1bAa4mU0lvB58IeHRQHdHZccArxao53J3vy6qpw6YBuxbwvKTyv5K8nhH\n19jYmHf8ypUrWbjwrQ3mk66rsbGRxsZGVq9eDcCrr87LTovHPfvsM6xatYoFC95g0KAdqKsbQn39\nnSxY8AarV39IU1MTmUyGhQvfyu444+lx+faor78T6D475e62vl1dKT2hS4GDgD+7+9tRD+c2oJQk\nVJX4vDmwLvpHVMcY4G4zqwL2AWaXUM8AYAWAmR0HfCuq81F3n2hm2wFXA70IryE/393jRIeZnQkc\nBpwY98Lao6OPrNuy3ClTpqey7K4mre8oyf3l7P+zZs3MDseamppwfzl7FJ8sP3funI3eoc6dOydb\nV3vq6Sq62/puCkpJQhl3f8fMAHD3OfHnEowws/sJvZA1wNnuvjKafzZwrJnVEHpY9wOFtphvm9kJ\nhB/JNgBfNbNaYBKwr7t/aGa3RL02gJ+5+8NmdmBU5m5CIhsP7AUc5+5l7xmV24QJ46muztDYWN7f\nDjc0LKUp03HN1bSmkYYPlzJhwvgOq7Mrqa+/q8i0O/OO29idabK+9tTTVXS39d0UlJKE/m1m/wM0\nm9kWwFmEB5uWIns6Lo9mwjPpPk/omfwI+HGBstnTcTEzGwoMBGZGPam+hOtWjwLnm9lpUdGeidkO\nBdZtCglIRGRTUEoSOgO4Cvg48BpwH/C1Dlr+bcCVQJO7v96GHhbAfEIyPNzdG83sFOCfhGR2nbvf\na2ZfAU5JzDMauN7MznD3aztkDVI0efIvGDiwH4sWrSjrciZMGE/DqmUdVl9ms2pqa7Zg8uRfdFid\nxVTC6bik0aOP4fbbby0w7VgALrvs4g3Gbdyyjs3W1Z56uorutr6bglZv0Xb399z9RHcf6O793f04\nd3+7ncttjup2YEvW35yQr4eSt9cSPcPuCuBhM3scOAL4F3AHcLmZPQgcTriGlKznm8B3zGyXdq4D\nQGrXZXQ9qHSV0FZmg6mp6Y3ZYEaOHJUdzmTCn2Amk8FsMHV1Q6irG9KifHtOKcV1tbeerqK7re+m\noJS7474IfB+oTY5v7c2q0S3UeW+jdvdxic9DE583OGR194uKLGM6kLuHuT36l1s2jncN8IlisW8s\n3aLd/WzMLdoQjtLz3aIdi6fH5duju/UIutv6dnWlnI67HBgLvFHmWLqszjzS7q4X89trwIAtAYqe\nAiz3qc3kkXnc4ylWtqOO5Ltbj6C7rW9XV0oSeoVw+7Ne3yAiIh2q1J7QA2b2EOt/41P0NJmIiEgp\nSnl23CWEJxk0En5rE/8TERFpl1J6Qj2TNxKIiIh0lFKS0D1mdjbwF8KdZQC4e6k/WBUREcmrlCR0\nQvT/dxLjmoGit2iLiIi0ppQ3q+7UGYGIiEj3UzAJmdkkd59kZlPyTdd1IhERaa9iPaGno//b9PI4\nERGRUhVLQnsAf3T3mzsrGBER6V6KJaHjKPxqBUnJ0KHD0g6hS1K7iVSmUu6Okwpy/PEnpR1Cl6R2\nE6lMxZLQ3mbWmGd8FdDs7tVliklERLqJYknoOXf/r06LREREup1Snh0nIiJSFsWS0B2dFoWIiHRL\nBZOQu1/amYGIiEj3o9NxIiKSGiUhERFJTau/EzKznsBZwAjCm1X/DFzv7s1ljk1ERDZxpfxY9Xqg\nBvgNoef0ZWB34JwyxiUiIt1AKUlomLvXxQNm9kfgxfKFJCIi3UUp14TeNLNdE8NbA2+VKR4REelG\nSukJ9QSeM7OHgUbgIGChmd0P4O4jyhifiIhswkpJQhfmDE8uRyBS2ZpWraPhLwtaLQOUVq6mw0IT\nkS6s2JtV93H3Z4C8d8G5+8Nli0oqSm1t/5LKfdD0AQB9avq0GF9dnaGxsWn9iJrS6xSRTVuxntCZ\nwNeAH+aZ1ky4ZVu6gYkTJ7Vr/oED+7Fo0YqOCUZENikFk5C7fy36OMPdr+6keEREpBsp5e64s8oe\nhYiIdEul3JjwZnQn3BPAqniku19UtqhERKRbKCUJPZ74XFWuQEREpPspdnfcKe5+s7vnuzFBRESk\n3YpdE/pmp0UhIiLdkl7lICIiqSl2TWh3M3stz/gqoNnddy5TTCIi0k0US0KvAKM6KxAREel+iiWh\nNe7+RqdFIiIi3U6xa0KPdVoUIiLSLRVMQu5+dmcGIiIi3U8pP1YVabNLL51EQ8NSIDxFe/ny8ADT\nPn36bFC2trZ/ux+SKiJdk5KQlEVDw1KWLllM30zobK9uCq9y6LlmdYty7zc1bTCviHQfSkJSNn0z\nGU7+aHhv0K3/Cb2ieDgWjxeR7kk/VhURkdQoCYmISGqUhEREJDVKQiIikholIRERSY2SkIiIpEZJ\nSEREUqMkJCIiqVESEhGR1CgJiYhIapSEREQkNUpCIiKSGiUhKWjGjGnMmDEt7TAKqvT4RKR1SkJS\n0OzZTzB79hNph1FQpccnIq1TEhIRkdQoCYmISGqUhEREJDVKQiIikholIRERSY2SkIiIpEZJSERE\nUqMkJCIiqVESEhGR1CgJiYhIanp0xkLM7LvAOcCO7r7GzG4CRgNbufvaqMw+wFPAIe7+cGLeC4Ex\nwFvRqM2A89z9oTbGMB8wd1/T3vWpNLNmzQRg5MhRLcZPnToFgLFjxwEwd+4cZs9+nAUL3gBg6NBh\nLcrPnv0E/fr14803F7BsWQNNTU3069ev3OFvtCVLFgMwYcJ4evXqxfvvr6Bnz83o1asXAFtttTXv\nvfcuq1ev5uMfH0RtbX8A3F8G4OSTT6Wubghz584B4J13erNs2Urq6oaksDYi3VOnJCHgJOA24ETg\nZqAZWAgcCdwdlRkDvFpg/svd/ToAM6sDpgH7tjGG5jaW7zLq6+8CNkxCDz10P7A+CdXX38m8eU5T\nUxMACxe+1aL8qlUrN6h75coNx1WaOBnlSq7fkiWLyWRCxz9e//r6O6mrG0J9/Z0A9OxZzdq1jUpC\nIp2o7EnIzIYDrwDXALcSkhCEpDQGuNvMqoB9gNkFqqlKfB4ArIjqPg74FrAOeNTdJ5rZdsDVQC9g\nG+B8d48THWZ2JnAYcGLcC+vK6uvrs8lj1qyZ2UQ0deqU7M526tQpDB16QLYHEMuXdHKtW7eOuXPn\nVNyO+ZxzzmzzPHF7xNxfZtasmRu0SyWur8imqjN6QqcD17v7PDNbY2b7E3ols4FjzawG+BRwP1Do\nL//bZnYk+mHbAAAOMUlEQVQC0AQ0AF81s1pgErCvu39oZreY2aFR+Z+5+8NmdmBU5m5CIhsP7AUc\n5+6bRM9o+vTp2c/19Xdlk1DcC4o/5/Z62uKKK37CRz+6RZvmaWhYSnXOTj+fD5ua+KBhKRMmjG9T\n/cuXL29T+ULiXmTLcXcqCYl0krImITPbAhgFDDSz8cDmwNmEnkszUA98ntAz+RHw4wJVZU/HJeoe\nCgwEZkY9qb7ALsCjwPlmdlpUtGditkOBdZtKAhIR6erK3RMaS+gF/S9A1OuZz/rTbrcBVwJN7v66\nmbWl7vnAAuBwd280s1OAfxKS2XXufq+ZfQU4JTHPaOB6MzvD3a9tx3pVjDFjxnD99dcDMHr0Mdnx\nw4eP4IEH/pb9PHToAVx22cUbtYxvf/t7be4ZTJgwnjUNS1st95FMhs1q+zN58i/aVP8555zZIb2h\n0aOP4fbbb80Zd2y76xWR0pT7Fu1xwNR4wN1XAXcCh0fDDmzJ+psT8vVQ8vZa3H0xcAXwsJk9DhwB\n/Au4A7jczB6MljMgp55vAt8xs102eq0qyOjRo6mp6U1NTe8WNyaMHTuOTCZDJpNh7Nhx1NUNwWxw\n9uI8kJ0v/pdPjx49KvLU1JVXXtPmeeL2iJkNZuTIUZgNxmwwe+yxB2aDK3J9RTZVZe0Juft/5Rl3\nFnBWYnho4vOYPOUvKlL/dGB6zujbo3+5ZXeOPq4BPtFa7F1JsgeUNHz4iJxyx7b5Fu3evfMnp0oy\nYMCWG3WLdtzjif/fYotwi7aIdJ7OukVbyij31uxYfGt2rK5uSNGj/Nx62nqzQGcbMGBLgDafyssV\nt8nAgf1YtGhFu+MSkdLpiQkiIpIaJSEREUmNkpCIiKRGSUhERFKjJCQiIqlREhIRkdQoCYmISGqU\nhEREJDVKQiIikho9MUEKyn2sT6Wp9PhEpHVKQlLQ8ceflHYIRVV6fCLSOp2OExGR1CgJiYhIapSE\nREQkNUpCIiKSGiUhERFJjZKQiIikRklIRERSoyQkIiKpURISEZHUKAmJiEhqlIRERCQ1SkIiIpIa\nPcBUyub9piZu/c/S7GcgO5ws07/TIxORSqEkJGVRW7s+tVRXZ1i7fAUAm/Xp06Jc/5yyItK9KAlJ\nWUycOCn7eeDAfixatCK9YESkYumakIiIpEZJSEREUqMkJCIiqVESEhGR1CgJiYhIapSEREQkNVXN\nzc1pxyAiIt2UekIiIpIaJSEREUmNkpCIiKRGSUhERFKjJCQiIqlREhIRkdToKdp5mNkXgC+6+0nR\n8DDgKmAt8Fd3vyjN+JLM7N/Av6LBf7j7eWnGk2RmVcD/AXsBHwKnu/tr6UaVn5k9DfwnGpzv7qel\nGU+uaBv8ibt/xsx2AW4CmoAX3f2sVIOL5MS4N3AP67fNq939jvSiAzPrAUwBdgQ2Ay4B5lBhbVkg\nzjepvPbMAL8BjNB+ZwKraWN7qieUw8yuJHzpVYnR1wBfcvdPA8PMbK9UgssR7YyedvcR0b+KSUCR\nzwO93P1TwPeBK1KOJy8z6wWQaMdKS0ATCH/svaJRVwAT3X04kDGz0akFF8kT477A5Yk2TXWHGTkZ\nWOzuBwNHAL+iAtuSlnEeSYhzHyqvPT8HNLv7QcAFwKVsRHsqCW3oMeDr8YCZ9QM2c/fXo1H3Aoel\nEFc++wLbm9n9ZnaPmX0i7YByHAT8BcDdnwD2SzecgvYC+pjZvWb2t+iIvpK8AnwhMbyvuz8Sff4z\nlbE9bhAjcJSZPWRm15tZnwLzdaYZhJ0lQDWwDtinAtsyGWeGcAZmX+B/Kqk93b0e+Fo0uAPQwEa0\nZ7dNQmY2zsxeMLPnE//vm+cIY3NgeWJ4BfDRzos0yBcv8DZwqbuPAH4M3NrZcbVic9af4gJYF3Xh\nK81KYLK7f5ZwADKtkuJ0998TdpixZC89le0xV54YnwAmREfErwGT0ogryd1XuvsH0YHlHcB5VGZb\n5sZ5PvAkcG4ltSeAuzeZ2U3AL4DpbER7dttrQu4+hXDetTXLCTvTWD9gWVmCKiJfvGZWQ/SH7+6P\nmdk2nR1XK5YT2iuWcfemtIIp4l+EI3ncfZ6ZLQG2Ad5KNarCkm2YyvZYgj+4e3wA8nvCTip1ZvZx\n4C7gV+5+u5ldlphcMW2ZJ86PVmJ7Arj7V8xsK2A2UJOYVFJ7VszRXqVy9xXAajPbKbrQ/lngkVZm\n6ywXAucARNep3kw3nA08BowCMLMDgBfSDaegccDlAGa2LeGP5+1UIyruGTM7OPp8JJWzPSbda2bx\n6ddDgafTDAbAzLYmnE7/rrvfHI3+Z6W1ZYE4K7E9Tzaz70WDHwKNwFNmNjwaV1J7dtueUBudSehq\nZoBZ7j475XhiPwFuNbOjCOeNv5JuOBv4PXC4mT0WDZ+aZjBF3ADcaGaPEHoZ4yq0xxY7F/iNmfUE\nXgZ+l3I8+Xwd+KWZrQHeYf21gzR9H9gCuMDMfgA0A98kxFlJbZkvzm8BV1ZYe95F+Lt5iJBLxgNz\ngevb0p56iraIiKRGp+NERCQ1SkIiIpIaJSEREUmNkpCIiKRGSUhERFKjJCQiIqnR74QkdWa2A+Gp\nBS8RHvuRIfxg9BZ3n7QR9V0HXOPuz+SMvxF4wN1v2Yg65wPD3X1BYlyG8HysbYGz3P3httZb4rKb\ngGcJvxfpQXgSxdfd/cWNrG84MCl62vVvCE9kfqZA2a8Cy939t22of4O2isafBZxO+I6bgJ+7+9SN\nWYc2xNLk7jrYrmBKQlIp3nL3feKB6BFE88zsNnf3tlTk7uX4IV++H9RtD+zu7tuXYXktlp3TNl8D\nbiE8WXmj6wRw96+2Uu5TwAMbU3eSme0PnAYMc/c1ZjYQmG1mz7p7OZ+koR9CVjglIalU20b/rwAw\ns/8Fjif0ku519+9FD3i8Ddg6KvtDd7/HzB4ALnT3h83sCuAoYCHhyckPRD2vB919p6juCwk7+ovM\n7GzCo/R7E47WT4iSYPLBjLE/Alua2ZPABGByVO5F4BuEVxvsRXicyeXuPtXMToni2S76dxUwCBgB\nLAaOdPc1rbTNw6x/zNADwFJgCHAC4Zl3FxH+tucDX3X3BjMbSXjM/iogm9Rz2uqnhNdvrAWuI/RM\njwY+Y2ZvA88B1xKSbxPhkf33mVkt4eG52xN+Jf+RPDF/LPq/L7DU3ReZ2ReBRVEc7xHel7Mvoad3\nkrsviB5V83PCM8kWA2e4+xvRa0yuBvoTHkA73t2fjb7bW4E+hIeoSoVTN1UqxXZm9oyZvWxmiwg7\n0s+7+0Iz+yxh57Qf4eh/ezM7ifDqgPnuPhQYC3w6WaGZHUtIAoOB44BdE5PzHa33I+x0h7v7J4F6\nQjIp5GhgobvvHw3vCnzG3U8lPOV4sbvvSXjW1yQz2yMqNxQYCRxMSCZ/cve9CAnss620E4Qk+Whi\n+Dl3H0xItD8BRrr7vsAs4DIz24zworFjorZalWfdvwgcCOwODCM8AuoF4G7gB+7+V0LCvCGqYzRw\nbfRKgYsI77XaC/g16w8Kkv4MvAG8bWYPRol/qbu/E03fErg/quO3rH+UzvXAie6+HyGJXh+Vv5nw\nlO79gDOA26PxvwKmRD3H+HFRUsHUE5JKkT0dZ2aXA59k/Wmgw4D9CQ9trCIcab9BeKr4JWa2PfAn\n4Ec5dR4C3BU9B26xmc0sFoC7r4iS24nRu5mOAP7ZhnVwd38/+jyC8GBU3H2Jmf0himcF8Ji7fwB8\nYGbNwP3RPG8AtXnqrTKzZwjrvhmht3FGYnp8xD+M0Kt6IHrYbgZYAuxJaN/4rZw3ExJH0nBghruv\nI3rPDoCZJcscFkZZ3M7VhMR7CPClaF0fMbMN3p7r7muBL5jZzoQEPAo418wOdfcngVXuHr+K5GbC\nq0k+AewC3B2tD0DfKPENJTy3LB7f28z6J2MBprE+aUmFUhKSSvRdwoX4c4GfEnZ2V7r7lQBmtjmw\nzt1XmlkdIVkcDXyH0OuJNdOyt7+uwPiewJoomT0I/BKYSXhQ5N5tiDvZw8g9y5Bh/d9bi9NtJTws\ntcU1oSLLrQYecffPA0Q9oH6EF45VJ8qvY0NrkwPRaa1FOWUywAh3XxaV+RjwHhu2Z2Nu5WY2lpAI\n7ye8qfgaM7uY0IN9kpY90/hFbhngtcTBSRWhl1VNSFrJ62TbuvvS6CaODIC7N0fDUsF0Ok4qRfaa\ni7s3EhLQedF7Su4HxppZHzPrQThN9sXobquL3P1O4CxgYJSgYn8DjjOzzaLrFkdE45cBW5jZAAuv\n9o7HDwXmuftVhHejHEnLnXfRuHPcR7gQj5ltSTh99WAb5m9rGQg9ogPNbLdo+ELCdarnCW2zZzT+\nxDzzPgwcY2Y9zKw34Y242xISVpw87ye0M2Y2hHC6robQzidH44fS8rRnrBq41MwGROV6EHo68V15\nvaOnwUPoQc4kXLuqNbODovGnA9PdfTnhppWToroOj+InimVsNP5Y1r9uXCqUkpBUihbXaNz9XuAf\nwMXufg/hsfFPEHaoz0S3Wd9COD30PGEHf2G0g4rv/LobeIhwo8AfCBfaicr8DHiKcN0kPp11L1Bt\nZi8Bfydc2N8pX3yF4k74ETAgEdvF7v5sK/O3dRktprn7u4Qd+Awze47Qi/tOdIrtRMJrP56i5YvH\n4rb6A2GdnyG0x8/d/RXCTn2imR0D/D/ggKju2wg3D3xASHa7mtkLhF7sq7lBuvtNhLeEPha173PA\nC+5+Y6LYcVHdhwPfim7QOA643MyeJSSXcVHZk4HTo/KXEG5aIYrx2Kj8EbR8K7JUIL3KQURSp9/z\ndF/60kWkEuhouJtST0hERFKjnpCIiKRGSUhERFKjJCQiIqlREhIRkdQoCYmISGqUhEREJDX/HziF\nIpnxBSfXAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2a4f20846d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x=\"residual\", y=\"time_of_day\", data=data)\n",
    "sns.plt.ylabel(\"Trip Time of Day\")\n",
    "sns.plt.xlabel(\"Residual from Predicted Speed\")\n",
    "sns.plt.title(\"Plot of distributions of model residuals based on trip time\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Andrew\\Anaconda3\\envs\\uoft2016\\lib\\site-packages\\statsmodels\\nonparametric\\kdetools.py:20: VisibleDeprecationWarning: using a non-integer number instead of an integer will result in an error in the future\n",
      "  y = X[:m/2+1] + np.r_[0,X[m/2+1:],0]*1j\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0, 2000)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAERCAYAAABsNEDqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XmUXOV95vHvraX36m5JXZKQAC1I+klCSAIBFkJYDIYY\nsBhwFjPOMIlJMDmBOImZZBKYEI9zQsg5JNhmfLAHQ7Bx5iTYDNiOnWA7IWBJDgKDViRe7eySet+7\nq7rqzh91W5Ra3eq6UvX+fM7pU1X3ve+9b70q1VN3e6/n+z4iIiKFiox1A0REZGJRcIiISCgKDhER\nCUXBISIioSg4REQkFAWHiIiEEhtuBjPzgEeBVUAPcIdz7lBe+U3A/UAaeNI59/hQdcxsNfAI0Af0\nAr/hnKs3sy8DVwLtwWJvds71PxcRkXGkkC2OW4BS59w64F7g4f4CM4sFr68FrgbuNLPkaep8Gbjb\nOXcN8BzwJ8H0NcDHnXPXBH8KDRGRcaqQ4FgPPA/gnNsKXJpXtgzY75xrc86lgU3AhkHqrAnmv9U5\ntyt4HgN6gq2TxcBjZrbZzG4/y/ckIiIjqJDgqAZa8173mVlkiLIOoAZIDJieMbOIc+4YgJmtA+4G\nvgRUktt9dRtwPXCXma04g/ciIiKjoJDgaCMXBCfqOOeyeWXVeWUJoPl0dczsVnLHP250zjUCXcAj\nzrke51wH8AK5YyMiIjIODXtwHNgCbASeMbO1wK68sr3AIjOrJRcAVwEPBWWn1DGz24A7gaudcy3B\nfEuAp4MD5zFyu7m+eboG+b7ve55XQNNFRCRPUb44veEGOcw7Q2plMOl2cscsKoMzqD4BfCFo0BPO\nua8PUuczwEGgHniL3G4sH3jJOfdFM/vvwK1ACnjKOffYMO326+t1/LxYkskE6s/iUF8Wl/qzuJLJ\nxOgExzil4Cgi/ecsHvVlcak/i6tYwaELAEVEJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgo\nCg4REQlFwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEo\nOEREJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhKLg\nEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhBIb6wacieee/zk9KW+s\nmzGiSiN9XPWR1WPdDBGRU0zI4PCjFcTKy8a6GSMqm2od6yaIiAxKu6pERCSUYbc4zMwDHgVWAT3A\nHc65Q3nlNwH3A2ngSefc40PVMbPVwCNAH9AL/IZzrt7MPgvcGSzjAefcj4r5JkVEpHgK2eK4BSh1\nzq0D7gUe7i8ws1jw+lrgauBOM0ueps6Xgbudc9cAzwF/YmazgM8BVwDXAw+aWbwI701EREZAIcGx\nHngewDm3Fbg0r2wZsN851+acSwObgA2D1FkTzH+rc25X8DxGbmvkcmCzc67POdcG7AdWntW7EhGR\nEVNIcFQD+Udq+8wsMkRZB1ADJAZMz5hZxDl3DMDM1gF3A186zTJERGQcKuSsqjZyQdAv4pzL5pVV\n55UlgObT1TGzW8ntvrrROddoZoMto2W4RiWqJvdZVV66l2QyMfyMRTKa65rs1JfFpf4cfwoJji3A\nRuAZM1sL7Mor2wssMrNaoAu4CngoKDuljpndRu4g+NXOuf5weAX4SzMrAcqBpcDu4RrV3tFTQNMn\nLi/VQ319+6isK5lMjNq6Jjv1ZXGpP4urWCFcSHA8B1xnZluC17eb2aeByuAMqnuAnwAe8IRz7gMz\nG1jnM8Hura8AbwHPmZkPvOSc+6KZPQJsDpZxn3MuVZR3JyIiRef5vj/WbQjt2Z9u9zNM8l1VqVY2\nXL58VNalX3XFo74sLvVncSWTiaIMuaELAEVEJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgo\nCg4REQlFwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEo\nOEREJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhKLg\nEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISSmy4\nGczMAx4FVgE9wB3OuUN55TcB9wNp4Enn3OMF1HkYeNM591jw+svAlUB7MMvNzrn+5yIiMo4MGxzA\nLUCpc26dmX0EeDiYhpnFgtdrgG5gi5l9H1g/WB0zqwOeAhYDb+atYw3wcedcU5Hel4iIjJBCdlWt\nB54HcM5tBS7NK1sG7HfOtTnn0sAmYMMgddYE81cBXwC+3b+AYOtkMfCYmW02s9vP6h2JiMiIKiQ4\nqoHWvNd9ZhYZoqwDqAESA6ZnzCzinDvinHsV8PLKKoFHgNuA64G7zGxFuLchIiKjpZDgaCMXBCfq\nOOeyeWXVeWUJoHmYOgN1AY8453qccx3AC+SOjYiIyDhUyDGOLcBG4BkzWwvsyivbCywys1pyAXAV\n8FBQNlSdgZYAT5vZ6qA964FvDteoRFVZAU2fuLx0L8lkYvgZi2Q01zXZqS+LS/05/hQSHM8B15nZ\nluD17Wb2aaAyOIPqHuAn5HY/PeGc+8DMTqkzYJl+/xPn3Jtm9hSwFUgB33LO7R2uUe0dPQU0feLy\nUj3U14/OiWXJZGLU1jXZqS+LS/1ZXMUKYc/3/eHnGmee/el2P8Mk3+JItbLh8uWjsi795ywe9WVx\nqT+LK5lMeMPPNTxdACgiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlF\nwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQF\nh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQc\nIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISSmy4GczMAx4F\nVgE9wB3OuUN55TcB9wNp4Enn3OMF1HkYeNM591jw+rPAncEyHnDO/ahI709ERIqskC2OW4BS59w6\n4F7g4f4CM4sFr68FrgbuNLPkUHXMrM7M/hm4KW8Zs4DPAVcA1wMPmln87N+aiIiMhEKCYz3wPIBz\nbitwaV7ZMmC/c67NOZcGNgEbBqmzJpi/CvgC8O28ZVwObHbO9Tnn2oD9wMozfkciIjKiCgmOaqA1\n73WfmUWGKOsAaoDEgOkZM4s45444514FvNMsv38ZIiIyDhUSHG3kguBEHedcNq+sOq8sATQPU2ew\n5Q9cRksB7RIRkTEw7MFxYAuwEXjGzNYCu/LK9gKLzKwW6AKuAh4KyoaqM9ArwF+aWQlQDiwFdg/X\nqERVWQFNn7i8dC/JZGL4GYtkNNc12akvi0v9Of4UEhzPAdeZ2Zbg9e1m9mmgMjiD6h7gJ+R2Pz3h\nnPvAzE6pM2CZfv8T59wxM3sE2Bws4z7nXGq4RrV39BTQ9InLS/VQX98+KutKJhOjtq7JTn1ZXOrP\n4ipWCHu+7w8/1zjz7E+3+xkm+RZHqpUNly8flXXpP2fxqC+LS/1ZXMlkwht+ruHpAkAREQlFwSEi\nIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iI\nhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiIS\nioJDRERCUXCIiEgoCg4REQlFwSEiIqEoOEREJBQFh4iIhKLgEBGRUBQcIiISioJjAqhv6ebpF/bz\n1tH2sW6KiAixsW6ADK03leFHL7/F81vfpi+TZcuuo9x72yWcM6NyrJsmIlOYtjjGId/3OXysm/u+\n8TI//PkREhVxPnbJuXR0p/nSd3bQ0tE71k0UkSlMWxzjTFNbD6/uPc6x5m5iUY+N6+Zx49p5lJXE\nSFTE+d7mw3zpOzv40/96CeWl+ucTkdGnb55xoieVYceBBva93YIPnFdXyt2/cjEzp1WcmOemK+fT\n3NHLS9vf56vP7uLzn1pFLKqNRhEZXQqOMZbN+ux7t4Xt+xtIpbNUV5Zw2dKZnFvTd1JoAHiex22/\ntIS2zhTb9jfwxI/28tmblhPxvDFqvYhMRQqOMXS0qYtX9x6nub2XeDTCpZbE5k0jGvEg1TponWgk\nwu/85wv5m3/cztY9x6itKuHWaxaPcstFZCobNjjMzAMeBVYBPcAdzrlDeeU3AfcDaeBJ59zjQ9Ux\nswuAbwJZYLdz7u5gGV8GrgT6zze92Tk3ac897exO85qr50hweu2iuTVcvKSu4GMWJfEov/+rK3nw\n71/jx6+8Q21VKR+//PyRbLKIyAmF7CC/BSh1zq0D7gUe7i8ws1jw+lrgauBOM0ueps7DwH3OuQ1A\nxMxuDqavAT7unLsm+JuUodGXybLzQAPf23SYI0fbqasp48a157PuotmhD3RXlcf5/KdWUVtVwtMv\nHODlPUdHqNUiIicrJDjWA88DOOe2ApfmlS0D9jvn2pxzaWATsGGQOmuC+dc45zYFz/8FuDbYOlkM\nPGZmm83s9rN8T+OO7/u8faydH2w+wvYDjcRjEa68aDY3rD2futryM15uXU05n//UaspLozzxw73s\nOdJUxFaLiAyukOCoBvJ3uPeZWWSIsg6gBkgMmJ4xsyiQfxS3PZi3AngEuA24HrjLzFaEeRPjWUtH\nL//6i3d5cdv7dPakWT5/Grd8dAEXzK3BK8JB7fNmVvF7v7wSz4OvPruLt49Nyo01ERlHCtk/0kYu\nCPpFnHPZvLLqvLIE0DxEnYyZZQfM2wJ0AY8453oAzOwFcsdGdp+uUYmqsgKaPnZ6Uxle3XOUnQcb\n8H04f1aC9avmMK26sHZ76V6SycTwMwLJZAIvFuWhv/8FX3lmJw/9/keZNb1i+IoDliHFob4sLvXn\n+FNIcGwBNgLPmNlaYFde2V5gkZnVkguAq4CHgrLB6rxuZh91zv0MuAF4ATDgaTNbHbRnPbkD6KfV\n3tFTQNNHn+/7HHivlW37GuhJZUhUxLls6UzmJivxvMLb7aV6qK8vfOth6dxq/ss1i/mHf9vPn31t\nC/f9tzVUlccLqptMJkKtS4amviwu9WdxFSuECwmO54DrzGxL8Pp2M/s0UBmcQXUP8BNyu6GecM59\nYGan1Ake/wj4hpnFyYXOM84538yeArYCKeBbzrm9RXl3o6yhpZute47R2NZLLOpx8ZI6ls+fRjQy\nOhfpXXfZeTR39PL81rf5ynd38EefvpjSeHRU1i0iU4fn+/5YtyG0Z3+63c8wvnZVdXan+d6mw2Sy\nPgvOSbDGklSUFfaLfzBeqpUNly8PXS/r+zz+wz28/MYxVl0wg9/7lYuGDS79qise9WVxqT+LK5lM\nFOVqYY1XUSTb9jeQyfqsvXAWV62ac1ahcTYinsdv3biMC+dPY8fBRr79Y8dE/HEgIuOXgqMIGtt6\nOPR+G9MSpSw6t2asm0MsGuGuT17E+bOq+NmOD/j+5sNj3SQRmUQUHGfJ931ee7MegDWWHDfjRpWX\nxvj8r62irqaMH2w5wovb3xvrJonIJKHgOEvvNXRytKmLuXWVzKkbXzdYqqkq5Z5bV1NVHufbP3Zs\n218/1k0SkUlAwXEWslmf11w9HnCJJce6OYOaPb2CP/y1VcRjEb7+/Tc48O7ggyeKiBRKwXEWDrzX\nSmtHigvOrWFaonSsmzOkhXOqueuWFWQyPl95ZgcfNHaOdZNEZAJTcJyhdF+WHQcaiEU9Vi+qG+vm\nDGvlBXX85g1GZ08fDz+9neZ23X5WRM6MguMMvXG4ie7eDMvnT6eibGLc1uSqlXP45FULaGzr5Uvf\n2UFXT99YN0lEJiAFxxno6uljz5EmykujXLhg+lg3J5SN6+Zz9cVzebe+g68+u5N0X3b4SiIieSbG\nT+VxZvuBBvoyPpcurSMeG5ns9X2f9vaRuWL25rXn0NjSya7DLXz9ezu559dX0t4+Nsc9qqqqijJK\nsIiMHgVHSM3tvRx8t5WaqhIWzR25i/16urv4t1cPUlp25vfrOJ3Fcyp4v7GLbQeaePCpbSw/r3LU\nv8B7e7r52GUXkEho9FORiUTBEdLrrh4fWLMkSSQysl+0pWXllFdUjdjyr720gue3vs2eI60ca+5l\nxYLpnDeratxcxCgi45OOcYTwfkMn7zV0Mnt6BXOT4+tivzNRWhLl2svOZeHcGhpbe3hp+/t8f9Nh\n9r3TQiajYx8iMjhtcRTI93MX+0FuaJHJsl++sizODVfM592jbbxxpIlD77Xx8hvH2HGggaXzpmHn\n1VKiodlFJI+Co0CH3m+jub2XhXOqmVEzvoZ0L4aaqhLWrZjN6kV17H2rmX3vtLBtXwO7Dzax5Pwa\nls2bOKcdi8jI0jdBAfoyWbbtbyAS8Vi9ePxf7Hc2KspirLEkFy2czr53Wtj7VjNvHG5m75FmFs6p\n4cIF06ipGr9XyYvIyFNwFGDvkWa6evpYsWB6wbdjnehK4lFWLJzBsvnTOPReG28cbuLAe60ceK+V\n82ZWsWLhdJK1I3PGl4iMbwqOYfSk+th9qInSeJQVCyfWxX7FEI1EWHxeLRecW8M7xzp443AT7xzv\n4J3jHcyaVs6FC6czt270T+UVkbGj4BjGjgONpDNZLl8yc0ofJI54HvNmJzh/VhXHmrrZfbiJ9xs6\nOfbae9RWlbBi4XTmz64e8VOURWTsKThOo7Ujxb53WkhUxFlyXu1YN2dc8DyP2TMqmD2jgqa2Ht44\n3MSRo+1s3nmUbfsaWD5/OovOrRmxK+pFZOwpOE7j9X31+H5wZz/9kj7F9Ooyrlo1h4sXp9lzpIn9\n77by6pvH2XGwgaXnT2PpvFrKSvQRE5ls9L96CMeaunjneAfJ2nLOmzlyV29PBlUVcS5fPouVi2bw\n5lstvPl2MzsPNvLG4SYWzKlm1rRyZtSUUVNZomMhIpOAgmMQ+Rf7Xbp08lzsN9LKSmKsXlzHhQum\nc+DdVvYcaeLAu60n7joYj0WYUV3GjJoy6mrKqCrJ4Pv+GLdaRMJScAziyNF2Glp7mDc7oVNOz0A8\nFmHZ/GnY+bU0tvXQ2NpDQ2vu8WhTF0ebuk7M+9KuRhbOqWHBOdUsOCfB/HOqqa4oGcPWF19vOkND\nSzf1LT20daVYsWA606sn30WkMnUoOAbIZLNs29dAxINLlkzui/1GWiTikawtPyl8U30Zmlp7aWjt\n5lhTB53dfew82MjOg40n5pmeKOH8mZWcP7OSebMqOTdZSVnJ+D2jLev7tHWmaWzrpXcvHHmvjca2\n3hN/bV3pk+aPRT3Wr5zDJ9bOm5SjEMjkp+AYwL3VQkd3mmXzppGYZL98x4OSWPTEWVlza/pIpdKU\nlNfR3J6mqT1FU0eKpvY02w82s/1g84l61RUxpiVKmJ6IMz1RQk1lnOgonrCQ7svS2ZOhs6ePzp4+\nOnoydHb3nZiWHWSPmwdUlEWpq46x5NxpzJlZTSwa4YXX3uXFbe+xacf7rF95Dp9YO486bdnKBKLg\nyNObyrDzYCMlsQgrL5gx1s2ZEkrLyqmtrWFaLSwMpvm+T2d3Hw1tPTS2dp/YzdXW1cVbx3LzRDyo\nKIsT8XJbNpGIR8Tz8DyPSCR33Un/tNxjbpoXTItGPLz+uvnzRHLL6Onto707TUdXmo7uND2pzKDt\nL4lHmJYoo6oiTqI8Tt20CuJRqCqPU1kWJxLx6O7qYP1Fc07cd+Rja+by8hvH+KefH+Gl7e+zeecH\nrFsxm0+sm89MBYhMAAqOPLsONZLqy7LGkpSO410jk53neVRVxKmqiDN/du7LNrc7KHXS8ZKu3j76\nMj5Z38fP5ubJ+j7FPN4e8aCyPM706lKqyktIVMSpKo+feBx4UWiiqoz2jp7TLjMaiXDlReew9sJZ\nvLLnOP/08yNs2vkBW3Yd5YoVs9h4xXxmTa8o3psQKTIFR6C9K8WbbzVTVR5n6Txd7DfeRDyP2qpS\naqtKuWCYOy/6QYBk+8Mk65949H0+fN1flhc6/fOUxqNUVcSpKIuN2I2topEIV6yYzUeWz+KVN4/x\nT1uOsGXXUX6++yhrl89m47p5nDNj4t/3RSYfBUdg274Gsj5cvKSOaERXPU9knucR9TyiE+SfMRLx\nWLt8Npcvm8Vrrp4fbDnMf7xxlJf3HOUjy2axcd185tQpQGT8UHAA9S3dHDnaTl1N2YldIyKjLeJ5\nXLZ0JmssybZ99fxgyxFe3nOMrXuOcdmymWxcN59zk7oYVcbelA+OyXpnP5m4Ip7HGpvJxUuS7Njf\nwPe3HOaVvcd5Ze9xLrUkN125QKMZyJia8sHxzvEOjjd3c97MKh2QlHEl4nlcvCTJ6sV17DjYyA82\nH+YXrp5fuHouWZLkpnXzmactZBkDUzo4stnc1obnwSVLkmPdHJmkfN+no6PjrJZxwaxS/vCXjb1v\nt/L8q+/z+r56Xt9Xz4r5tVx/2RzOm3l2x0AyWZ+eVOaUv9507tHzPKYnSphRXcq0qhKigxxAqqqq\n0hb7FDGlg2PfOy20d6Wx82upqdLFfjIyenu6+Nn2VqprinO23mVLajg/Wcaet9vZfaSF3UdaOGd6\nKYvnVhHxPNKZLH0Zn3SfTzqTJd3X/zpLOuMPeJ4l3eeTGewKxiF4QHlplKryGJVlUSrLYpRGM2y4\n+HzmzZlBVXlcATLJTdngSKUz7DjQSDwaYdUiXewnI6u0rJzyiuIdl1hQmWD+3Bkcbepix4FGPmjq\n5oOm3oLrRyMe8ViEkliUyrIo8VjkpL+SE89zZdmsT0d37mLI9q40Hd0pjrecvL7XDu7NvdeSKMma\ncpK1ZSeGnEnWllFXU05dTVnRbojm+7kATPVlSaUzgz6WxiJUlMUpL4tRWRajvCSmWyQUwbDBYWYe\n8CiwCugB7nDOHcorvwm4H0gDTzrnHh+qjpldAHwTyAK7nXN3B8v4LHBnsIwHnHM/Kt5bHNzuw030\npjNcvLhO94yQCcnzPM6ZUcns6RUca+rm3foOotHgiz+aFwTxIAiiHwZEMb48+zLZXJh0pWlqaaey\nLEZbd5aGtl6ON3fxbv3gu+dqKnMXVNZVlzKjupSSWIRUX/bDEOh/ns5CxKOzO006nT1pnnTwdybX\nepaVRCkviVJeGqW8NJb3PEp5SezD56UxKgbMV1YSpbo6MeQWVdb3yQRbcpmsTybY+stk+x8HlAWP\nmYx/4nnW93OBHY2cEujxWOSU6WNx+UAh35i3AKXOuXVm9hHg4WAaZhYLXq8BuoEtZvZ9YP0QdR4G\n7nPObTKzr5nZzcDLwOeAS4AKYLOZ/cQ5d/LIcHle3ttEdWU5NZWl1FSVUFYSDbVp3NmdZu+RZirK\nYiybP63geiLjUf5dGUdTLBo5cVFmZbSTVKqXOXW1QCW+75NKZ3NjevV8OKZXRzC+1+EPOjj8QeHH\nfWLR3DAx0YhHLOpRUhYlGokRjXrEIh7RAeX9zzNZPwgaPy9wfHrTGTp60qT7wkdPaTxCWUkU388d\nG+oPgUzWH3TMspEW8XL/FvFYhFjUIx6NEItFiEe94DGYHovwl3d9tCjrLCQ41gPPAzjntprZpXll\ny4D9zrk2ADPbBGwArhhQZ00w/xrn3Kbg+b8Av0Ru62Ozc64PaDOz/cBK4LWhGrT3rXag/cTrkliE\nmqoSaipLqa4qobayhJqqEirL44Ne9bttfwOZrM/Fi+uITZSrxETGuYG74yqAoY7qZLM+nT253V5Z\n3ycWieRCIPiSi0Zyj7U15XR29o7YMZOhdnel+7L0pjOk0v1bP7myru5uUuncVoPnBbv8ohEipZHc\nOGn545+dGPusf+y04DF4fsr8QXl///SHUjYLGb//ef50n0yWk173ZXLtzi8bCYUERzXQmve6z8wi\nzrnsIGUdQA2QGDA9Y2ZRcsfV+rUH9QfO27+MIW28YjbNHdDamaK1I0VrZ4qG1h7qW04eIygS8aiu\niFNTVUpNECbRiMeh99uYlihl4ZzqAt6+iBRbJOKRqCgZdgTqaCQyogfaPc+jJB7NHXcpjw87f3Pj\nMbxIjNppE+O4qB+M3da/i6xYCgmONnJf7v36Q6O/LP/bNwE0D1EnY2bZAfO2DLGMltM1aFpZispo\nhHNrIkAZUEYm69PZk6GtK0Nb94eP7V0pWjpSpyxj5bxSUp1Np1vNmMpmekn1dI/KuqJeH91dpx+Y\nbyT09vTgRaJ0d53dqarjyWB9ORnf50Aj9R7H6rM5lKnwb1mIQoJjC7AReMbM1gK78sr2AovMrBbo\nAq4CHgrKBqvzupl91Dn3M+AG4AXgVeABMysByoGlwO7TNehTG6/SaREiImPEG+6ez3lnSK0MJt1O\n7mB4ZXAG1SeAL5DbDfWEc+7rg9Vxzu0zs8XAN4A4udD5rHPON7PfBn4nWMYDzrnvFfVdiohI0Qwb\nHCIiIvl0SpGIiISi4BARkVAUHCIiEoqCQ0REQplQgzQNN26WDM3MXuPDCy0PA3/FOBk3bKIIhs/5\na+fcfwoz7pqZlQF/D8wkd93SbzrnGsfiPYwnA/pzNfBDYF9Q/DXn3HfVn8MLhn76O2A+UAI8AOxh\nBD+fE22L48S4WcC95Ma+kmGYWSmAc+6a4O+3+XDcsA1AxMxuNrNZ5MYNuwK4HnjQzIa/nHYKMLM/\nJncqeWkwKUz//S6w0zn3UeDb5AYFndIG6c81wN/mfUa/q/4s2G1AQ9Af1wNfZYQ/nxMtOE4aNwu4\n9PSzS2AVUGlmPzazfw1+6V0yYNyw64DLCcYNC8Yf6x83TOAA8Mm81wPHXRuq/1aR97kN5r12dJo8\nrp3Sn8AnzOwlM/uGmVWh/izUd/jwyz4K9FH4/+8z6s+JFhyDjps1Vo2ZQLqAh5xzHyf36+L/UqRx\nw6YK59zqI/8QAAAEGElEQVRz5P5D9gvTf/nT++ed0gbpz63AHwe/kA+Ru6i4kLHwpnx/Oue6nHOd\nZpYAvgv8T0b48znRvnRPN26WDG0fubDAObcfaARm5ZWf8bhhU1ih464NHLtNfTq47znntvU/B1aT\n+zJTfxbAzM4jN4TTt5xz/8gIfz4nWnBsAW4EGGTcLBnabwF/C2Bmc8h9eH5iZhuC8huATeTGDVtv\nZiVmVkMB44ZNYa+bWf/NDYbrv58TfG6Dx00DFyb8OO+WDR8jd1sF9WcBgmMXPwb+h3PuW8HkbSP5\n+ZxQZ1UBzwHXmdmW4PXtY9mYCeQJ4MngfilZ4DPktjoeDw6O7QWeCcYNewTYTG5T9z7n3KlDCwvA\nHwHfKKT/zOxrwLeC/u8Ffn3MWj1+/S7wv80sBRwF7nTOdag/C3IvuVuf3G9mfw74wB+Q688R+Xxq\nrCoREQllou2qEhGRMabgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQllol3HIXLWzOxXgT8l9/n3\ngKecc39rZk8C/+6ceyqY7xrgz8ldZR8FtgOfd869N8zys8G8HrlB/N4GPhdctY+ZHQY2OOfeNrOr\nyY1UXBGs45/JnZe/nNyAcz4wj9zwEE1Aj3PuiiJ1hcgZUXDIlBJcOf83wGrnXIuZVQAvmZkbMN9V\n5L64b3HOvRpMuwt4FvjIMKvxnXOX5C3rd4B/MbOlzrk+cmGAmZWQGwrmiiBEYsD/A+5yzn0VuDiY\n7++AF/sDTWSsaVeVTDV15H4wVUFugDjgN8ldXZvvz4C/6A+NYN5HgafDDjXvnPs/5K7IvT6Y1D8A\nXQUfDj5HECp/ALw4YBEeIuOIgkOmFOfcTuAHwCEz22pmfw3EnHMHB8y6lkHG7HHOPeycS5/BqneT\nGxsof1kt5HZTvW5m283sy8Bc55zGB5NxTcEhU45z7i5yxw0eDR7/w8w+Ocis/buU4ma2Lfh7Kxhg\nMywf6B6kLQ8C5wAPktvy+Gcz+/0zWL7IqNExDplSzOxGoMo59x3gW+QGd7sD+G3geN6sr5C7wc3e\nYAuj/3jDv5O7PWdYK4GvD2hL/w21vgY8TW432D8AXwIeOYN1iIwKbXHIVNMF/JWZzYMT97FfDrzO\nyccS/he50UYv659gZiuBBUBmmHWcdEzCzH4XyDrnXhwwXxPwhWC5/S4EtiEyjmmLQ6YU59yLZvZF\n4IfBWUyQu23mXwCP5c23xcxuBR4ws5nkfmQ1Avc457YMXO4Avpn1B5FH7o52N+SXB+vYb2afAZ4w\ns2pyQ95vBX5v4PLCv1ORkaNh1UVEJBRtcYiEZGYLyV1vkf+rywte3+Gce31MGiYySrTFISIioejg\nuIiIhKLgEBGRUBQcIiISioJDRERCUXCIiEgoCg4REQnl/wO6QJWp3YmH4wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2a4bb91acf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(data[\"SIG_DIST\"])\n",
    "plt.xlim(0, 2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['XCoord',\n",
       " 'YCoord',\n",
       " 'LONGITUDE',\n",
       " 'LATITUDE',\n",
       " 'ALTITUDE',\n",
       " 'SPEED',\n",
       " 'HORT_ACCUR',\n",
       " 'VERT_ACCUR',\n",
       " 'STARTED_AT',\n",
       " 'RECORDED_A',\n",
       " 'APP_USER_I',\n",
       " 'WINTER',\n",
       " 'RIDER_HIST',\n",
       " 'WORKZIP',\n",
       " 'INCOME',\n",
       " 'CYCLINGFRE',\n",
       " 'AGE',\n",
       " 'CYCLING_LE',\n",
       " 'GENDER',\n",
       " 'RIDER_TYPE',\n",
       " 'SCHOOLZIP',\n",
       " 'HOMEZIP',\n",
       " 'CYCLINGEXP',\n",
       " 'PURPOSE',\n",
       " 'FID',\n",
       " 'CUMUL_METE',\n",
       " 'LF_NAME',\n",
       " 'ONE_WAY_DI',\n",
       " 'SIG_DIST',\n",
       " 'SOURCEOID',\n",
       " 'SLOPE_TF',\n",
       " 'SHAPE_LENG',\n",
       " 'RDCLASS',\n",
       " 'BIKE_CLASS',\n",
       " 'BIKE_CODE',\n",
       " 'EMME_MATCH',\n",
       " 'EMME_CONTR',\n",
       " 'LINK_DIR',\n",
       " 'trip_length',\n",
       " 'TRIP_ID',\n",
       " 'bike_lanes',\n",
       " 'sharrows',\n",
       " 'bike_path',\n",
       " 'time_of_day',\n",
       " 'EMME_ID',\n",
       " 'LINK_ID',\n",
       " 'AM_VOL',\n",
       " 'MID_VOL',\n",
       " 'PM_VOL',\n",
       " 'EVE_VOL',\n",
       " 'volume',\n",
       " 'ID',\n",
       " 'LANES',\n",
       " 'VDF',\n",
       " 'speed_limit',\n",
       " 'lane_cap']"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<3511876x57 sparse matrix of type '<class 'numpy.object_'>'\n",
       "\twith 176373853 stored elements in COOrdinate format>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sps.csr_matrix(data.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, -79.415536, 43.665914, ..., nan, 40.0, 250],\n",
       "       [1, -79.415564, 43.665991, ..., nan, 40.0, 250],\n",
       "       [2, -79.41561899999999, 43.665989, ..., nan, 40.0, 250],\n",
       "       ..., \n",
       "       [3511873, -79.39794499999999, 43.664221999999995, ..., 50.0, 40.0,\n",
       "        600],\n",
       "       [3511874, -79.397883, 43.664235999999995, ..., 50.0, 40.0, 600],\n",
       "       [3511875, -79.397816, 43.66425, ..., 50.0, 40.0, 600]], dtype=object)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [uoft2016]",
   "language": "python",
   "name": "Python [uoft2016]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}