Filter Arrests.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import geopandas as gpd\n",
    "import pandas as pd \n",
    "import numpy as np \n",
    "import seaborn as sns \n",
    "from matplotlib import pyplot as plt \n",
    "import process as fe\n",
    "import pickle as pkl"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Filtering Arrests \n",
    "\n",
    "This notebook will walk through the process of geographically filtering a data set of 'theft of services' arrests and turn it into a data set of 'fare evasion' arrests by comparing their locations to that of subway stations."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading Data\n",
    "\n",
    "Using a few helper functions in `process` to load the datasets we'll be using"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ARREST_KEY</th>\n",
       "      <th>ARREST_DATE</th>\n",
       "      <th>PD_CD</th>\n",
       "      <th>PD_DESC</th>\n",
       "      <th>KY_CD</th>\n",
       "      <th>OFNS_DESC</th>\n",
       "      <th>LAW_CODE</th>\n",
       "      <th>LAW_CAT_CD</th>\n",
       "      <th>ARREST_BORO</th>\n",
       "      <th>ARREST_PRECINCT</th>\n",
       "      <th>JURISDICTION_CODE</th>\n",
       "      <th>AGE_GROUP</th>\n",
       "      <th>PERP_SEX</th>\n",
       "      <th>PERP_RACE</th>\n",
       "      <th>X_COORD_CD</th>\n",
       "      <th>Y_COORD_CD</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>173114463</td>\n",
       "      <td>2017-12-31</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>Q</td>\n",
       "      <td>114</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>ASIAN / PACIFIC ISLANDER</td>\n",
       "      <td>1009113.0</td>\n",
       "      <td>219613.0</td>\n",
       "      <td>40.769437</td>\n",
       "      <td>-73.910241</td>\n",
       "      <td>POINT (-73.91024 40.76944)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>173113521</td>\n",
       "      <td>2017-12-31</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>F</td>\n",
       "      <td>WHITE HISPANIC</td>\n",
       "      <td>995369.0</td>\n",
       "      <td>221138.0</td>\n",
       "      <td>40.773650</td>\n",
       "      <td>-73.959857</td>\n",
       "      <td>POINT (-73.95986 40.77365)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>173114460</td>\n",
       "      <td>2017-12-31</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>24</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>992586.0</td>\n",
       "      <td>227883.0</td>\n",
       "      <td>40.792167</td>\n",
       "      <td>-73.969896</td>\n",
       "      <td>POINT (-73.96990 40.79217)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>165</th>\n",
       "      <td>173114446</td>\n",
       "      <td>2017-12-31</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>40</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>WHITE HISPANIC</td>\n",
       "      <td>1011485.0</td>\n",
       "      <td>237929.0</td>\n",
       "      <td>40.819702</td>\n",
       "      <td>-73.901603</td>\n",
       "      <td>POINT (-73.90160 40.81970)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>365</th>\n",
       "      <td>173093896</td>\n",
       "      <td>2017-12-30</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>52</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>1013465.0</td>\n",
       "      <td>254827.0</td>\n",
       "      <td>40.866076</td>\n",
       "      <td>-73.894376</td>\n",
       "      <td>POINT (-73.89438 40.86608)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012081</th>\n",
       "      <td>191748723</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>45-64</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>980772.0</td>\n",
       "      <td>194845.0</td>\n",
       "      <td>40.701489</td>\n",
       "      <td>-74.012545</td>\n",
       "      <td>POINT (-74.01254 40.70149)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012138</th>\n",
       "      <td>191775988</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>40</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18-24</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>1007034.0</td>\n",
       "      <td>236608.0</td>\n",
       "      <td>40.816089</td>\n",
       "      <td>-73.917689</td>\n",
       "      <td>POINT (-73.91769 40.81609)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012222</th>\n",
       "      <td>191744335</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>48</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK HISPANIC</td>\n",
       "      <td>1019444.0</td>\n",
       "      <td>245654.0</td>\n",
       "      <td>40.840877</td>\n",
       "      <td>-73.872807</td>\n",
       "      <td>POINT (-73.87281 40.84088)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012436</th>\n",
       "      <td>191735303</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>23</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>999617.0</td>\n",
       "      <td>225524.0</td>\n",
       "      <td>40.785682</td>\n",
       "      <td>-73.944510</td>\n",
       "      <td>POINT (-73.94451 40.78568)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012772</th>\n",
       "      <td>191731987</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>&lt;18</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>980877.0</td>\n",
       "      <td>199041.0</td>\n",
       "      <td>40.713006</td>\n",
       "      <td>-74.012168</td>\n",
       "      <td>POINT (-74.01217 40.71301)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>295861 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         ARREST_KEY ARREST_DATE  PD_CD                          PD_DESC  \\\n",
       "1         173114463  2017-12-31  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "48        173113521  2017-12-31  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "50        173114460  2017-12-31  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "165       173114446  2017-12-31  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "365       173093896  2017-12-30  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "...             ...         ...    ...                              ...   \n",
       "5012081   191748723  2019-01-02  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012138   191775988  2019-01-02  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012222   191744335  2019-01-02  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012436   191735303  2019-01-01  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012772   191731987  2019-01-01  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "\n",
       "         KY_CD                        OFNS_DESC    LAW_CODE LAW_CAT_CD  \\\n",
       "1        343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "48       343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "50       343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "165      343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "365      343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "...        ...                              ...         ...        ...   \n",
       "5012081  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012138  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012222  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012436  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012772  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "\n",
       "        ARREST_BORO  ARREST_PRECINCT  JURISDICTION_CODE AGE_GROUP PERP_SEX  \\\n",
       "1                 Q              114                0.0     25-44        M   \n",
       "48                M               19                1.0     25-44        F   \n",
       "50                M               24                0.0     25-44        M   \n",
       "165               B               40                1.0     25-44        M   \n",
       "365               B               52                1.0     25-44        M   \n",
       "...             ...              ...                ...       ...      ...   \n",
       "5012081           M                1                1.0     45-64        M   \n",
       "5012138           B               40                1.0     18-24        M   \n",
       "5012222           B               48                1.0     25-44        M   \n",
       "5012436           M               23                0.0     25-44        M   \n",
       "5012772           M                1                3.0       <18        M   \n",
       "\n",
       "                        PERP_RACE  X_COORD_CD  Y_COORD_CD   Latitude  \\\n",
       "1        ASIAN / PACIFIC ISLANDER   1009113.0    219613.0  40.769437   \n",
       "48                 WHITE HISPANIC    995369.0    221138.0  40.773650   \n",
       "50                          BLACK    992586.0    227883.0  40.792167   \n",
       "165                WHITE HISPANIC   1011485.0    237929.0  40.819702   \n",
       "365                         BLACK   1013465.0    254827.0  40.866076   \n",
       "...                           ...         ...         ...        ...   \n",
       "5012081                     BLACK    980772.0    194845.0  40.701489   \n",
       "5012138                     BLACK   1007034.0    236608.0  40.816089   \n",
       "5012222            BLACK HISPANIC   1019444.0    245654.0  40.840877   \n",
       "5012436                     BLACK    999617.0    225524.0  40.785682   \n",
       "5012772                     BLACK    980877.0    199041.0  40.713006   \n",
       "\n",
       "         Longitude                    geometry  \n",
       "1       -73.910241  POINT (-73.91024 40.76944)  \n",
       "48      -73.959857  POINT (-73.95986 40.77365)  \n",
       "50      -73.969896  POINT (-73.96990 40.79217)  \n",
       "165     -73.901603  POINT (-73.90160 40.81970)  \n",
       "365     -73.894376  POINT (-73.89438 40.86608)  \n",
       "...            ...                         ...  \n",
       "5012081 -74.012545  POINT (-74.01254 40.70149)  \n",
       "5012138 -73.917689  POINT (-73.91769 40.81609)  \n",
       "5012222 -73.872807  POINT (-73.87281 40.84088)  \n",
       "5012436 -73.944510  POINT (-73.94451 40.78568)  \n",
       "5012772 -74.012168  POINT (-74.01217 40.71301)  \n",
       "\n",
       "[295861 rows x 19 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thefts_of_service = fe.load_tos_arrests()\n",
    "\n",
    "thefts_of_service"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set include arrests to false because we will be saving arrests later. \n",
    "stations, _, census, lines, nyc = fe.load_all_data(include_arrests=False)\n",
    "del _"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One of the key parameters is the size of the buffer around each subway station. This will determine which arrests make it into the final fare evasion data set, and therefore is a very important assumption. `get_buffer` handles projecting from latitude-longitude to feet and back. \n",
    "\n",
    "I also plot the arrests over New York City to make sure the data looks good."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "BUFFER_SIZE = 475\n",
    "\n",
    "station_buffers = fe.get_buffer(stations, feet=BUFFER_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10 ,10))\n",
    "\n",
    "nyc.plot(ax=ax, alpha=0.1)\n",
    "station_buffers.geometry.plot(ax=ax)\n",
    "\n",
    "ax.axis('off');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For my research, I am only using 2010-present data, so I save some time by filtering to an earlier date\n",
    "\n",
    "tos = thefts_of_service.copy()\n",
    "\n",
    "tos = tos[tos['ARREST_DATE'] > pd.datetime(2010, 1, 1)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One of the most computationally expensive steps is assigning each arrest to a subway station. We can speed up the process significantly by filtering out arrests which did not occur within any of the buffer points first."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Geographic Filtering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a multipoint of all station buffers\n",
    "\n",
    "mp = station_buffers.unary_union"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the indices of arrests which occur near a subway station\n",
    "\n",
    "fare_evasion_idx = tos.geometry.intersects(mp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "evasions = tos[fare_evasion_idx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ARREST_KEY</th>\n",
       "      <th>ARREST_DATE</th>\n",
       "      <th>PD_CD</th>\n",
       "      <th>PD_DESC</th>\n",
       "      <th>KY_CD</th>\n",
       "      <th>OFNS_DESC</th>\n",
       "      <th>LAW_CODE</th>\n",
       "      <th>LAW_CAT_CD</th>\n",
       "      <th>ARREST_BORO</th>\n",
       "      <th>ARREST_PRECINCT</th>\n",
       "      <th>JURISDICTION_CODE</th>\n",
       "      <th>AGE_GROUP</th>\n",
       "      <th>PERP_SEX</th>\n",
       "      <th>PERP_RACE</th>\n",
       "      <th>X_COORD_CD</th>\n",
       "      <th>Y_COORD_CD</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>173113521</td>\n",
       "      <td>2017-12-31</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>F</td>\n",
       "      <td>WHITE HISPANIC</td>\n",
       "      <td>995369.0</td>\n",
       "      <td>221138.0</td>\n",
       "      <td>40.773650</td>\n",
       "      <td>-73.959857</td>\n",
       "      <td>POINT (-73.95986 40.77365)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>165</th>\n",
       "      <td>173114446</td>\n",
       "      <td>2017-12-31</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>40</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>WHITE HISPANIC</td>\n",
       "      <td>1011485.0</td>\n",
       "      <td>237929.0</td>\n",
       "      <td>40.819702</td>\n",
       "      <td>-73.901603</td>\n",
       "      <td>POINT (-73.90160 40.81970)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>365</th>\n",
       "      <td>173093896</td>\n",
       "      <td>2017-12-30</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>52</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>1013465.0</td>\n",
       "      <td>254827.0</td>\n",
       "      <td>40.866076</td>\n",
       "      <td>-73.894376</td>\n",
       "      <td>POINT (-73.89438 40.86608)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>420</th>\n",
       "      <td>173102770</td>\n",
       "      <td>2017-12-30</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>30</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>999539.0</td>\n",
       "      <td>239499.0</td>\n",
       "      <td>40.824040</td>\n",
       "      <td>-73.944760</td>\n",
       "      <td>POINT (-73.94476 40.82404)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>531</th>\n",
       "      <td>173107288</td>\n",
       "      <td>2017-12-30</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIED</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEFT</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>45-64</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>983985.0</td>\n",
       "      <td>205857.0</td>\n",
       "      <td>40.731715</td>\n",
       "      <td>-74.000958</td>\n",
       "      <td>POINT (-74.00096 40.73171)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012058</th>\n",
       "      <td>191753702</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>46</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>WHITE HISPANIC</td>\n",
       "      <td>1010257.0</td>\n",
       "      <td>248767.0</td>\n",
       "      <td>40.849453</td>\n",
       "      <td>-73.905998</td>\n",
       "      <td>POINT (-73.90600 40.84945)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012081</th>\n",
       "      <td>191748723</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>45-64</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>980772.0</td>\n",
       "      <td>194845.0</td>\n",
       "      <td>40.701489</td>\n",
       "      <td>-74.012545</td>\n",
       "      <td>POINT (-74.01254 40.70149)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012138</th>\n",
       "      <td>191775988</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>40</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18-24</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>1007034.0</td>\n",
       "      <td>236608.0</td>\n",
       "      <td>40.816089</td>\n",
       "      <td>-73.917689</td>\n",
       "      <td>POINT (-73.91769 40.81609)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012222</th>\n",
       "      <td>191744335</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>B</td>\n",
       "      <td>48</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25-44</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK HISPANIC</td>\n",
       "      <td>1019444.0</td>\n",
       "      <td>245654.0</td>\n",
       "      <td>40.840877</td>\n",
       "      <td>-73.872807</td>\n",
       "      <td>POINT (-73.87281 40.84088)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5012772</th>\n",
       "      <td>191731987</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>478.0</td>\n",
       "      <td>THEFT OF SERVICES, UNCLASSIFIE</td>\n",
       "      <td>343.0</td>\n",
       "      <td>OTHER OFFENSES RELATED TO THEF</td>\n",
       "      <td>PL 1651503</td>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>&lt;18</td>\n",
       "      <td>M</td>\n",
       "      <td>BLACK</td>\n",
       "      <td>980877.0</td>\n",
       "      <td>199041.0</td>\n",
       "      <td>40.713006</td>\n",
       "      <td>-74.012168</td>\n",
       "      <td>POINT (-74.01217 40.71301)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>181227 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         ARREST_KEY ARREST_DATE  PD_CD                          PD_DESC  \\\n",
       "48        173113521  2017-12-31  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "165       173114446  2017-12-31  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "365       173093896  2017-12-30  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "420       173102770  2017-12-30  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "531       173107288  2017-12-30  478.0  THEFT OF SERVICES, UNCLASSIFIED   \n",
       "...             ...         ...    ...                              ...   \n",
       "5012058   191753702  2019-01-02  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012081   191748723  2019-01-02  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012138   191775988  2019-01-02  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012222   191744335  2019-01-02  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "5012772   191731987  2019-01-01  478.0   THEFT OF SERVICES, UNCLASSIFIE   \n",
       "\n",
       "         KY_CD                        OFNS_DESC    LAW_CODE LAW_CAT_CD  \\\n",
       "48       343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "165      343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "365      343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "420      343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "531      343.0  OTHER OFFENSES RELATED TO THEFT  PL 1651503          M   \n",
       "...        ...                              ...         ...        ...   \n",
       "5012058  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012081  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012138  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012222  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "5012772  343.0   OTHER OFFENSES RELATED TO THEF  PL 1651503          M   \n",
       "\n",
       "        ARREST_BORO  ARREST_PRECINCT  JURISDICTION_CODE AGE_GROUP PERP_SEX  \\\n",
       "48                M               19                1.0     25-44        F   \n",
       "165               B               40                1.0     25-44        M   \n",
       "365               B               52                1.0     25-44        M   \n",
       "420               M               30                1.0     25-44        M   \n",
       "531               M                6                1.0     45-64        M   \n",
       "...             ...              ...                ...       ...      ...   \n",
       "5012058           B               46                1.0     25-44        M   \n",
       "5012081           M                1                1.0     45-64        M   \n",
       "5012138           B               40                1.0     18-24        M   \n",
       "5012222           B               48                1.0     25-44        M   \n",
       "5012772           M                1                3.0       <18        M   \n",
       "\n",
       "              PERP_RACE  X_COORD_CD  Y_COORD_CD   Latitude  Longitude  \\\n",
       "48       WHITE HISPANIC    995369.0    221138.0  40.773650 -73.959857   \n",
       "165      WHITE HISPANIC   1011485.0    237929.0  40.819702 -73.901603   \n",
       "365               BLACK   1013465.0    254827.0  40.866076 -73.894376   \n",
       "420               BLACK    999539.0    239499.0  40.824040 -73.944760   \n",
       "531               BLACK    983985.0    205857.0  40.731715 -74.000958   \n",
       "...                 ...         ...         ...        ...        ...   \n",
       "5012058  WHITE HISPANIC   1010257.0    248767.0  40.849453 -73.905998   \n",
       "5012081           BLACK    980772.0    194845.0  40.701489 -74.012545   \n",
       "5012138           BLACK   1007034.0    236608.0  40.816089 -73.917689   \n",
       "5012222  BLACK HISPANIC   1019444.0    245654.0  40.840877 -73.872807   \n",
       "5012772           BLACK    980877.0    199041.0  40.713006 -74.012168   \n",
       "\n",
       "                           geometry  \n",
       "48       POINT (-73.95986 40.77365)  \n",
       "165      POINT (-73.90160 40.81970)  \n",
       "365      POINT (-73.89438 40.86608)  \n",
       "420      POINT (-73.94476 40.82404)  \n",
       "531      POINT (-74.00096 40.73171)  \n",
       "...                             ...  \n",
       "5012058  POINT (-73.90600 40.84945)  \n",
       "5012081  POINT (-74.01254 40.70149)  \n",
       "5012138  POINT (-73.91769 40.81609)  \n",
       "5012222  POINT (-73.87281 40.84088)  \n",
       "5012772  POINT (-74.01217 40.71301)  \n",
       "\n",
       "[181227 rows x 19 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evasions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save to csv here because the initial filtering is expensive\n",
    "\n",
    "evasions.to_csv('evasions.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100.0% complete/42 Stitylace\r"
     ]
    }
   ],
   "source": [
    "evasion_station = fe.get_station_arrests(evasions, stations, station_buffers)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'm saving the \"thefts of service\" data set just in case I want to change buffer/filtering parameters later on without having to load every single arrest into memory."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "evasion_station.to_csv('evasions.csv')\n",
    "thefts_of_service.to_csv('tos.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "stations, evasions, census, lines, nyc = fe.load_all_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We reload the data on the line above, since if you already have the data sets saved it is quite time-intensive to recreate them. Notice that the `stations` DataFrame has a `arrests` column - this is because by default `load_all_data` populates the `stations` DataFrame every time it's called. This behavior can be changed by passing `include_arrests=False`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>complex_nm</th>\n",
       "      <th>trains</th>\n",
       "      <th>station_ct</th>\n",
       "      <th>bcode</th>\n",
       "      <th>stop_lat</th>\n",
       "      <th>stop_lon</th>\n",
       "      <th>tot2007</th>\n",
       "      <th>tot2008</th>\n",
       "      <th>tot2009</th>\n",
       "      <th>tot2010</th>\n",
       "      <th>...</th>\n",
       "      <th>avwken11</th>\n",
       "      <th>avwken12</th>\n",
       "      <th>avwken13</th>\n",
       "      <th>avwken14</th>\n",
       "      <th>avwken15</th>\n",
       "      <th>avwken16</th>\n",
       "      <th>avwken17</th>\n",
       "      <th>srv_notes</th>\n",
       "      <th>geometry</th>\n",
       "      <th>arrests</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>complex_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>bx001</th>\n",
       "      <td>138 St-Grand Concourse</td>\n",
       "      <td>4 5</td>\n",
       "      <td>1</td>\n",
       "      <td>36005</td>\n",
       "      <td>40.813224</td>\n",
       "      <td>-73.929849</td>\n",
       "      <td>800595.0</td>\n",
       "      <td>864369.0</td>\n",
       "      <td>887662.0</td>\n",
       "      <td>898509.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2526.0</td>\n",
       "      <td>2648.0</td>\n",
       "      <td>2540.0</td>\n",
       "      <td>3017.0</td>\n",
       "      <td>3136.0</td>\n",
       "      <td>3123.0</td>\n",
       "      <td>2793</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.92985 40.81322)</td>\n",
       "      <td>163.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bx002</th>\n",
       "      <td>149 St-Grand Concourse</td>\n",
       "      <td>2 4 5</td>\n",
       "      <td>2</td>\n",
       "      <td>36005</td>\n",
       "      <td>40.818375</td>\n",
       "      <td>-73.927351</td>\n",
       "      <td>3112547.0</td>\n",
       "      <td>3454530.0</td>\n",
       "      <td>3660150.0</td>\n",
       "      <td>3979328.0</td>\n",
       "      <td>...</td>\n",
       "      <td>14022.0</td>\n",
       "      <td>13024.0</td>\n",
       "      <td>14202.0</td>\n",
       "      <td>15326.0</td>\n",
       "      <td>13683.0</td>\n",
       "      <td>12797.0</td>\n",
       "      <td>12623</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.92735 40.81838)</td>\n",
       "      <td>593.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bx003</th>\n",
       "      <td>161 St-Yankee Stadium</td>\n",
       "      <td>B D 4</td>\n",
       "      <td>2</td>\n",
       "      <td>36005</td>\n",
       "      <td>40.827905</td>\n",
       "      <td>-73.925651</td>\n",
       "      <td>7836990.0</td>\n",
       "      <td>8576546.0</td>\n",
       "      <td>8410256.0</td>\n",
       "      <td>8434247.0</td>\n",
       "      <td>...</td>\n",
       "      <td>34258.0</td>\n",
       "      <td>35820.0</td>\n",
       "      <td>35071.0</td>\n",
       "      <td>36176.0</td>\n",
       "      <td>36969.0</td>\n",
       "      <td>35171.0</td>\n",
       "      <td>33398</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.92565 40.82791)</td>\n",
       "      <td>3637.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bx004</th>\n",
       "      <td>167 St</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>36005</td>\n",
       "      <td>40.835537</td>\n",
       "      <td>-73.921400</td>\n",
       "      <td>2715327.0</td>\n",
       "      <td>2876058.0</td>\n",
       "      <td>2892398.0</td>\n",
       "      <td>2931947.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11089.0</td>\n",
       "      <td>11188.0</td>\n",
       "      <td>10176.0</td>\n",
       "      <td>10417.0</td>\n",
       "      <td>11942.0</td>\n",
       "      <td>11579.0</td>\n",
       "      <td>9981</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.92140 40.83554)</td>\n",
       "      <td>680.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bx005</th>\n",
       "      <td>167 St</td>\n",
       "      <td>B D</td>\n",
       "      <td>1</td>\n",
       "      <td>36005</td>\n",
       "      <td>40.833771</td>\n",
       "      <td>-73.918440</td>\n",
       "      <td>2834640.0</td>\n",
       "      <td>2920517.0</td>\n",
       "      <td>2874428.0</td>\n",
       "      <td>2907900.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11956.0</td>\n",
       "      <td>11924.0</td>\n",
       "      <td>12887.0</td>\n",
       "      <td>13220.0</td>\n",
       "      <td>12890.0</td>\n",
       "      <td>12775.0</td>\n",
       "      <td>12682</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.91844 40.83377)</td>\n",
       "      <td>572.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qn074</th>\n",
       "      <td>Sutphin Blvd-Archer Av-JFK Airport</td>\n",
       "      <td>E J Z</td>\n",
       "      <td>1</td>\n",
       "      <td>36081</td>\n",
       "      <td>40.700486</td>\n",
       "      <td>-73.807969</td>\n",
       "      <td>6064612.0</td>\n",
       "      <td>6433286.0</td>\n",
       "      <td>6418547.0</td>\n",
       "      <td>6447287.0</td>\n",
       "      <td>...</td>\n",
       "      <td>24498.0</td>\n",
       "      <td>25388.0</td>\n",
       "      <td>26184.0</td>\n",
       "      <td>28862.0</td>\n",
       "      <td>28383.0</td>\n",
       "      <td>28587.0</td>\n",
       "      <td>25234</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.80797 40.70049)</td>\n",
       "      <td>1722.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qn075</th>\n",
       "      <td>Vernon Blvd-Jackson Av</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>36081</td>\n",
       "      <td>40.742626</td>\n",
       "      <td>-73.953581</td>\n",
       "      <td>2414808.0</td>\n",
       "      <td>2967358.0</td>\n",
       "      <td>2939995.0</td>\n",
       "      <td>3110952.0</td>\n",
       "      <td>...</td>\n",
       "      <td>9441.0</td>\n",
       "      <td>8612.0</td>\n",
       "      <td>9491.0</td>\n",
       "      <td>10347.0</td>\n",
       "      <td>12582.0</td>\n",
       "      <td>14981.0</td>\n",
       "      <td>14901</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.95358 40.74263)</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qn076</th>\n",
       "      <td>Woodhaven Blvd</td>\n",
       "      <td>J Z</td>\n",
       "      <td>1</td>\n",
       "      <td>36081</td>\n",
       "      <td>40.693879</td>\n",
       "      <td>-73.851576</td>\n",
       "      <td>1180476.0</td>\n",
       "      <td>1269407.0</td>\n",
       "      <td>1285768.0</td>\n",
       "      <td>1285323.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4227.0</td>\n",
       "      <td>3835.0</td>\n",
       "      <td>4255.0</td>\n",
       "      <td>5226.0</td>\n",
       "      <td>5388.0</td>\n",
       "      <td>5099.0</td>\n",
       "      <td>3884</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.85158 40.69388)</td>\n",
       "      <td>104.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qn077</th>\n",
       "      <td>Woodhaven Blvd</td>\n",
       "      <td>M R</td>\n",
       "      <td>1</td>\n",
       "      <td>36081</td>\n",
       "      <td>40.733106</td>\n",
       "      <td>-73.869229</td>\n",
       "      <td>7443133.0</td>\n",
       "      <td>7695845.0</td>\n",
       "      <td>7447147.0</td>\n",
       "      <td>7307038.0</td>\n",
       "      <td>...</td>\n",
       "      <td>28011.0</td>\n",
       "      <td>27847.0</td>\n",
       "      <td>27194.0</td>\n",
       "      <td>27105.0</td>\n",
       "      <td>26745.0</td>\n",
       "      <td>25550.0</td>\n",
       "      <td>23929</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.86923 40.73311)</td>\n",
       "      <td>653.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qn078</th>\n",
       "      <td>Woodside-61 St</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>36081</td>\n",
       "      <td>40.745630</td>\n",
       "      <td>-73.902984</td>\n",
       "      <td>5169510.0</td>\n",
       "      <td>5517813.0</td>\n",
       "      <td>5379712.0</td>\n",
       "      <td>5394103.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20841.0</td>\n",
       "      <td>20030.0</td>\n",
       "      <td>20432.0</td>\n",
       "      <td>19669.0</td>\n",
       "      <td>20127.0</td>\n",
       "      <td>20360.0</td>\n",
       "      <td>21067</td>\n",
       "      <td>NaN</td>\n",
       "      <td>POINT (-73.90298 40.74563)</td>\n",
       "      <td>80.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>425 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    complex_nm trains  station_ct  bcode  \\\n",
       "complex_id                                                                 \n",
       "bx001                   138 St-Grand Concourse    4 5           1  36005   \n",
       "bx002                   149 St-Grand Concourse  2 4 5           2  36005   \n",
       "bx003                    161 St-Yankee Stadium  B D 4           2  36005   \n",
       "bx004                                   167 St      4           1  36005   \n",
       "bx005                                   167 St    B D           1  36005   \n",
       "...                                        ...    ...         ...    ...   \n",
       "qn074       Sutphin Blvd-Archer Av-JFK Airport  E J Z           1  36081   \n",
       "qn075                   Vernon Blvd-Jackson Av      7           1  36081   \n",
       "qn076                           Woodhaven Blvd    J Z           1  36081   \n",
       "qn077                           Woodhaven Blvd    M R           1  36081   \n",
       "qn078                           Woodside-61 St      7           1  36081   \n",
       "\n",
       "             stop_lat   stop_lon    tot2007    tot2008    tot2009    tot2010  \\\n",
       "complex_id                                                                     \n",
       "bx001       40.813224 -73.929849   800595.0   864369.0   887662.0   898509.0   \n",
       "bx002       40.818375 -73.927351  3112547.0  3454530.0  3660150.0  3979328.0   \n",
       "bx003       40.827905 -73.925651  7836990.0  8576546.0  8410256.0  8434247.0   \n",
       "bx004       40.835537 -73.921400  2715327.0  2876058.0  2892398.0  2931947.0   \n",
       "bx005       40.833771 -73.918440  2834640.0  2920517.0  2874428.0  2907900.0   \n",
       "...               ...        ...        ...        ...        ...        ...   \n",
       "qn074       40.700486 -73.807969  6064612.0  6433286.0  6418547.0  6447287.0   \n",
       "qn075       40.742626 -73.953581  2414808.0  2967358.0  2939995.0  3110952.0   \n",
       "qn076       40.693879 -73.851576  1180476.0  1269407.0  1285768.0  1285323.0   \n",
       "qn077       40.733106 -73.869229  7443133.0  7695845.0  7447147.0  7307038.0   \n",
       "qn078       40.745630 -73.902984  5169510.0  5517813.0  5379712.0  5394103.0   \n",
       "\n",
       "            ...  avwken11  avwken12  avwken13  avwken14  avwken15  avwken16  \\\n",
       "complex_id  ...                                                               \n",
       "bx001       ...    2526.0    2648.0    2540.0    3017.0    3136.0    3123.0   \n",
       "bx002       ...   14022.0   13024.0   14202.0   15326.0   13683.0   12797.0   \n",
       "bx003       ...   34258.0   35820.0   35071.0   36176.0   36969.0   35171.0   \n",
       "bx004       ...   11089.0   11188.0   10176.0   10417.0   11942.0   11579.0   \n",
       "bx005       ...   11956.0   11924.0   12887.0   13220.0   12890.0   12775.0   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "qn074       ...   24498.0   25388.0   26184.0   28862.0   28383.0   28587.0   \n",
       "qn075       ...    9441.0    8612.0    9491.0   10347.0   12582.0   14981.0   \n",
       "qn076       ...    4227.0    3835.0    4255.0    5226.0    5388.0    5099.0   \n",
       "qn077       ...   28011.0   27847.0   27194.0   27105.0   26745.0   25550.0   \n",
       "qn078       ...   20841.0   20030.0   20432.0   19669.0   20127.0   20360.0   \n",
       "\n",
       "            avwken17  srv_notes                    geometry  arrests  \n",
       "complex_id                                                            \n",
       "bx001           2793        NaN  POINT (-73.92985 40.81322)    163.0  \n",
       "bx002          12623        NaN  POINT (-73.92735 40.81838)    593.0  \n",
       "bx003          33398        NaN  POINT (-73.92565 40.82791)   3637.0  \n",
       "bx004           9981        NaN  POINT (-73.92140 40.83554)    680.0  \n",
       "bx005          12682        NaN  POINT (-73.91844 40.83377)    572.0  \n",
       "...              ...        ...                         ...      ...  \n",
       "qn074          25234        NaN  POINT (-73.80797 40.70049)   1722.0  \n",
       "qn075          14901        NaN  POINT (-73.95358 40.74263)     11.0  \n",
       "qn076           3884        NaN  POINT (-73.85158 40.69388)    104.0  \n",
       "qn077          23929        NaN  POINT (-73.86923 40.73311)    653.0  \n",
       "qn078          21067        NaN  POINT (-73.90298 40.74563)     80.0  \n",
       "\n",
       "[425 rows x 42 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stations"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}