Midterm Project Work.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chicago Crash Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# anna poon's analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('fivethirtyeight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash = pd.read_csv(\"traffic_crashes.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>CRASH_RECORD_ID</th>\n",
       "      <th>RD_NO</th>\n",
       "      <th>CRASH_DATE_EST_I</th>\n",
       "      <th>CRASH_DATE</th>\n",
       "      <th>POSTED_SPEED_LIMIT</th>\n",
       "      <th>TRAFFIC_CONTROL_DEVICE</th>\n",
       "      <th>DEVICE_CONDITION</th>\n",
       "      <th>WEATHER_CONDITION</th>\n",
       "      <th>LIGHTING_CONDITION</th>\n",
       "      <th>FIRST_CRASH_TYPE</th>\n",
       "      <th>...</th>\n",
       "      <th>INJURIES_NON_INCAPACITATING</th>\n",
       "      <th>INJURIES_REPORTED_NOT_EVIDENT</th>\n",
       "      <th>INJURIES_NO_INDICATION</th>\n",
       "      <th>INJURIES_UNKNOWN</th>\n",
       "      <th>CRASH_HOUR</th>\n",
       "      <th>CRASH_DAY_OF_WEEK</th>\n",
       "      <th>CRASH_MONTH</th>\n",
       "      <th>LATITUDE</th>\n",
       "      <th>LONGITUDE</th>\n",
       "      <th>LOCATION</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>016eaf0b25d64bf0917f77ad2c7468196178eaea750a5f...</td>\n",
       "      <td>JE287447</td>\n",
       "      <td>NaN</td>\n",
       "      <td>07/02/2021 03:40:00 PM</td>\n",
       "      <td>20</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>CLEAR</td>\n",
       "      <td>DAYLIGHT</td>\n",
       "      <td>SIDESWIPE SAME DIRECTION</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>06ab00f94c073b8bd07ccf085935a800077ad9cee09490...</td>\n",
       "      <td>JE192048</td>\n",
       "      <td>NaN</td>\n",
       "      <td>04/07/2021 09:00:00 AM</td>\n",
       "      <td>20</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>CLEAR</td>\n",
       "      <td>DAYLIGHT</td>\n",
       "      <td>PARKED MOTOR VEHICLE</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>41.743068</td>\n",
       "      <td>-87.693816</td>\n",
       "      <td>POINT (-87.693815614874 41.743067832139)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>176fbd45c70226ad9a63ff015d87090dd51b2674d0f23e...</td>\n",
       "      <td>JD181795</td>\n",
       "      <td>NaN</td>\n",
       "      <td>03/11/2020 10:00:00 AM</td>\n",
       "      <td>10</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>CLEAR</td>\n",
       "      <td>DAYLIGHT</td>\n",
       "      <td>REAR TO REAR</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>41.684308</td>\n",
       "      <td>-87.680775</td>\n",
       "      <td>POINT (-87.680774721691 41.68430795958)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>242250246f9ea909eb5805f45125840489a1ea37fab7dd...</td>\n",
       "      <td>JC128630</td>\n",
       "      <td>NaN</td>\n",
       "      <td>01/24/2019 07:40:00 PM</td>\n",
       "      <td>30</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>NO CONTROLS</td>\n",
       "      <td>CLEAR</td>\n",
       "      <td>DARKNESS, LIGHTED ROAD</td>\n",
       "      <td>SIDESWIPE SAME DIRECTION</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>19</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>41.886422</td>\n",
       "      <td>-87.644413</td>\n",
       "      <td>POINT (-87.64441278977 41.886421852271)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4 rows × 49 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     CRASH_RECORD_ID     RD_NO  \\\n",
       "0  016eaf0b25d64bf0917f77ad2c7468196178eaea750a5f...  JE287447   \n",
       "1  06ab00f94c073b8bd07ccf085935a800077ad9cee09490...  JE192048   \n",
       "2  176fbd45c70226ad9a63ff015d87090dd51b2674d0f23e...  JD181795   \n",
       "3  242250246f9ea909eb5805f45125840489a1ea37fab7dd...  JC128630   \n",
       "\n",
       "  CRASH_DATE_EST_I              CRASH_DATE  POSTED_SPEED_LIMIT  \\\n",
       "0              NaN  07/02/2021 03:40:00 PM                  20   \n",
       "1              NaN  04/07/2021 09:00:00 AM                  20   \n",
       "2              NaN  03/11/2020 10:00:00 AM                  10   \n",
       "3              NaN  01/24/2019 07:40:00 PM                  30   \n",
       "\n",
       "  TRAFFIC_CONTROL_DEVICE DEVICE_CONDITION WEATHER_CONDITION  \\\n",
       "0            NO CONTROLS      NO CONTROLS             CLEAR   \n",
       "1            NO CONTROLS      NO CONTROLS             CLEAR   \n",
       "2            NO CONTROLS      NO CONTROLS             CLEAR   \n",
       "3            NO CONTROLS      NO CONTROLS             CLEAR   \n",
       "\n",
       "       LIGHTING_CONDITION          FIRST_CRASH_TYPE  ...  \\\n",
       "0                DAYLIGHT  SIDESWIPE SAME DIRECTION  ...   \n",
       "1                DAYLIGHT      PARKED MOTOR VEHICLE  ...   \n",
       "2                DAYLIGHT              REAR TO REAR  ...   \n",
       "3  DARKNESS, LIGHTED ROAD  SIDESWIPE SAME DIRECTION  ...   \n",
       "\n",
       "  INJURIES_NON_INCAPACITATING  INJURIES_REPORTED_NOT_EVIDENT  \\\n",
       "0                         0.0                            0.0   \n",
       "1                         0.0                            0.0   \n",
       "2                         0.0                            0.0   \n",
       "3                         0.0                            0.0   \n",
       "\n",
       "  INJURIES_NO_INDICATION INJURIES_UNKNOWN CRASH_HOUR CRASH_DAY_OF_WEEK  \\\n",
       "0                    2.0              0.0         15                 6   \n",
       "1                    1.0              0.0          9                 4   \n",
       "2                    2.0              0.0         10                 4   \n",
       "3                    2.0              0.0         19                 5   \n",
       "\n",
       "  CRASH_MONTH   LATITUDE  LONGITUDE                                  LOCATION  \n",
       "0           7        NaN        NaN                                       NaN  \n",
       "1           4  41.743068 -87.693816  POINT (-87.693815614874 41.743067832139)  \n",
       "2           3  41.684308 -87.680775   POINT (-87.680774721691 41.68430795958)  \n",
       "3           1  41.886422 -87.644413   POINT (-87.64441278977 41.886421852271)  \n",
       "\n",
       "[4 rows x 49 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.head(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>POSTED_SPEED_LIMIT</th>\n",
       "      <th>LANE_CNT</th>\n",
       "      <th>STREET_NO</th>\n",
       "      <th>BEAT_OF_OCCURRENCE</th>\n",
       "      <th>NUM_UNITS</th>\n",
       "      <th>INJURIES_TOTAL</th>\n",
       "      <th>INJURIES_FATAL</th>\n",
       "      <th>INJURIES_INCAPACITATING</th>\n",
       "      <th>INJURIES_NON_INCAPACITATING</th>\n",
       "      <th>INJURIES_REPORTED_NOT_EVIDENT</th>\n",
       "      <th>INJURIES_NO_INDICATION</th>\n",
       "      <th>INJURIES_UNKNOWN</th>\n",
       "      <th>CRASH_HOUR</th>\n",
       "      <th>CRASH_DAY_OF_WEEK</th>\n",
       "      <th>CRASH_MONTH</th>\n",
       "      <th>LATITUDE</th>\n",
       "      <th>LONGITUDE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>565687.000000</td>\n",
       "      <td>1.903210e+05</td>\n",
       "      <td>565687.000000</td>\n",
       "      <td>565682.000000</td>\n",
       "      <td>565687.000000</td>\n",
       "      <td>564517.000000</td>\n",
       "      <td>564517.000000</td>\n",
       "      <td>564517.000000</td>\n",
       "      <td>564517.000000</td>\n",
       "      <td>564517.000000</td>\n",
       "      <td>564517.000000</td>\n",
       "      <td>564517.0</td>\n",
       "      <td>565687.000000</td>\n",
       "      <td>565687.000000</td>\n",
       "      <td>565687.000000</td>\n",
       "      <td>562352.000000</td>\n",
       "      <td>562352.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>28.342368</td>\n",
       "      <td>1.382022e+01</td>\n",
       "      <td>3675.132524</td>\n",
       "      <td>1238.706687</td>\n",
       "      <td>2.035155</td>\n",
       "      <td>0.184064</td>\n",
       "      <td>0.001151</td>\n",
       "      <td>0.020030</td>\n",
       "      <td>0.103481</td>\n",
       "      <td>0.059401</td>\n",
       "      <td>2.015652</td>\n",
       "      <td>0.0</td>\n",
       "      <td>13.231306</td>\n",
       "      <td>4.125094</td>\n",
       "      <td>6.792265</td>\n",
       "      <td>41.854660</td>\n",
       "      <td>-87.673299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.286897</td>\n",
       "      <td>3.028444e+03</td>\n",
       "      <td>2906.837601</td>\n",
       "      <td>705.983650</td>\n",
       "      <td>0.451937</td>\n",
       "      <td>0.556900</td>\n",
       "      <td>0.036770</td>\n",
       "      <td>0.165469</td>\n",
       "      <td>0.415107</td>\n",
       "      <td>0.312451</td>\n",
       "      <td>1.167690</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.538002</td>\n",
       "      <td>1.979108</td>\n",
       "      <td>3.402583</td>\n",
       "      <td>0.331965</td>\n",
       "      <td>0.674209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>111.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-87.935877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>30.000000</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>1227.000000</td>\n",
       "      <td>712.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>41.781140</td>\n",
       "      <td>-87.721396</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>30.000000</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>3200.000000</td>\n",
       "      <td>1135.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>41.874790</td>\n",
       "      <td>-87.673621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>30.000000</td>\n",
       "      <td>4.000000e+00</td>\n",
       "      <td>5600.000000</td>\n",
       "      <td>1822.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>41.924051</td>\n",
       "      <td>-87.632968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>70.000000</td>\n",
       "      <td>1.191625e+06</td>\n",
       "      <td>451100.000000</td>\n",
       "      <td>6100.000000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>42.022780</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       POSTED_SPEED_LIMIT      LANE_CNT      STREET_NO  BEAT_OF_OCCURRENCE  \\\n",
       "count       565687.000000  1.903210e+05  565687.000000       565682.000000   \n",
       "mean            28.342368  1.382022e+01    3675.132524         1238.706687   \n",
       "std              6.286897  3.028444e+03    2906.837601          705.983650   \n",
       "min              0.000000  0.000000e+00       0.000000          111.000000   \n",
       "25%             30.000000  2.000000e+00    1227.000000          712.000000   \n",
       "50%             30.000000  2.000000e+00    3200.000000         1135.000000   \n",
       "75%             30.000000  4.000000e+00    5600.000000         1822.000000   \n",
       "max             70.000000  1.191625e+06  451100.000000         6100.000000   \n",
       "\n",
       "           NUM_UNITS  INJURIES_TOTAL  INJURIES_FATAL  INJURIES_INCAPACITATING  \\\n",
       "count  565687.000000   564517.000000   564517.000000            564517.000000   \n",
       "mean        2.035155        0.184064        0.001151                 0.020030   \n",
       "std         0.451937        0.556900        0.036770                 0.165469   \n",
       "min         1.000000        0.000000        0.000000                 0.000000   \n",
       "25%         2.000000        0.000000        0.000000                 0.000000   \n",
       "50%         2.000000        0.000000        0.000000                 0.000000   \n",
       "75%         2.000000        0.000000        0.000000                 0.000000   \n",
       "max        18.000000       21.000000        4.000000                 7.000000   \n",
       "\n",
       "       INJURIES_NON_INCAPACITATING  INJURIES_REPORTED_NOT_EVIDENT  \\\n",
       "count                564517.000000                  564517.000000   \n",
       "mean                      0.103481                       0.059401   \n",
       "std                       0.415107                       0.312451   \n",
       "min                       0.000000                       0.000000   \n",
       "25%                       0.000000                       0.000000   \n",
       "50%                       0.000000                       0.000000   \n",
       "75%                       0.000000                       0.000000   \n",
       "max                      21.000000                      15.000000   \n",
       "\n",
       "       INJURIES_NO_INDICATION  INJURIES_UNKNOWN     CRASH_HOUR  \\\n",
       "count           564517.000000          564517.0  565687.000000   \n",
       "mean                 2.015652               0.0      13.231306   \n",
       "std                  1.167690               0.0       5.538002   \n",
       "min                  0.000000               0.0       0.000000   \n",
       "25%                  1.000000               0.0       9.000000   \n",
       "50%                  2.000000               0.0      14.000000   \n",
       "75%                  2.000000               0.0      17.000000   \n",
       "max                 61.000000               0.0      23.000000   \n",
       "\n",
       "       CRASH_DAY_OF_WEEK    CRASH_MONTH       LATITUDE      LONGITUDE  \n",
       "count      565687.000000  565687.000000  562352.000000  562352.000000  \n",
       "mean            4.125094       6.792265      41.854660     -87.673299  \n",
       "std             1.979108       3.402583       0.331965       0.674209  \n",
       "min             1.000000       1.000000       0.000000     -87.935877  \n",
       "25%             2.000000       4.000000      41.781140     -87.721396  \n",
       "50%             4.000000       7.000000      41.874790     -87.673621  \n",
       "75%             6.000000      10.000000      41.924051     -87.632968  \n",
       "max             7.000000      12.000000      42.022780       0.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.fillna(value=0, inplace=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "565687"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['CRASH_RECORD_ID', 'RD_NO', 'CRASH_DATE_EST_I', 'CRASH_DATE',\n",
       "       'POSTED_SPEED_LIMIT', 'TRAFFIC_CONTROL_DEVICE', 'DEVICE_CONDITION',\n",
       "       'WEATHER_CONDITION', 'LIGHTING_CONDITION', 'FIRST_CRASH_TYPE',\n",
       "       'TRAFFICWAY_TYPE', 'LANE_CNT', 'ALIGNMENT', 'ROADWAY_SURFACE_COND',\n",
       "       'ROAD_DEFECT', 'REPORT_TYPE', 'CRASH_TYPE', 'INTERSECTION_RELATED_I',\n",
       "       'NOT_RIGHT_OF_WAY_I', 'HIT_AND_RUN_I', 'DAMAGE', 'DATE_POLICE_NOTIFIED',\n",
       "       'PRIM_CONTRIBUTORY_CAUSE', 'SEC_CONTRIBUTORY_CAUSE', 'STREET_NO',\n",
       "       'STREET_DIRECTION', 'STREET_NAME', 'BEAT_OF_OCCURRENCE',\n",
       "       'PHOTOS_TAKEN_I', 'STATEMENTS_TAKEN_I', 'DOORING_I', 'WORK_ZONE_I',\n",
       "       'WORK_ZONE_TYPE', 'WORKERS_PRESENT_I', 'NUM_UNITS',\n",
       "       'MOST_SEVERE_INJURY', 'INJURIES_TOTAL', 'INJURIES_FATAL',\n",
       "       'INJURIES_INCAPACITATING', 'INJURIES_NON_INCAPACITATING',\n",
       "       'INJURIES_REPORTED_NOT_EVIDENT', 'INJURIES_NO_INDICATION',\n",
       "       'INJURIES_UNKNOWN', 'CRASH_HOUR', 'CRASH_DAY_OF_WEEK', 'CRASH_MONTH',\n",
       "       'LATITUDE', 'LONGITUDE', 'LOCATION'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.columns\n",
    "\n",
    "# use posted speed limit, weather condition, lighting condition, road defect, crash hour, roadway surface conditions\n",
    "# predict injuries vs no injuries\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NO INJURY / DRIVE AWAY              418500\n",
       "INJURY AND / OR TOW DUE TO CRASH    147187\n",
       "Name: CRASH_TYPE, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.CRASH_TYPE.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6    92128\n",
       "7    83877\n",
       "5    80950\n",
       "3    80659\n",
       "4    80154\n",
       "2    78343\n",
       "1    69576\n",
       "Name: CRASH_DAY_OF_WEEK, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.CRASH_DAY_OF_WEEK.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16    43298\n",
       "15    43077\n",
       "17    42332\n",
       "14    38153\n",
       "18    35165\n",
       "13    34894\n",
       "12    33528\n",
       "8     29343\n",
       "11    29067\n",
       "9     26111\n",
       "10    25910\n",
       "19    25858\n",
       "7     23663\n",
       "20    20666\n",
       "21    18466\n",
       "22    17045\n",
       "23    14492\n",
       "6     12462\n",
       "0     11948\n",
       "1     10211\n",
       "2      8700\n",
       "5      7747\n",
       "3      7116\n",
       "4      6435\n",
       "Name: CRASH_HOUR, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.CRASH_HOUR.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CLEAR                       449627\n",
       "RAIN                         49376\n",
       "UNKNOWN                      25291\n",
       "SNOW                         20252\n",
       "CLOUDY/OVERCAST              16824\n",
       "OTHER                         1719\n",
       "FOG/SMOKE/HAZE                 867\n",
       "FREEZING RAIN/DRIZZLE          724\n",
       "SLEET/HAIL                     720\n",
       "BLOWING SNOW                   169\n",
       "SEVERE CROSS WIND GATE         116\n",
       "BLOWING SAND, SOIL, DIRT         2\n",
       "Name: WEATHER_CONDITION, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.WEATHER_CONDITION.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0     489964\n",
       "1.0      57265\n",
       "2.0      12473\n",
       "3.0       3787\n",
       "4.0       1368\n",
       "5.0        490\n",
       "6.0        197\n",
       "7.0         73\n",
       "8.0         26\n",
       "9.0         16\n",
       "10.0         9\n",
       "15.0         5\n",
       "11.0         5\n",
       "21.0         3\n",
       "12.0         2\n",
       "13.0         2\n",
       "19.0         1\n",
       "16.0         1\n",
       "Name: INJURIES_TOTAL, dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash[\"INJURIES_TOTAL\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'whiskers': [<matplotlib.lines.Line2D at 0x7fe5109be6a0>,\n",
       "  <matplotlib.lines.Line2D at 0x7fe5109bea30>],\n",
       " 'caps': [<matplotlib.lines.Line2D at 0x7fe5109bedc0>,\n",
       "  <matplotlib.lines.Line2D at 0x7fe5109ce190>],\n",
       " 'boxes': [<matplotlib.lines.Line2D at 0x7fe5109be2e0>],\n",
       " 'medians': [<matplotlib.lines.Line2D at 0x7fe5109ce520>],\n",
       " 'fliers': [],\n",
       " 'means': []}"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbQAAAEJCAYAAAD8YNLtAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAATsklEQVR4nO3dcUzU5+HH8c/9Tm3YmnD2PEBGleqYPyA6bOddcJ2UI21S1sI0ntqlzYpNM8u5blEs0G5SndsktMZUW9pMLqlrTUY7HOCcSxOvgRYHNBtZApFeUmusChdQalmKWrzfH8bb8ABxu5PfPb5fiX/4fJ/n7vkakne+3/tyWoaGhkICACDO/c90bwAAgGggaAAAIxA0AIARCBoAwAgEDQBgBIIGADACQQMAGIGgAQCMQNCAOBcIBKZ7C8D/CwQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGCEmAVt3759WrJkiZKTk5WXl6e2trZJ53d3d6uwsFApKSnKzMxUdXW1QqHQuHOPHTsmu92u3NzcWGwdABCHYhK0hoYGVVRUaPPmzWppaZHT6ZTH49GpU6fGnX/hwgWtXLlSSUlJOnr0qHbu3Kk9e/Zo7969EXOHhoa0YcMG5eXlxWLrAIA4FZOgvfrqq/rhD3+oH/3oR1q0aJFqamqUnJwsn8837vx33nlHX375pWpra5WVlaXi4mL99Kc/1WuvvRZxlbZx40Y99thjWrZsWSy2DgCIU1EP2qVLl9TV1SW32z1m3O12q729fdw1HR0dys3NVUJCQnisoKBAZ8+e1cmTJ8Nj+/btUzAY1JYtW6K9bQBAnJsR7RccHBzU6OioHA7HmHGHw6FgMDjummAwqNTU1Ij5146lp6eru7tb1dXVeu+992S1Wqe8n0AgcJNnAMQffs5xO8jIyJj0eNSDdo3FYhnz91AoFDF2o/nXxi9evKinnnpKv/zlL5Wenn5T+7jRPwAQ7wKBAD/ngGIQNLvdLqvVGnE1NjAwEHHVdk1SUtK486WrV2p9fX06fvy4vF6vvF6vJOnKlSsKhUKy2+165513Im5xAgBuL1H/DG3WrFnKycmR3+8fM+73++VyucZd43Q6dezYMY2MjIyZP3fuXM2fP1+pqalqa2tTa2tr+M/69eu1YMECtba2yul0Rvs0AABxJiZPOXq9Xh04cED79+9Xb2+vysvL1dfXp5KSEknStm3bVFRUFJ6/evVqJSQkqLS0VD09PWpqatLu3btVWloqi8WimTNnKisra8yfOXPm6I477lBWVpbuvPPOWJwGACCOxOQztFWrVuncuXOqqalRf3+/MjMzVV9fr3nz5kmS+vr6dOLEifD8xMREHTx4UGVlZcrPz5fNZpPX69XGjRtjsT0AgIEsQ0ND438dB4C4wEMhwFV8lyMAwAgEDQBgBIIGADACQQMAGIGgAQCMQNAAAEYgaAAAIxA0AIARCBoAwAgEDQBgBIIGADACQQMAGIGgAQCMQNAAAEYgaAAAIxA0AIARCBoAwAgEDQBgBIIGADACQQMAGIGgAQCMQNAAAEYgaAAAIxA0AIARCBoAwAgEDQBgBIIGADACQQMAGIGgAQCMQNAAAEYgaAAAI8QsaPv27dOSJUuUnJysvLw8tbW1TTq/u7tbhYWFSklJUWZmpqqrqxUKhcLHm5qatHLlSi1cuFBpaWkqKCjQ4cOHY7V9AECciUnQGhoaVFFRoc2bN6ulpUVOp1Mej0enTp0ad/6FCxe0cuVKJSUl6ejRo9q5c6f27NmjvXv3hud8+OGHWrFiherr69XS0qIHH3xQjz/++A1DCQC4PViGhoZCN552cwoKCpSdna1XXnklPHbvvfequLhYVVVVEfPr6ur04osv6uOPP1ZCQoIkqaamRj6fTz09PbJYLOO+j9vtVm5urn71q19F+xSAuBEIBJSRkTHd2wCmXdSv0C5duqSuri653e4x4263W+3t7eOu6ejoUG5ubjhm0tUonj17VidPnpzwvYaHh2Wz2aKybwBAfJsR7RccHBzU6OioHA7HmHGHw6FgMDjummAwqNTU1Ij5146lp6dHrPntb3+rM2fOaO3atZPuJxAI3MTugfjEzzluBze6ExH1oF1z/W3CUCg04a3DieaPNy5JjY2N2rp1q+rq6jRv3rxJ98GtGJiOW47AVVG/5Wi322W1WiOuxgYGBiKu2q5JSkoad76kiDWNjY3asGGDXn/9dRUWFkZx5wCAeBb1oM2aNUs5OTny+/1jxv1+v1wu17hrnE6njh07ppGRkTHz586dq/nz54fHDh48qB//+Md67bXXVFxcHO2tAwDiWEwe2/d6vTpw4ID279+v3t5elZeXq6+vTyUlJZKkbdu2qaioKDx/9erVSkhIUGlpqXp6etTU1KTdu3ertLQ0fMvxD3/4g55++mlVVVVp+fLl6u/vV39/v86fPx+LUwAAxJmYfIa2atUqnTt3TjU1Nerv71dmZqbq6+vDn3f19fXpxIkT4fmJiYk6ePCgysrKlJ+fL5vNJq/Xq40bN4bn+Hw+ffXVV6qsrFRlZWV4/Lvf/a7+9Kc/xeI0AABxJCa/hwbg1uGhEOAqvssRAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBFiFrR9+/ZpyZIlSk5OVl5entra2iad393drcLCQqWkpCgzM1PV1dUKhUJj5nzwwQfKy8tTcnKyvv3tb8vn88Vq+wCAOBOToDU0NKiiokKbN29WS0uLnE6nPB6PTp06Ne78CxcuaOXKlUpKStLRo0e1c+dO7dmzR3v37g3P+fTTT7VmzRo5nU61tLRo06ZNeu6559TY2BiLUwAAxBnL0NBQ6MbTbk5BQYGys7P1yiuvhMfuvfdeFRcXq6qqKmJ+XV2dXnzxRX388cdKSEiQJNXU1Mjn86mnp0cWi0VVVVVqbm7W3/72t/C6n/zkJzp+/Ljee++9aJ8CEDcCgYAyMjKmexvAtIv6FdqlS5fU1dUlt9s9Ztztdqu9vX3cNR0dHcrNzQ3HTLoaxbNnz+rkyZPhOde/ZkFBgf7+97/r8uXLUT4LAEC8iXrQBgcHNTo6KofDMWbc4XAoGAyOuyYYDI47/9qxyeZ89dVXGhwcjNb2AQBxakasXthisYz5eygUihi70fzrx6cy53qBQGBqGwaipPnxR7X1f1Nu2fstvWXv9C/bj/fp0beap+GdcTu70a31qAfNbrfLarVGXI0NDAxEXGFdk5SUNO586V9XahPNmTFjhu66664J98NnC7jVNrUf1/AtfL/p+Axt0y19N2Bqon7LcdasWcrJyZHf7x8z7vf75XK5xl3jdDp17NgxjYyMjJk/d+5czZ8/Pzzn/fffj3jNpUuXaubMmdE9CQBA3InJY/ter1cHDhzQ/v371dvbq/LycvX19amkpESStG3bNhUVFYXnr169WgkJCSotLVVPT4+ampq0e/dulZaWhm8nlpSU6MyZM6qoqFBvb6/279+vAwcOaOPGjbE4BQBAnInJZ2irVq3SuXPnVFNTo/7+fmVmZqq+vl7z5s2TJPX19enEiRPh+YmJiTp48KDKysqUn58vm80mr9c7Jlbp6emqr6/X888/L5/Pp5SUFFVXV6u4uDgWpwAAiDMx+T00ALcOv4cGXMV3OQIAjEDQAABGIGgAACMQNACAEQgaAMAIBA0AYASCBgAwAkEDABiBoAEAjEDQAABGIGgAACMQNACAEQgaAMAIBA0AYASCBgAwAkEDABiBoAEAjEDQAABGIGgAACMQNACAEQgaAMAIBA0AYASCBgAwAkEDABiBoAEAjEDQAABGIGgAACMQNACAEQgaAMAIBA0AYASCBgAwAkEDABgh6kG7ePGitmzZogULFig1NVXr1q3T6dOnb7iusbFRLpdLSUlJcrlcam5uHnN8165dys/P1913362FCxdq7dq16unpifb2AQBxKupBq6ysVHNzs+rq6nT48GF98cUXWrt2rUZHRydc09HRofXr18vj8ai1tVUej0dPPvmkPvroo/CcDz74QE899ZT+8pe/qKmpSTNmzNAPfvADnT9/PtqnAACIQ5ahoaFQtF7s888/1ze/+U29+uqrWrNmjSTps88+0+LFi/Xuu++qoKBg3HUlJSU6f/68/vjHP4bHiouLNWfOHNXV1Y27Znh4WPPmzdPbb7+thx9+OFqnAMSdQCCgjIyM6d4GMO2ieoXW1dWly5cvy+12h8fS0tK0aNEitbe3T7ius7NzzBpJKigomHTN8PCwrly5IpvN9l/vGwAQ/2ZE88WCwaCsVqvsdvuYcYfDoWAwOOG6/v5+ORyOm1pTUVGhxYsXy+l0TrqnQCAwhZ0D8Y2fc9wObnQnYkpB27Fjh1566aVJ51z/EMe/C4VCslgsk66//vhka55//nn99a9/1ZEjR2S1Wid9XW7FwHTccgSumlLQnnnmmfBnYhNJS0tTZ2enRkdHNTg4qDlz5oSPDQwMaPny5ROuTU5OjrgaGxgYiLhqk64+dNLQ0KDm5malp6dPZfsAgNvAlIJmt9sjbiOOJycnRzNnzpTf75fH45EknT59Wr29vXK5XBOuW7Zsmfx+v5599tnwmN/vj1hTXl6uhoYGHTp0SN/61remsnUAwG0iqp+hJSYm6oknntDWrVvlcDg0e/ZsvfDCC8rOztYDDzwQnldUVKT77rtPVVVVkqQNGzaosLBQu3bt0iOPPKJDhw6ptbVVR44cCa8pKyvT73//e7311luy2Wzq7++XJH3961/XnXfeGc3TAADEoag+ti9JIyMj+sUvfqF3331XIyMjWrFihV5++WWlpaWF5yxevFj333+/amtrw2ONjY3asWOHPv30U91zzz36+c9/rqKiovDxiZ5mLC8vV2VlZTRPAYgrfIYGXBX1oAG4tQgacBXf5QgAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACFEP2sWLF7VlyxYtWLBAqampWrdunU6fPn3DdY2NjXK5XEpKSpLL5VJzc/OEc19++WXZbDZt2bIlmlsHAMSxqAetsrJSzc3Nqqur0+HDh/XFF19o7dq1Gh0dnXBNR0eH1q9fL4/Ho9bWVnk8Hj355JP66KOPIuZ2dnbqzTffVHZ2drS3DgCIY1EN2ueff67f/e532r59u/Lz85WTk6M33nhD3d3dev/99ydcV1tbq+9973sqKyvTokWLVFZWpvvvv1+1tbURr//0009rz549stls0dw6ACDORTVoXV1dunz5stxud3gsLS1NixYtUnt7+4TrOjs7x6yRpIKCgog1P/vZz1RcXKy8vLxobhsAYIAZ0XyxYDAoq9Uqu90+ZtzhcCgYDE64rr+/Xw6HY9I1b775pj755BO98cYbN7WnQCBwU/OBeMTPOW4HGRkZkx6fUtB27Nihl156adI5kz3EEQqFZLFYJl1//fF/XxMIBLR9+3b9+c9/1qxZs6ay5bAb/QMA8S4QCPBzDmiKQXvmmWe0Zs2aSeekpaWps7NTo6OjGhwc1Jw5c8LHBgYGtHz58gnXJicnR1zBDQwMhK/aOjo6NDg4qNzc3PDx0dFRtbW1yefz6cyZM7rjjjumcioAAENNKWh2uz3iNuJ4cnJyNHPmTPn9fnk8HknS6dOn1dvbK5fLNeG6ZcuWye/369lnnw2P+f3+8Jrvf//7Wrp06Zg1Xq9XCxcu1KZNm276qg0AYJ6ofoaWmJioJ554Qlu3bpXD4dDs2bP1wgsvKDs7Ww888EB4XlFRke677z5VVVVJkjZs2KDCwkLt2rVLjzzyiA4dOqTW1lYdOXJEkmSz2SKeavza176m2bNnKysrK5qnAACIU1ENmiT9+te/ltVqVUlJiUZGRrRixQq9/vrrslqt4TknTpzQN77xjfDfXS6XfD6fduzYod/85je655575PP59J3vfCfa2wMAGMoyNDQUmu5NAPjP8VAIcBXf5QgAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEy9DQUGi6NwEAwH+LKzQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABcerDDz/UunXrlJmZKZvNprfffnu6twRMK4IGxKl//vOfysrK0s6dO5WQkDDd2wGm3Yzp3gCA/8xDDz2khx56SJJUWlo6zbsBph9XaAAAIxA0AIARCBoAwAgEDQBgBIIGADACTzkCcWp4eFiffPKJJOnKlSv67LPP9I9//EOzZ8/W3XffPc27A249/vsYIE61trbq0UcfjRh/7LHHVFtbOw07AqYXQQMAGIHP0AAARiBoAAAjEDQAgBEIGgDACAQNAGAEggYAMAJBAwAYgaABAIxA0AAARvg/5uhcfEc59RcAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.boxplot(crash['INJURIES_TOTAL'], showfliers = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0     489964\n",
       "1.0      57265\n",
       "2.0      12473\n",
       "3.0       3787\n",
       "4.0       1368\n",
       "5.0        490\n",
       "6.0        197\n",
       "7.0         73\n",
       "8.0         26\n",
       "9.0         16\n",
       "10.0         9\n",
       "15.0         5\n",
       "11.0         5\n",
       "21.0         3\n",
       "12.0         2\n",
       "13.0         2\n",
       "19.0         1\n",
       "16.0         1\n",
       "Name: INJURIES_TOTAL, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crash.INJURIES_TOTAL.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Does rain lead to more injuries in crashes caused by weather?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'not_winter' 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/var/folders/wz/7k6v05bx5gvffw4vkwfkmkn40000gn/T/ipykernel_85313/1678534244.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mnot_winter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mCRASH_TYPE\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue_counts\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnot_winter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mCRASH_TYPE\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue_counts\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'not_winter' is not defined"
     ]
    }
   ],
   "source": [
    "not_winter.CRASH_TYPE.value_counts()/sum(not_winter.CRASH_TYPE.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "winter.CRASH_TYPE.value_counts()/sum(winter.CRASH_TYPE.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crashes.iloc[np.where(crashes.WEATHER_CONDITION == \"CLEAR\")]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "crashes = crash.iloc[np.where(crash.INJURIES_TOTAL <= 3)]\n",
    "\n",
    "plt.hist(crashes.iloc[np.where(crashes.WEATHER_CONDITION == \"CLEAR\")].INJURIES_TOTAL, alpha=0.5, label=\"Above Median MPG\")\n",
    "plt.title(\"Auto Displacement Value and MPG\")\n",
    "plt.xlabel(\"Auto Displacement\")\n",
    "plt.ylabel(\"Number of Cars\")\n",
    "plt.legend()\n",
    "\n",
    "print()\n",
    "\n",
    "# visible separation between groups; displacement value is probably a good indicator\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crashes = crash.iloc[np.where(crash.INJURIES_TOTAL <= 3)]\n",
    "#plt.bar(crashes.iloc[np.where(crashes.WEATHER_CONDITION == \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).index[:] - 0.2, crashes.iloc[np.where(crashes.WEATHER_CONDITION == \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).values[:], alpha=0.3, label=\"Clear\")\n",
    "#plt.bar(crashes.iloc[np.where(crashes.WEATHER_CONDITION != \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).index[:] + 0.2, crashes.iloc[np.where(crashes.WEATHER_CONDITION != \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).values[:], alpha=0.3, label=\"Inclement Weather\")\n",
    "\n",
    "plt.bar(crashes.iloc[np.where(crashes.WEATHER_CONDITION == \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).index[:] - 0.2, crashes.iloc[np.where(crashes.WEATHER_CONDITION == \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).values[:], 0.4, alpha=1, label=\"Clear\", color=\"#1E88E5\")\n",
    "plt.bar(crashes.iloc[np.where(crashes.WEATHER_CONDITION != \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).index[:] + 0.2, crashes.iloc[np.where(crashes.WEATHER_CONDITION != \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True).values[:], 0.4, alpha=1, label=\"Inclement Weather\", color=\"#D81B60\")\n",
    "\n",
    "plt.title(\"Weather vs. Clear: Injury Count\")\n",
    "plt.xlabel(\"Number of Injuries\")\n",
    "plt.ylabel(\"Percent of Total Injuries\")\n",
    "plt.legend()\n",
    "\n",
    "plt.savefig(\"weather_injury.png\", dpi=300, bbox_inches=\"tight\")\n",
    "\n",
    "print()\n",
    "\n",
    "# visible separation between groups; displacement value is probably a good indicator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "winter = crash.iloc[np.where(np.logical_or(crash.CRASH_MONTH == 1, crash.CRASH_MONTH == 12))]\n",
    "not_winter = crash.iloc[np.where(np.logical_and(crash.CRASH_MONTH != 1, crash.CRASH_MONTH != 12))]\n",
    "rain = crash.iloc[np.where(crash.WEATHER_CONDITION == \"RAIN\")]\n",
    "clear = crash.iloc[np.where(crash.WEATHER_CONDITION == \"CLEAR\")]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rain.DAMAGE.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "clear.DAMAGE.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.hist(rain.DAMAGE.value_counts(), alpha=0.5)\n",
    "plt.bar(rain.DAMAGE.value_counts(normalize=True).index[:], rain.DAMAGE.value_counts(normalize=True).values[:], alpha=0.3)\n",
    "plt.bar(clear.DAMAGE.value_counts(normalize=True).index[:], clear.DAMAGE.value_counts(normalize=True).values[:], alpha=0.3)\n",
    "\n",
    "#plt.hist(clear.DAMAGE.value_counts(), alpha=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "winter.DAMAGE.value_counts()/sum(winter.DAMAGE.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "not_winter.DAMAGE.value_counts()/sum(not_winter.DAMAGE.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.INJURIES_TOTAL.value_counts()/sum(crash.INJURIES_TOTAL.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.CRASH_HOUR.value_counts()/sum(crash.CRASH_HOUR.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "winter.CRASH_HOUR.value_counts()/sum(winter.CRASH_HOUR.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#weather = crash.iloc[np.where(crash.PRIM_CONTRIBUTORY_CAUSE == \"WEATHER\")]\n",
    "weather = crash.iloc[np.where(crash.WEATHER_CONDITION != \"CLEAR\")]\n",
    "weather.WEATHER_CONDITION.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.INJURIES_TOTAL.value_counts()/sum(crash.INJURIES_TOTAL.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather.INJURIES_TOTAL.value_counts()/sum(weather.INJURIES_TOTAL.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cw = crash.iloc[np.where(crash.PRIM_CONTRIBUTORY_CAUSE == \"WEATHER\")]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.CRASH_MONTH.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cw.WEATHER_CONDITION.value_counts(normalize=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cw.ROADWAY_SURFACE_COND.value_counts(normalize=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sum(cw.ROADWAY_SURFACE_COND.value_counts(normalize=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.INJURIES_TOTAL.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "8984-(3970+3453+1268)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(6,6)) \n",
    "plt.rcParams.update({'font.size': 16})\n",
    "plt.title(\"Road Condition of Crashes Caused by Weather 2016-2021\")\n",
    "\n",
    "labels = 'Wet', 'Snow/Slush', 'Ice', 'Other'\n",
    "sizes = [3970/8984, 3453/8984, 1268/8984, 777/8984]\n",
    "#explode = (0, 0, 0, 0.05, 0.05)  # only \"explode\" the 2nd slice (i.e. 'Hogs')\n",
    "\n",
    "colors = ['#C3E7AD','#D7ADE7','#B5D0F0', '#F3B19A', '#F3E89A']\n",
    "\n",
    "plt.pie(sizes, labels=labels, colors=colors,\n",
    "autopct='%1.1f%%', shadow=False, startangle=60)\n",
    "\n",
    "plt.savefig(\"weather_causes\", dpi=300, bbox_inches=\"tight\")\n",
    "\n",
    "plt.show()\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#weather = crash.iloc[np.where(crash.PRIM_CONTRIBUTORY_CAUSE == \"WEATHER\")]\n",
    "weather = crash.iloc[np.where(np.logical_or(weather.WEATHER_CONDITION == \"CLEAR\", weather.WEATHER_CONDITION ==\"RAIN\"))]\n",
    "#weather = weather.iloc[np.where(weather.INJURIES_TOTAL >= 1)]\n",
    "weather = weather.reset_index()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather.WEATHER_CONDITION.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.INJURIES_INCAPACITATING.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather.INJURIES_INCAPACITATING.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def regression(x, y):\n",
    "    xcol = x\n",
    "    ycol = y\n",
    "    \n",
    "    meanx = np.mean(xcol)\n",
    "    meany = np.mean(ycol)\n",
    "    \n",
    "    stdx = np.std(xcol)\n",
    "    stdy = np.std(ycol)\n",
    "    \n",
    "    numer = 0\n",
    "    denom = 0\n",
    "    \n",
    "    for i in range(len(x)):\n",
    "        numer += (xcol[i] - meanx) * (ycol[i] - meany)\n",
    "        denom += (xcol[i] - meanx) ** 2\n",
    "\n",
    "    slope = numer / denom\n",
    "    intercept = meany - (slope * meanx)\n",
    "    \n",
    "    return \"Slope:\", slope, \"Intercept:\", intercept\n",
    "\n",
    "def standard_units(x):\n",
    "    \"Convert any array of numbers to standard units.\"\n",
    "    return (x - np.average(x))/np.std(x,ddof=1)  \n",
    "\n",
    "def correlation(x, y):\n",
    "    x_in_standard_units = standard_units(x)\n",
    "    y_in_standard_units = standard_units(y)\n",
    "    if len(x)!=len(y):\n",
    "        raise ValueError('arrays are of different lengths')\n",
    "    return sum(x_in_standard_units * y_in_standard_units)/(len(x)-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "condition = np.array([])\n",
    "for i in range(len(weather)):\n",
    "    if weather.WEATHER_CONDITION[i] == \"CLEAR\":\n",
    "        condition = np.append(condition, 0)\n",
    "    elif weather.WEATHER_CONDITION[i] == \"RAIN\":\n",
    "        condition = np.append(condition, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather.WEATHER_CONDITION\n",
    "np.count_nonzero(condition == 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "injury = np.array([])\n",
    "\n",
    "for i in range(len(weather)):\n",
    "    if weather.INJURIES_TOTAL[i] == 1:\n",
    "        injury = np.append(injury, 0)\n",
    "    else:\n",
    "        injury = np.append(injury, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(injury), len(condition)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "regression((condition), (weather.INJURIES_TOTAL))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(condition, weather.INJURIES_TOTAL)\n",
    "plt.title(\"Weather Condition vs. Total Injuries \\n Regression 2016-2021\")\n",
    "plt.xlabel(\"Clear vs. Rain\")\n",
    "plt.ylabel(\"# Injuries\")\n",
    "\n",
    "x = np.linspace(np.min(condition), np.max(condition),60)\n",
    "y = 0.03895057220501825*x + 0.20695106713925018\n",
    "plt.plot(x, y, '-r')\n",
    "\n",
    "plt.savefig(\"rain_regression\", dpi=300, bbox_inches=\"tight\")\n",
    "\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather_snow = crash.iloc[np.where(np.logical_or(crash.WEATHER_CONDITION == \"CLEAR\", crash.WEATHER_CONDITION ==\"SNOW\"))]\n",
    "snow_condition = np.array([])\n",
    "for i in range(len(weather_snow)):\n",
    "    if weather_snow.WEATHER_CONDITION[i] == \"CLEAR\":\n",
    "        snow_condition = np.append(snow_condition, 0)\n",
    "    elif weather_snow.WEATHER_CONDITION[i] == \"SNOW\":\n",
    "        snow_condition = np.append(snow_condition, 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(condition, weather.INJURIES_TOTAL)\n",
    "plt.title(\"Weather Condition vs. Total Injuries \\n Regression 2016-2021\")\n",
    "plt.xlabel(\"Clear vs. Rain\")\n",
    "plt.ylabel(\"# Injuries\")\n",
    "\n",
    "x = np.linspace(np.min(condition), np.max(condition),60)\n",
    "y = 0.03895057220501825*x + 0.20695106713925018\n",
    "plt.plot(x, y, '-r')\n",
    "\n",
    "plt.savefig(\"rain_regression\", dpi=300, bbox_inches=\"tight\")\n",
    "\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather.INJURIES_TOTAL.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.CRASH_TYPE.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "types = np.array([])\n",
    "\n",
    "CRASH_TYPE\n",
    "for i in range(len(weather)):\n",
    "    if weather.CRASH_TYPE[i] == \"NO INJURY / DRIVE AWAY\":\n",
    "        cost = np.append(cost, 0)\n",
    "    elif weather.CRASH_TYPE[i] == \"$501 - $1,500\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cost = np.array([])\n",
    "\n",
    "for i in range(len(weather)):\n",
    "    if weather.DAMAGE[i] == \"$500 OR LESS\":\n",
    "        cost = np.append(cost, 0)\n",
    "    elif weather.DAMAGE[i] == \"$501 - $1,500\":\n",
    "        cost = np.append(cost, 1)\n",
    "    elif weather.DAMAGE[i] == \"OVER $1,500\":\n",
    "        cost = np.append(cost, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(cost) == len(condition)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "regression(condition, injury)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "correlation(condition, injury)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict = pd.DataFrame({})\n",
    "# use posted speed limit, weather condition, lighting condition, road defect, crash hour, roadway surface conditions\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.ROAD_DEFECT.value_counts()\n",
    "\n",
    "defects = np.array([])\n",
    "\n",
    "for i in range(len(crash)):\n",
    "    if crash.ROAD_DEFECT[i] == \"NO DEFECTS\":\n",
    "        defects = np.append(defects, 0)\n",
    "    else:\n",
    "        defects = np.append(defects, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.WEATHER_CONDITION.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather = np.array([])\n",
    "\n",
    "for i in range(len(crash)):\n",
    "    if crash.WEATHER_CONDITION[i] == \"RAIN\":\n",
    "        weather = np.append(weather, 1)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"SNOW\":\n",
    "        weather = np.append(weather, 2)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"CLOUDY/OVERCAST\":\n",
    "        weather = np.append(weather, 3)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"FREEZING RAIN/DRIZZLE\":\n",
    "        weather = np.append(weather, 4)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"FOG/SMOKE/HAZE\":\n",
    "        weather = np.append(weather, 5)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"SLEET/HAIL\":\n",
    "        weather = np.append(weather, 6)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"BLOWING SNOW\":\n",
    "        weather = np.append(weather, 7)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"SEVERE CROSS WIND GATE\":\n",
    "        weather = np.append(weather, 8)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"BLOWING SAND, SOIL, DIRT\":\n",
    "        weather = np.append(weather, 9)   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.LIGHTING_CONDITION.value_counts()\n",
    "\n",
    "lighting = np.array([])\n",
    "\n",
    "for i in range(len(crash)):\n",
    "    if crash.LIGHTING_CONDITION[i] == \"DAYLIGHT\":\n",
    "        lighting = np.append(lighting, 0)\n",
    "    else:\n",
    "        lighting = np.append(lighting, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.ROADWAY_SURFACE_COND.value_counts()\n",
    "\n",
    "conditions = np.array([])\n",
    "\n",
    "for i in range(len(crash)):\n",
    "    if crash.ROADWAY_SURFACE_COND[i] == \"DRY\":\n",
    "        conditions = np.append(conditions, 0)\n",
    "    else:\n",
    "        conditions = np.append(conditions, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.INJURIES_TOTAL.value_counts()\n",
    "\n",
    "injuries = np.array([])\n",
    "\n",
    "for i in range(len(crash)):\n",
    "    if crash.INJURIES_TOTAL[i] >= 1:\n",
    "        injuries = np.append(injuries, 0)\n",
    "    else:\n",
    "        injuries = np.append(injuries, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def su(column):\n",
    "    \"Convert a dataframe column to standard units.\"\n",
    "    x = predict[column].tolist()\n",
    "    return (x - np.average(x))/np.std(x,ddof=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict = predict.iloc[:500]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict[\"defects\"] = defects\n",
    "predict[\"conditions\"] = conditions\n",
    "predict[\"lighting\"] = lighting\n",
    "predict[\"weather\"] = weather\n",
    "predict[\"speed_limit\"] = crash.POSTED_SPEED_LIMIT\n",
    "predict[\"hour\"] = crash.iloc[:500].CRASH_HOUR\n",
    "predict[\"injuries\"] = injuries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict.defects = su(\"defects\")\n",
    "predict.conditions = su(\"conditions\")\n",
    "predict.lighting = su(\"lighting\")\n",
    "predict.weather = su(\"weather\")\n",
    "predict.speed_limit = su(\"speed_limit\")\n",
    "predict.hour = su(\"hour\")\n",
    "predict.injuries = su(\"injuries\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def standard_units(x):\n",
    "    \"Convert any array of numbers to standard units.\"\n",
    "    return (x - np.average(x))/np.std(x,ddof=1)  \n",
    "\n",
    "def correlation(x, y):\n",
    "    x_in_standard_units = standard_units(x)\n",
    "    y_in_standard_units = standard_units(y)\n",
    "    if len(x)!=len(y):\n",
    "        raise ValueError('arrays are of different lengths')\n",
    "    return sum(x_in_standard_units * y_in_standard_units)/(len(x)-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "correlation(predict.defects, predict.injuries), \\\n",
    "correlation(predict.conditions, predict.injuries), \\\n",
    "correlation(predict.lighting, predict.injuries), \\\n",
    "correlation(predict.weather, predict.injuries), \\\n",
    "correlation(predict.speed_limit, predict.injuries), \\\n",
    "correlation(predict.hour, predict.injuries)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(predict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sim = predict.sample(frac=1)\n",
    "\n",
    "x_train=sim.iloc[1000:4000].drop(columns=['injuries'])\n",
    "x_test=sim.iloc[0:1000].drop(columns=['injuries'])\n",
    "y_train=sim.iloc[1000:4000].injuries\n",
    "y_test=sim.iloc[0:1000].injuries\n",
    "\n",
    "x_train.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def distance(pt1, pt2):\n",
    "    return np.sqrt(sum((pt1 - pt2)**2))\n",
    "\n",
    "def all_distances(X_train, point):\n",
    "    attributes = X_train.copy()\n",
    "    def distance_from_point(row):\n",
    "        return distance(point, np.array(row))\n",
    "    return attributes.apply(distance_from_point,axis=1)\n",
    "\n",
    "def closest(X_train, Y_train, point, k):\n",
    "    attributes=X_train.copy()\n",
    "    attributes['Distance']=all_distances(attributes, point)\n",
    "    attributes['Class']=Y_train\n",
    "    attributes=attributes.sort_values('Distance')    \n",
    "    topk = attributes.head(k)\n",
    "    return topk\n",
    "\n",
    "def majority(topkclasses):\n",
    "    ones = topkclasses[topkclasses.Class==1].shape[0]\n",
    "    zeros = topkclasses[topkclasses.Class==0].shape[0]\n",
    "    if ones > zeros:\n",
    "        return 1\n",
    "    else:\n",
    "        return 0\n",
    "\n",
    "def classify(X_train,Y_train, X_test, k):\n",
    "    n=X_test.shape[0]\n",
    "    \n",
    "    pred_class = np.array([])\n",
    "    for i in np.arange(n):\n",
    "        topkclasses = closest(X_train,Y_train, X_test.iloc[i], k)\n",
    "        pred_class = np.append(pred_class,majority(topkclasses))\n",
    "    \n",
    "    report_df=X_test.copy()\n",
    "    report_df['Class']=pred_class\n",
    "    \n",
    "    return report_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "class_df=classify(x_train,y_train, x_test, 5)\n",
    "class_df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "1 - 527616/len(crash)\n",
    "# 13.5% of the crashes involved at least one injury"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.dt_col = pd.to_datetime(crash.CRASH_DATE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.groupby(pd.to_datetime(crash.CRASH_DATE).year).CRASH_DAY_OF_WEEK.size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.barh(crash.CRASH_DAY_OF_WEEK.value_counts().index[:], crash.CRASH_DAY_OF_WEEK.value_counts().values[:], color=\"#6bb0ff\")\n",
    "\n",
    "plt.yticks(ticks=[1,2,3,4,5,6,7], labels=[\"Sunday\",\"Monday\", \"Tuesday\", \"Wednesday\", \"Thursday\", \"Friday\", \"Saturday\"])\n",
    "plt.ylabel(\"Day of Week\", labelpad=10)\n",
    "plt.xlabel(\"Number of Crashes\", labelpad=10)\n",
    "plt.title(\"Total Number of Crashes 2016-2021 vs Day of Week\", pad=15)\n",
    "\n",
    "plt.savefig('day_of_week_dist.png', dpi=300, bbox_inches=\"tight\")\n",
    "\n",
    "# I want the average number of crashes per day of week \n",
    "# groupby day, then average the number of occurences by the number of unique Sundays, etc.\n",
    "\n",
    "print()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.iloc[np.where(crash.WEATHER_CONDITION == \"RAIN\")].INJURIES_TOTAL.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sum(crash.CRASH_DAY_OF_WEEK.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.iloc[np.where(crash.WEATHER_CONDITION == \"CLEAR\")].INJURIES_TOTAL.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rain = np.array([])\n",
    "\n",
    "for i in range(len(crash)):\n",
    "    if crash.WEATHER_CONDITION[i] == \"CLEAR\":\n",
    "        rain = np.append(rain, 0)\n",
    "    elif crash.WEATHER_CONDITION[i] == \"RAIN\":\n",
    "        rain = np.append(rain, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather = crash.iloc[np.where(np.logical_or(crash.WEATHER_CONDITION == \"CLEAR\", crash.WEATHER_CONDITION ==\"RAIN\"))]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather = crash.iloc[np.where(np.logical_or(crash.WEATHER_CONDITION == \"CLEAR\", crash.WEATHER_CONDITION ==\"RAIN\"))]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "injuries = np.array([])\n",
    "\n",
    "for i in range(len(weather)):\n",
    "    if weather.INJURIES_TOTAL[i] >= 1:\n",
    "        injuries = np.append(injuries, 1)\n",
    "    else:\n",
    "        injuries = np.append(injuries, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(injure)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "correlation(rain, injure)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Most frequent: hour, month, day of the week, street, type of crash, weather + road condition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# average crashes per year\n",
    "\n",
    "year = pd.DatetimeIndex(crash['CRASH_DATE']).year\n",
    "date = pd.to_datetime(crash['CRASH_DATE']).dt.date\n",
    "crash[\"date\"] = date\n",
    "day = crash.CRASH_DAY_OF_WEEK\n",
    "month = crash.CRASH_MONTH\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "year.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.barh((crash.CRASH_DAY_OF_WEEK.value_counts()/per_day.CRASH_DAY_OF_WEEK.value_counts()).index[:], crash.CRASH_DAY_OF_WEEK.value_counts()/per_day.CRASH_DAY_OF_WEEK.value_counts().values[:], color=\"#ff907d\")\n",
    "\n",
    "plt.yticks(ticks=[1,2,3,4,5,6,7], labels=[\"Sunday\",\"Monday\", \"Tuesday\", \"Wednesday\", \"Thursday\", \"Friday\", \"Saturday\"])\n",
    "plt.ylabel(\"Day of Week\", labelpad=10)\n",
    "plt.xlabel(\"Average Number of Crashes per Day\", labelpad=10)\n",
    "plt.title(\"Average Number of Crashes 2016-2021 per Week Day\", pad=15)\n",
    "\n",
    "plt.savefig('ave_day_of_week_dist.png', dpi=300, bbox_inches=\"tight\")\n",
    "\n",
    "# I want the average number of crashes per day of week \n",
    "# groupby day, then average the number of occurences by the number of unique Sundays, etc.\n",
    "\n",
    "print()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.barh(crash.CRASH_DAY_OF_WEEK.value_counts().index[:], crash.CRASH_DAY_OF_WEEK.value_counts().values[:], color=\"#6bb0ff\")\n",
    "\n",
    "plt.yticks(ticks=[1,2,3,4,5,6,7], labels=[\"Sunday\",\"Monday\", \"Tuesday\", \"Wednesday\", \"Thursday\", \"Friday\", \"Saturday\"])\n",
    "plt.ylabel(\"Day of Week\", labelpad=10)\n",
    "plt.xlabel(\"Total Number of Crashes per Week Day Label\", labelpad=10)\n",
    "plt.title(\"Total Number of Crashes 2016-2021 vs Day of Week\", pad=15)\n",
    "\n",
    "plt.savefig('day_of_week_dist.png', dpi=300, bbox_inches=\"tight\")\n",
    "\n",
    "# I want the average number of crashes per day of week \n",
    "# groupby day, then average the number of occurences by the number of unique Sundays, etc.\n",
    "\n",
    "print()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "per_day = crash.drop_duplicates(subset=\"date\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "sum(year.value_counts())/len(np.unique(year))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "day.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "year.value_counts().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sum(year.value_counts())/len(np.unique(date))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "date = date.sort_values()\n",
    "date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dated = pd.DataFrame({\n",
    "    \"date\": date.sort_values(ascending=True),\n",
    "    \"count\": \n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(date, lake_levels)   \n",
    "plt.xlabel('Year')   \n",
    "plt.ylabel('Level (in meters)')\n",
    "plt.title(\"Timeseries of Monthly Lake Levels of Lake Michigan\")\n",
    "\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "date.value_counts(sort=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "date."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.plot(date.value_counts().index[:], date.value_counts().values[:], color=\"#6bb0ff\")\n",
    "plt.plot(per_day.date, per.value_counts())\n",
    "\n",
    "plt.ylabel(\"Number on Date\")\n",
    "plt.xlabel(\"Date\")\n",
    "plt.title(\"Number of Crashes vs Day of Week\")\n",
    "\n",
    "# I want the average number of crashes per day of week \n",
    "# groupby day, then average the number of occurences by the number of unique Sundays, etc.\n",
    "\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "date.value_counts().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sum(date.value_counts())/len(np.unique(date))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "month.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash.LOCATION.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}