all_imports.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "8d26efd4-ad62-4d61-b9f7-94d04ca1a9f9",
   "metadata": {},
   "source": [
    "# Project 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e5b16494-353a-405b-b16c-9edb17384485",
   "metadata": {},
   "outputs": [],
   "source": [
    "#First I will bring in my imports\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c8a5b76e-5b14-416f-b006-d35d641443e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# function for merging data\n",
    "\n",
    "def get_merged_dataframes(main_df, main_df_columns, csv_files, year, merge_columns):\n",
    "    \n",
    "    df1 = pd.read_csv(f'./data/{year}_fars/{main_df}.csv', encoding='unicode_escape')\n",
    "    df1.columns = df1.columns.str.lower()\n",
    "    merged_dataframe = df1[df1.columns.intersection(main_df_columns)]\n",
    "    \n",
    "    for csv in csv_files:\n",
    "        try:\n",
    "            df2 = pd.read_csv(f'./data/{year}_fars/{csv}.csv')\n",
    "            df2.columns = df2.columns.str.lower() \n",
    "            merged_dataframe = pd.merge(merged_dataframe, df2, how='outer', on=merge_columns, suffixes=('', '_remove'))\n",
    "            merged_dataframe = merged_dataframe.drop([column for column in merged_dataframe.columns if 'remove' in column], axis=1)\n",
    "        except FileNotFoundError:\n",
    "            print(f'The {csv} file does not exist for the year {year}.')           \n",
    "    \n",
    "    merged_dataframe.loc[:,'data_year'] = year\n",
    "    \n",
    "    return merged_dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "88c5e6bd-3c8b-4d19-95b5-dad6256f4fa1",
   "metadata": {},
   "outputs": [],
   "source": [
    "def combine_person_data(start_year, end_year=None):  \n",
    "    \n",
    "    columns = ['state', 'peds', 've_total', 'persons', 'county', 'day', 'month', 'hour', 'minute', 'week', 'tway_id', 'route', 'rur_urb', \n",
    "               'rd_owner','nhs', 'sp_jur', 'milept', 'latitude', 'longitude', 'harm', \n",
    "               'man_coll', 'reljct1', 'reljct2', 'typ_int', 'rel_road', 'wrk_zone', 'lgt_cond', \n",
    "               'sch_bus', 'rail', 'not_hour', 'arr_hour', 'hosp_hr',\n",
    "              'eventnum', 'vnumber1', 'aoi1', 'soe', 'vnumber2', 'aoi2',\n",
    "              'weather', 'crashrf', 'st_case', 'drunk_dr', 'inj_sev', 'inj_sevname']\n",
    "    \n",
    "    csvs_to_merge = ['acc_aux', 'weather', 'crashrf', 'cevent', 'person']\n",
    "    \n",
    "    merge_columns = ['st_case']\n",
    "    \n",
    "    if end_year is None:\n",
    "        end_year = start_year\n",
    "    \n",
    "    full_df_list = [get_merged_dataframes('accident', columns, csvs_to_merge, year, merge_columns) for year in range(start_year, end_year + 1)]\n",
    "    \n",
    "    return pd.concat(full_df_list, ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "57b8ded1-f32c-405a-b539-285425bf8194",
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df = combine_person_data(2020)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1a4d6fae-384c-4ef5-8575-b25553f07683",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ab5048b1-9a36-44e0-8283-33c3e5641b7e",
   "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>state</th>\n",
       "      <th>st_case</th>\n",
       "      <th>ve_total</th>\n",
       "      <th>peds</th>\n",
       "      <th>persons</th>\n",
       "      <th>county</th>\n",
       "      <th>day</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>minute</th>\n",
       "      <th>nhs</th>\n",
       "      <th>route</th>\n",
       "      <th>tway_id</th>\n",
       "      <th>rur_urb</th>\n",
       "      <th>rd_owner</th>\n",
       "      <th>milept</th>\n",
       "      <th>latitude</th>\n",
       "      <th>sp_jur</th>\n",
       "      <th>man_coll</th>\n",
       "      <th>reljct1</th>\n",
       "      <th>reljct2</th>\n",
       "      <th>typ_int</th>\n",
       "      <th>wrk_zone</th>\n",
       "      <th>rel_road</th>\n",
       "      <th>lgt_cond</th>\n",
       "      <th>weather</th>\n",
       "      <th>sch_bus</th>\n",
       "      <th>rail</th>\n",
       "      <th>not_hour</th>\n",
       "      <th>arr_hour</th>\n",
       "      <th>hosp_hr</th>\n",
       "      <th>drunk_dr</th>\n",
       "      <th>year</th>\n",
       "      <th>fatals</th>\n",
       "      <th>a_crainj</th>\n",
       "      <th>a_region</th>\n",
       "      <th>a_ru</th>\n",
       "      <th>a_inter</th>\n",
       "      <th>a_relrd</th>\n",
       "      <th>a_intsec</th>\n",
       "      <th>a_roadfc</th>\n",
       "      <th>a_junc</th>\n",
       "      <th>a_mancol</th>\n",
       "      <th>a_tod</th>\n",
       "      <th>a_dow</th>\n",
       "      <th>a_ct</th>\n",
       "      <th>a_weather</th>\n",
       "      <th>a_lt</th>\n",
       "      <th>a_mc</th>\n",
       "      <th>a_spcra</th>\n",
       "      <th>a_ped</th>\n",
       "      <th>a_ped_f</th>\n",
       "      <th>a_pedal</th>\n",
       "      <th>a_pedal_f</th>\n",
       "      <th>a_roll</th>\n",
       "      <th>a_polpur</th>\n",
       "      <th>a_posbac</th>\n",
       "      <th>a_d15_19</th>\n",
       "      <th>a_d16_19</th>\n",
       "      <th>a_d15_20</th>\n",
       "      <th>a_d16_20</th>\n",
       "      <th>a_d65pls</th>\n",
       "      <th>a_d21_24</th>\n",
       "      <th>a_d16_24</th>\n",
       "      <th>a_rd</th>\n",
       "      <th>a_hr</th>\n",
       "      <th>a_dist</th>\n",
       "      <th>a_drowsy</th>\n",
       "      <th>bia</th>\n",
       "      <th>spj_indian</th>\n",
       "      <th>indian_res</th>\n",
       "      <th>statename</th>\n",
       "      <th>weathername</th>\n",
       "      <th>crashrf</th>\n",
       "      <th>crashrfname</th>\n",
       "      <th>eventnum</th>\n",
       "      <th>vnumber1</th>\n",
       "      <th>aoi1</th>\n",
       "      <th>aoi1name</th>\n",
       "      <th>soe</th>\n",
       "      <th>soename</th>\n",
       "      <th>vnumber2</th>\n",
       "      <th>vnumber2name</th>\n",
       "      <th>aoi2</th>\n",
       "      <th>aoi2name</th>\n",
       "      <th>ve_forms</th>\n",
       "      <th>veh_no</th>\n",
       "      <th>per_no</th>\n",
       "      <th>str_veh</th>\n",
       "      <th>dayname</th>\n",
       "      <th>monthname</th>\n",
       "      <th>hourname</th>\n",
       "      <th>minutename</th>\n",
       "      <th>rur_urbname</th>\n",
       "      <th>func_sys</th>\n",
       "      <th>func_sysname</th>\n",
       "      <th>harm_ev</th>\n",
       "      <th>harm_evname</th>\n",
       "      <th>man_collname</th>\n",
       "      <th>sch_busname</th>\n",
       "      <th>make</th>\n",
       "      <th>makename</th>\n",
       "      <th>mak_mod</th>\n",
       "      <th>body_typ</th>\n",
       "      <th>body_typname</th>\n",
       "      <th>mod_year</th>\n",
       "      <th>mod_yearname</th>\n",
       "      <th>tow_veh</th>\n",
       "      <th>tow_vehname</th>\n",
       "      <th>spec_use</th>\n",
       "      <th>spec_usename</th>\n",
       "      <th>emer_use</th>\n",
       "      <th>emer_usename</th>\n",
       "      <th>rollover</th>\n",
       "      <th>rollovername</th>\n",
       "      <th>impact1</th>\n",
       "      <th>impact1name</th>\n",
       "      <th>fire_exp</th>\n",
       "      <th>fire_expname</th>\n",
       "      <th>age</th>\n",
       "      <th>agename</th>\n",
       "      <th>sex</th>\n",
       "      <th>sexname</th>\n",
       "      <th>per_typ</th>\n",
       "      <th>per_typname</th>\n",
       "      <th>inj_sev</th>\n",
       "      <th>inj_sevname</th>\n",
       "      <th>seat_pos</th>\n",
       "      <th>seat_posname</th>\n",
       "      <th>rest_use</th>\n",
       "      <th>rest_usename</th>\n",
       "      <th>rest_mis</th>\n",
       "      <th>rest_misname</th>\n",
       "      <th>air_bag</th>\n",
       "      <th>air_bagname</th>\n",
       "      <th>ejection</th>\n",
       "      <th>ejectionname</th>\n",
       "      <th>ej_path</th>\n",
       "      <th>ej_pathname</th>\n",
       "      <th>extricat</th>\n",
       "      <th>extricatname</th>\n",
       "      <th>drinking</th>\n",
       "      <th>drinkingname</th>\n",
       "      <th>alc_det</th>\n",
       "      <th>alc_detname</th>\n",
       "      <th>alc_status</th>\n",
       "      <th>alc_statusname</th>\n",
       "      <th>atst_typ</th>\n",
       "      <th>atst_typname</th>\n",
       "      <th>alc_res</th>\n",
       "      <th>alc_resname</th>\n",
       "      <th>drugs</th>\n",
       "      <th>drugsname</th>\n",
       "      <th>drug_det</th>\n",
       "      <th>drug_detname</th>\n",
       "      <th>dstatus</th>\n",
       "      <th>dstatusname</th>\n",
       "      <th>hospital</th>\n",
       "      <th>hospitalname</th>\n",
       "      <th>doa</th>\n",
       "      <th>doaname</th>\n",
       "      <th>death_da</th>\n",
       "      <th>death_daname</th>\n",
       "      <th>death_mo</th>\n",
       "      <th>death_moname</th>\n",
       "      <th>death_yr</th>\n",
       "      <th>death_yrname</th>\n",
       "      <th>death_hr</th>\n",
       "      <th>death_hrname</th>\n",
       "      <th>death_mn</th>\n",
       "      <th>death_mnname</th>\n",
       "      <th>death_tm</th>\n",
       "      <th>death_tmname</th>\n",
       "      <th>lag_hrs</th>\n",
       "      <th>lag_hrsname</th>\n",
       "      <th>lag_mins</th>\n",
       "      <th>lag_minsname</th>\n",
       "      <th>work_inj</th>\n",
       "      <th>work_injname</th>\n",
       "      <th>hispanic</th>\n",
       "      <th>hispanicname</th>\n",
       "      <th>location</th>\n",
       "      <th>locationname</th>\n",
       "      <th>helm_use</th>\n",
       "      <th>helm_usename</th>\n",
       "      <th>helm_mis</th>\n",
       "      <th>helm_misname</th>\n",
       "      <th>vpicmake</th>\n",
       "      <th>vpicmakename</th>\n",
       "      <th>vpicmodel</th>\n",
       "      <th>vpicmodelname</th>\n",
       "      <th>vpicbodyclass</th>\n",
       "      <th>vpicbodyclassname</th>\n",
       "      <th>icfinalbody</th>\n",
       "      <th>icfinalbodyname</th>\n",
       "      <th>data_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>10001</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>51</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>cr-4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>32.433133</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0000000</td>\n",
       "      <td>99</td>\n",
       "      <td>3</td>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>2020</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>Clear</td>\n",
       "      <td>0</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>63</td>\n",
       "      <td>Ran Off Roadway - Right</td>\n",
       "      <td>5555</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2:00am-2:59am</td>\n",
       "      <td>58</td>\n",
       "      <td>Rural</td>\n",
       "      <td>5</td>\n",
       "      <td>Major Collector</td>\n",
       "      <td>42</td>\n",
       "      <td>Tree (Standing Only)</td>\n",
       "      <td>The First Harmful Event was Not a Collision wi...</td>\n",
       "      <td>No</td>\n",
       "      <td>59.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>Lexus ES-250/300/300h/330/ 350</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4-door sedan, hardtop</td>\n",
       "      <td>1997.0</td>\n",
       "      <td>1997</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Trailing Units</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Special Use</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rollover, Tripped by Object/Vehicle</td>\n",
       "      <td>12.0</td>\n",
       "      <td>12 Clock Point</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No or Not Reported</td>\n",
       "      <td>24</td>\n",
       "      <td>24 Years</td>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>1</td>\n",
       "      <td>Driver of a Motor Vehicle In-Transport</td>\n",
       "      <td>4</td>\n",
       "      <td>Fatal Injury (K)</td>\n",
       "      <td>11</td>\n",
       "      <td>Front Seat, Left Side</td>\n",
       "      <td>20</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>1</td>\n",
       "      <td>Deployed- Front</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Ejected</td>\n",
       "      <td>0</td>\n",
       "      <td>Ejection Path Not Applicable</td>\n",
       "      <td>1</td>\n",
       "      <td>Extricated</td>\n",
       "      <td>9</td>\n",
       "      <td>Reported as Unknown</td>\n",
       "      <td>9</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>1</td>\n",
       "      <td>Blood</td>\n",
       "      <td>294</td>\n",
       "      <td>0.294 % BAC</td>\n",
       "      <td>9</td>\n",
       "      <td>Reported as Unknown</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Transported for Treatment</td>\n",
       "      <td>7</td>\n",
       "      <td>Died at Scene</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2020</td>\n",
       "      <td>2020</td>\n",
       "      <td>2</td>\n",
       "      <td>2:00-2:59</td>\n",
       "      <td>58</td>\n",
       "      <td>58</td>\n",
       "      <td>258</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>7</td>\n",
       "      <td>Non-Hispanic</td>\n",
       "      <td>0</td>\n",
       "      <td>Occupant of a Motor Vehicle</td>\n",
       "      <td>20</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>515.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>2475.0</td>\n",
       "      <td>ES</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Sedan/Saloon</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>10001</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>51</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>cr-4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>32.433133</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0000000</td>\n",
       "      <td>99</td>\n",
       "      <td>3</td>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>2020</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>Clear</td>\n",
       "      <td>0</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>63</td>\n",
       "      <td>Ran Off Roadway - Right</td>\n",
       "      <td>5555</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2:00am-2:59am</td>\n",
       "      <td>58</td>\n",
       "      <td>Rural</td>\n",
       "      <td>5</td>\n",
       "      <td>Major Collector</td>\n",
       "      <td>42</td>\n",
       "      <td>Tree (Standing Only)</td>\n",
       "      <td>The First Harmful Event was Not a Collision wi...</td>\n",
       "      <td>No</td>\n",
       "      <td>59.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>Lexus ES-250/300/300h/330/ 350</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4-door sedan, hardtop</td>\n",
       "      <td>1997.0</td>\n",
       "      <td>1997</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Trailing Units</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Special Use</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rollover, Tripped by Object/Vehicle</td>\n",
       "      <td>12.0</td>\n",
       "      <td>12 Clock Point</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No or Not Reported</td>\n",
       "      <td>21</td>\n",
       "      <td>21 Years</td>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>2</td>\n",
       "      <td>Passenger of a Motor Vehicle In-Transport</td>\n",
       "      <td>3</td>\n",
       "      <td>Suspected Serious Injury (A)</td>\n",
       "      <td>13</td>\n",
       "      <td>Front Seat, Right Side</td>\n",
       "      <td>3</td>\n",
       "      <td>Shoulder and Lap Belt Used</td>\n",
       "      <td>0</td>\n",
       "      <td>No Indication of Mis-Use</td>\n",
       "      <td>1</td>\n",
       "      <td>Deployed- Front</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Ejected</td>\n",
       "      <td>0</td>\n",
       "      <td>Ejection Path Not Applicable</td>\n",
       "      <td>1</td>\n",
       "      <td>Extricated</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>9</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>0</td>\n",
       "      <td>Test Not Given</td>\n",
       "      <td>0</td>\n",
       "      <td>Test Not Given</td>\n",
       "      <td>996</td>\n",
       "      <td>Test Not Given</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>0</td>\n",
       "      <td>Test Not Given</td>\n",
       "      <td>5</td>\n",
       "      <td>EMS Ground</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>88</td>\n",
       "      <td>Not Applicable (Non-Fatal)</td>\n",
       "      <td>88</td>\n",
       "      <td>Not Applicable (Non-Fatal)</td>\n",
       "      <td>8888</td>\n",
       "      <td>Not Applicable (Non-fatal)</td>\n",
       "      <td>88</td>\n",
       "      <td>Not Applicable (Non-fatal)</td>\n",
       "      <td>88</td>\n",
       "      <td>Not Applicable (Non-fatal)</td>\n",
       "      <td>8888</td>\n",
       "      <td>Not Applicable (Non-fatal)</td>\n",
       "      <td>999</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>99</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Applicable (not a fatality)</td>\n",
       "      <td>0</td>\n",
       "      <td>Not A Fatality (not Applicable)</td>\n",
       "      <td>0</td>\n",
       "      <td>Occupant of a Motor Vehicle</td>\n",
       "      <td>20</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>515.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>2475.0</td>\n",
       "      <td>ES</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Sedan/Saloon</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>10001</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>51</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>cr-4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>32.433133</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0000000</td>\n",
       "      <td>99</td>\n",
       "      <td>3</td>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>2020</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>Clear</td>\n",
       "      <td>0</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>63</td>\n",
       "      <td>Ran Off Roadway - Right</td>\n",
       "      <td>5555</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2:00am-2:59am</td>\n",
       "      <td>58</td>\n",
       "      <td>Rural</td>\n",
       "      <td>5</td>\n",
       "      <td>Major Collector</td>\n",
       "      <td>42</td>\n",
       "      <td>Tree (Standing Only)</td>\n",
       "      <td>The First Harmful Event was Not a Collision wi...</td>\n",
       "      <td>No</td>\n",
       "      <td>59.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>Lexus ES-250/300/300h/330/ 350</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4-door sedan, hardtop</td>\n",
       "      <td>1997.0</td>\n",
       "      <td>1997</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Trailing Units</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Special Use</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rollover, Tripped by Object/Vehicle</td>\n",
       "      <td>12.0</td>\n",
       "      <td>12 Clock Point</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No or Not Reported</td>\n",
       "      <td>22</td>\n",
       "      <td>22 Years</td>\n",
       "      <td>2</td>\n",
       "      <td>Female</td>\n",
       "      <td>2</td>\n",
       "      <td>Passenger of a Motor Vehicle In-Transport</td>\n",
       "      <td>4</td>\n",
       "      <td>Fatal Injury (K)</td>\n",
       "      <td>21</td>\n",
       "      <td>Second Seat, Left Side</td>\n",
       "      <td>20</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>20</td>\n",
       "      <td>Not Deployed</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Ejected</td>\n",
       "      <td>0</td>\n",
       "      <td>Ejection Path Not Applicable</td>\n",
       "      <td>1</td>\n",
       "      <td>Extricated</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>9</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>1</td>\n",
       "      <td>Blood</td>\n",
       "      <td>69</td>\n",
       "      <td>0.069 % BAC</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Transported for Treatment</td>\n",
       "      <td>7</td>\n",
       "      <td>Died at Scene</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2020</td>\n",
       "      <td>2020</td>\n",
       "      <td>2</td>\n",
       "      <td>2:00-2:59</td>\n",
       "      <td>58</td>\n",
       "      <td>58</td>\n",
       "      <td>258</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>7</td>\n",
       "      <td>Non-Hispanic</td>\n",
       "      <td>0</td>\n",
       "      <td>Occupant of a Motor Vehicle</td>\n",
       "      <td>20</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>515.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>2475.0</td>\n",
       "      <td>ES</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Sedan/Saloon</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>10001</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>51</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>cr-4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>32.433133</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0000000</td>\n",
       "      <td>99</td>\n",
       "      <td>3</td>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>2020</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>Clear</td>\n",
       "      <td>0</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>63</td>\n",
       "      <td>Ran Off Roadway - Right</td>\n",
       "      <td>5555</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>55</td>\n",
       "      <td>Non-Harmful Event</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2:00am-2:59am</td>\n",
       "      <td>58</td>\n",
       "      <td>Rural</td>\n",
       "      <td>5</td>\n",
       "      <td>Major Collector</td>\n",
       "      <td>42</td>\n",
       "      <td>Tree (Standing Only)</td>\n",
       "      <td>The First Harmful Event was Not a Collision wi...</td>\n",
       "      <td>No</td>\n",
       "      <td>59.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>Lexus ES-250/300/300h/330/ 350</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4-door sedan, hardtop</td>\n",
       "      <td>1997.0</td>\n",
       "      <td>1997</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Trailing Units</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Special Use</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rollover, Tripped by Object/Vehicle</td>\n",
       "      <td>12.0</td>\n",
       "      <td>12 Clock Point</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No or Not Reported</td>\n",
       "      <td>22</td>\n",
       "      <td>22 Years</td>\n",
       "      <td>2</td>\n",
       "      <td>Female</td>\n",
       "      <td>2</td>\n",
       "      <td>Passenger of a Motor Vehicle In-Transport</td>\n",
       "      <td>4</td>\n",
       "      <td>Fatal Injury (K)</td>\n",
       "      <td>23</td>\n",
       "      <td>Second Seat, Right Side</td>\n",
       "      <td>20</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>20</td>\n",
       "      <td>Not Deployed</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Ejected</td>\n",
       "      <td>0</td>\n",
       "      <td>Ejection Path Not Applicable</td>\n",
       "      <td>1</td>\n",
       "      <td>Extricated</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>9</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>1</td>\n",
       "      <td>Blood</td>\n",
       "      <td>120</td>\n",
       "      <td>0.120 % BAC</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Transported for Treatment</td>\n",
       "      <td>7</td>\n",
       "      <td>Died at Scene</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2020</td>\n",
       "      <td>2020</td>\n",
       "      <td>2</td>\n",
       "      <td>2:00-2:59</td>\n",
       "      <td>58</td>\n",
       "      <td>58</td>\n",
       "      <td>258</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>7</td>\n",
       "      <td>Non-Hispanic</td>\n",
       "      <td>0</td>\n",
       "      <td>Occupant of a Motor Vehicle</td>\n",
       "      <td>20</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>515.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>2475.0</td>\n",
       "      <td>ES</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Sedan/Saloon</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>10001</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>51</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>cr-4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>32.433133</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0000000</td>\n",
       "      <td>99</td>\n",
       "      <td>3</td>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>2020</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>Clear</td>\n",
       "      <td>0</td>\n",
       "      <td>None</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>12 Clock Point</td>\n",
       "      <td>42</td>\n",
       "      <td>Tree (Standing Only)</td>\n",
       "      <td>9999</td>\n",
       "      <td>Not a Motor Vehicle</td>\n",
       "      <td>77</td>\n",
       "      <td>Not a Motor Vehicle</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2:00am-2:59am</td>\n",
       "      <td>58</td>\n",
       "      <td>Rural</td>\n",
       "      <td>5</td>\n",
       "      <td>Major Collector</td>\n",
       "      <td>42</td>\n",
       "      <td>Tree (Standing Only)</td>\n",
       "      <td>The First Harmful Event was Not a Collision wi...</td>\n",
       "      <td>No</td>\n",
       "      <td>59.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>Lexus ES-250/300/300h/330/ 350</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4-door sedan, hardtop</td>\n",
       "      <td>1997.0</td>\n",
       "      <td>1997</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Trailing Units</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No Special Use</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rollover, Tripped by Object/Vehicle</td>\n",
       "      <td>12.0</td>\n",
       "      <td>12 Clock Point</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No or Not Reported</td>\n",
       "      <td>24</td>\n",
       "      <td>24 Years</td>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>1</td>\n",
       "      <td>Driver of a Motor Vehicle In-Transport</td>\n",
       "      <td>4</td>\n",
       "      <td>Fatal Injury (K)</td>\n",
       "      <td>11</td>\n",
       "      <td>Front Seat, Left Side</td>\n",
       "      <td>20</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>1</td>\n",
       "      <td>Deployed- Front</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Ejected</td>\n",
       "      <td>0</td>\n",
       "      <td>Ejection Path Not Applicable</td>\n",
       "      <td>1</td>\n",
       "      <td>Extricated</td>\n",
       "      <td>9</td>\n",
       "      <td>Reported as Unknown</td>\n",
       "      <td>9</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>1</td>\n",
       "      <td>Blood</td>\n",
       "      <td>294</td>\n",
       "      <td>0.294 % BAC</td>\n",
       "      <td>9</td>\n",
       "      <td>Reported as Unknown</td>\n",
       "      <td>8</td>\n",
       "      <td>Not Reported</td>\n",
       "      <td>2</td>\n",
       "      <td>Test Given</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Transported for Treatment</td>\n",
       "      <td>7</td>\n",
       "      <td>Died at Scene</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>January</td>\n",
       "      <td>2020</td>\n",
       "      <td>2020</td>\n",
       "      <td>2</td>\n",
       "      <td>2:00-2:59</td>\n",
       "      <td>58</td>\n",
       "      <td>58</td>\n",
       "      <td>258</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>7</td>\n",
       "      <td>Non-Hispanic</td>\n",
       "      <td>0</td>\n",
       "      <td>Occupant of a Motor Vehicle</td>\n",
       "      <td>20</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>7</td>\n",
       "      <td>None Used/Not Applicable</td>\n",
       "      <td>515.0</td>\n",
       "      <td>Lexus</td>\n",
       "      <td>2475.0</td>\n",
       "      <td>ES</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Sedan/Saloon</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Not Applicable</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   state  st_case  ve_total  peds  persons  county  day  month  hour  minute  \\\n",
       "0      1    10001         1     0        4      51    1      1     2      58   \n",
       "1      1    10001         1     0        4      51    1      1     2      58   \n",
       "2      1    10001         1     0        4      51    1      1     2      58   \n",
       "3      1    10001         1     0        4      51    1      1     2      58   \n",
       "4      1    10001         1     0        4      51    1      1     2      58   \n",
       "\n",
       "   nhs  route tway_id  rur_urb  rd_owner  milept   latitude  sp_jur  man_coll  \\\n",
       "0    0      4    cr-4        1         2       0  32.433133       0         0   \n",
       "1    0      4    cr-4        1         2       0  32.433133       0         0   \n",
       "2    0      4    cr-4        1         2       0  32.433133       0         0   \n",
       "3    0      4    cr-4        1         2       0  32.433133       0         0   \n",
       "4    0      4    cr-4        1         2       0  32.433133       0         0   \n",
       "\n",
       "   reljct1  reljct2  typ_int  wrk_zone  rel_road  lgt_cond  weather  sch_bus  \\\n",
       "0        0        1        1         0         4         2        1        0   \n",
       "1        0        1        1         0         4         2        1        0   \n",
       "2        0        1        1         0         4         2        1        0   \n",
       "3        0        1        1         0         4         2        1        0   \n",
       "4        0        1        1         0         4         2        1        0   \n",
       "\n",
       "      rail  not_hour  arr_hour  hosp_hr  drunk_dr  year  fatals  a_crainj  \\\n",
       "0  0000000        99         3       99         1  2020       3         1   \n",
       "1  0000000        99         3       99         1  2020       3         1   \n",
       "2  0000000        99         3       99         1  2020       3         1   \n",
       "3  0000000        99         3       99         1  2020       3         1   \n",
       "4  0000000        99         3       99         1  2020       3         1   \n",
       "\n",
       "   a_region  a_ru  a_inter  a_relrd  a_intsec  a_roadfc  a_junc  a_mancol  \\\n",
       "0         4     1        2        4         2         5       2         1   \n",
       "1         4     1        2        4         2         5       2         1   \n",
       "2         4     1        2        4         2         5       2         1   \n",
       "3         4     1        2        4         2         5       2         1   \n",
       "4         4     1        2        4         2         5       2         1   \n",
       "\n",
       "   a_tod  a_dow  a_ct  a_weather  a_lt  a_mc  a_spcra  a_ped  a_ped_f  \\\n",
       "0      2      1     1          1     2     2        1      2        2   \n",
       "1      2      1     1          1     2     2        1      2        2   \n",
       "2      2      1     1          1     2     2        1      2        2   \n",
       "3      2      1     1          1     2     2        1      2        2   \n",
       "4      2      1     1          1     2     2        1      2        2   \n",
       "\n",
       "   a_pedal  a_pedal_f  a_roll  a_polpur  a_posbac  a_d15_19  a_d16_19  \\\n",
       "0        2          2       1         2         1         2         2   \n",
       "1        2          2       1         2         1         2         2   \n",
       "2        2          2       1         2         1         2         2   \n",
       "3        2          2       1         2         1         2         2   \n",
       "4        2          2       1         2         1         2         2   \n",
       "\n",
       "   a_d15_20  a_d16_20  a_d65pls  a_d21_24  a_d16_24  a_rd  a_hr  a_dist  \\\n",
       "0         2         2         2         1         1     1     2       2   \n",
       "1         2         2         2         1         1     1     2       2   \n",
       "2         2         2         2         1         1     1     2       2   \n",
       "3         2         2         2         1         1     1     2       2   \n",
       "4         2         2         2         1         1     1     2       2   \n",
       "\n",
       "   a_drowsy  bia  spj_indian  indian_res statename weathername  crashrf  \\\n",
       "0         2    0           0           0   Alabama       Clear        0   \n",
       "1         2    0           0           0   Alabama       Clear        0   \n",
       "2         2    0           0           0   Alabama       Clear        0   \n",
       "3         2    0           0           0   Alabama       Clear        0   \n",
       "4         2    0           0           0   Alabama       Clear        0   \n",
       "\n",
       "  crashrfname  eventnum  vnumber1  aoi1           aoi1name  soe  \\\n",
       "0        None         1         1    55  Non-Harmful Event   63   \n",
       "1        None         1         1    55  Non-Harmful Event   63   \n",
       "2        None         1         1    55  Non-Harmful Event   63   \n",
       "3        None         1         1    55  Non-Harmful Event   63   \n",
       "4        None         2         1    12     12 Clock Point   42   \n",
       "\n",
       "                   soename  vnumber2         vnumber2name  aoi2  \\\n",
       "0  Ran Off Roadway - Right      5555    Non-Harmful Event    55   \n",
       "1  Ran Off Roadway - Right      5555    Non-Harmful Event    55   \n",
       "2  Ran Off Roadway - Right      5555    Non-Harmful Event    55   \n",
       "3  Ran Off Roadway - Right      5555    Non-Harmful Event    55   \n",
       "4     Tree (Standing Only)      9999  Not a Motor Vehicle    77   \n",
       "\n",
       "              aoi2name  ve_forms  veh_no  per_no  str_veh  dayname monthname  \\\n",
       "0    Non-Harmful Event         1       1       1        0        1   January   \n",
       "1    Non-Harmful Event         1       1       2        0        1   January   \n",
       "2    Non-Harmful Event         1       1       3        0        1   January   \n",
       "3    Non-Harmful Event         1       1       4        0        1   January   \n",
       "4  Not a Motor Vehicle         1       1       1        0        1   January   \n",
       "\n",
       "        hourname minutename rur_urbname  func_sys     func_sysname  harm_ev  \\\n",
       "0  2:00am-2:59am         58       Rural         5  Major Collector       42   \n",
       "1  2:00am-2:59am         58       Rural         5  Major Collector       42   \n",
       "2  2:00am-2:59am         58       Rural         5  Major Collector       42   \n",
       "3  2:00am-2:59am         58       Rural         5  Major Collector       42   \n",
       "4  2:00am-2:59am         58       Rural         5  Major Collector       42   \n",
       "\n",
       "            harm_evname                                       man_collname  \\\n",
       "0  Tree (Standing Only)  The First Harmful Event was Not a Collision wi...   \n",
       "1  Tree (Standing Only)  The First Harmful Event was Not a Collision wi...   \n",
       "2  Tree (Standing Only)  The First Harmful Event was Not a Collision wi...   \n",
       "3  Tree (Standing Only)  The First Harmful Event was Not a Collision wi...   \n",
       "4  Tree (Standing Only)  The First Harmful Event was Not a Collision wi...   \n",
       "\n",
       "  sch_busname  make makename                         mak_mod  body_typ  \\\n",
       "0          No  59.0    Lexus  Lexus ES-250/300/300h/330/ 350       4.0   \n",
       "1          No  59.0    Lexus  Lexus ES-250/300/300h/330/ 350       4.0   \n",
       "2          No  59.0    Lexus  Lexus ES-250/300/300h/330/ 350       4.0   \n",
       "3          No  59.0    Lexus  Lexus ES-250/300/300h/330/ 350       4.0   \n",
       "4          No  59.0    Lexus  Lexus ES-250/300/300h/330/ 350       4.0   \n",
       "\n",
       "            body_typname  mod_year mod_yearname  tow_veh        tow_vehname  \\\n",
       "0  4-door sedan, hardtop    1997.0         1997      0.0  No Trailing Units   \n",
       "1  4-door sedan, hardtop    1997.0         1997      0.0  No Trailing Units   \n",
       "2  4-door sedan, hardtop    1997.0         1997      0.0  No Trailing Units   \n",
       "3  4-door sedan, hardtop    1997.0         1997      0.0  No Trailing Units   \n",
       "4  4-door sedan, hardtop    1997.0         1997      0.0  No Trailing Units   \n",
       "\n",
       "   spec_use    spec_usename  emer_use    emer_usename  rollover  \\\n",
       "0       0.0  No Special Use       0.0  Not Applicable       1.0   \n",
       "1       0.0  No Special Use       0.0  Not Applicable       1.0   \n",
       "2       0.0  No Special Use       0.0  Not Applicable       1.0   \n",
       "3       0.0  No Special Use       0.0  Not Applicable       1.0   \n",
       "4       0.0  No Special Use       0.0  Not Applicable       1.0   \n",
       "\n",
       "                          rollovername  impact1     impact1name  fire_exp  \\\n",
       "0  Rollover, Tripped by Object/Vehicle     12.0  12 Clock Point       0.0   \n",
       "1  Rollover, Tripped by Object/Vehicle     12.0  12 Clock Point       0.0   \n",
       "2  Rollover, Tripped by Object/Vehicle     12.0  12 Clock Point       0.0   \n",
       "3  Rollover, Tripped by Object/Vehicle     12.0  12 Clock Point       0.0   \n",
       "4  Rollover, Tripped by Object/Vehicle     12.0  12 Clock Point       0.0   \n",
       "\n",
       "         fire_expname  age   agename  sex sexname  per_typ  \\\n",
       "0  No or Not Reported   24  24 Years    1    Male        1   \n",
       "1  No or Not Reported   21  21 Years    1    Male        2   \n",
       "2  No or Not Reported   22  22 Years    2  Female        2   \n",
       "3  No or Not Reported   22  22 Years    2  Female        2   \n",
       "4  No or Not Reported   24  24 Years    1    Male        1   \n",
       "\n",
       "                                 per_typname  inj_sev  \\\n",
       "0     Driver of a Motor Vehicle In-Transport        4   \n",
       "1  Passenger of a Motor Vehicle In-Transport        3   \n",
       "2  Passenger of a Motor Vehicle In-Transport        4   \n",
       "3  Passenger of a Motor Vehicle In-Transport        4   \n",
       "4     Driver of a Motor Vehicle In-Transport        4   \n",
       "\n",
       "                    inj_sevname  seat_pos             seat_posname  rest_use  \\\n",
       "0              Fatal Injury (K)        11    Front Seat, Left Side        20   \n",
       "1  Suspected Serious Injury (A)        13   Front Seat, Right Side         3   \n",
       "2              Fatal Injury (K)        21   Second Seat, Left Side        20   \n",
       "3              Fatal Injury (K)        23  Second Seat, Right Side        20   \n",
       "4              Fatal Injury (K)        11    Front Seat, Left Side        20   \n",
       "\n",
       "                 rest_usename  rest_mis              rest_misname  air_bag  \\\n",
       "0    None Used/Not Applicable         7  None Used/Not Applicable        1   \n",
       "1  Shoulder and Lap Belt Used         0  No Indication of Mis-Use        1   \n",
       "2    None Used/Not Applicable         7  None Used/Not Applicable       20   \n",
       "3    None Used/Not Applicable         7  None Used/Not Applicable       20   \n",
       "4    None Used/Not Applicable         7  None Used/Not Applicable        1   \n",
       "\n",
       "       air_bagname  ejection ejectionname  ej_path  \\\n",
       "0  Deployed- Front         0  Not Ejected        0   \n",
       "1  Deployed- Front         0  Not Ejected        0   \n",
       "2     Not Deployed         0  Not Ejected        0   \n",
       "3     Not Deployed         0  Not Ejected        0   \n",
       "4  Deployed- Front         0  Not Ejected        0   \n",
       "\n",
       "                    ej_pathname  extricat extricatname  drinking  \\\n",
       "0  Ejection Path Not Applicable         1   Extricated         9   \n",
       "1  Ejection Path Not Applicable         1   Extricated         8   \n",
       "2  Ejection Path Not Applicable         1   Extricated         8   \n",
       "3  Ejection Path Not Applicable         1   Extricated         8   \n",
       "4  Ejection Path Not Applicable         1   Extricated         9   \n",
       "\n",
       "          drinkingname  alc_det   alc_detname  alc_status  alc_statusname  \\\n",
       "0  Reported as Unknown        9  Not Reported           2      Test Given   \n",
       "1         Not Reported        9  Not Reported           0  Test Not Given   \n",
       "2         Not Reported        9  Not Reported           2      Test Given   \n",
       "3         Not Reported        9  Not Reported           2      Test Given   \n",
       "4  Reported as Unknown        9  Not Reported           2      Test Given   \n",
       "\n",
       "   atst_typ    atst_typname  alc_res     alc_resname  drugs  \\\n",
       "0         1           Blood      294     0.294 % BAC      9   \n",
       "1         0  Test Not Given      996  Test Not Given      8   \n",
       "2         1           Blood       69     0.069 % BAC      8   \n",
       "3         1           Blood      120     0.120 % BAC      8   \n",
       "4         1           Blood      294     0.294 % BAC      9   \n",
       "\n",
       "             drugsname  drug_det  drug_detname  dstatus     dstatusname  \\\n",
       "0  Reported as Unknown         8  Not Reported        2      Test Given   \n",
       "1         Not Reported         8  Not Reported        0  Test Not Given   \n",
       "2         Not Reported         8  Not Reported        2      Test Given   \n",
       "3         Not Reported         8  Not Reported        2      Test Given   \n",
       "4  Reported as Unknown         8  Not Reported        2      Test Given   \n",
       "\n",
       "   hospital                   hospitalname  doa         doaname  death_da  \\\n",
       "0         0  Not Transported for Treatment    7   Died at Scene         1   \n",
       "1         5                     EMS Ground    0  Not Applicable        88   \n",
       "2         0  Not Transported for Treatment    7   Died at Scene         1   \n",
       "3         0  Not Transported for Treatment    7   Died at Scene         1   \n",
       "4         0  Not Transported for Treatment    7   Died at Scene         1   \n",
       "\n",
       "                 death_daname  death_mo                death_moname  death_yr  \\\n",
       "0                           1         1                     January      2020   \n",
       "1  Not Applicable (Non-Fatal)        88  Not Applicable (Non-Fatal)      8888   \n",
       "2                           1         1                     January      2020   \n",
       "3                           1         1                     January      2020   \n",
       "4                           1         1                     January      2020   \n",
       "\n",
       "                 death_yrname  death_hr                death_hrname  death_mn  \\\n",
       "0                        2020         2                   2:00-2:59        58   \n",
       "1  Not Applicable (Non-fatal)        88  Not Applicable (Non-fatal)        88   \n",
       "2                        2020         2                   2:00-2:59        58   \n",
       "3                        2020         2                   2:00-2:59        58   \n",
       "4                        2020         2                   2:00-2:59        58   \n",
       "\n",
       "                 death_mnname  death_tm                death_tmname  lag_hrs  \\\n",
       "0                          58       258                         258        0   \n",
       "1  Not Applicable (Non-fatal)      8888  Not Applicable (Non-fatal)      999   \n",
       "2                          58       258                         258        0   \n",
       "3                          58       258                         258        0   \n",
       "4                          58       258                         258        0   \n",
       "\n",
       "  lag_hrsname  lag_mins lag_minsname  work_inj  \\\n",
       "0           0         0            0         0   \n",
       "1     Unknown        99      Unknown         8   \n",
       "2           0         0            0         0   \n",
       "3           0         0            0         0   \n",
       "4           0         0            0         0   \n",
       "\n",
       "                      work_injname  hispanic                     hispanicname  \\\n",
       "0                               No         7                     Non-Hispanic   \n",
       "1  Not Applicable (not a fatality)         0  Not A Fatality (not Applicable)   \n",
       "2                               No         7                     Non-Hispanic   \n",
       "3                               No         7                     Non-Hispanic   \n",
       "4                               No         7                     Non-Hispanic   \n",
       "\n",
       "   location                 locationname  helm_use    helm_usename  helm_mis  \\\n",
       "0         0  Occupant of a Motor Vehicle        20  Not Applicable         7   \n",
       "1         0  Occupant of a Motor Vehicle        20  Not Applicable         7   \n",
       "2         0  Occupant of a Motor Vehicle        20  Not Applicable         7   \n",
       "3         0  Occupant of a Motor Vehicle        20  Not Applicable         7   \n",
       "4         0  Occupant of a Motor Vehicle        20  Not Applicable         7   \n",
       "\n",
       "               helm_misname  vpicmake vpicmakename  vpicmodel vpicmodelname  \\\n",
       "0  None Used/Not Applicable     515.0        Lexus     2475.0            ES   \n",
       "1  None Used/Not Applicable     515.0        Lexus     2475.0            ES   \n",
       "2  None Used/Not Applicable     515.0        Lexus     2475.0            ES   \n",
       "3  None Used/Not Applicable     515.0        Lexus     2475.0            ES   \n",
       "4  None Used/Not Applicable     515.0        Lexus     2475.0            ES   \n",
       "\n",
       "   vpicbodyclass vpicbodyclassname  icfinalbody icfinalbodyname  data_year  \n",
       "0           13.0      Sedan/Saloon          0.0  Not Applicable       2020  \n",
       "1           13.0      Sedan/Saloon          0.0  Not Applicable       2020  \n",
       "2           13.0      Sedan/Saloon          0.0  Not Applicable       2020  \n",
       "3           13.0      Sedan/Saloon          0.0  Not Applicable       2020  \n",
       "4           13.0      Sedan/Saloon          0.0  Not Applicable       2020  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets take a look at the columns to see if there are any interesting takeaways\n",
    "pd.set_option('display.max.columns', None)\n",
    "accidents_df.head(5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5fcc0c2a-91e5-4b68-837f-5b26cf92570d",
   "metadata": {},
   "source": [
    "# More data exploration and cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "69798a17-5744-4b4d-a102-09e28f42882a",
   "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>fatals</th>\n",
       "      <th>persons</th>\n",
       "      <th>peds</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1249</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1250</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1251</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1252</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1253</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>256954</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>256955</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>256956</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258219</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>264564</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1318 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        fatals  persons  peds\n",
       "1249         2        1     2\n",
       "1250         2        1     2\n",
       "1251         2        1     2\n",
       "1252         2        1     2\n",
       "1253         2        1     2\n",
       "...        ...      ...   ...\n",
       "256954       2        1     2\n",
       "256955       2        1     2\n",
       "256956       2        1     2\n",
       "258219       1        0     1\n",
       "264564       1        0     1\n",
       "\n",
       "[1318 rows x 3 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# First lets take a look at the fatals, persons, oeds columns\n",
    "# To note- not everyone involved in the crashes died\n",
    "\n",
    "accidents_df.loc[(accidents_df['fatals'] > accidents_df['persons']), ['fatals', 'persons', 'peds']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "77638a87-c1e3-4cad-9472-c4e51ec5fcc0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Died at Scene', 'Not Applicable', 'Died En Route', 'Unknown'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This column represents those who died at the scene or en route\n",
    "# Taking a deeper look - I would need to binarize this column to make use of it\n",
    "accidents_df['doaname'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "bba55bde-6dcb-4608-9cf3-7e4c868ef618",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Not Applicable    190244\n",
       "Died at Scene      71342\n",
       "Unknown             2852\n",
       "Died En Route        753\n",
       "Name: doaname, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### far too many not applicable- will have to drop column\n",
    "accidents_df['doaname'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "838f7247-4ad3-4a5b-8d4a-edfba3dc8b39",
   "metadata": {},
   "outputs": [],
   "source": [
    "#the doaname column has been dropped\n",
    "accidents_df = accidents_df.drop(columns=['doaname'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "33994994-1037-48e3-bb29-2b646d4c923c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    225889\n",
       "2     30537\n",
       "3      6284\n",
       "4      1766\n",
       "5       547\n",
       "6       104\n",
       "7        40\n",
       "8        24\n",
       "Name: fatals, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's take a look at the total number of fatals in a unique state case\n",
    "# As we can see the majority of fatal accidents that occur in this data set involve one person... why?\n",
    "accidents_df['fatals'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9aee8a96-d6f2-4c37-9095-1c7de6ebe192",
   "metadata": {},
   "outputs": [],
   "source": [
    "#here I will bin the columns to make them binary- either fatals = 1 or fatal > 1\n",
    "\n",
    "## Make a seasons function that bins fatals\n",
    "def binary_fatals(x):\n",
    "    if x == 1:\n",
    "        return 0\n",
    "    elif x > 0:\n",
    "        return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "24005dd5-77de-4933-afda-47e99525d670",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Lets save the previous equation to the df\n",
    "accidents_df['binary_fatals'] = accidents_df['fatals'].map(binary_fatals)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "fb725397-c197-445a-b7fd-320cfc6f1933",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    225889\n",
       "1     39302\n",
       "Name: binary_fatals, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# now lets take a look at the totals\n",
    "accidents_df['binary_fatals'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "916422ff-b33c-451c-8edb-b6593b3fb065",
   "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>binary_fatals</th>\n",
       "      <th>inj_sev</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>binary_fatals</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.082831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inj_sev</th>\n",
       "      <td>0.082831</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               binary_fatals   inj_sev\n",
       "binary_fatals       1.000000  0.082831\n",
       "inj_sev             0.082831  1.000000"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# I want to check the correlation between inj_sev and binary_fatals\n",
    "# The correlation is weak\n",
    "# Lets take a closer look inj_sev\n",
    "accidents_df[['binary_fatals', 'inj_sev']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d656f94a-5662-40b1-b644-6b0b1dc1edae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4    117315\n",
       "0     57797\n",
       "3     33335\n",
       "2     31109\n",
       "1     21740\n",
       "9      2743\n",
       "5      1141\n",
       "6        11\n",
       "Name: inj_sev, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The majority of inj_sev are denoted as fatal\n",
    "# In order to get a netter understanding I need to drop some rows from this column\n",
    "# I also need to binarize it\n",
    "accidents_df['inj_sev'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4afa8e73-e048-4dda-8d67-aac3f4106b5b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#here I will bin the columns to make them binary- either fatals = 1 or fatal > 1\n",
    "\n",
    "## Make a seasons function that bins fatals\n",
    "def inj_sev_binary(x):\n",
    "    if x == 4:\n",
    "        return 0\n",
    "    elif x != 4:\n",
    "        return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "fb8e9b9f-0697-48ba-896e-3e076eda6d1d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving the changes to the dataframe\n",
    "accidents_df['inj_sev_binary'] = accidents_df['inj_sev'].map(inj_sev_binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "10b1e357-a323-428f-8d76-24bfb96dfaa5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1], dtype=int64)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets take a look to make sure the changes were made\n",
    "accidents_df['inj_sev_binary'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "099d09a8-d23e-4a4a-b72b-6a5b0f810dbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    147876\n",
       "0    117315\n",
       "Name: inj_sev_binary, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's see how many there are of each\n",
    "accidents_df['inj_sev_binary'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "120ff0c4-d374-4321-b914-1a4bc5e67456",
   "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>binary_fatals</th>\n",
       "      <th>inj_sev</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>binary_fatals</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.082831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inj_sev</th>\n",
       "      <td>0.082831</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               binary_fatals   inj_sev\n",
       "binary_fatals       1.000000  0.082831\n",
       "inj_sev             0.082831  1.000000"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The correlation has improved\n",
    "accidents_df[['binary_fatals', 'inj_sev']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ffb0feec-f16e-48ea-be5d-0c49741a48e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df.loc[accidents_df['st_case'] == 80224].duplicated().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "96716896-5026-49cf-b70d-800f3f2529b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df.loc[accidents_df['st_case'] == 80224, ['inj_sev', 'fatals', 'st_case', 'persons', 'peds', 'per_no', 'veh_no', 'per_typ']].duplicated().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d7bc7433-c43d-44bb-90ad-51b6170e20ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Lets take a closer look at the persons column along with a few others\n",
    "accidents_df_persons = accidents_df.loc[accidents_df['persons'] > 5, ['inj_sev', 'fatals', 'st_case', 'persons', 'peds', 'veh_no','inj_sev_binary']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "09552ef2-d4bb-4ad1-95d8-da50b6448605",
   "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>inj_sev</th>\n",
       "      <th>fatals</th>\n",
       "      <th>st_case</th>\n",
       "      <th>peds</th>\n",
       "      <th>veh_no</th>\n",
       "      <th>inj_sev_binary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>persons</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>6</th>\n",
       "      <td>1.998068</td>\n",
       "      <td>1.340397</td>\n",
       "      <td>264587.453601</td>\n",
       "      <td>0.074805</td>\n",
       "      <td>1.963352</td>\n",
       "      <td>0.778482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.729460</td>\n",
       "      <td>1.338920</td>\n",
       "      <td>257677.038560</td>\n",
       "      <td>0.101183</td>\n",
       "      <td>2.198560</td>\n",
       "      <td>0.811105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.792552</td>\n",
       "      <td>1.428592</td>\n",
       "      <td>275054.414039</td>\n",
       "      <td>0.096590</td>\n",
       "      <td>2.451134</td>\n",
       "      <td>0.823085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.770030</td>\n",
       "      <td>1.649713</td>\n",
       "      <td>312112.722177</td>\n",
       "      <td>0.168991</td>\n",
       "      <td>2.654908</td>\n",
       "      <td>0.818430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.584937</td>\n",
       "      <td>1.552301</td>\n",
       "      <td>322338.958159</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.744770</td>\n",
       "      <td>0.844770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.609426</td>\n",
       "      <td>1.713976</td>\n",
       "      <td>234937.588841</td>\n",
       "      <td>0.058505</td>\n",
       "      <td>2.957746</td>\n",
       "      <td>0.845070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.459677</td>\n",
       "      <td>1.346774</td>\n",
       "      <td>344525.532258</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.906586</td>\n",
       "      <td>0.887769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.428054</td>\n",
       "      <td>1.341176</td>\n",
       "      <td>118945.976471</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.866968</td>\n",
       "      <td>0.896833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.186023</td>\n",
       "      <td>1.143885</td>\n",
       "      <td>313422.302158</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.373073</td>\n",
       "      <td>0.918294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1.863768</td>\n",
       "      <td>1.173913</td>\n",
       "      <td>127950.152174</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.272464</td>\n",
       "      <td>0.921739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.656250</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>239625.062500</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.597656</td>\n",
       "      <td>0.937500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.588235</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>61128.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.705882</td>\n",
       "      <td>0.882353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1.319838</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>78156.307692</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.587045</td>\n",
       "      <td>0.947368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2.300000</td>\n",
       "      <td>1.636364</td>\n",
       "      <td>475154.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.604545</td>\n",
       "      <td>0.918182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2.104762</td>\n",
       "      <td>2.200000</td>\n",
       "      <td>41036.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.380952</td>\n",
       "      <td>0.895238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0.181818</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>220483.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>0.954545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1.125000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>80575.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.791667</td>\n",
       "      <td>0.958333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2.076923</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>180021.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.269231</td>\n",
       "      <td>0.884615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>1.071429</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>370149.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.035714</td>\n",
       "      <td>0.964286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>2.161290</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>482487.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.967742</td>\n",
       "      <td>0.967742</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>4.983607</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>420105.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.983607</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          inj_sev    fatals        st_case      peds    veh_no  inj_sev_binary\n",
       "persons                                                                       \n",
       "6        1.998068  1.340397  264587.453601  0.074805  1.963352        0.778482\n",
       "7        1.729460  1.338920  257677.038560  0.101183  2.198560        0.811105\n",
       "8        1.792552  1.428592  275054.414039  0.096590  2.451134        0.823085\n",
       "9        1.770030  1.649713  312112.722177  0.168991  2.654908        0.818430\n",
       "10       1.584937  1.552301  322338.958159  0.000000  2.744770        0.844770\n",
       "11       1.609426  1.713976  234937.588841  0.058505  2.957746        0.845070\n",
       "12       1.459677  1.346774  344525.532258  0.000000  3.906586        0.887769\n",
       "13       1.428054  1.341176  118945.976471  0.000000  3.866968        0.896833\n",
       "14       1.186023  1.143885  313422.302158  0.000000  4.373073        0.918294\n",
       "15       1.863768  1.173913  127950.152174  0.000000  3.272464        0.921739\n",
       "16       0.656250  1.000000  239625.062500  0.000000  4.597656        0.937500\n",
       "17       0.588235  2.000000   61128.000000  0.000000  4.705882        0.882353\n",
       "19       1.319838  1.000000   78156.307692  0.000000  5.587045        0.947368\n",
       "20       2.300000  1.636364  475154.000000  0.000000  1.604545        0.918182\n",
       "21       2.104762  2.200000   41036.000000  0.000000  1.380952        0.895238\n",
       "22       0.181818  1.000000  220483.000000  0.000000  1.500000        0.954545\n",
       "24       1.125000  1.000000   80575.000000  0.000000  8.791667        0.958333\n",
       "26       2.076923  3.000000  180021.000000  0.000000  5.269231        0.884615\n",
       "28       1.071429  1.000000  370149.000000  0.000000  1.035714        0.964286\n",
       "31       2.161290  1.000000  482487.000000  0.000000  1.967742        0.967742\n",
       "61       4.983607  1.000000  420105.000000  0.000000  1.000000        0.983607"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The injury severity binary column is ti;tinh towards one for every row\n",
    "# Most accidents above 5 persons seems to be moderately severe\n",
    "accidents_df_persons.groupby('persons').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "6fefd65c-6cb5-404e-8e23-a3d7279cdaf4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Fatal Injury (K)                117315\n",
       "No Apparent Injury (O)           57797\n",
       "Suspected Serious Injury (A)     33335\n",
       "Suspected Minor Injury (B)       31109\n",
       "Possible Injury (C)              21740\n",
       "Unknown/Not Reported              2743\n",
       "Injured, Severity Unknown         1141\n",
       "Died Prior to Crash*                11\n",
       "Name: inj_sevname, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['inj_sevname'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "c8a541b4-8cfa-4708-a438-3220839e29fc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Fatal Injury (K)', 'Suspected Serious Injury (A)',\n",
       "       'No Apparent Injury (O)', 'Suspected Minor Injury (B)',\n",
       "       'Possible Injury (C)', 'Unknown/Not Reported',\n",
       "       'Injured, Severity Unknown'], dtype=object)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df = accidents_df[accidents_df[\"inj_sevname\"].str.contains(\"Died Prior to Crash*\") == False]\n",
    "accidents_df['inj_sevname'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b841a49d-cccc-4597-ae01-9b69b891777c",
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df = accidents_df[accidents_df[\"inj_sevname\"].str.contains(\"Unknown/Not Reported\") == False]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "1ea704c7-a827-4c7d-88d9-1248dfaa1ac5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Fatal Injury (K)', 'Suspected Serious Injury (A)',\n",
       "       'No Apparent Injury (O)', 'Suspected Minor Injury (B)',\n",
       "       'Possible Injury (C)', 'Injured, Severity Unknown'], dtype=object)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['inj_sevname'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "4c193c9e-5d4f-4cb1-887f-bce13e58a8a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Fatal Injury (K)', 'Suspected Serious Injury (A)',\n",
       "       'No Apparent Injury (O)', 'Suspected Minor Injury (B)',\n",
       "       'Possible Injury (C)'], dtype=object)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df = accidents_df[accidents_df[\"inj_sevname\"].str.contains(\"Injured, Severity Unknown\") == False]\n",
    "accidents_df['inj_sevname'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "e8e5ca5f-1ee7-428d-a322-153c4c434e39",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 2, 1, 3], dtype=int64)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df = accidents_df[accidents_df['wrk_zone'] != 4]\n",
    "accidents_df['wrk_zone'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "735ec071-4b5a-47fd-bee5-4e72ee9a433c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#here I will bin the columns to make them binary- either drunk or not drunk\n",
    "\n",
    "## Make a seasons function that bins months into seasons\n",
    "def work_zone_binary(x):\n",
    "    if x == 0:\n",
    "        return 0\n",
    "    elif x != 0:\n",
    "        return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5ce81f6c-8a3d-4d00-a796-f2e2ed69a7eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "accidents_df['work_zone_binary'] = accidents_df['wrk_zone'].map(work_zone_binary) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "e49e4cb0-f355-44f2-a210-fb6e2420bab9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1], dtype=int64)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['work_zone_binary'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "0ee1a62e-bce1-42aa-a118-fec0ce4004b7",
   "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>binary_fatals</th>\n",
       "      <th>work_zone_binary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>binary_fatals</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.007718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>work_zone_binary</th>\n",
       "      <td>0.007718</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  binary_fatals  work_zone_binary\n",
       "binary_fatals          1.000000          0.007718\n",
       "work_zone_binary       0.007718          1.000000"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[['binary_fatals', 'work_zone_binary']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "ca6cc456-b78b-45bd-96f0-76edbb102789",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3, 1, 5, 4, 9, 6, 8, 7], dtype=int64)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['lgt_cond'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "7e533b94-a37c-4fd3-933c-fbea8dd01a77",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3, 1, 5, 4, 6, 8, 7], dtype=int64)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df = accidents_df[accidents_df['lgt_cond'] != 9]\n",
    "accidents_df['lgt_cond'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "6fe68b63-1d57-4b3a-b5c6-d257651d4e95",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3, 1, 5, 4, 6, 7], dtype=int64)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df = accidents_df[accidents_df['lgt_cond'] != 8]\n",
    "accidents_df['lgt_cond'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "038d30ad-765a-40e1-8aa7-1d5160d141dd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3, 1, 5, 4, 6], dtype=int64)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df = accidents_df[accidents_df['lgt_cond'] != 7]\n",
    "accidents_df['lgt_cond'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "e347d676-d808-4257-9cdf-a4af31cd68a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#here I will bin the columns to make them binary- either drunk or not drunk\n",
    "\n",
    "## Make a seasons function that bins months into seasons\n",
    "def lgt_cond_binary(x):\n",
    "    if x == 1:\n",
    "        return 0\n",
    "    elif x != 1:\n",
    "        return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "08212b0c-136e-4b32-b0c3-1c3d0a0d2ef4",
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df['lgt_cond_binary'] = accidents_df['lgt_cond'].map(lgt_cond_binary) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "507233a9-a113-4797-8693-0ee2f0bcad62",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1], dtype=int64)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['work_zone_binary'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "8ff808ff-ee69-4f27-887f-c695dfecc129",
   "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>binary_fatals</th>\n",
       "      <th>lgt_cond_binary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>binary_fatals</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lgt_cond_binary</th>\n",
       "      <td>0.000918</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 binary_fatals  lgt_cond_binary\n",
       "binary_fatals         1.000000         0.000918\n",
       "lgt_cond_binary       0.000918         1.000000"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[['binary_fatals', 'lgt_cond_binary']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "82ca9c9b-b447-4aad-94cd-26a7de4ca304",
   "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>binary_fatals</th>\n",
       "      <th>hour</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>binary_fatals</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.000175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hour</th>\n",
       "      <td>-0.000175</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               binary_fatals      hour\n",
       "binary_fatals       1.000000 -0.000175\n",
       "hour               -0.000175  1.000000"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[['binary_fatals', 'hour']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "2143b257-e0ea-4f97-a3eb-68ab73cd3f59",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "inj_sevname\n",
       "Fatal Injury (K)                1.248047\n",
       "No Apparent Injury (O)          1.109554\n",
       "Possible Injury (C)             1.147678\n",
       "Suspected Minor Injury (B)      1.157776\n",
       "Suspected Serious Injury (A)    1.237976\n",
       "Name: fatals, dtype: float64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.groupby('inj_sevname')['fatals'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "d8a154ee-43b0-4f0e-8ecd-694913621dfd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "fatals\n",
       "1    2.428551\n",
       "2    2.828426\n",
       "3    3.066913\n",
       "4    3.370392\n",
       "5    3.233397\n",
       "6    3.840000\n",
       "7    3.400000\n",
       "8    4.000000\n",
       "Name: inj_sev, dtype: float64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.groupby('fatals')['inj_sev'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "43fca73b-734b-45f2-8363-2209a5cc66bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3, 1, 2, 4, 8, 6, 5, 7], dtype=int64)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['fatals'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "4506251f-929c-47c3-bbf7-6204fd9b3d0d",
   "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>drunk_dr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265186</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265187</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265188</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265189</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265190</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>257072 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        drunk_dr\n",
       "0              1\n",
       "1              1\n",
       "2              1\n",
       "3              1\n",
       "4              1\n",
       "...          ...\n",
       "265186         0\n",
       "265187         0\n",
       "265188         0\n",
       "265189         0\n",
       "265190         0\n",
       "\n",
       "[257072 rows x 1 columns]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[['drunk_dr']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4664b256-6982-4ca3-8007-684a5362216a",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "0e4d7ef9-22ec-4b6c-a622-5e5461d974f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 2, 3, 4], dtype=int64)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['drunk_dr'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "7121efb8-ae8a-48d3-b501-bcd87a61a566",
   "metadata": {},
   "outputs": [],
   "source": [
    "#here I will bin the columns to make them binary- either drunk or not drunk\n",
    "\n",
    "## Make a seasons function that bins months into seasons\n",
    "def drunk(x):\n",
    "    if x == 0:\n",
    "        return 0\n",
    "    elif x != 0:\n",
    "        return 1\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "029d9bec-aee8-47b2-b3be-3bfc8d9ffead",
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df['drunk'] = accidents_df['drunk_dr'].map(drunk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "8aa1e76b-a34e-41ee-a8fc-19108c005b71",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0], dtype=int64)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['drunk'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "7a3d1166-7c8c-41b3-a733-b0ab90c2f571",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    186553\n",
       "1     70519\n",
       "Name: drunk, dtype: int64"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['drunk'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "e6c38d42-9866-4ce3-9acf-b9cbaadcfd4b",
   "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>drunk</th>\n",
       "      <th>inj_sev</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>drunk</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.101006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inj_sev</th>\n",
       "      <td>0.101006</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            drunk   inj_sev\n",
       "drunk    1.000000  0.101006\n",
       "inj_sev  0.101006  1.000000"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[['drunk', 'inj_sev']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "2c563da3-ced1-404a-9c79-8a110ac6a190",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Fatal Injury (K)', 'Suspected Serious Injury (A)',\n",
       "       'No Apparent Injury (O)', 'Suspected Minor Injury (B)',\n",
       "       'Possible Injury (C)'], dtype=object)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['inj_sevname'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "c282e925-cc21-46f6-b58d-41649b48afa6",
   "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>Null</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>make</th>\n",
       "      <td>16113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fire_expname</th>\n",
       "      <td>16113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>impact1name</th>\n",
       "      <td>16113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>impact1</th>\n",
       "      <td>16113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rollovername</th>\n",
       "      <td>16113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>spj_indian</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>indian_res</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>statename</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>weathername</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>drunk</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>200 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               Null\n",
       "make          16113\n",
       "fire_expname  16113\n",
       "impact1name   16113\n",
       "impact1       16113\n",
       "rollovername  16113\n",
       "...             ...\n",
       "spj_indian        0\n",
       "indian_res        0\n",
       "statename         0\n",
       "weathername       0\n",
       "drunk             0\n",
       "\n",
       "[200 rows x 1 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets take a look at the missing values\n",
    "missin_values_count = pd.DataFrame({'Null': accidents_df.isnull().sum()})\n",
    "missin_values_count.sort_values(by= 'Null', ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "d64110ac-e3c0-44e1-864c-f3525b17bdc3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3, 1, 5, 4, 6], dtype=int64)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['lgt_cond'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f92a492-08cd-4db6-9c9b-16b4ef629a34",
   "metadata": {},
   "source": [
    "# Great! There are no null values! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "8f46a7a5-27cb-42c0-a223-0a84d3a0bb22",
   "metadata": {},
   "outputs": [],
   "source": [
    "# sns.set(font_scale=.4, style='white')\n",
    "# corr = accidents_df.drop(columns = ['fatals']).corr()\n",
    "# mask = np.zeros_like(corr, dtype= bool)\n",
    "# mask[np.triu_indices_from(mask)] = True\n",
    "# f, ax = plt.subplots(figsize=(25,25))\n",
    "# cmap = sns.diverging_palette(220, 10, as_cmap=True)\n",
    "# sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0, square=True, linewidth=.5,\n",
    "#            cbar_kws={'shrink': .5}, annot = True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0c763a3f-fcc3-4809-b556-ddc960b2f6e1",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "b2cbcbfc-0ac6-462e-a565-553b1e38dd62",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "79119246-df9b-41a1-93f2-dc63c211bf98",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "28813226-e0cd-40de-bb99-e6423ce178e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df['crashrf'].isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "c04dddf8-0d50-44ef-8bcf-d0952517bcbd",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df.drop(columns='crashrfname', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "3f6a5642-1659-412d-a753-a7587752a3ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "f87dba96-b5a4-486e-8b7b-c676caf6387d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df['county'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "75551210-e16e-48f6-9066-877b4fba13a5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "023b0674-9a2c-46f5-9af5-f312ac3ac84b",
   "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>fatals</th>\n",
       "      <th>weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>fatals</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.007273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>weather</th>\n",
       "      <td>-0.007273</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           fatals   weather\n",
       "fatals   1.000000 -0.007273\n",
       "weather -0.007273  1.000000"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[['fatals', 'weather']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "90a13e7e-f01d-4a20-a7da-dc81d3532d9a",
   "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>weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265186</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265187</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265188</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265189</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265190</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>257072 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        weather\n",
       "0             1\n",
       "1             1\n",
       "2             1\n",
       "3             1\n",
       "4             1\n",
       "...         ...\n",
       "265186        6\n",
       "265187        6\n",
       "265188        6\n",
       "265189        6\n",
       "265190        6\n",
       "\n",
       "[257072 rows x 1 columns]"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[['weather']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "6f6354d7-7dec-49df-9742-dd673eb7fd15",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df[['vnumber2', 'fatals', 'aoi2']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "2abc6474-ec9d-4bf5-a940-29d3ebba2753",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df.groupby('state')['fatals'].agg(['count', 'mean'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "59c199e7-3bd1-4802-aaf1-2af351ddfa07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 2, 1, 3], dtype=int64)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Here I want to clean the work zone column and see if there is a corr between that and fatals\n",
    "accidents_df['wrk_zone'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8b701917-0a7b-4cf3-b1a9-e3e2a0803127",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6eec8323-c7fa-430f-aff1-4bbfe45cbada",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "cabae443-3ef5-4d49-adbb-a089e2606c72",
   "metadata": {},
   "source": [
    "#### Let's take a look at the correlations between X features and Y(fatals)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "38fb70e2-77ab-43b5-82cc-feaf75042844",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "binary_fatals    1.000000\n",
       "fatals           0.872888\n",
       "persons          0.201581\n",
       "per_no           0.146894\n",
       "doa              0.117993\n",
       "                   ...   \n",
       "lag_hrs         -0.080950\n",
       "lag_mins        -0.091552\n",
       "year                  NaN\n",
       "a_crainj              NaN\n",
       "data_year             NaN\n",
       "Name: binary_fatals, Length: 136, dtype: float64"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correlations = accidents_df.corr()['binary_fatals'].sort_values(ascending = False)\n",
    "correlations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "7a69fd41-a7e8-43e0-a991-2e3a7139d805",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "binary_fatals    1.000000\n",
       "fatals           0.872888\n",
       "persons          0.201581\n",
       "per_no           0.146894\n",
       "doa              0.117993\n",
       "fire_exp         0.108002\n",
       "seat_pos         0.106745\n",
       "a_ct             0.105935\n",
       "inj_sev          0.103747\n",
       "ve_total         0.100664\n",
       "ve_forms         0.092218\n",
       "eventnum         0.091395\n",
       "a_mancol         0.082711\n",
       "a_mc             0.079400\n",
       "peds             0.059763\n",
       "drinking         0.058062\n",
       "veh_no           0.057839\n",
       "a_hr             0.049823\n",
       "man_coll         0.047103\n",
       "per_typ          0.044444\n",
       "a_ped            0.044441\n",
       "vnumber1         0.043959\n",
       "drugs            0.042447\n",
       "helm_mis         0.042149\n",
       "a_ped_f          0.040071\n",
       "extricat         0.039228\n",
       "bia              0.032168\n",
       "hispanic         0.030579\n",
       "a_pedal          0.030061\n",
       "Name: binary_fatals, dtype: float64"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.set_option('display.max.columns', None)\n",
    "correlations.head(29)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "972940c5-0c3a-4bc2-a875-5ebc605bfc6f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "inj_sev_binary    1.000000\n",
       "death_mo          0.956085\n",
       "death_yr          0.955728\n",
       "death_da          0.943833\n",
       "lag_hrs           0.920259\n",
       "                    ...   \n",
       "doa              -0.699278\n",
       "inj_sev          -0.833264\n",
       "year                   NaN\n",
       "a_crainj               NaN\n",
       "data_year              NaN\n",
       "Name: inj_sev_binary, Length: 136, dtype: float64"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correlations2 = accidents_df.corr()['inj_sev_binary'].sort_values(ascending = False)\n",
    "correlations2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "d6285fdc-68d7-427d-b4e1-11229973a981",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "inj_sev_binary      1.000000\n",
       "death_mo            0.956085\n",
       "death_yr            0.955728\n",
       "death_da            0.943833\n",
       "lag_hrs             0.920259\n",
       "death_tm            0.901124\n",
       "death_hr            0.901080\n",
       "lag_mins            0.891601\n",
       "death_mn            0.838143\n",
       "work_inj            0.828424\n",
       "alc_res             0.424592\n",
       "persons             0.345023\n",
       "a_ct                0.337104\n",
       "veh_no              0.334827\n",
       "ve_forms            0.283374\n",
       "ve_total            0.282724\n",
       "a_mancol            0.212092\n",
       "a_rd                0.206076\n",
       "per_no              0.197930\n",
       "vnumber1            0.156433\n",
       "seat_pos            0.145476\n",
       "hospital            0.138576\n",
       "helm_mis            0.125556\n",
       "drug_det            0.123566\n",
       "a_posbac            0.121060\n",
       "alc_det             0.114681\n",
       "sex                 0.103914\n",
       "man_coll            0.097483\n",
       "tow_veh             0.093987\n",
       "peds                0.072816\n",
       "eventnum            0.068544\n",
       "a_ru                0.068477\n",
       "a_roll              0.058994\n",
       "age                 0.047806\n",
       "rur_urb             0.041842\n",
       "work_zone_binary    0.039742\n",
       "nhs                 0.037889\n",
       "wrk_zone            0.035363\n",
       "emer_use            0.034716\n",
       "a_mc                0.033027\n",
       "Name: inj_sev_binary, dtype: float64"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.set_option('display.max.columns', None)\n",
    "correlations2.head(40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "84866c47-5a32-4070-82be-d3d410f89c18",
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df = accidents_df.drop(columns=['tway_id', 'statename', 'weathername', 'year', 'aoi1name', 'soename', 'vnumber2name', 'aoi2name', 'rail', 'crashrfname'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6005e8d3-9dce-48ba-abde-0af2f060ffe2",
   "metadata": {},
   "source": [
    "# Let's check to see which features would be the best to use to predict fatality "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "789aaac9-8e81-4bc0-89e2-bd5f16bf6aa6",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_selection import SelectKBest\n",
    "from sklearn.feature_selection import chi2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "4b3dbfb0-79e2-417d-a6fd-f7fe681c2793",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['inj_sev', 'persons', 'peds', 've_total', 'drunk_dr', 'inj_sev_binary', 'lgt_cond_binary', 'work_zone_binary' ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "383a52c5-6e0d-4f5a-81a5-e8060e113b7a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              Specs         Score\n",
      "1           persons  26074.760656\n",
      "0           inj_sev   2927.346604\n",
      "3          ve_total   2886.277084\n",
      "2              peds   2084.111283\n",
      "5    inj_sev_binary    741.583297\n",
      "4          drunk_dr    157.526996\n",
      "7  work_zone_binary     14.987448\n",
      "6   lgt_cond_binary      0.110774\n"
     ]
    }
   ],
   "source": [
    "X = accidents_df[features]\n",
    "y = accidents_df['binary_fatals']\n",
    "\n",
    "bestfeatures = SelectKBest(score_func=chi2, k=2)\n",
    "fit = bestfeatures.fit(X,y)\n",
    "dfscores = pd.DataFrame(fit.scores_)\n",
    "dfcolumns = pd.DataFrame(X.columns)\n",
    "\n",
    "featureScores = pd.concat([dfcolumns, dfscores], axis=1)\n",
    "featureScores.columns = ['Specs', 'Score']\n",
    "print(featureScores.nlargest(10, 'Score'))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7300b13-e5f4-4ba4-81e1-cadb33656a5f",
   "metadata": {},
   "source": [
    "### Now we will try a Feature Importance Method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "43783716-4f92-475f-9f7e-07819f7d457e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.06360422 0.4350662  0.11512054 0.19147199 0.06976901 0.04817425\n",
      " 0.04929495 0.02749885]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import ExtraTreesRegressor\n",
    "\n",
    "plt.figure(figsize=(15,7))\n",
    "\n",
    "plt.xlabel('Totals', size = 20)\n",
    "plt.ylabel('Inputs', size = 30)\n",
    "plt.title('Graph of  Best Coefficients', size = 40)\n",
    "plt.xticks(size = 10)\n",
    "plt.yticks(size = 10)\n",
    "\n",
    "model = ExtraTreesRegressor()\n",
    "model.fit(X,y)\n",
    "print(model.feature_importances_)\n",
    "feat_importances = pd.Series(model.feature_importances_, index=X.columns)\n",
    "feat_importances.nlargest(10).plot(kind='barh')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "ed1ec96d-edc7-4621-a9a1-81a7c6e47ba2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#[['st_case', 'vnumber2', 'milept', 'crashrf', 'persons', 'man_coll', 'rd_owner', 'weather', 'a_weather', 'county', 'persons', 'hosp_hr', 'county', 'day', 'minute', \n",
    " #               'month', 'hour', 'latitude', 'peds']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a1a2cb1-50c0-4e28-942b-92a29dc2687e",
   "metadata": {},
   "source": [
    "#### Lets create our mini Databases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "21f9d08e-177d-48a9-97c7-62564d1d5383",
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import entropy\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "3e1be802-767c-4c86-815a-e47c36300ec2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# THis db is for the chi squared version\n",
    "\n",
    "X= accidents_df[features]\n",
    "\n",
    "y = accidents_df[['binary_fatals']]\n",
    "\n",
    "# drop state case from model\n",
    "# add workzone/ light condition / weather/ \n",
    "# make sure to clean data from columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "ec098180-8d4b-46aa-ab9d-9c1922722c2c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#split data into train test split\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = .2, random_state = 42, stratify=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "6b855bb5-ce78-4255-ad4f-e54e2f11d000",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "425ac7cb-2af0-4b73-91d4-c2752688f974",
   "metadata": {},
   "outputs": [],
   "source": [
    "sc_X = StandardScaler()\n",
    "\n",
    "X_train2 = pd.DataFrame(sc_X.fit_transform(X_train))\n",
    "\n",
    "X_test2 = pd.DataFrame(sc_X.transform(X_test)) #we do the feature scaling\n",
    "\n",
    "X_train2.columns = X_train.columns.values\n",
    "X_test2.columns = X_test.columns.values\n",
    "\n",
    "X_train2.index = X_train.index.values\n",
    "X_test2.index = X_test.index.values\n",
    "\n",
    "X_train = X_train2\n",
    "X_test = X_test2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "d06c6552-cd6a-4041-b5cc-2aa2d7976aef",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "forest = RandomForestRegressor(n_estimators=10, #criterion='crossentropy',\n",
    "                              random_state=0)\n",
    "tree = DecisionTreeRegressor(#criterion='crossentropy', \n",
    "                             random_state=0)\n",
    "bayes = GaussianNB()\n",
    "k_nn = KNeighborsRegressor(n_neighbors=5, metric='minkowski', p =2)\n",
    "log = LogisticRegression(random_state=0)\n",
    "\n",
    "models = [forest, tree, bayes, k_nn, log]\n",
    "scores = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "47a94ccc-0422-44fc-8446-b40cbf92a32a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr = LinearRegression()\n",
    "lr.fit(X_train,y_train)\n",
    "#try this as a regressor "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "46fa9f6b-6e10-46b3-bcd6-cfa7c1c76008",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.08743924024804095"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train score\n",
    "lr.score(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "56523be8-5c2e-49d6-9af5-d4f4e1cd9f24",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ricardo\\anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg = LogisticRegression()\n",
    "lg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "9f9f9810-dae4-4295-a52d-c850745d6035",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8498713877961849"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "fb91617f-5866-4a84-abc7-b0174c71ad9c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.45818437,  0.81304226,  0.29086116, -0.17238768,  0.11252729,\n",
       "        -0.16811878]])"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "448a29f8-6d6d-48f1-8871-c4c6d3f227cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-89-e0147d7dbc7c>:1: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  forest.fit(X_train, y_train)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(n_estimators=10, random_state=0)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "34970de3-e29f-4f0b-b080-ffacf085a0ee",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3107465905411846"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "f3c26926-cb41-43a5-8aab-50be3aee6287",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3106402357096866"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "73c9ab28-34fb-4767-a87a-ccc6d7e965c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "505534a8-973d-45e4-a6b5-8f26209a8db3",
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNeighborsClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "fad48edb-5384-404f-8ef7-8227652fc49d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(X_train, y_train.values.ravel())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "b5005ebd-2aec-4fb5-8734-faea1277f303",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8629708689711509"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "86559015-c68c-4b5e-8318-3765629f7f1e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8601964407274142"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "6ef84bef-802e-4546-ab01-6da924eda1a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "30b4dcf4-7e8c-472e-b05c-fa5b54290636",
   "metadata": {},
   "outputs": [],
   "source": [
    "rf = RandomForestClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "20c9e334-3085-4059-b2da-33ff2a2058af",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-99-168a6fc83696>:1: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  rf.fit(X_train, y_train)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier()"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "24219bcd-5b53-48c6-929f-e33920ba3082",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8771564718467373"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "4fecaa22-abc1-4d37-be43-539f756d7f4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df['binary_fatals'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "51e03248-13c5-4aa1-b0c6-f21b7aed9646",
   "metadata": {},
   "outputs": [],
   "source": [
    "#accidents_df['binary_fatals'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d780c602-fd57-46a7-bb80-1fb76c78bd57",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "cd2dab49-7338-4f2f-87da-53d5d1cd3e82",
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.model_selection import cross_val_score\n",
    "# from sklearn.model_selection import StratifiedKFold\n",
    "\n",
    "# kfold = StratifiedKFold(n_splits=10, shuffle=True, random_state=1)\n",
    "\n",
    "# for mod in models:\n",
    "#     mod.fit(X_train, y_train.values.ravel())\n",
    "#     acc = cross_val_score(mod, X_train, y_train.values.ravel(),\n",
    "#                          scoring = 'accuracy', cv=kfold, n_jobs=-1)\n",
    "#     scores.append(acc.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "898d8aac-3d27-46a4-a29f-427766fc1e55",
   "metadata": {},
   "outputs": [],
   "source": [
    "# results = pd.DataFrame({\n",
    "#     'Model': ['Random Forest', 'Decision Tree', 'Naive Bayes',\n",
    "#              'K Nearest Neighbor', 'Logistic Regression'],\n",
    "#     'Accuracy (Chi-Squared)': scores})\n",
    "\n",
    "# result_df = results.sort_values(by='Accuracy (Chi-Squared)',\n",
    "#                                ascending=False).reset_index(drop=True)\n",
    "# result_df = result_df.transpose()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "772e99bc-1e4e-4057-bc37-86a3d026a0fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "85133fe8-c86a-40df-911b-2e751fbf86d9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86d9d431-cbc6-4a9a-a77d-0d9535d9369d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "26a7c92a-d725-4a18-8713-675aeaffd231",
   "metadata": {},
   "source": [
    "# Lets run the models again on injury severity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "915035c2-79b9-44d3-a493-f4c0f6292840",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['persons', 'peds', 've_total', 'drunk_dr', 'lgt_cond_binary', 'work_zone_binary']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "9719288d-51e6-451e-b8f6-70fa0d82da3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# THis db is for the chi squared version\n",
    "\n",
    "X= accidents_df[features]\n",
    "\n",
    "y = accidents_df[['inj_sev_binary']]\n",
    "\n",
    "# drop state case from model\n",
    "# add workzone/ light condition / weather/ \n",
    "# make sure to clean data from columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "c2af8065-b8e1-45d1-ad95-bf0fb1a33436",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              Specs         Score\n",
      "0           persons  76386.771434\n",
      "2          ve_total  22767.693429\n",
      "1              peds   3093.861243\n",
      "3          drunk_dr    920.511683\n",
      "4   lgt_cond_binary    426.749213\n",
      "5  work_zone_binary    396.459942\n"
     ]
    }
   ],
   "source": [
    "X = accidents_df[features]\n",
    "y = accidents_df['inj_sev_binary']\n",
    "\n",
    "bestfeatures = SelectKBest(score_func=chi2, k=2)\n",
    "fit = bestfeatures.fit(X,y)\n",
    "dfscores = pd.DataFrame(fit.scores_)\n",
    "dfcolumns = pd.DataFrame(X.columns)\n",
    "\n",
    "featureScores = pd.concat([dfcolumns, dfscores], axis=1)\n",
    "featureScores.columns = ['Specs', 'Score']\n",
    "print(featureScores.nlargest(10, 'Score'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "68d5f5c3-0063-44f2-acfd-0d082b47bda5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = .2, random_state = 42, stratify=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "86b5d8f4-c918-4cf4-aa16-30a458398437",
   "metadata": {},
   "outputs": [],
   "source": [
    "sc_X = StandardScaler()\n",
    "\n",
    "X_train2 = pd.DataFrame(sc_X.fit_transform(X_train))\n",
    "\n",
    "X_test2 = pd.DataFrame(sc_X.transform(X_test)) #we do the feature scaling\n",
    "\n",
    "X_train2.columns = X_train.columns.values\n",
    "X_test2.columns = X_test.columns.values\n",
    "\n",
    "X_train2.index = X_train.index.values\n",
    "X_test2.index = X_test.index.values\n",
    "\n",
    "X_train = X_train2\n",
    "X_test = X_test2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "dbbbb966-2c5a-4be5-9b57-f6059bd5fd95",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.55038046 0.12645585 0.30519365 0.01039741 0.00579426 0.00177838]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.figure(figsize=(15,7))\n",
    "\n",
    "plt.xlabel('Totals', size = 20)\n",
    "plt.ylabel('Inputs', size = 30)\n",
    "plt.title('Graph of  Best Coefficients', size = 40)\n",
    "plt.xticks(size = 10)\n",
    "plt.yticks(size = 10)\n",
    "\n",
    "model = ExtraTreesRegressor()\n",
    "model.fit(X,y)\n",
    "print(model.feature_importances_)\n",
    "feat_importances = pd.Series(model.feature_importances_, index=X.columns)\n",
    "feat_importances.nlargest(10).plot(kind='barh')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "1459a692-f289-49a4-9c3b-06e9d355f97c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg = LogisticRegression()\n",
    "lg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "7c611141-d9c9-4547-ab68-8de577ee8b8a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6923858657862363"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "e7d8da76-c079-46f3-b791-d9d7c6fcae94",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.20703697,  0.34558234,  0.16227215, -0.06590143, -0.01986731,\n",
       "        -0.0045333 ]])"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "c7bc2075-adea-4329-b149-c5c35b3a6b38",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6936886122726831"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "8cd8bd6e-a464-4de3-b693-0e856331943f",
   "metadata": {},
   "outputs": [],
   "source": [
    "rf = RandomForestClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "3a0798c4-52b9-427b-acc9-fa5e2fda181a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier()"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "e7a5a9c5-1040-41c6-bb66-82bc7ac1d9ee",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.711251580278129"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "42759c56-4b73-41f7-9f7e-eb003a30fd23",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7124727094142188"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ceb1ece2-9a10-4eeb-b7ed-3805a1c5a135",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "e00e3d48-5d32-4499-bb45-feeb6d5b51df",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate predictions.\n",
    "preds = lg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b4ffab9b-7298-4eec-a919-604b1ba2fb92",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b4326d84-ece8-482d-9e01-aa43be411d81",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e67fb754-1426-4a80-8633-92d320a3c485",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "6274e68e-3ba8-4d72-ab25-75f7aa7d3482",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(51415,)"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "90ddab10-e767-4b40-a32b-1504233c9ba5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(51415, 6)"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "58a913f7-2cf7-442b-90f5-2ba6b54f5688",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix, plot_confusion_matrix, accuracy_score, plot_roc_curve, roc_auc_score, precision_score, f1_score, recall_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "02da622f-ab35-4d4a-88b5-405f628d10ce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[15933,  7161],\n",
       "       [ 8588, 19733]], dtype=int64)"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_test,preds)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "id": "ceeb900a-a0ed-4d85-ab93-08d68136e5bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEHCAYAAACEKcAKAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAqu0lEQVR4nO3deZxVdf3H8dd7hl0EBEQRXBNUwERBMxdEyXD5GWaoqKWlZhlp2c9KzNz6WZpZP7VcyAU1FwTMfV8ISwEBIRAXMFBQfyqCiiLb8Pn9cc7gZZy5c2fu3Dszl/eTx3nMud/zPd/zvXN53M98l/M9igjMzMzqq6yxK2BmZs2bA4mZmeXFgcTMzPLiQGJmZnlxIDEzs7y0aOwKWP21aNcxWnXasrGrYXWwS/dNG7sKVkczZkxfEhGb51NGeYdtI9Z+llPe+Oz9xyLikHyuV2wOJM1Yq05b0uv71zR2NawO/vXrIY1dBaujti31Rr5lxNrPaL3TMTnlXTnzL13zvV6xOZCYmRWcQKU7kuBAYmZWaALKyhu7FgXjQGJmVgxSY9egYBxIzMwKzl1bZmaWL7dIzMys3oRbJGZmlg+5RWJmZnnyrC0zM6s/D7abmVk+hLu2zMwsT26RmJlZ/blry8zM8lXmri0zM6svr7VlZmb5cdeWmZnly7O2zMwsL26RmJlZvclLpJiZWb482G5mZvXnwXYzM8uXu7bMzKze/DwSMzPLj7u2zMwsX+7aMjOzvJTwrK3SbWuZmTUVSru2ctlqLUo3SXpP0pyMtP6SJkuaKWmapL0yjo2SNF/Sq5KGZqQPkDQ7PXaVlDSZJLWWNDZNnyJpu9rq5EBiZlYMlTcl1rbVbgxwSJW03wMXRUR/4Pz0NZL6ACOAvuk510iqbBpdC5wG9Eq3yjJPAZZFxI7An4DLaquQA4mZWRFIymmrTURMApZWTQY6pPsdgbfT/WHAXRGxKiIWAPOBvSR1BzpExPMREcCtwJEZ59yS7o8HhqiWinmMxMyswJIn7eY82N5V0rSM16MjYnQt5/wUeEzSH0gaCPuk6T2AyRn5Fqdpa9L9qumV5ywCiIi1kj4CugBLarq4A4mZWaEp3XKzJCIG1vEKpwNnRcQESccANwJfq+GqkSWdWo5Vy11bZmYFJ8rKynLa6ukk4J50fxxQOdi+GNg6I19Pkm6vxel+1fQNzpHUgqSrrGpX2gYcSMzMiqChxkhq8DZwQLp/EDAv3b8fGJHOxNqeZFB9akS8AyyXtHc6/nEicF/GOSel+8OBp9NxlBq5a8vMrAjyCBJVy7kTGEwylrIYuAD4PnBl2oJYSTIbi4h4SdLdwFxgLTAyIirSok4nmQHWFngk3SDpFrtN0nySlsiI2urkQGJmVmh1GyPJKiKOq+HQgBryXwJcUk36NKBfNekrgaPrUicHEjOzAhN5dVs1eQ4kZmZFkMdAepPnQGJmVgRukZiZWf014BhJU+RAYmZWBG6RmJlZvXmw3czM8uZAYmZm9SdQmQOJmZnlwS0SMzPLiwOJmZnVmwfbzcwsf6UbRxxIzMwKTu7aMjOzPHmtLTMzy0/pNkgcSKxwzjtiF/bt1ZVln67m+OunAHDqoO0ZtvtWfLhiDQDXPvM6z83/gBZlYtThO7PzVh2ICP742GvMeONDAP73uP503bQV5WVi5psfcvkjr7Iu4Jt79GD4nj1Zty74bHUFv3voFRYs+bSx3m7JmbfwXU4+96b1r994+wNGnXY43bt15LLRD/Pqwnd5aszZ7N5n2/V55sx7i5/97k6Wf7ISlYmnb/kFbVq35DfX3M9dD03lo+UrWDzpj43xdhqdu7aaCEnPRcQ+teS5AfhjRMwtUrUqr7szcBcQwPCIeL2GfOdGxG9zKG8hMDAiljRoRYvowVnvMO6FxVwwrM8G6XdNWcTtk9/cIO3IPXoAcML1U9isXUv+9/j+fPeGFwjgVxNm8+nq5KFulw7flSF9tuCJl97l8Tn/x99nvAXA/r278pODe/HTO2cW/H1tLHpttwXP3jEKgIqKdfQ57FccfuBufLZyNbf+/vuc9bs7N8i/dm0FPzj/Fq676ER27d2TpR9+QssW5QAcsv+ufP+YAxh41EVFfx9NQZ6P0W3ymlUgqS2IpHlOLUZdqnEkcF9EXFBLvnOBWgNJKZj55od079gmp7zbd92EFxYuA2DZijUsX7mWXbbqwNy3P14fRMrLRItyUfn46Mp0gLYtywmyPlba8vCPF15lu56bs033zjXmeXrKK/TdsQe79u4JQOdO7dcf23PX7Qtex6aulANJsxr9kfRJ+nOwpImSxkt6RdLt6QPsSdMH1nB+uaQxkuZImi3prDT9S5IelTRd0rOSdpbUUdJCSWVpnnaSFklqWU25hwE/BU6V9Eyadm9a3kuSTkvTLgXaSpop6faa8pW64Xv25G+n7cV5R+zCpm2Sv2XmvbucQb27Ui7RvVMbdu6+KVt0aL3+nCuP78+jP9ufFasrePrl9z4va2BPJoz8Kj8esiN/fOy1or+XjcU9j0/nW0OrfZLreq+/8R4SfOuMP3PAty/lylufKFLtmgeVKaetOWpWLZIqdgf6Am8D/wL2Bf5Zyzn9gR4R0Q9AUqc0fTTww4iYJ+krwDURcZCkWcABwDPAEcBjEbGmaqER8bCk64BPIuIPafLJEbFUUlvgBUkTIuIcST+OiP4Zp1eX74Oa3kAabE4DaNmxWy1vt+m5Z/pb3PTsAiLgBwfuwE8O7sX/PPAyD8x8h+26bsKYU/fk/z5ayexFH1Gx7vMWxk/umEmr8jIu/mZfBm7XmakLlgIwftpixk9bzNf7bcH39tuei+8vao/mRmH1mrU8Mmk254/8RtZ8aysqmDzrPzx9y89p26YVR/7oKvrvvA0H7LVTkWratLlF0jRNjYjFEbEOmAlsl8M5/wF2kHS1pEOAjyW1B/YBxkmaCVwPdE/zjwWOTfdHpK9zdWYaiCYDWwO98swHQESMjoiBETGwRbtOdahO07D009Wsi2Qg6b4Zb9Nnqw4AVETwv0/M4zt/ncrP7/437du0YNHSzzY4d3XFOia99j6Ddur6hXKfmPMuB+y0eTHewkbnyefmstvOW9OtS4es+bbaohP77r4jXTq1p12bVhy8T19mvbqoSLVs4vT5OEltW3PUnAPJqoz9CnJoXUXEMmA3YCIwEriB5HfwYUT0z9h2SU+5HzhUUmdgAPB0LhWTNBj4GvDViNgNeBH4wmBBrvlKSZf2rdbvH7Dz5vzn/WSWVesWZbRpmfx33Gv7zlSsCxYs+ZS2LcvXn1Musc+OXVm4ZAUAW3duu76sfXt1ZdHSFcV6GxuV8Y9N41tfz96tBTBk7z68NP8tVqxczdq1Ffxrxnx22n7LItSw6RMg5bY1R825a6vOJHUFVkfEBEmvA2Mi4mNJCyQdHRHj0rGWL0fErIj4RNJU4ErgwYioyHqBz3UElkXEinQ2194Zx9ZIapl2kWXL1+z95pt92WPbzejUriUP/GRfRv/jPwzYdjN6bbkpEcE7H63k0odeAaDzJq248oT+rAt4/+NVXHhf0kXVtlU5fzh2N1qWi/IyMW3BMv4+PZmpdfTArdlzh81YWxEsX7mWi9yt1eBWrFzNxKmv8Kdzj1uf9uAzs/jlH8axZNknHHvWdezauwcTrv4xnTq040fHH8SQE38PEgfv25eh+/UD4Pyr7mXCY9NYsXINfQ8/j+8M+yrnnHZ4Y72tRtB8Wxu5UOUMmOZA0icR0T79S/7siPivNP3PwLSIGCNpIvDfETG9mvN3A27m85bYqIh4RNL2wLUkXVotgbsi4uL0nOHAOGBwRPwjS90uJB0jkdQauBfoAbwKbA5cGBETJV0GfAOYAZycJd9Capn+226rnaLX96/J/kuzJmXKr4c0dhWsjtq21PSIqHYCT67abNk7tjnxqpzyzrv80LyvV2zNqkUSEe3TnxNJuqcq03+cka0LsLSG82cBe1STvgA4pIZzxpPDPakRcWHG/irg0Bry/RL4ZUZSTfm2q+2aZtZMCMqa6YysXDSrQFIbSU8As9PAYGbWJIjSDiTNebD9CyLi4Ig4HkDSlPR+jcxt13yvIekv1ZT7vfxrb2alrKEG2yXdJOk9SXOqpJ8h6dX0nrTfZ6SPkjQ/PTY0I31Aej/dfElXZdyL11rS2DR9iqTtaqtTSbVIMkXEVwpU7shClGtmpa0BB9vHAH8Gbs0o+0BgGMlEoVWSuqXpfUhuXegLbAU8Kal3OnHoWpJ70iYDD5N07z8CnEIyCWhHSSOAy/j8NohqlVSLxMysScqxNZJLrImISXxxHPh04NJ0fJaIqFz+YRjJ5KFVaZf/fGAvSd2BDhHxfCQzrm4lWeap8pxb0v3xwBDVEgUdSMzMCiy5jyTnGxK7SpqWseWydFJvYP+0K+ofkvZM03sAmXeFLk7TeqT7VdM3OCci1gIfkUxiqlHJdm2ZmTUdqstg+5J6TP9tAWxGci/ansDdknag+hmnkSWdWo5Vyy0SM7MiKPASKYuBeyIxFVgHdE3Tt87I15NkfcLF6X7VdDLPkdSC5Mbpam+pqORAYmZWaA04RlKDe4GDACT1BloBS0iWeRqRzsTanmQtv6kR8Q6wXNLe6fjHicB9aVn3Ayel+8OBp6OWO9fdtWVmVmCVYyQNUpZ0JzCYZCxlMXABcBNwUzoleDVwUvrl/5Kku4G5wFpgZMZST6eTzABrSzJb65E0/UbgNknzSVoiI2qrkwOJmVkRNNTs34g4roZD364h/yXAJdWkTwP6VZO+Eji6LnVyIDEzK4JSvrPdgcTMrNBU2g+2ciAxMyuwyueRlCoHEjOzgivt55E4kJiZFUEJxxEHEjOzYnCLxMzM6k1+sJWZmeXLLRIzM8tLCccRBxIzs2Jwi8TMzOovvwUZmzwHEjOzAlPdnkfS7DiQmJkVQVkJN0kcSMzMiqCE44gDiZlZocmLNpqZWb5KeIik5kAi6WqyPPA9Is4sSI3MzErQxtoimVa0WpiZlTCxkQ62R8Qtma8lbRIRnxa+SmZmpaeUu7bKassg6auS5gIvp693k3RNwWtmZlYqlDyPJJetOao1kAD/CwwFPgCIiFnAoALWycys5Ei5bc1RTrO2ImJRlUhZUZjqmJmVno12jCTDIkn7ACGpFXAmaTeXmZnlpoTjSE6B5IfAlUAP4C3gMWBkIStlZlZKNvoHW0XEEuCEItTFzKxklXLXVi6ztnaQ9ICk9yW9J+k+STsUo3JmZqVCOW61liPdlH4Xz6nm2NmSQlLXjLRRkuZLelXS0Iz0AZJmp8euUjoQLqm1pLFp+hRJ29VWp1xmbd0B3A10B7YCxgF35nCemZmlGnD67xjgkGrK3xo4GHgzI60PMALom55zjaTy9PC1wGlAr3SrLPMUYFlE7Aj8CbistgrlEkgUEbdFxNp0+xtZlk4xM7MNJbO2cttqExGTgKXVHPoT8As2/H4eBtwVEasiYgEwH9hLUnegQ0Q8HxEB3AocmXFO5Q3p44EhqiXCZVtrq3O6+4ykc4C70goeCzyUrVAzM8ugwj7YStI3gLciYlaV7/wewOSM14vTtDXpftX0ynMWAUTEWkkfAV2AJTVdP9tg+3SSwFFZqx9kHAvgN1nONTOzDHW4a72rpMy1DkdHxOgs5bYDfgV8vbrD1aRFlvRs59Qo21pb22c70czMclPZtZWjJRExsA7FfwnYHqhsjfQEZkjai6SlsXVG3p7A22l6z2rSyThnsaQWQEeq70pbL6c72yX1A/oAbSrTIuLWXM41M7PCLSMfEbOBbhnXWQgMjIglku4H7pD0R5LJUr2AqRFRIWm5pL2BKcCJwNVpEfcDJwHPA8OBp9NxlBrVGkgkXQAMJgkkDwOHAv8kGZwxM7McNFQYkXQnyXdyV0mLgQsi4sbq8kbES5LuBuYCa4GREVG5xNXpJDPA2gKPpBvAjcBtkuaTtERG1FanXFokw4HdgBcj4nuStgBuyOE8MzMjvbO9gVokEXFcLce3q/L6EuCSavJNA/pVk74SOLoudcolkHwWEeskrZXUAXgP8A2JZmZ1sFEvkQJMk9QJ+CvJTK5PgKmFrJSZWakp4RVSclpr60fp7nWSHiW5ieXfha2WmVnpECrptbay3ZC4R7ZjETGjMFUyMysxzfihVbnI1iK5IsuxAA5q4LpYHe3YrT33n7lvY1fD6mCzPX/c2FWwRtJcH6Obi2w3JB5YzIqYmZWyXBY2bK5yuiHRzMzqT0D5Rj5ry8zM8lTCccSBxMys0KTSHiPJ5QmJkvRtSeenr7dJFwMzM7McNdTzSJqiXMZ/rgG+ClTelr8c+EvBamRmVoKk3LbmKJeura9ExB6SXgSIiGWSWhW4XmZmJUNAi+YaJXKQSyBZkz7jNwAkbQ6sK2itzMxKTAnHkZwCyVXA34Fuki4hWQ34vILWysyshEgb6RIplSLidknTgSEkLbQjI+LlgtfMzKyElHAcyenBVtsAK4AHMtMi4s1CVszMrJQ01xlZucila+shPn9YfBuSZwO/CvQtYL3MzEpG8sz20o0kuXRt7Zr5Ol0V+AcFq5GZWakRlJfwYlt1vrM9ImZI2rMQlTEzK1VqsKe2Nz25jJH8LONlGbAH8H7BamRmVmKSrq3GrkXh5NIi2TRjfy3JmMmEwlTHzKw0bbSBJL0RsX1E/LxI9TEzK0mlvGhjtkfttoiItdkeuWtmZrXbmLu2ppKMh8yUdD8wDvi08mBE3FPgupmZlQb5wVadgQ9IntFeeT9JAA4kZmY52JhbJN3SGVtz+DyAVIqC1srMrMSU8BBJ1ueRlAPt023TjP3KzczMciLKctxqLUm6SdJ7kuZkpF0u6RVJ/5b0d0mdMo6NkjRf0quShmakD5A0Oz12ldLZAJJaSxqbpk+RtF1tdcrWInknIi6u9V2ZmVlWokFbJGOAPwO3ZqQ9AYxKJ0hdBowCfimpDzCCZEmrrYAnJfWOiArgWuA0YDLwMHAI8AhwCrAsInaUNAK4DDg2W4WytUhKuCFmZlZEghZlymmrTURMApZWSXs8ItamLycDPdP9YcBdEbEqIhYA84G9JHUHOkTE8xERJEHpyIxzbkn3xwNDKlsrNckWSIbU+o7MzKxWlS2SHB+121XStIzttDpe7mSSlgVAD2BRxrHFaVqPdL9q+gbnpMHpI6BLtgvW2LUVEUtrOmZmZnVTh9V/l0TEwPpcQ9KvSFYgub0yqZpsVSdPZaZnO6dGdV600czM6q7Qs7YknQT8FzAk7a6CpKWxdUa2nsDbaXrPatIzz1ksqQXQkSpdaVWV8MLGZmZNg0i+bHPZ6lW+dAjwS+AbEbEi49D9wIh0Jtb2QC9gakS8AyyXtHc6/nEicF/GOSel+8OBpzMCU7XcIjEzKzQ13Fpbku4EBpOMpSwGLiCZpdUaeCK9zuSI+GFEvCTpbmAuSZfXyHTGFsDpJDPA2pKMqVSOq9wI3CZpPklLZERtdXIgMTMrMAHlDRRIIuK4apJvzJL/EuCSatKnAf2qSV8JHF2XOjmQmJkVQSnfT+FAYmZWBKW8RIoDiZlZwWnjfB6JmZk1jMpZW6XKgcTMrAjqcENis+NAYmZWaA04/bcpciAxMyswd22ZmVne3CIxM7O8lG4YcSAxMyuKEm6QOJCYmRVaQy6R0hQ5kJiZFZxQCXduOZCYmRVBCTdIHEjMzAotmf5bupHEgcTMrNDkFomZmeXJgcTMzOrNs7bMzCxvnrVlZmZ5KeEGiQOJFc+Y8ZMY/8gUJOi1fXd+9/NjGX3n04x7eAqdO7UH4KyTD+WAr+zCmrUVnHfF3cyd9xYV69Yx7GsD+MHxQwB48OkXuf6Op5BEty4duHzU8WzWcZPGfGsl4+pfn8DQ/fqxZNly9hnxWwD69erBFeeMoH271rz5zgec9utbWP7pSo4+ZCBnfOdr68/tu+NWHPCdy5jz2luMu+pHbNmlA+Utypn84uuc/fuxrFsXfO+o/Tj16EFUrFvHpytW8dPf3smrC/6vsd5uUblFYpand5d8xG33PstDN/6CNq1b8tOLb+WhZ2YCcNK3BnHKMYM3yP/oP2axZk0FD9xwNp+tXM3hp1zO4Qftzpabd+S319zLQzf+gs06bsLlox/kb/f+kzNOGlr8N1WC7nxwMn+9+x9cd9GJ69OuPO94fn3l33luxnxOOGJvzvjOEH573UOMe3Qa4x6dBkCfL23F7VecxpzX3gLg5FE3sfzTlQDcctmpHDlkD+55YjrjH5vGzff8E4BDB+3K/5x1FEefeU2R32XxCSgr3ThSuJWNJT2XQ54bJPUpVB0aiqT+kg6rJU9rSU9Kminp2Cz5vitpqxyuOUbS8PrUt6mqqFjHylVrWFtRwWer1tCtS4ca80pixcpVrK2oYOWqNbRsUU77dm2IgAhYsXIVEcEnK1bSrUvHIr6L0vbci6+z7OMVG6TtuE03npsxH4CJU1/hiAP7f+G8bw0dwITHpq9/XRlEWpSX0aplOUFskA7Qrk2r5MPcGEiU5bg1RwVrkUTEPjnkObVQ129g/YGBwMNZ8uwOtIyI/rWU9V1gDvB2Q1Ssudiia0dOPnowBx3/P7Ru3ZJ9B/Rmv4E78eJLC7n9vn9x3xPT6de7J7/84RF03LQdQwd9maefe4n9j7mYlatWc84Ph9GpQzsALvjJUXzj+1fQrk0rtu3RlfPPOKqR311pe+U/73DooF15ZNJshg3Zgx5bbPaFPN88eA9OOHv0BmnjrxrJgL7b8uRzc7nvqRfXp5969CB+dPyBtGrZgm+cflXB699UNM8QkZtCtkg+SX8OljRR0nhJr0i6XenC/Gn6wBrOL0//Kp8jabaks9L0L0l6VNJ0Sc9K2llSR0kLJZWledpJWiSpZQ1lT5R0maSpkl6TtH+a3kbSzen1XpR0oKRWwMXAsTW1NiR1A/4G9E/zfEnS+ZJeSOs/WonhJAHp9jRf2+ry5fu7b4o+Wr6Cp56bw5N/O5dJY8/ns5Wruf/J6Rz3jX144tZR3Hv9WWzepQOXXfcAALNfeZOyMjFp7Pk8edu53Dz+Hyx6+wPWrK3grgee5+/XncWksefTe4fujL7z6UZ+d6XtxxffzqlHD+KZW39B+3atWbOmYoPjA/puy2cr1/Dy6+9skD78zL+w86Hn0qpVCwYN3Gl9+g3jJrHHNy/iwqvv4+yTDynKe2hsSddW6bZIivXQrt2BnwJ9gB2AfXM4pz/QIyL6RcSuwM1p+mjgjIgYAJwNXBMRHwGzgAPSPEcAj0XEmizlt4iIvdJ6XZCmjQRIr3cccAvJ7+h8YGxE9I+IsVULioj3gFOBZ9M8rwN/jog9I6If0Bb4r4gYD0wDTkjzfVZdvmy/FEmnSZomadrSD97PlrVJeX7GPHpu2YXOndrTskU5B++3Ky++tJCum21KeXkZZWVlHH3YV5j96ptAMqC+/54707JFOV0225Q9+m7HnNcW8cr8pA9+m626IolDD9iNF+cubMR3VvrmvfEu3zrjLxx44u+Z8Ph0Fry14f+7o74+gAmPTav23FWr1/LIpNkcdsCuXzg24fHpHD74ywWpc1OkHLfmqFiBZGpELI6IdcBMYLsczvkPsIOkqyUdAnwsqT2wDzBO0kzgeqB7mn8sUNlaGJG+zuae9Of0jPrsB9wGEBGvAG8AvXOoa3UOlDRF0mzgIKBvnvlI6zU6IgZGxMDOXTavZ9WKr3u3Tsx6+Q0+W7maiOD5F+exwzZb8N4HH6/P8+Q/59Bru+7r80+eOY+IYMVnq5j18hvssE03unXtyOtvvMvSDz8B4Lnp89hhm26N8p42Fl03S2bUSeLsk4dy84R/rj8miWFDdmfCE5+Pj2zSthVbpONf5eVlHLxvH+YtfBeAHbb+/P/s0P368vqbzeePobyVcCQp1qytVRn7FblcNyKWSdoNGErSUjiGpPXwYQ3jEPcDv5PUGRgA1NbfUVmnzPo0yMcoqQ1wDTAwIhZJuhBoU998pWC3Xbbl64O+zFGn/4kW5WXssmMPjj18b8774928PP9tJNFjy8246KfJ/ILjh+3LuZeP5YhT/0BEcNTQPdlph2SOwsjvHMy3f3YNLcrL2WqLTvzu5yMa862VlBv+57vsO6AXXTq1Z86Dv+HS0Q+zSbvWnDp8EAAPTpzJ7Q9MXp9/n9135O33PuSNtz5Yn9aubWvu+OMPaN2yBWXlZTz7wmvclM7U+v4xgzhgr51Zu7aCDz9ewY8uurW4b7ARNdT0X0k3kfRcvJf2ZJB+740l+aN4IXBMRCxLj40CTiH5rjszIh5L0wcAY0h6Qh4GfhIRIak1cCvJ9+gHwLERsTBbnZrs9F9JXYHVETFB0uvAmIj4WNICSUdHxLh0POHLETErIj6RNBW4EngwIiqyXqB6k4ATgKcl9Qa2AV4FegGb1qGcymCwJG1FDQfGp2nLM8rKlq/knHnSUM6sMk339+ccX23eTdq25srzT6z22Igj9mHEEbXO5bB6OPW8MdWmX3/XxGrT/zVjHl8/+YoN0t5fupwhJ11ebf5RV0zIp3rNWgNO/x0D/Jnky77SOcBTEXGppHPS179MZ8WOIOnp2Ap4UlLv9PvxWuA0YDJJIDkEeIQk6CyLiB0ljQAu4/PenurfW4O9tfqraf5fD2Bi2oU1BhiVpp8AnCJpFvASMCzjnLHAt6m9W6sm1wDlaTfTWOC7EbEKeAboU9vU3koR8SHwV2A2cC/wQsbhMcB16ftalSWfmZWSBuraiohJwNIqycNIxnRJfx6ZkX5XRKyKiAXAfGAvSd2BDhHxfEQESVA6spqyxgNDapsEVMjpv+3TnxOBiRnpP87I1oUv/kIq880C9qgmfQFJ5KzunPHk8FFExOCM/SWkYyQRsZJkem7V/EuBPWspcyIbvs/zgPOqyTcByPyzrKZ8X6iHmTVPSYzIuUnSVVLm7IXRETG6xtyJLSLiHYCIeCedSQrJH+STM/ItTtPWpPtV0yvPWZSWtVbSRyTf1UtqunijdW1JegKYnQYGM7PSVbfnkSyJiGpvi6jflb8gsqRnO6dGjRZIIuLgyn1JU4DWVbJ8JyJm53MNSX/hi1ONr4yIm6vLn2OZ3wN+UiX5XxExsr5lmlnpK/CErHcldU9bI92B99L0xcDWGfl6ktwMvTjdr5qeec5iSS2AjtTQc1SpSQy2R8RXClRug3+5p0Go3oHIzDZShY0k9wMnAZemP+/LSL9D0h9JBtt7kdyOUSFpuaS9gSnAicDVVcp6nmQC0NPpOEqNmkQgMTMrbQ1317qkO4HBJGMpi0luqL4UuFvSKcCbwNEAEfGSpLuBucBaYGTGjNbT+Xz67yPpBnAjcJuk+SQtkVrn1zuQmJkVWEPeaxgRx9VwaEgN+S8BLqkmfRrQr5r0laSBKFcOJGZmxdBM71rPhQOJmVkR+MFWZmaWl2a6sG9OHEjMzAqtbveRNDsOJGZmReCuLTMzqzfhFomZmeWphOOIA4mZWVGUcCRxIDEzKwKPkZiZWV4a8MFWTY4DiZlZMTiQmJlZfdXxwVbNjgOJmVmh+YZEMzPLVwnHEQcSM7OiKOFI4kBiZlZwDfdgq6bIgcTMrMAa8sFWTZEDiZlZMZRwJHEgMTMrAk//NTOzvJTwEIkDiZlZwclLpJiZWd5KN5I4kJiZFZgfbGVmZnkr4TjiQGJmVgyl3CIpa+wKmJltDJTjv1rLkc6S9JKkOZLulNRGUmdJT0ial/7cLCP/KEnzJb0qaWhG+gBJs9NjV0n1D3UOJGZmRSDltmUvQz2AM4GBEdEPKAdGAOcAT0VEL+Cp9DWS+qTH+wKHANdIKk+LuxY4DeiVbofU9705kJiZFViuQSTHNkELoK2kFkA74G1gGHBLevwW4Mh0fxhwV0SsiogFwHxgL0ndgQ4R8XxEBHBrxjl15kBiZlYEdeja6ippWsZ2WmUZEfEW8AfgTeAd4KOIeBzYIiLeSfO8A3RLT+kBLMqoxuI0rUe6XzW9XjzYbmZWDLmPQCyJiIHVFpGMfQwDtgc+BMZJ+nYdrxpZ0uvFLRIzsyJQjlstvgYsiIj3I2INcA+wD/Bu2l1F+vO9NP9iYOuM83uSdIUtTverpteLA4mZWRE00BjJm8Dektqls6yGAC8D9wMnpXlOAu5L9+8HRkhqLWl7kkH1qWn313JJe6flnJhxTp25a8vMrMDUQA+2iogpksYDM4C1wIvAaKA9cLekU0iCzdFp/pck3Q3MTfOPjIiKtLjTgTFAW+CRdKsXBxIzs2YkIi4ALqiSvIqkdVJd/kuAS6pJnwb0a4g6OZCYmRVBKd/Z7kBiZlYEfrCVmZnVX+43GzZLDiRmZgXmZeTNzCxv7toyM7O8uEViZmZ5KeE44kBiZlYUJRxJHEjMzIqglMdIlCxFb82RpPeBNxq7HgXSFVjS2JWwOinVz2zbiNg8nwIkPUry+8nFkoio90OmGoMDiTVJkqbVtJS2NU3+zDZeXv3XzMzy4kBiZmZ5cSCxpmp0Y1fA6syf2UbKYyRmZpYXt0jMzCwvDiRmZpYXBxIzM8uLA4llJem5HPLcIKlPMepT5bo7S5op6UVJX8qS79wcy1soKdebxoqqKX8OdSWpv6TDasnTWtKT6ed7bJZ835W0VQ7XHCNpeH3qa7VzILGsImKfHPKcGhFzi1GfKo4E7ouI3SPi9Sz5cgokTVkT/xzqqj+QNZAAuwMtI6J/RIzNku+7QK2BxArLgcSykvRJ+nOwpImSxkt6RdLtUrIwdppe7R3NksrTvwbnSJot6aw0/UuSHpU0XdKzaeuiY9oqKEvztJO0SFLLaso9DPgpcKqkZ9K0e9PyXpJ0Wpp2KdA2/cv29pryNXVN9XPIuO5lkqZKek3S/ml6G0k3p9d7UdKBkloBFwPH1tTakNQN+BvQP83zJUnnS3ohrf9oJYYDA4Hb03xtq8uX7+/echAR3rzVuAGfpD8HAx8BPUn+AHke2C89NhEYWMP5A4AnMl53Sn8+BfRK978CPJ3u3wccmO4fC9yQpW4XAmdnvO6c/mwLzAG6ZL6HHPItBLo29u+8GX4OE4Er0v3DgCfT/f8Gbk73dwbeBNqQtCL+XMv7HQw8WPUzS/dvA46o7j1nyTcGGN7Yn2Opbm6RWF1MjYjFEbEOmAlsl8M5/wF2kHS1pEOAjyW1B/YBxkmaCVwPdE/zjyX54gIYkb7O1ZmSZgGTga2BXnnma6qa4udwT/pzekZ99iP5MiciXiFZYLR3DnWtzoGSpkiaDRwE9M0znzUgLyNvdbEqY7+CHP7/RMQySbsBQ4GRwDEkXVIfRkT/ak65H/idpM4kf0U/nUvFJA0GvgZ8NSJWSJpI8tdvvfI1cU3xc6isU2Z9GqRbSVIb4BqSlsciSRdS/WebUz5reG6RWEEpmQVVFhETgF8De0TEx8ACSUeneZR+yRERnwBTgStJujYqcrxUR2BZGhx2BvbOOLYmo38/W76SVcTPIdMk4IS07N7ANsCrwHJg0zqUUxkMlqStqMzZV5llZctnBeRAYg2lprV2egAT066TMcCoNP0E4JS0i+klYFjGOWOBb1O3bq1HgRaS/g38hqTbqtJo4N/pYHu2fKWgsT+HTNcA5Wk301jguxGxCngG6FPTYHtVEfEh8FdgNnAv8ELG4THAden7WpUlnxWQ19qyvKVfFN+IiAWNXZeNmT8HayxukVheJD0BzPaXV+Py52CNyS0SazCSpgCtqyR/JyJm51nuX4B9qyRfGRE351NuqWpOn4Ok7wE/qZL8r4gYWd8yrfgcSMzMLC/u2jIzs7w4kJiZWV4cSKykSapIp5nOkTROUrs8ylq/gqxqWWlXyZpYtS60WM151a5AXFN6lTyf1PFaF0o6u651NKvKgcRK3WeRrCDbD1gN/DDzoKTy+hQata+0O5hk+RGzkudAYhuTZ4Ed09bCM5LuAGYrWRn38nTV2H9L+gGsv9P7z5LmSnoI6FZZkDJW2pV0iKQZkmZJekrSdiQB66y0NbS/pM0lTUiv8YKkfdNzu0h6XMnquNeTw7IiyrJ6saQr0ro8JWnzNO0LK/w2yG/TLOW1tmyjIKkFcCjJne0AewH9ImJB+mX8UUTsKak18C9Jj5M8E2MnYFdgC2AucFOVcjcnuZt6UFpW54hYKuk6khV7/5DmuwP4U0T8U9I2wGPALsAFwD8j4mJJhwO5LGt/cnqNtsALkiZExAfAJsCMiPhvSeenZf+Y5M7+H0bEPElfIbnj/KB6/BrNquVAYqWubbp8BiQtkhtJupymZty893Xgy/r8CXodSVYEHgTcma4z9bak6hYu3BuYVFlWRCytoR5fI1kWpPJ1B0mbptc4Kj33IUnLcnhPZ0r6ZrpfuXrxB8A6Pl/O5G/APdpwhd/K86veY2KWFwcSK3WfVV3dNv1C/TQzCTgjIh6rku8wal67KvPcXG7GKiNZcfizauqS881cqtvqxZFet6YVfs0ahMdIzJJuptMrVwiW1FvSJiSr145Ix1C6AwdWc+7zwAGStk/P7ZymV13h9nGSbibSfP3T3cwVcg8FNqulrtlWLy7j8xVvjyfpMqtxhV+zhuJAYgY3kIx/zJA0h+QBTy2AvwPzSFaTvRb4R9UTI+J9knGNe9IVdCu7lh4Avlk52A6cCQxMB/Pn8vnssYuAQZJmkHSxvVlLXbOtXvwp0FfSdJIxkIvT9Gwr/JrlzUukmJlZXtwiMTOzvDiQmJlZXhxIzMwsLw4kZmaWFwcSMzPLiwOJmZnlxYHEzMzy8v8EAjURupuqpAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(lg, X_test, y_test, cmap='Blues', \n",
    "                      values_format='d', display_labels=['inj_sev_fatal', 'inj_sev_not_fatal']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "id": "c3f2f44d-8e0d-4c83-ba17-7bbb021ea8d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    0.550826\n",
       "0    0.449174\n",
       "Name: inj_sev_binary, dtype: float64"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['inj_sev_binary'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e82c0775-2cdc-478d-953a-668aacc99bf5",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}