NYC_CrashData_Severity_Classifier.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9aaa1618-6656-4a2d-b53e-ca53e9480126",
   "metadata": {
    "id": "9aaa1618-6656-4a2d-b53e-ca53e9480126",
    "tags": []
   },
   "source": [
    "### NYC Car Accident Project: Machine Learning\n",
    "Predicting accident severity of car accidents in NYC with supervised machine learning/classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "129b99e6-31b5-4874-bd33-e36fa34a0770",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting xgboost\n",
      "  Downloading xgboost-1.6.2-py3-none-macosx_10_15_x86_64.macosx_11_0_x86_64.macosx_12_0_x86_64.whl (1.7 MB)\n",
      "\u001b[K     |████████████████████████████████| 1.7 MB 2.1 MB/s eta 0:00:01     |███████▋                        | 409 kB 2.1 MB/s eta 0:00:01\n",
      "\u001b[?25hRequirement already satisfied: scipy in /Users/jenarriaza/opt/anaconda3/lib/python3.8/site-packages (from xgboost) (1.5.4)\n",
      "Requirement already satisfied: numpy in /Users/jenarriaza/opt/anaconda3/lib/python3.8/site-packages (from xgboost) (1.22.3)\n",
      "Installing collected packages: xgboost\n",
      "Successfully installed xgboost-1.6.2\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6415ff11",
   "metadata": {
    "id": "6415ff11",
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from numpy import linalg\n",
    "import sys\n",
    "import matplotlib.animation as animation\n",
    "from IPython.display import HTML\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import os\n",
    "import cv2\n",
    "#import pycaret\n",
    "# from pycaret.classification import *\n",
    "\n",
    "import plotly.offline as py\n",
    "import plotly.graph_objs as go\n",
    "import plotly.io as pio\n",
    "pio.renderers.default = \"png\"\n",
    "\n",
    "# Ignore printing warnings for general readability\n",
    "import warnings \n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "# pip install scikit-surprise\n",
    "# Importing libraries for model building & evaluation\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.utils import resample\n",
    "# from surprise import Reader, Dataset\n",
    "# from surprise.model_selection import train_test_split, cross_validate, GridSearchCV\n",
    "# from surprise import accuracy\n",
    "from sklearn.utils import resample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "193d80b5-731a-4542-ad45-be442f3ea4a0",
   "metadata": {
    "id": "193d80b5-731a-4542-ad45-be442f3ea4a0",
    "outputId": "260e36ba-d990-44ea-c955-45ae002296a4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/Users/jenarriaza/Applied Data Analytics-II'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "os.getcwd()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "973a917d-5673-4fb6-b94c-9643b8c88da7",
   "metadata": {
    "id": "973a917d-5673-4fb6-b94c-9643b8c88da7"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CRASH DATE</th>\n",
       "      <th>CRASH TIME</th>\n",
       "      <th>BOROUGH</th>\n",
       "      <th>ZIP CODE</th>\n",
       "      <th>LATITUDE</th>\n",
       "      <th>LONGITUDE</th>\n",
       "      <th>LOCATION</th>\n",
       "      <th>ON STREET NAME</th>\n",
       "      <th>CROSS STREET NAME</th>\n",
       "      <th>OFF STREET NAME</th>\n",
       "      <th>...</th>\n",
       "      <th>CONTRIBUTING FACTOR VEHICLE 2</th>\n",
       "      <th>CONTRIBUTING FACTOR VEHICLE 3</th>\n",
       "      <th>CONTRIBUTING FACTOR VEHICLE 4</th>\n",
       "      <th>CONTRIBUTING FACTOR VEHICLE 5</th>\n",
       "      <th>COLLISION_ID</th>\n",
       "      <th>VEHICLE TYPE CODE 1</th>\n",
       "      <th>VEHICLE TYPE CODE 2</th>\n",
       "      <th>VEHICLE TYPE CODE 3</th>\n",
       "      <th>VEHICLE TYPE CODE 4</th>\n",
       "      <th>VEHICLE TYPE CODE 5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>06/29/2022</td>\n",
       "      <td>6:55</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>THROGS NECK BRIDGE</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4541903</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Pick-up Truck</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>06/30/2022</td>\n",
       "      <td>0:17</td>\n",
       "      <td>BROOKLYN</td>\n",
       "      <td>11207</td>\n",
       "      <td>40.669792</td>\n",
       "      <td>-73.89240</td>\n",
       "      <td>(40.669792, -73.8924)</td>\n",
       "      <td>WYONA STREET</td>\n",
       "      <td>SUTTER AVENUE</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4542062</td>\n",
       "      <td>Taxi</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>04/14/2021</td>\n",
       "      <td>5:32</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>BRONX WHITESTONE BRIDGE</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4407480</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>06/29/2022</td>\n",
       "      <td>7:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Wood hill avenue</td>\n",
       "      <td>Sagamore avenue</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4542269</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Bus</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>04/13/2021</td>\n",
       "      <td>21:35</td>\n",
       "      <td>BROOKLYN</td>\n",
       "      <td>11217</td>\n",
       "      <td>40.683580</td>\n",
       "      <td>-73.97617</td>\n",
       "      <td>(40.68358, -73.97617)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>620       ATLANTIC AVENUE</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4407147</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   CRASH DATE CRASH TIME   BOROUGH ZIP CODE   LATITUDE  LONGITUDE  \\\n",
       "0  06/29/2022       6:55       NaN      NaN        NaN        NaN   \n",
       "1  06/30/2022       0:17  BROOKLYN    11207  40.669792  -73.89240   \n",
       "2  04/14/2021       5:32       NaN      NaN        NaN        NaN   \n",
       "3  06/29/2022       7:00       NaN      NaN        NaN        NaN   \n",
       "4  04/13/2021      21:35  BROOKLYN    11217  40.683580  -73.97617   \n",
       "\n",
       "                LOCATION           ON STREET NAME CROSS STREET NAME  \\\n",
       "0                    NaN       THROGS NECK BRIDGE               NaN   \n",
       "1  (40.669792, -73.8924)             WYONA STREET     SUTTER AVENUE   \n",
       "2                    NaN  BRONX WHITESTONE BRIDGE               NaN   \n",
       "3                    NaN         Wood hill avenue   Sagamore avenue   \n",
       "4  (40.68358, -73.97617)                      NaN               NaN   \n",
       "\n",
       "                            OFF STREET NAME  ...  \\\n",
       "0                                       NaN  ...   \n",
       "1                                       NaN  ...   \n",
       "2                                       NaN  ...   \n",
       "3                                       NaN  ...   \n",
       "4  620       ATLANTIC AVENUE                 ...   \n",
       "\n",
       "   CONTRIBUTING FACTOR VEHICLE 2  CONTRIBUTING FACTOR VEHICLE 3  \\\n",
       "0                    Unspecified                            NaN   \n",
       "1                            NaN                            NaN   \n",
       "2                    Unspecified                            NaN   \n",
       "3                    Unspecified                    Unspecified   \n",
       "4                            NaN                            NaN   \n",
       "\n",
       "   CONTRIBUTING FACTOR VEHICLE 4  CONTRIBUTING FACTOR VEHICLE 5  COLLISION_ID  \\\n",
       "0                            NaN                            NaN       4541903   \n",
       "1                            NaN                            NaN       4542062   \n",
       "2                            NaN                            NaN       4407480   \n",
       "3                            NaN                            NaN       4542269   \n",
       "4                            NaN                            NaN       4407147   \n",
       "\n",
       "   VEHICLE TYPE CODE 1  VEHICLE TYPE CODE 2  VEHICLE TYPE CODE 3  \\\n",
       "0                Sedan        Pick-up Truck                  NaN   \n",
       "1                 Taxi                  NaN                  NaN   \n",
       "2                Sedan                Sedan                  NaN   \n",
       "3                Sedan                Sedan                  Bus   \n",
       "4                Sedan                  NaN                  NaN   \n",
       "\n",
       "  VEHICLE TYPE CODE 4 VEHICLE TYPE CODE 5  \n",
       "0                 NaN                 NaN  \n",
       "1                 NaN                 NaN  \n",
       "2                 NaN                 NaN  \n",
       "3                 NaN                 NaN  \n",
       "4                 NaN                 NaN  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"Motor_Vehicle_Collisions_-_Crashes.csv\", low_memory=False) # KEPLER.GL CAN READ GeoJSON, JSON, CSV\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c67d7096-68ea-4479-a5a9-1abaffa1bd7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CRASH DATE                             0\n",
       "CRASH TIME                             0\n",
       "BOROUGH                           595905\n",
       "ZIP CODE                          596138\n",
       "LATITUDE                          222330\n",
       "LONGITUDE                         222330\n",
       "LOCATION                          222330\n",
       "ON STREET NAME                    398501\n",
       "CROSS STREET NAME                 705089\n",
       "OFF STREET NAME                  1615548\n",
       "NUMBER OF PERSONS INJURED             18\n",
       "NUMBER OF PERSONS KILLED              31\n",
       "NUMBER OF PEDESTRIANS INJURED          0\n",
       "NUMBER OF PEDESTRIANS KILLED           0\n",
       "NUMBER OF CYCLIST INJURED              0\n",
       "NUMBER OF CYCLIST KILLED               0\n",
       "NUMBER OF MOTORIST INJURED             0\n",
       "NUMBER OF MOTORIST KILLED              0\n",
       "CONTRIBUTING FACTOR VEHICLE 1       5766\n",
       "CONTRIBUTING FACTOR VEHICLE 2     285421\n",
       "CONTRIBUTING FACTOR VEHICLE 3    1787552\n",
       "CONTRIBUTING FACTOR VEHICLE 4    1891928\n",
       "CONTRIBUTING FACTOR VEHICLE 5    1913724\n",
       "COLLISION_ID                           0\n",
       "VEHICLE TYPE CODE 1                11201\n",
       "VEHICLE TYPE CODE 2               344445\n",
       "VEHICLE TYPE CODE 3              1791910\n",
       "VEHICLE TYPE CODE 4              1892882\n",
       "VEHICLE TYPE CODE 5              1913945\n",
       "dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "11be05b6-d15b-4f7f-882b-ccbfcdd3d1a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.columns = [x.lower() for x in df.columns]\n",
    "df.columns = df.columns.str.replace(' ', '_')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "17ed686f-e69f-4fac-ac8d-149c7f015bb3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['crash_date', 'crash_time', 'borough', 'zip_code', 'latitude',\n",
       "       'longitude', 'location', 'on_street_name', 'cross_street_name',\n",
       "       'off_street_name', 'number_of_persons_injured',\n",
       "       'number_of_persons_killed', 'number_of_pedestrians_injured',\n",
       "       'number_of_pedestrians_killed', 'number_of_cyclist_injured',\n",
       "       'number_of_cyclist_killed', 'number_of_motorist_injured',\n",
       "       'number_of_motorist_killed', 'contributing_factor_vehicle_1',\n",
       "       'contributing_factor_vehicle_2', 'contributing_factor_vehicle_3',\n",
       "       'contributing_factor_vehicle_4', 'contributing_factor_vehicle_5',\n",
       "       'collision_id', 'vehicle_type_code_1', 'vehicle_type_code_2',\n",
       "       'vehicle_type_code_3', 'vehicle_type_code_4', 'vehicle_type_code_5'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3f53a2d3-b125-4684-a9ee-e51c075c8c2c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop duplicates.\n",
    "df.drop_duplicates(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "dcafdec2-906a-4346-8d0a-3eea4eb51837",
   "metadata": {
    "id": "dcafdec2-906a-4346-8d0a-3eea4eb51837",
    "outputId": "a5481ed3-a646-4333-f82c-a358aaaa72d8"
   },
   "outputs": [],
   "source": [
    "# filling a null values using fillna()\n",
    "df[\"borough\"].fillna(\"Not Available\", inplace = True)\n",
    "df[\"zip_code\"].fillna(\"0\", inplace = True)\n",
    "df[\"on_street_name\"].fillna(\"Not Available\", inplace = True)\n",
    "df[\"cross_street_name\"].fillna(\"Not Available\", inplace = True)\n",
    "df[\"off_street_name\"].fillna(\"Not Available\", inplace = True)\n",
    "df[\"contributing_factor_vehicle_1\"].fillna(\"Not Available\", inplace = True)\n",
    "df[\"contributing_factor_vehicle_2\"].fillna(\"Not Available\", inplace = True)\n",
    "df[\"vehicle_type_code_1\"].fillna(\"Not Available\", inplace = True)\n",
    "df[\"vehicle_type_code_2\"].fillna(\"Not Available\", inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ea7dbde5-2848-497e-a514-b524bca1978c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fill missing lat/long with borough mean\n",
    "df[['latitude', 'longitude']]=df[['borough','latitude', 'longitude']].groupby('borough').ffill().bfill()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "de6d9c34-5c08-4719-a2a7-0c0d1979a283",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "crash_date                             0\n",
       "crash_time                             0\n",
       "borough                                0\n",
       "zip_code                               0\n",
       "latitude                               0\n",
       "longitude                              0\n",
       "location                          222330\n",
       "on_street_name                         0\n",
       "cross_street_name                      0\n",
       "off_street_name                        0\n",
       "number_of_persons_injured             18\n",
       "number_of_persons_killed              31\n",
       "number_of_pedestrians_injured          0\n",
       "number_of_pedestrians_killed           0\n",
       "number_of_cyclist_injured              0\n",
       "number_of_cyclist_killed               0\n",
       "number_of_motorist_injured             0\n",
       "number_of_motorist_killed              0\n",
       "contributing_factor_vehicle_1          0\n",
       "contributing_factor_vehicle_2          0\n",
       "contributing_factor_vehicle_3    1787552\n",
       "contributing_factor_vehicle_4    1891928\n",
       "contributing_factor_vehicle_5    1913724\n",
       "collision_id                           0\n",
       "vehicle_type_code_1                    0\n",
       "vehicle_type_code_2                    0\n",
       "vehicle_type_code_3              1791910\n",
       "vehicle_type_code_4              1892882\n",
       "vehicle_type_code_5              1913945\n",
       "dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59ab7c54-4ee0-40f6-ab40-ba4801f49140",
   "metadata": {
    "id": "59ab7c54-4ee0-40f6-ab40-ba4801f49140"
   },
   "source": [
    "# Cleanup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "ef9313f5-db00-420b-97a7-257645b73286",
   "metadata": {},
   "outputs": [],
   "source": [
    "# What to do with date/time column?\n",
    "# Change data type of date column\n",
    "# convert to datetime format, if needed\n",
    "df.crash_date = pd.to_datetime(df.crash_date)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "8b3f190e-cb45-46e4-9ee0-8d55bd6bb1cf",
   "metadata": {
    "id": "8b3f190e-cb45-46e4-9ee0-8d55bd6bb1cf"
   },
   "outputs": [],
   "source": [
    "# Renaming columns\n",
    "df.rename(columns={'number_of_cyclist_injured': 'amt_cyc_inj', 'veh_type_code1':'veh_type_1','veh_type_code2':'veh_type_2','cont_fact_veh1':'contributing_factor1','cont_fact_veh2':'contributing_factor2'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "6f61287c-58ae-44da-98e1-fa998be0ed8c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['crash_date', 'crash_time', 'borough', 'zip_code', 'latitude',\n",
       "       'longitude', 'location', 'on_street_name', 'cross_street_name',\n",
       "       'off_street_name', 'number_of_persons_injured',\n",
       "       'number_of_persons_killed', 'number_of_pedestrians_injured',\n",
       "       'number_of_pedestrians_killed', 'amt_cyc_inj',\n",
       "       'number_of_cyclist_killed', 'number_of_motorist_injured',\n",
       "       'number_of_motorist_killed', 'contributing_factor_vehicle_1',\n",
       "       'contributing_factor_vehicle_2', 'contributing_factor_vehicle_3',\n",
       "       'contributing_factor_vehicle_4', 'contributing_factor_vehicle_5',\n",
       "       'collision_id', 'vehicle_type_code_1', 'vehicle_type_code_2',\n",
       "       'vehicle_type_code_3', 'vehicle_type_code_4', 'vehicle_type_code_5'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d41e8d0-e2f5-4db4-b4c6-5f463d108fc7",
   "metadata": {},
   "source": [
    "#### Choose variables of interest for ML run"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "67f3368c-c89a-4247-bd0e-b010e54ddb35",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Drop unnecessary columns\n",
    "\n",
    "# df.drop(['collision_id','long_lat', 'amt_ped_inj','amt_ped_killed', 'amt_cyc_inj', 'amt_cyc_killed', 'amt_mot_inj','amt_mot_killed','vehicle_type_code_3', 'vehicle_type_code_4', 'vehicle_type_code_5'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e6a17cc6-3f3d-47ee-8a50-2c15e18fb369",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['crash_date', 'crash_time', 'borough', 'zip_code', 'latitude',\n",
       "       'longitude', 'location', 'on_street_name', 'cross_street_name',\n",
       "       'off_street_name', 'number_of_persons_injured',\n",
       "       'number_of_persons_killed', 'number_of_pedestrians_injured',\n",
       "       'number_of_pedestrians_killed', 'amt_cyc_inj',\n",
       "       'number_of_cyclist_killed', 'number_of_motorist_injured',\n",
       "       'number_of_motorist_killed', 'contributing_factor_vehicle_1',\n",
       "       'contributing_factor_vehicle_2', 'contributing_factor_vehicle_3',\n",
       "       'contributing_factor_vehicle_4', 'contributing_factor_vehicle_5',\n",
       "       'collision_id', 'vehicle_type_code_1', 'vehicle_type_code_2',\n",
       "       'vehicle_type_code_3', 'vehicle_type_code_4', 'vehicle_type_code_5'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "444001af-8d80-42eb-af2d-a7f1870fc404",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[['crash_date','crash_time','borough', 'zip_code', 'latitude', 'longitude', 'on_street_name',\n",
    "       'cross_street_name', 'off_street_name', 'number_of_persons_injured','contributing_factor_vehicle_1',\n",
    "       'contributing_factor_vehicle_2', 'vehicle_type_code_1', 'vehicle_type_code_2']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "90faf4ff-0e1e-4edd-ae30-ba6b94a7e00f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extract more data of interest \n",
    "\n",
    "df['hour'] = df['crash_time'].str[:-3]\n",
    "df['year'] = df['crash_date'].dt.year\n",
    "df['month'] = df['crash_date'].dt.month\n",
    "df['day_of_week'] = df['crash_date'].dt.dayofweek"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1941ccd1-41cd-4590-b0de-974200ae393b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>crash_date</th>\n",
       "      <th>crash_time</th>\n",
       "      <th>borough</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>on_street_name</th>\n",
       "      <th>cross_street_name</th>\n",
       "      <th>off_street_name</th>\n",
       "      <th>number_of_persons_injured</th>\n",
       "      <th>contributing_factor_vehicle_1</th>\n",
       "      <th>contributing_factor_vehicle_2</th>\n",
       "      <th>vehicle_type_code_1</th>\n",
       "      <th>vehicle_type_code_2</th>\n",
       "      <th>hour</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day_of_week</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-06-29</td>\n",
       "      <td>6:55</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>0</td>\n",
       "      <td>40.669792</td>\n",
       "      <td>-73.89240</td>\n",
       "      <td>THROGS NECK BRIDGE</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Following Too Closely</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Pick-up Truck</td>\n",
       "      <td>6</td>\n",
       "      <td>2022</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-06-30</td>\n",
       "      <td>0:17</td>\n",
       "      <td>BROOKLYN</td>\n",
       "      <td>11207</td>\n",
       "      <td>40.669792</td>\n",
       "      <td>-73.89240</td>\n",
       "      <td>WYONA STREET</td>\n",
       "      <td>SUTTER AVENUE</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>Taxi</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>0</td>\n",
       "      <td>2022</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2021-04-14</td>\n",
       "      <td>5:32</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>0</td>\n",
       "      <td>40.683580</td>\n",
       "      <td>-73.97617</td>\n",
       "      <td>BRONX WHITESTONE BRIDGE</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Following Too Closely</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>5</td>\n",
       "      <td>2021</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-06-29</td>\n",
       "      <td>7:00</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>0</td>\n",
       "      <td>40.683580</td>\n",
       "      <td>-73.97617</td>\n",
       "      <td>Wood hill avenue</td>\n",
       "      <td>Sagamore avenue</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Turning Improperly</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>7</td>\n",
       "      <td>2022</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2021-04-13</td>\n",
       "      <td>21:35</td>\n",
       "      <td>BROOKLYN</td>\n",
       "      <td>11217</td>\n",
       "      <td>40.683580</td>\n",
       "      <td>-73.97617</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>620       ATLANTIC AVENUE</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Unspecified</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>Sedan</td>\n",
       "      <td>Not Available</td>\n",
       "      <td>21</td>\n",
       "      <td>2021</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  crash_date crash_time        borough zip_code   latitude  longitude  \\\n",
       "0 2022-06-29       6:55  Not Available        0  40.669792  -73.89240   \n",
       "1 2022-06-30       0:17       BROOKLYN    11207  40.669792  -73.89240   \n",
       "2 2021-04-14       5:32  Not Available        0  40.683580  -73.97617   \n",
       "3 2022-06-29       7:00  Not Available        0  40.683580  -73.97617   \n",
       "4 2021-04-13      21:35       BROOKLYN    11217  40.683580  -73.97617   \n",
       "\n",
       "            on_street_name cross_street_name  \\\n",
       "0       THROGS NECK BRIDGE     Not Available   \n",
       "1             WYONA STREET     SUTTER AVENUE   \n",
       "2  BRONX WHITESTONE BRIDGE     Not Available   \n",
       "3         Wood hill avenue   Sagamore avenue   \n",
       "4            Not Available     Not Available   \n",
       "\n",
       "                            off_street_name  number_of_persons_injured  \\\n",
       "0                             Not Available                        0.0   \n",
       "1                             Not Available                        1.0   \n",
       "2                             Not Available                        0.0   \n",
       "3                             Not Available                        0.0   \n",
       "4  620       ATLANTIC AVENUE                                       1.0   \n",
       "\n",
       "  contributing_factor_vehicle_1 contributing_factor_vehicle_2  \\\n",
       "0         Following Too Closely                   Unspecified   \n",
       "1                   Unspecified                 Not Available   \n",
       "2         Following Too Closely                   Unspecified   \n",
       "3            Turning Improperly                   Unspecified   \n",
       "4                   Unspecified                 Not Available   \n",
       "\n",
       "  vehicle_type_code_1 vehicle_type_code_2 hour  year  month  day_of_week  \n",
       "0               Sedan       Pick-up Truck    6  2022      6            2  \n",
       "1                Taxi       Not Available    0  2022      6            3  \n",
       "2               Sedan               Sedan    5  2021      4            2  \n",
       "3               Sedan               Sedan    7  2022      6            2  \n",
       "4               Sedan       Not Available   21  2021      4            1  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "5b6ed2b7-2010-4eb9-b71f-37541f366a17",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "crash_date                        0\n",
       "crash_time                        0\n",
       "borough                           0\n",
       "zip_code                          0\n",
       "latitude                          0\n",
       "longitude                         0\n",
       "on_street_name                    0\n",
       "cross_street_name                 0\n",
       "off_street_name                   0\n",
       "number_of_persons_injured        18\n",
       "contributing_factor_vehicle_1     0\n",
       "contributing_factor_vehicle_2     0\n",
       "vehicle_type_code_1               0\n",
       "vehicle_type_code_2               0\n",
       "hour                              0\n",
       "year                              0\n",
       "month                             0\n",
       "day_of_week                       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eede8402-50b4-4228-9dd6-b21ad5048f6b",
   "metadata": {},
   "source": [
    "### Dropping rows where lat/long out of bounds, if needed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "295d8cff-c015-497a-873b-e514f70b924f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# df1 = df[df.apply(lambda row: (40.554168<=row['latitude']<= 40.889596) and (-74.163230<=row['longitude']<=-73.726940), axis=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "604afe5e-ea2a-49c8-ad5b-ea3db6644fd7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1.921679e+06\n",
       "mean     4.064191e+01\n",
       "std      1.834246e+00\n",
       "min      0.000000e+00\n",
       "25%      4.066865e+01\n",
       "50%      4.072258e+01\n",
       "75%      4.076954e+01\n",
       "max      4.334444e+01\n",
       "Name: latitude, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.latitude.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d3da154f-83fd-4515-b989-be41def7ebfa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    430199.000000\n",
       "mean        -73.634437\n",
       "std           4.530717\n",
       "min         -74.253006\n",
       "25%         -73.962510\n",
       "50%         -73.918940\n",
       "75%         -73.861670\n",
       "max           0.000000\n",
       "Name: longitude, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.longitude.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf2e0c58-7c6b-45e0-aac2-04f7d5fb41f9",
   "metadata": {},
   "source": [
    "#### Data Transformations "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1e70f16e-7dc6-457d-a2bf-5bb912862de8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fix data types\n",
    "\n",
    "#df[\"zip_code\"] = df['zip_code'].astype('float64')\n",
    "#df[\"latitude\"] = df['latitude'].astype('float64')\n",
    "#df[\"longitude\"] = df['longitude'].astype('float64')\n",
    "#df[\"hour\"] = df['hour'].astype('int64')\n",
    "#df1[\"on_street\"] = df1['on_street'].astype('float64')\n",
    "#df1[\"veh_type\"] = df1['veh_type'].astype('float64')\n",
    "#df1.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5e71a4dc-5c56-4cc3-899c-cca267ae41ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Manipulate vehicle type\n",
    "\n",
    "# df1['veh_type'] = df1['veh_type'].apply(lambda x: float(x.split()[0].replace('Station Wagon/Sport Utility Vehicle', 'SUV')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "460f7d44-779f-492d-9764-bf11192b73e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "crash_date                        0\n",
       "crash_time                        0\n",
       "borough                           0\n",
       "zip_code                          0\n",
       "latitude                          0\n",
       "longitude                         0\n",
       "on_street_name                    0\n",
       "cross_street_name                 0\n",
       "off_street_name                   0\n",
       "number_of_persons_injured        18\n",
       "contributing_factor_vehicle_1     0\n",
       "contributing_factor_vehicle_2     0\n",
       "vehicle_type_code_1               0\n",
       "vehicle_type_code_2               0\n",
       "hour                              0\n",
       "year                              0\n",
       "month                             0\n",
       "day_of_week                       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6013c5ad-0baf-4f0e-88cc-4605862e49f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Run clean-up on NaNs, if needed\n",
    "\n",
    "# df1 = df[df['number_of_persons_injured'].notna()]\n",
    "# df1 = df1.replace(np.nan, '', regex=True)\n",
    "# df1 = df1[df1['on_street'].notna()]\n",
    "# df1 = df1[df1['off_street_name'].notna()]\n",
    "# df1 = df1[df1['veh_type'].notna()]\n",
    "# df1 = df[df['amt_injured'].notna()]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bea66812-d0c7-4f48-b095-a6058aa6fc44",
   "metadata": {},
   "source": [
    "# Machine Learning: Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "292d08de-a192-4c0f-929c-e51b0c04a184",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluator(x):\n",
    "   if x == 0:\n",
    "      return False\n",
    "   elif x == 1:\n",
    "      return True\n",
    "   else:\n",
    "      return True\n",
    "df['serious_accident'] = df['number_of_persons_injured'].map(lambda x: evaluator(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "5d19f9cb-d17e-45ff-b79f-9ca440b7a0b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['crash_date', 'crash_time', 'borough', 'zip_code', 'latitude',\n",
       "       'longitude', 'on_street_name', 'cross_street_name', 'off_street_name',\n",
       "       'number_of_persons_injured', 'contributing_factor_vehicle_1',\n",
       "       'contributing_factor_vehicle_2', 'vehicle_type_code_1',\n",
       "       'vehicle_type_code_2', 'hour', 'year', 'month', 'day_of_week',\n",
       "       'serious_accident'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0d32e45-3b33-4398-a656-a1c11b7d8373",
   "metadata": {},
   "source": [
    "#### Filter years/top vehicles, etc. if needed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d9b099bd-7486-475b-9f85-508b78d86410",
   "metadata": {},
   "outputs": [],
   "source": [
    "# df1 = df1[df1.year.isin([2020,2021])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "069bd370-0265-4f01-bd60-36f6c416383b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Select top vehicle types\n",
    "#veh_type_top5 = ['Sedan','Station Wagon/Sport Utility Vehicle','Taxi','Pick-up Truck','Box Truck']\n",
    "            \n",
    "\n",
    "#df1 = df1[df1.veh_type.isin(veh_type_top5)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8baac161-337f-4adc-8c90-4a7981a08eec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    1510524\n",
       "True      411137\n",
       "Name: serious_accident, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.serious_accident.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de2a2741-b18d-4ed9-89d2-2254193457d9",
   "metadata": {},
   "source": [
    "### Remove the target columns for ML run"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "f218e0b5-7d43-4051-b772-be983b8b41d0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop columns to be exlcuded from machine learning\n",
    "df.drop(['crash_date','crash_time','number_of_persons_injured'], axis=1, inplace=True) # 'on_street', 'cross_street_name', 'off_street_name'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb2b2b7a-542b-43cd-b562-2a29ad49ac76",
   "metadata": {},
   "source": [
    "#### Retitle, reindex columns if needed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "34bf8c58-eef8-438b-a4bc-8624f8c7ea43",
   "metadata": {},
   "outputs": [],
   "source": [
    "#columnsTitles = ['latitude', 'longitude','cont_fact_veh1', 'veh_type','hour', 'serious_accident']\n",
    "\n",
    "#df1 = df1.reindex(columns=columnsTitles)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "978569b1-b5b1-46c8-bb64-de78593d39c3",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Classification\n",
    "- use next 2 cells to write code for an ML classification problem, if needed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "af518783-2602-4032-b4d8-2a1d15c93bff",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "#df2 = df\n",
    "conditions = [\n",
    "    (df['amt_injured'] >=1), # 1 or more injuries\n",
    "    (df['amt_injured'] ==0)] # property damage (only)\n",
    "choices = ['Injury', 'Property Damage']\n",
    "df['Severity'] = np.select(conditions, choices, default='Property Damage')\n",
    "#print(df3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7ef754c6-00fb-462b-b918-c29b8aa647c5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# map severity to code: \n",
    "df['Severity'] = df['Severity'].map({'Property Damage':1, 'Injury':2})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "389d7410-3284-47e9-9bde-c3e3fc9cc67a",
   "metadata": {},
   "source": [
    "### Handling Class Imbalance\n",
    "Applying down/up-sambling to imbalances in data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "b34eedfa-bbaa-4c9a-9e3b-51be2376a06b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    1510524\n",
       "True      411137\n",
       "Name: serious_accident, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.serious_accident.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "167bb546-e135-4102-b51c-7526a144e917",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='serious_accident', ylabel='count'>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x='serious_accident',data=df,\n",
    "               order = df['serious_accident'].value_counts().index)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dda79d8-7908-4312-9230-29eb292598e7",
   "metadata": {},
   "source": [
    "### Upsampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "69f5d410-4852-4a10-a0fd-c70c8a370590",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_majority1 = df[df['serious_accident']==False]         \n",
    "df_minority1 = df[df['serious_accident']==True] \n",
    "\n",
    "df_minority_upsampled1 = resample(df_minority1,\n",
    "                                   replace=True, # sample without replacement\n",
    "                                   n_samples=1510524, # to match majority class\n",
    "                                   random_state=123) # reproducable results\n",
    "df_upsampled1 = pd.concat([df_minority_upsampled1, df_majority1])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7625ae44-40c0-43c9-878d-2f43fbfe7120",
   "metadata": {},
   "source": [
    "### Downsampling (if needed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d001cad4-f32d-4c3a-93a4-a79ec834b869",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_majority1 = df[df['serious_accident']==False]         \n",
    "df_minority1 = df[df['serious_accident']==True] \n",
    "\n",
    "df_majority_downsampled1 = resample(df_majority1,\n",
    "                                   replace=False, # sample without replacement\n",
    "                                   n_samples=411137, # to match minority class\n",
    "                                   random_state=123) # reproducable results\n",
    "\n",
    "df_downsampled1 = pd.concat([df_majority_downsampled1, df_minority1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "036a6f1a-218a-4f1d-903c-417c92a1f093",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='serious_accident', ylabel='count'>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Check cless imbalances\n",
    "\n",
    "sns.countplot(x='serious_accident',data=df_downsampled1,\n",
    "               order = df_downsampled1['serious_accident'].value_counts().index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "9dab69eb-3157-47a6-8af2-8a88a3378555",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    411137\n",
       "True     411137\n",
       "Name: serious_accident, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check downsampled counts\n",
    "df_downsampled1.serious_accident.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "80f7fa61-8efc-40f3-84d7-5627b00bb267",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['borough', 'zip_code', 'latitude', 'longitude', 'on_street_name',\n",
       "       'cross_street_name', 'off_street_name', 'contributing_factor_vehicle_1',\n",
       "       'contributing_factor_vehicle_2', 'vehicle_type_code_1',\n",
       "       'vehicle_type_code_2', 'hour', 'year', 'month', 'day_of_week',\n",
       "       'serious_accident'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_downsampled1.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92c513ea-d9dd-4313-beb8-029222f7f43d",
   "metadata": {},
   "source": [
    "# AdaBoost Classification Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "563a21f2-7ec8-47fc-bedc-743874649727",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "On Street unique value:  ['' 'WEST 145 STREET                 ' 'VAN WYCK EXPWY                  '\n",
      " ... 'SUMMERFIELD PLACE               ' 'KNOLLS CRESCENT                 '\n",
      " 'SANDS PLACE                     ']\n",
      "Contact Factor (Veh 1) unique value:  ['Passing or Lane Usage Improper' 'Driver Inattention/Distraction'\n",
      " 'Unsafe Lane Changing' 'Unspecified' 'Unsafe Speed' 'Oversized Vehicle'\n",
      " 'Driverless/Runaway Vehicle' 'Following Too Closely' 'Other Vehicular'\n",
      " 'Pedestrian/Bicyclist/Other Pedestrian Error/Confusion'\n",
      " 'Failure to Yield Right-of-Way' 'Steering Failure' 'Alcohol Involvement'\n",
      " 'Reaction to Uninvolved Vehicle' 'Passenger Distraction'\n",
      " 'View Obstructed/Limited' 'Backing Unsafely' 'Outside Car Distraction'\n",
      " 'Passing Too Closely' 'Turning Improperly' 'Driver Inexperience'\n",
      " 'Brakes Defective' 'Pavement Slippery' 'Animals Action'\n",
      " 'Traffic Control Disregarded' 'Aggressive Driving/Road Rage'\n",
      " 'Fell Asleep' 'Cell Phone (hand-Held)' 'Failure to Keep Right'\n",
      " 'Pavement Defective' 'Physical Disability' 'Tinted Windows'\n",
      " 'Traffic Control Device Improper/Non-Working' 'Fatigued/Drowsy'\n",
      " 'Accelerator Defective' 'Tire Failure/Inadequate' 'Drugs (illegal)'\n",
      " 'Illnes' 'Lane Marking Improper/Inadequate' 'Glare' 'Obstruction/Debris'\n",
      " 'Lost Consciousness' 'Headlights Defective'\n",
      " 'Shoulders Defective/Improper' 'Tow Hitch Defective'\n",
      " 'Other Electronic Device' 'Vehicle Vandalism' '' 'Eating or Drinking'\n",
      " 'Other Lighting Defects' 'Prescription Medication' 'Texting'\n",
      " 'Using On Board Navigation Device' 'Cell Phone (hands-free)'\n",
      " 'Listening/Using Headphones' 'Windshield Inadequate']\n",
      "Vehicle type unique value:  ['Station Wagon/Sport Utility Vehicle' 'Sedan' 'Bus' 'Pick-up Truck'\n",
      " 'Box Truck' 'Taxi' 'Tractor Truck Diesel' 'Dump' 'Bike' 'Motorcycle'\n",
      " 'E-Bike']\n",
      "Month unique value:  [ 7 11  3 12  1  4  8 10  5  9  2  6]\n"
     ]
    }
   ],
   "source": [
    "print('On Street unique value: ', df_downsampled1['on_street'].unique())\n",
    "print('Contact Factor (Veh 1) unique value: ', df_downsampled1['cont_fact_veh1'].unique())\n",
    "print('Vehicle type unique value: ', df_downsampled1['veh_type'].unique())\n",
    "print('Month unique value: ', df_downsampled1['month'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "8a15e772-c463-4073-a79d-ff5c59654765",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop target predictors \n",
    "# df_upsampled1.drop(['location'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "b4ea1297-a172-4e45-bd64-f22f1983553b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['borough', 'zip_code', 'latitude', 'longitude', 'on_street_name',\n",
       "       'cross_street_name', 'off_street_name', 'contributing_factor_vehicle_1',\n",
       "       'contributing_factor_vehicle_2', 'vehicle_type_code_1',\n",
       "       'vehicle_type_code_2', 'hour', 'year', 'month', 'day_of_week',\n",
       "       'serious_accident'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "93ef2a57-9d8d-4202-ad5b-f1e301d0ca8a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1921661 entries, 0 to 1921678\n",
      "Data columns (total 16 columns):\n",
      " #   Column                         Dtype  \n",
      "---  ------                         -----  \n",
      " 0   borough                        object \n",
      " 1   zip_code                       object \n",
      " 2   latitude                       float64\n",
      " 3   longitude                      float64\n",
      " 4   on_street_name                 object \n",
      " 5   cross_street_name              object \n",
      " 6   off_street_name                object \n",
      " 7   contributing_factor_vehicle_1  object \n",
      " 8   contributing_factor_vehicle_2  object \n",
      " 9   vehicle_type_code_1            object \n",
      " 10  vehicle_type_code_2            object \n",
      " 11  hour                           object \n",
      " 12  year                           int64  \n",
      " 13  month                          int64  \n",
      " 14  day_of_week                    int64  \n",
      " 15  serious_accident               bool   \n",
      "dtypes: bool(1), float64(2), int64(3), object(10)\n",
      "memory usage: 236.4+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0179d7e8-4c9c-4746-ba41-235dee53d14c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_downsampled1['borough'] = df_downsampled1['borough'].astype('str')\n",
    "df_downsampled1['zip_code'] = df_downsampled1['zip_code'].astype('str')\n",
    "df_downsampled1['on_street_name'] = df_downsampled1['on_street_name'].astype('str')\n",
    "df_downsampled1['off_street_name'] = df_downsampled1['off_street_name'].astype('str')\n",
    "df_downsampled1['cross_street_name'] = df_downsampled1['cross_street_name'].astype('str')\n",
    "df_downsampled1['vehicle_type_code_1'] = df_downsampled1['vehicle_type_code_1'].astype('str')\n",
    "df_downsampled1['vehicle_type_code_2'] = df_downsampled1['vehicle_type_code_2'].astype('str')\n",
    "df_downsampled1['contributing_factor_vehicle_1'] = df_downsampled1['contributing_factor_vehicle_1'].astype('str')\n",
    "df_downsampled1['contributing_factor_vehicle_2'] = df_downsampled1['contributing_factor_vehicle_2'].astype('str')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "7bf9101d-01b2-41cd-9e6b-df4f83731fa1",
   "metadata": {},
   "outputs": [],
   "source": [
    "LE = LabelEncoder()\n",
    "df_downsampled1['borough'] = LE.fit_transform(df_downsampled1['borough'])\n",
    "df_downsampled1['zip_code'] = LE.fit_transform(df_downsampled1['zip_code'])\n",
    "df_downsampled1['on_street_name'] = LE.fit_transform(df_downsampled1['on_street_name'])\n",
    "df_downsampled1['off_street_name'] = LE.fit_transform(df_downsampled1['off_street_name'])\n",
    "df_downsampled1['cross_street_name'] = LE.fit_transform(df_downsampled1['cross_street_name'])\n",
    "df_downsampled1['vehicle_type_code_1'] = LE.fit_transform(df_downsampled1['vehicle_type_code_1'])\n",
    "df_downsampled1['vehicle_type_code_2'] = LE.fit_transform(df_downsampled1['vehicle_type_code_2'])\n",
    "df_downsampled1['contributing_factor_vehicle_1'] = LE.fit_transform(df_downsampled1['contributing_factor_vehicle_1'])\n",
    "df_downsampled1['contributing_factor_vehicle_2'] = LE.fit_transform(df_downsampled1['contributing_factor_vehicle_2'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "dede4910-7f0a-4199-a1a3-5d6103290d6a",
   "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>borough</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>on_street_name</th>\n",
       "      <th>cross_street_name</th>\n",
       "      <th>off_street_name</th>\n",
       "      <th>contributing_factor_vehicle_1</th>\n",
       "      <th>contributing_factor_vehicle_2</th>\n",
       "      <th>vehicle_type_code_1</th>\n",
       "      <th>vehicle_type_code_2</th>\n",
       "      <th>hour</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>serious_accident</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>406910</th>\n",
       "      <td>1</td>\n",
       "      <td>145</td>\n",
       "      <td>40.627090</td>\n",
       "      <td>-73.944030</td>\n",
       "      <td>2171</td>\n",
       "      <td>9939</td>\n",
       "      <td>85540</td>\n",
       "      <td>18</td>\n",
       "      <td>57</td>\n",
       "      <td>528</td>\n",
       "      <td>604</td>\n",
       "      <td>11</td>\n",
       "      <td>2019</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>384051</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>40.725750</td>\n",
       "      <td>-73.831260</td>\n",
       "      <td>10801</td>\n",
       "      <td>10159</td>\n",
       "      <td>85540</td>\n",
       "      <td>11</td>\n",
       "      <td>57</td>\n",
       "      <td>538</td>\n",
       "      <td>618</td>\n",
       "      <td>17</td>\n",
       "      <td>2019</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187494</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>40.722584</td>\n",
       "      <td>-73.850900</td>\n",
       "      <td>11796</td>\n",
       "      <td>3616</td>\n",
       "      <td>85540</td>\n",
       "      <td>57</td>\n",
       "      <td>57</td>\n",
       "      <td>528</td>\n",
       "      <td>604</td>\n",
       "      <td>19</td>\n",
       "      <td>2021</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>165946</th>\n",
       "      <td>4</td>\n",
       "      <td>204</td>\n",
       "      <td>40.677550</td>\n",
       "      <td>-73.744660</td>\n",
       "      <td>8471</td>\n",
       "      <td>10159</td>\n",
       "      <td>32283</td>\n",
       "      <td>57</td>\n",
       "      <td>57</td>\n",
       "      <td>538</td>\n",
       "      <td>604</td>\n",
       "      <td>16</td>\n",
       "      <td>2020</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1716583</th>\n",
       "      <td>4</td>\n",
       "      <td>212</td>\n",
       "      <td>40.692518</td>\n",
       "      <td>-73.859097</td>\n",
       "      <td>6799</td>\n",
       "      <td>1810</td>\n",
       "      <td>85540</td>\n",
       "      <td>57</td>\n",
       "      <td>57</td>\n",
       "      <td>417</td>\n",
       "      <td>473</td>\n",
       "      <td>15</td>\n",
       "      <td>2013</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1921664</th>\n",
       "      <td>1</td>\n",
       "      <td>147</td>\n",
       "      <td>40.661997</td>\n",
       "      <td>-73.919593</td>\n",
       "      <td>5045</td>\n",
       "      <td>8399</td>\n",
       "      <td>85540</td>\n",
       "      <td>57</td>\n",
       "      <td>29</td>\n",
       "      <td>508</td>\n",
       "      <td>456</td>\n",
       "      <td>15</td>\n",
       "      <td>2012</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1921665</th>\n",
       "      <td>4</td>\n",
       "      <td>131</td>\n",
       "      <td>40.767228</td>\n",
       "      <td>-73.918907</td>\n",
       "      <td>1041</td>\n",
       "      <td>1041</td>\n",
       "      <td>85540</td>\n",
       "      <td>11</td>\n",
       "      <td>29</td>\n",
       "      <td>417</td>\n",
       "      <td>456</td>\n",
       "      <td>15</td>\n",
       "      <td>2012</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1921669</th>\n",
       "      <td>1</td>\n",
       "      <td>167</td>\n",
       "      <td>40.677288</td>\n",
       "      <td>-73.928227</td>\n",
       "      <td>2123</td>\n",
       "      <td>7911</td>\n",
       "      <td>85540</td>\n",
       "      <td>11</td>\n",
       "      <td>57</td>\n",
       "      <td>508</td>\n",
       "      <td>61</td>\n",
       "      <td>18</td>\n",
       "      <td>2012</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1921671</th>\n",
       "      <td>1</td>\n",
       "      <td>153</td>\n",
       "      <td>40.636133</td>\n",
       "      <td>-73.972736</td>\n",
       "      <td>4297</td>\n",
       "      <td>10215</td>\n",
       "      <td>85540</td>\n",
       "      <td>57</td>\n",
       "      <td>57</td>\n",
       "      <td>417</td>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "      <td>2012</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1921675</th>\n",
       "      <td>4</td>\n",
       "      <td>131</td>\n",
       "      <td>40.774711</td>\n",
       "      <td>-73.933386</td>\n",
       "      <td>942</td>\n",
       "      <td>1212</td>\n",
       "      <td>85540</td>\n",
       "      <td>18</td>\n",
       "      <td>57</td>\n",
       "      <td>417</td>\n",
       "      <td>580</td>\n",
       "      <td>17</td>\n",
       "      <td>2012</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>822274 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         borough  zip_code   latitude  longitude  on_street_name  \\\n",
       "406910         1       145  40.627090 -73.944030            2171   \n",
       "384051         3         1  40.725750 -73.831260           10801   \n",
       "187494         3         1  40.722584 -73.850900           11796   \n",
       "165946         4       204  40.677550 -73.744660            8471   \n",
       "1716583        4       212  40.692518 -73.859097            6799   \n",
       "...          ...       ...        ...        ...             ...   \n",
       "1921664        1       147  40.661997 -73.919593            5045   \n",
       "1921665        4       131  40.767228 -73.918907            1041   \n",
       "1921669        1       167  40.677288 -73.928227            2123   \n",
       "1921671        1       153  40.636133 -73.972736            4297   \n",
       "1921675        4       131  40.774711 -73.933386             942   \n",
       "\n",
       "         cross_street_name  off_street_name  contributing_factor_vehicle_1  \\\n",
       "406910                9939            85540                             18   \n",
       "384051               10159            85540                             11   \n",
       "187494                3616            85540                             57   \n",
       "165946               10159            32283                             57   \n",
       "1716583               1810            85540                             57   \n",
       "...                    ...              ...                            ...   \n",
       "1921664               8399            85540                             57   \n",
       "1921665               1041            85540                             11   \n",
       "1921669               7911            85540                             11   \n",
       "1921671              10215            85540                             57   \n",
       "1921675               1212            85540                             18   \n",
       "\n",
       "         contributing_factor_vehicle_2  vehicle_type_code_1  \\\n",
       "406910                              57                  528   \n",
       "384051                              57                  538   \n",
       "187494                              57                  528   \n",
       "165946                              57                  538   \n",
       "1716583                             57                  417   \n",
       "...                                ...                  ...   \n",
       "1921664                             29                  508   \n",
       "1921665                             29                  417   \n",
       "1921669                             57                  508   \n",
       "1921671                             57                  417   \n",
       "1921675                             57                  417   \n",
       "\n",
       "         vehicle_type_code_2 hour  year  month  day_of_week  serious_accident  \n",
       "406910                   604   11  2019      6            3             False  \n",
       "384051                   618   17  2019      7            1             False  \n",
       "187494                   604   19  2021     10            5             False  \n",
       "165946                   604   16  2020      4            3             False  \n",
       "1716583                  473   15  2013      7            5             False  \n",
       "...                      ...  ...   ...    ...          ...               ...  \n",
       "1921664                  456   15  2012      7            1              True  \n",
       "1921665                  456   15  2012      7            6              True  \n",
       "1921669                   61   18  2012      7            0              True  \n",
       "1921671                   41    0  2012      7            0              True  \n",
       "1921675                  580   17  2012      7            1              True  \n",
       "\n",
       "[822274 rows x 16 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_downsampled1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "9126b9f1-5b74-4b5e-ad23-4cc67eb0e2a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,9))\n",
    "ax = sns.heatmap(df_downsampled1.corr(), annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "29c02d5e-eb8f-4abf-b6b3-0db8d28ba60a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "borough                          0\n",
       "zip_code                         0\n",
       "latitude                         0\n",
       "longitude                        0\n",
       "on_street_name                   0\n",
       "cross_street_name                0\n",
       "off_street_name                  0\n",
       "contributing_factor_vehicle_1    0\n",
       "contributing_factor_vehicle_2    0\n",
       "vehicle_type_code_1              0\n",
       "vehicle_type_code_2              0\n",
       "hour                             0\n",
       "year                             0\n",
       "month                            0\n",
       "day_of_week                      0\n",
       "serious_accident                 0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_downsampled1.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "12bbd930-25d9-4eab-ae5e-d0c1aa2c301f",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = df_downsampled1.drop('serious_accident', axis=1)\n",
    "y = df_downsampled1['serious_accident']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "abcd34ce-55bc-4bdf-a1f6-05a81b06f941",
   "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>borough</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>on_street_name</th>\n",
       "      <th>cross_street_name</th>\n",
       "      <th>off_street_name</th>\n",
       "      <th>contributing_factor_vehicle_1</th>\n",
       "      <th>contributing_factor_vehicle_2</th>\n",
       "      <th>vehicle_type_code_1</th>\n",
       "      <th>vehicle_type_code_2</th>\n",
       "      <th>hour</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day_of_week</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>358617</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>40.808784</td>\n",
       "      <td>-73.934790</td>\n",
       "      <td>6251</td>\n",
       "      <td>10159</td>\n",
       "      <td>85540</td>\n",
       "      <td>45</td>\n",
       "      <td>60</td>\n",
       "      <td>538</td>\n",
       "      <td>604</td>\n",
       "      <td>15</td>\n",
       "      <td>2019</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>333653</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>40.806408</td>\n",
       "      <td>-73.940160</td>\n",
       "      <td>4591</td>\n",
       "      <td>10159</td>\n",
       "      <td>85540</td>\n",
       "      <td>21</td>\n",
       "      <td>57</td>\n",
       "      <td>528</td>\n",
       "      <td>604</td>\n",
       "      <td>19</td>\n",
       "      <td>2022</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>578291</th>\n",
       "      <td>2</td>\n",
       "      <td>62</td>\n",
       "      <td>40.779434</td>\n",
       "      <td>-73.947520</td>\n",
       "      <td>4</td>\n",
       "      <td>5907</td>\n",
       "      <td>85540</td>\n",
       "      <td>18</td>\n",
       "      <td>57</td>\n",
       "      <td>528</td>\n",
       "      <td>70</td>\n",
       "      <td>14</td>\n",
       "      <td>2018</td>\n",
       "      <td>9</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1803451</th>\n",
       "      <td>1</td>\n",
       "      <td>144</td>\n",
       "      <td>40.619749</td>\n",
       "      <td>-74.027715</td>\n",
       "      <td>1755</td>\n",
       "      <td>1338</td>\n",
       "      <td>85540</td>\n",
       "      <td>28</td>\n",
       "      <td>28</td>\n",
       "      <td>60</td>\n",
       "      <td>473</td>\n",
       "      <td>13</td>\n",
       "      <td>2013</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1495258</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>40.840827</td>\n",
       "      <td>-73.945958</td>\n",
       "      <td>9315</td>\n",
       "      <td>5212</td>\n",
       "      <td>85540</td>\n",
       "      <td>4</td>\n",
       "      <td>34</td>\n",
       "      <td>417</td>\n",
       "      <td>473</td>\n",
       "      <td>16</td>\n",
       "      <td>2014</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         borough  zip_code   latitude  longitude  on_street_name  \\\n",
       "358617         3         1  40.808784 -73.934790            6251   \n",
       "333653         3         1  40.806408 -73.940160            4591   \n",
       "578291         2        62  40.779434 -73.947520               4   \n",
       "1803451        1       144  40.619749 -74.027715            1755   \n",
       "1495258        3         1  40.840827 -73.945958            9315   \n",
       "\n",
       "         cross_street_name  off_street_name  contributing_factor_vehicle_1  \\\n",
       "358617               10159            85540                             45   \n",
       "333653               10159            85540                             21   \n",
       "578291                5907            85540                             18   \n",
       "1803451               1338            85540                             28   \n",
       "1495258               5212            85540                              4   \n",
       "\n",
       "         contributing_factor_vehicle_2  vehicle_type_code_1  \\\n",
       "358617                              60                  538   \n",
       "333653                              57                  528   \n",
       "578291                              57                  528   \n",
       "1803451                             28                   60   \n",
       "1495258                             34                  417   \n",
       "\n",
       "         vehicle_type_code_2 hour  year  month  day_of_week  \n",
       "358617                   604   15  2019      8            4  \n",
       "333653                   604   19  2022      4            5  \n",
       "578291                    70   14  2018      9            3  \n",
       "1803451                  473   13  2013      2            5  \n",
       "1495258                  473   16  2014      8            2  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train2, x_test2 , y_train2 ,y_test2 = train_test_split(x,y,test_size =0.3, random_state=5)\n",
    "\n",
    "# print the data\n",
    "x_train2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "19172a62-a846-476f-b165-eab64cfe9df3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train:  (575591, 15)\n",
      "x_test:  (246683, 15)\n",
      "y_train:  (575591,)\n",
      "y_test:  (246683,)\n"
     ]
    }
   ],
   "source": [
    "print('x_train: ', x_train2.shape)\n",
    "print('x_test: ', x_test2.shape)\n",
    "print('y_train: ', y_train2.shape)\n",
    "print('y_test: ', y_test2.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "f48a6ea0-5dfd-4126-8d2b-e2bed2cc59b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "ada3 = AdaBoostClassifier(n_estimators=1000, learning_rate=0.1, random_state=10)\n",
    "model3 = ada3.fit(x_train2,y_train2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "cec5bc03-38f7-4958-8acf-894737d22c8c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True, False, False, ...,  True, False,  True])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred3 = model3.predict(x_test2)\n",
    "pred3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "a73550dd-7a84-4b99-85e1-1ca6e28dc2ad",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "67.51377273667015"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_test2, pred3)*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "d5d619a7-d434-4c28-a0fc-2ac99f446ddf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       False       0.65      0.74      0.70    123286\n",
      "        True       0.70      0.61      0.65    123397\n",
      "\n",
      "    accuracy                           0.68    246683\n",
      "   macro avg       0.68      0.68      0.67    246683\n",
      "weighted avg       0.68      0.68      0.67    246683\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "#print classification report for model\n",
    "print(classification_report(y_test2, pred3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "097ca79c-ab04-43d1-9c75-2d5fe90d807e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfcAAAD4CAYAAADxVK9GAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAAAw9UlEQVR4nO3deZxcVZ338c+XsBNIWKKDCEQBWcKSmLATZMk44wbhIRggAgFHHhRZ5AVDRhhFlBFkBgYRDctAWDKCAcNEVAKyhLAEspCFAAFNcGR52CRg2Am/549zmtxUqrqruru6uqu/79erX1Sfe+6551ZrTp17b52vIgIzMzNrHqs1ugNmZmbWuTy4m5mZNRkP7mZmZk3Gg7uZmVmT8eBuZmbWZFZvdAfMNtlkkxg4cGCju2Fm1qPMnj37lYgYUG6bB3druIEDBzJr1qxGd8PMrEeR9OdK23xZ3szMrMl4cDczM2syHtzNzMyajAf3BpJ0jqTT63yM7STNlfSopK3qeazCMe+VNKwrjmVmZqvy4N78RgI3R8SQiPhToztjZmb158G9i0k6S9JTku4Hts1l35A0U9I8SbdIWlfS+pKWSFoj19mg+HuZdgdLmiFpvqTJkjaU9EXgVOCbku6psN8Zkk7Ory+WdHd+fYCkifn15yU9JGmOpEmS+ubyoZKmSZotaaqkTUvaXk3SBEk/6oz3zszMquPBvQtJGgocDgwGvgjsmjf9OiJ2jYhdgCeAr0fE34B7gS/lOofneu9XaP464MyI2BlYAHw/In4HjAcujoj9K+w3HRieXw8D+uYPEMOB+yRtApwNjIiIzwKzgNNynUuBURExFLgaOK/Q7urARODpiDi7zHtxvKRZkma9/PLLFbpmZmbt4e+5d63hwOSIeAtA0pRcvmOe3fYH+gJTc/lVwD8DtwLHAt8o16ikfkD/iJiWi64FJlXZp9nAUEkbAO8Cc0iD/HDgZGAPYAfgAUkAawIPka467Ajcmcv7AC8U2r0c+FVEFAf8j0TEFcAVAMOGDXPusJlZJ/Lg3j1MAEZGxDxJY4H9ACLiAUkDJe0H9ImIxzr7wBHxvqQlwFjgQWA+sD+wNekqwlbAnRFxRHE/STsBCyNizwpNPwjsL+k/IuKdzu63mZlV5svyXes+YKSkdSStD3wll68PvJAvdY8p2ec64L+Bayo1GhGvA69Jarm8fhQwrVL9MqYDp+f+TQdOAB6NiABmAHtL2hpA0nqSPgMsAgZI2jOXryFpUKHN/wJ+B/xKkj9Empl1IQ/uXSgi5gA3AfOA3wMz86Z/BR4GHgCeLNltIrAh8Ms2mj8GuFDSfNI9/XNr6Np0YFPgoYh4EXgnlxERL5Nm9b/MbT8EbBcR7wGjgAskzQPmAnuVnO9FwKPA9ZL8vzUzsy6iNDmz7krSKODgiDiq0X2pl2HDhoXXljczq42k2RFRdk0RXy7txiRdCnyB9GS9mZlZVTy4d2MRcVJpmaTLgL1Lii+JiIr35PN+GwN3ldl0YES82v5emplZd+PBvYeJiBPbud+rpHvxZmbW5PyQk5mZWZPx4G4Nt+C51xvdBTOzpuLB3czMrMl4cG9Ckn4nqX8XH9Mxr2Zm3YQfqGtCEeGvzpmZ9WKeufdwkk6QNDf/LJF0j6RnJG2S16V/UtJESU9IulnSuq20taukB3P07CM5dnZtSddIWiDpUUn757rrSLoxtzsZWKfQTtmI2JJjfZQKt/wt33M3M+tMHtx7uIgYHxGDSfGxzwIXlVTZFvh5RGwPvAF8q1w7ktYkLY17So6eHQG8DZyYDhM7AUcA10paG/gm8FZu9/vA0NxO2YjYMv2+IiKGRcSwPuv268hbYGZmJTy4N49LgLsj4jcl5X+JiAfy6xuAfSrsvy3wQkTMBIiINyLig1z/hlz2JPBn4DPAvoXy+aQ0OVg5InYuac37LTt8dmZmVjXfc28COSZ2S+DbZTaXhgfUO0xAlImINTOzruOZew8naSgprvVrEfFhmSpbtMSyAkcC91doahGwqaRdc7vr56jW6eQY2hz1ukWue19uD0k7AjvndipFxFa002a+LG9m1pk8uPd83wY2Au7JD9VdVbJ9EXCipCdI0bG/KNdIjnAdDVyaI1zvBNYGfg6sJmkB6Z782Ih4N7fTN7d7LjA7t1M2IrYTz9fMzNrgyNcmJmkgcFtE7NjovrTGka9mZrVrLfLVM3czM7Mm4wfqmlhEPAOsMmvP30v/VEnxmRExtSv6ZWZm9eXBvReKiEMa3QczM6sfX5Y3MzNrMh7cuwlJ3+3O7dXTgudeZ+C43za6G2ZmTcODewX5O95dqexgrKQ9f6ceM7ibmVnn6tWDu6SjJc3PQSnXS5ogabykh4GfSBosaUauM1nShnm/kyU9nstvzGWfKwS4PCpp/QrH3FTSfbneY5KGSzofWCeXTcyBL4skXQc8Bmwu6QxJM/Mxf1Bo72s55GWupMsl9Sltr0I/BubQlyslLZR0h6R18rZv5GPNk3RLS9hMfn9+kd+TxZL2k3R1bmdCoe02g2PMzKx+eu3gLmkQKeDkgByUckre9Elgr4g4DbiO9BT5zsACUkAKwDhgSC4/IZedDpyYQ1yGk0JXyjkSmJrr7QLMjYhxwNsRMTgixuR625ACXwaR1n3fBtgNGAwMlbSvpO1JC8/sndtbDoyp0F452wCX5WMsBQ7N5b+OiF3z+/IE8PXCPhsCewLfAaYAFwODgJ3yh6GqgmPMzKx+evPT8gcAkyLiFYCI+KskctlySf2A/hExLde/FpiUX88HJkq6Fbg1lz0AXJRnyr+OiGcrHHcmcLWkNYBbI2JuhXp/jogZ+fXn88+j+fe+pIF5Z1Ia28zc93WAl6o7fQCWFI4/GxiYX+8o6UdA/3ys4lfkfhMRkVesezEiFgBIWpj3/yQrgmMA1iStUrcSSccDxwP02WBADV02M7O29NqZeyverKLOl4DLgM+SBtbVI+J84J9IA+wDksouuRoR95ES1Z4DJkg6uop+CPhxnokPjoitI+K/cvm1hfJtI+Kcak4ye7fwejkrPuxNAL6dY15/QFqGtnSfD0v2/zDv3xIc09KnHSKiOPMHHPlqZlZPvXlwvxs4TNLGAJI2Km6MiNeB1yQNz0VHAdPyw22bR8Q9wJlAP9Ia61tFxIKIuIA0Oy87uEvakjTjvRK4ivQBAeD9PJsvZypwXMu9a0mbSfoYcBcwKr9G0ka5/bbaa8v6wAt5/9Yu65dTc3CMmZl1rl57WT4iFko6jzRgL2fFJe+iY4Dx+YGyxcCxQB/ghnzZXsBPI2KppB9K2p80g10I/L7CofcDzpD0PrAMaJm5XwHMlzQHOKukr3fk++sP5Uvdy0gpcI9LOhu4I3/oeB84kZS5/lF7bdx3L+dfgYeBl/N/yz4cWE5EvKwUQftLSWvl4rOBpyrts9Nm/Zh1/pdq7KKZmVXi4BhrOAfHmJnVTg6OMTMz6z167WX5epO0E3B9SfG7EbF7F/djY9K9+VIHRsSrXdkXMzPrGh7c6yR/RWxwN+jHq92hH2Zm1nV8Wd7MzKzJeHA3MzNrMr4sbw1XKRXuGX89zsysXTxzbxI5COaxbtCP/STd1uh+mJn1Zh7cDUl9Gt0HMzPrPB7cm8vqOTL2CUk3S1pX0oE5gnZBjmddC0DSM5IuyCviHSbpiFznMUkXtDQoaVnh9aiWaFdJW+Xo1wWSflSsR1qO92ZJT+b+qIvO38zM8ODebLYlxcRuD7xBilqdAIzOITCrA98s1H81x7LeB1xASsobDOwqaWQbx7oEuCS3W5qANwQ4lZQO92lg79KdJR0vaZakWcvfer2GUzQzs7Z4cG8uf4mIB/LrG4ADSbGuLeu6X0tKpGtxU/7vrsC9EfFyRHwATCypV86erIjA/e+SbY9ExLMR8SEwlxVRsh9xKpyZWf14cG8upUEBS9uoX028bbHNtSvWWlmlKFkzM+sCHtybyxaS9syvjwRmAQNb4lfJsbVl9nsE+JykTfLDdUcU6r0oafucOndIYZ8ZwKH59eGdeRJmZtYxnlE1l0XAiZKuBh4HTiYNwpMkrU7KmR9fulNEvCBpHHAPKcb2txHxP3nzOOA2UvzrLKBvLj+VFH17FnA70O4b5458NTPrXI58tXbJGfdvR0RIOhw4IiIObk9bjnw1M6tda5Gvnrlbew0Ffpa/5rYUOK6x3TEzsxYe3K1dImI6sEuj+2FmZqvyA3VmZmZNxoO7mZlZk/HgbmZm1mR8z906RFJ/4MiI+Hn+fT/g9Ij4crVtVIp8beHoVzOz2njmbh3VH/hWozthZmYreHDvRXLm+5OSJkh6Kie2jZD0gKSnJe0maSNJt0qan1Pfds77npNT5e6VtFjSybnZ84GtJM2VdGEucyqcmVkD+bJ877M1cBjpe+kzScvU7gMcBHwX+AvwaESMlHQAcB0pKQ5gO2B/YH1gkaRfkFaw2zEiBsNHl+WHAIOA54EHSKlw99f9zMzMDPDMvTdaEhELcmLbQuCuSMsULiClt+0DXA8QEXcDG0vaIO/724h4NyJeAV4CPl7hGG2mwjny1cysfjy49z7FxLYPC79/SNtXcqpNe2uzniNfzczqx4O7lZoOjIGPLrG/EhFvtFL/b6TL9GZm1k34nruVOge4WtJ84C3gmNYqR8Sr+YG8x4DfA5W/01aBU+HMzDqXU+Gs4ZwKZ2ZWu9ZS4XxZ3szMrMl4cDczM2syHtzNzMyajAd3MzOzJuPB3czMrMn4q3DWcG2lwoGT4czMauGZezcgaT9Je3XX9szMrGfx4N497AeUHYwltefqSsX2zMys+Xlwr5Gk0yQ9ln9OzTGqT0i6UtJCSXdIWqeV/U+W9HiOVL1R0kDgBOA7OTZ1eI5kHS/pYeAnkraSdLuk2ZKmS9outzVA0i2SZuafvcu1V6EfEyT9VNKDOcJ1VC7vK+kuSXMkLZB0cC5vMy4211svR8M+IunRlv3NzKzr+J57DSQNBY4FdgcEPAxMA7YBjoiIb0j6FXAocEOFZsYBn4qIdyX1j4ilksYDyyLi3/Nxvg58EtgrIpZLugs4ISKelrQ78HPgAOAS4OKIuF/SFsDUiNi+tL1WbEpKgdsOmALcDLwDHBIRb0jaBJghaUqu31Zc7EjgLODuiDhOUn/gEUl/iIg3S97L44HjAfpsMKCNbpqZWS08uNdmH2Byy0Al6dfAcFKM6txcZzZlIk4L5gMTJd0K3NpKvUl5YO9LusQ+SVLLtrXyf0cAOxTKN8j1q3VrjmV9XFJLfKuAf5O0LykpbjNWRLsuiYgFAJI+iouV1BIXC/B54CBJp+ff1wa2AJ4oHjgirgCuAFhr0228BrKZWSfy4N45SiNOK16WB74E7At8BThL0k4V6rXMdFcDlkbE4DJ1VgP2iIh3ioWFwb4txX637DQGGAAMjYj3JT1DGqBL61eKixVwaEQsqrYTZmbWuXzPvTbTgZGS1pW0HnBILquKpNWAzSPiHuBMoB/Ql1ZiU3Pc6hJJh+U2JGmXvPkO4KRC+4Pzy47EsPYDXsoD+/7AljXuPxU4SfkThqQh7eyHmZm1k2fuNYiIOZImAI/koquA12poog9wg6R+pBnuT/M9998AN+eHz04qs98Y4BeSzgbWAG4E5gEnA5fleNbVgftID9Ot1F5EVP0BBJgI/CZfap8FPFnDvgA/BP4TmJ8/zCwBvtzaDo58NTPrXI58tYZz5KuZWe0c+WpmZtaL+LJ8nUi6DNi7pPiSiLimi/txFunra0WTIuK8ruyHmZl1HQ/udRIRJza6DwB5EPdAbmbWi/iyvJmZWZPx4G5mZtZkfFm+h5G0LCJqWYWumjYPAnaIiPMljQSeiojHa2zjXuD0iKj5sfdqIl/by1GxZtYbeeZuRMSUiDg//zoS2KGB3TEzsw7y4N5D5ZXqLszpdAskjc7l+0m6V9LNOcVtYmG1uC/mstk5Ee62XD5W0s+UMuAPAi7MiXJb5baG5Xqb5OVokbSOUqrdE5ImU1hyV9LnJT2Uk+Um1bjevZmZdZAvy/dc/wcYDOwCbALMlHRf3jYEGAQ8DzwA7C1pFnA5sG9ELJH0y9IGI+LBnAB3W0TcDK2uU/9N4K2cQrczMCfX3wQ4GxgREW9KOhM4DTi3E87ZzMyq4MG959oH+GVELAdelDQN2BV4A3gkIp4FkDSXlNi2DFgcEUvy/r8kR662077ATwEiYn5eAhdgD9Jl/QfyB4M1gYdKd3bkq5lZ/Xhwb06lKXUd+Tt/wIrbN2u3VjETcGdEHNFaJUe+mpnVj++591zTgdGS+kgaQJpJP9JK/UXApyUNzL+PrlCvNFHuGWBofj2qUH4fcCSApB2BnXP5DNJtgK3ztvUkfaaaEzIzs87hmXvPNRnYk5QOF8A/R8T/k7RducoR8bakbwG3S3oTmFmh3RuBKyWdTBrM/x34Vb6MXvy+2i+AayQ9ATwBzM7HeVnSWOCXktbKdc8Gnqp0Ik6FMzPrXE6F60Uk9Y2IZfnp+cuApyPi4kb3y6lwZma1cyqctfhGfsBuIdCP9PS8mZk1GV+W70XyLL3hM3UzM6svz9zNzMyajAd3MzOzJuPB3czMrMn4nrs1XD1T4cpxUpyZNTvP3JuQpGVtbO+fv/Pe8vsnJLWsJT9Y0hfbccxzJJ1ee2/NzKyzeXDvnfoDHw3uEfF8RLSsPjcYqHlwNzOz7sODexOT1FfSXTl6dYGkg/Om84GtcqzrhZIG5ujYNUnpbaPzttGlM/Jcb2B+fZakpyTdD2xbqLOVpNtztOz0SqvmmZlZffiee3N7BzgkIt7IUawzcqTrOGDHiBgM0DJYR8R7kr4HDIuIb+dt55RrWNJQ4HDSTH91UuTr7Lz5CuCEiHha0u7Az4EDSvZ3KpyZWZ14cG9uAv5N0r7Ah8BmwMc7qe3hwOSIeAsgf2hAUl9gL2BSIQt+rdKdnQpnZlY/Htyb2xhgADA0It6X9AzVxbYWFSNfqWL/1YClLVcFzMys6/mee3PrB7yUB/b9gS1zeWmsa1G5yNfPAkj6LPCpXH4fMFLSOpLWB74CEBFvAEskHZb3kaRdOu+UzMysLZ65N7eJwG8kLQBmAU8CRMSrkh6Q9Bjwe1JCXIt7gHE5YObHwC3A0ZIWAg+To1sjYo6km0iRsy+xcoTsGOAXks4G1iDFyM6r1ElHvpqZdS5HvlrDOfLVzKx2jnw1MzPrRTy4m5mZNRkP7mZmZk3Gg7uZmVmT8eBuZmbWZPxVOFtJXor2tojYsauO2dWRr+DYVzNrbp65W91J8odIM7Mu5MHdyukj6UpJCyXdkVehGyxphqT5kiZL2hBA0r2ShuXXm+QlbpE0VtIUSXcDdzXuVMzMeh8P7lbONsBlETEIWAocClwHnBkROwMLgO9X0c5ngVER8bl6ddTMzFblwd3KWRIRc/Pr2cBWQP+ImJbLrgX2raKdOyPir+U2SDpe0ixJs5a/9XqHO2xmZit4cLdy3i28Xg70b6VuMTWuNDHuzUo7RcQVETEsIob1WbdfuzppZmbleXC3arwOvCZpeP79KKBlFv8MMDS/HtXF/TIzszL8FLNV6xhgvKR1gcXAsbn834FfSToeaNf32ZwKZ2bWuZwKZw3nVDgzs9o5Fc7MzKwX8eBuZmbWZDy4m5mZNRkP7mZmZk3Gg7uZmVmT8VfhrOEakQpXidPizKwZeOZuZmbWZDy4W91J6tPoPpiZ9SYe3G0lks6VdGrh9/MknSLpDEkzc+TrDwrbb5U0O8fDHl8oXybpPyTNA/bs2rMwM+vdPLhbqauBowEkrQYcDvw/UgzsbsBgYKikllS44yJiKDAMOFnSxrl8PeDhiNglIu4vPYhT4czM6scP1NlKIuIZSa9KGgJ8HHgU2BX4fH4N0Jc02N9HGtAPyeWb5/JXSWlyt7RynCuAKwDW2nQbr4FsZtaJPLhbOVcBY4G/I83kDwR+HBGXFytJ2g8YAewZEW9JupcVsa/vRMTyLuqvmZkV+LK8lTMZ+EfSjH1q/jlOUl8ASZtJ+hjQD3gtD+zbAXs0qsNmZraCZ+62ioh4T9I9wNI8+75D0vbAQ5IAlgFfA24HTpD0BLAImNGe4zny1cysc3lwt1XkB+n2AA5rKYuIS4BLylT/Qrk2IqJvfXpnZmZt8WV5W4mkHYA/AndFxNON7o+ZmdXOM3dbSUQ8Dny60f0wM7P288zdzMysyXhwNzMzazIe3M3MzJqM77lbwzny1cysc3nm3kGSTpb0hKSJktaS9AdJcyWNrqGNkfkp9c7qU6e2Z2ZmPYsH9477FvD3ETEGGAIQEYMj4qYa2hgJlB2MJbXn6krF9szMrPl5cK+BpNMkPZZ/TpU0nvS1sd9LOhO4Adg1z9y3qtDG+ZIez9Gp/y5pL+Ag4MKW/STdK+k/Jc0CTpE0VNK0HK06VdKmua2tJN2ey6dL2q5cexX6ca+kCyQ9IukpScNz+cDc1pz8s1cu3y/34X8kLc7nMSbvv6DlOJIGSLolx8POlLR3p/4RzMysTb7nXiVJQ4Fjgd0BAQ+TlmD9R2D/iHhF0sPA6RHx5QptbAwcAmwXESGpf0QslTQFuC0ibs71ANaMiGGS1gCmAQdHxMv5cv95wHGkVLUTIuJpSbsDP4+IA0rba8XqEbGbpC8C3yeFwLxEuhLxjqRtgF+S4lwBdgG2B/4KLAauyvufApwEnEpaxe7iiLhf0hakdem3L/NeHA8cD9BngwFtdNPMzGrhwb16+wCTI+JNAEm/BobX2MbrwDvAf0m6Dbitlbotl/W3BXYE7syDfh/ghRzishcwKZcDrFVjf36d/zsbGJhfrwH8TNJgUmzrZwr1Z0bECwCS/gTckcsXAPvn1yOAHQp92kBS34hYVjywI1/NzOrHg3sXiogPJO1GilAdBXwbOKBC9TfzfwUsjIg9ixslbUAKdhncgS69m/+7nBX/W/gO8CJplr4a6cNIaX2ADwu/f1jYfzVgj4go7mdmZl3Ig3v1pgMTJJ1PGnAPAY4CTqu2gTzbXjcififpAdKlbYC/AetX2G0RMEDSnhHxUL5M/5mIWChpiaTDImKS0lR554iY10Z7bekHPBsRH0o6hnSloBZ3kC7RXwggaXBEzG1tB6fCmZl1Lj9QV6WImANMAB4h3W+/KiIerbGZ9YHbJM0H7mfFB4MbgTMkPVr6AFxEvEea5V8gaR4wl3Q5HmAM8PVcvhA4uK32qvBz4Jjc5nasuIJQrZOBYfmBwceBE2rc38zMOkgRvt1pjTVs2LCYNWtWo7thZtajSJodEcPKbfPM3czMrMn4nnudSJoMfKqk+MyImNrF/bgMKP2u+SURcU1X9sPMzLqOB/c6iYhDGt0HgIg4sdF9MDOzruXL8mZmZk3GM3druO6UCtdbOQ3PrLl45m5mZtZkumRwz2EkR7ay/ROSWtZVHyvpZzW2P1bSJwq/X9XZkaeSLpS0UNKF7dj3u53ZlxqPvaxC+QmSjm5lv4GSHmvH8QZLeii/V/NVQ/StmZl1jq66LD8QOBL479INklaPiOdJC7W011jgMeB5gIj4pw60VcnxwEYRsbwd+34X+LdqK+fV5hQRH7bjWFWJiPF1avot4OgcZvMJYLakqRGxtE7HMzOzElXN3CUdnWdh8yRdn2d1d+eyu3L6F5ImSPqppAdzLGjLgH0+MDxHkH4nz7SnSLobuKvMLHHzHEn6tKTv57ZXqiPpdEnn5GMMAybm9tfJ+w7L9ZZJOi/3fYakj+fyrfLvCyT9qNIMN9edAvQlDVSjJX1F0sN5Bbg/FNrsK+ma3OZ8SYfm5WrXyX2bmOutFB1bOL9Fkq4jfVDZvEw/TiheOShe5ZD0NaX41bmSLpfUp1Cv3PmfI+n0/HrrfB7zlGJetyo5bp985WJmPq//W+m9ioinIuLp/Pp5UsrcKrFvko6XNEvSrOVvvV6pOTMza4c2B3dJg4CzgQMiYhfgFOBS4NqI2BmYCPy0sMumpAS1L5MGdYBxwPSIGBwRF+eyzwKjIuJzZQ67G3AosDNwWMtAXU6ONZ0FjMntv11SZT1gRu77fcA3cvklpO977wQ829p7EBEHAW/n9m8iLR27R0QMIS31+s+56r8Cr0fETvm9uTsixhX2HaOVo2P3AL4haUjefxtSbOugiPhzma7cQlrTvsVo4EZJ2+fXe+cgmeWkpWlbO/+iicBluc5ewAsl27+ez2tXYNfc59Lv8K9CKSRnTeBPpdsi4oqIGBYRw/qs26+tpszMrAbVzNwPACZFxCsAEfFXYE9WXGK/njSYt7g1Ij6MiMeBj7fS7p25rUrbXs0D9a9L2q/Ve6yIVi1Gm+4JTMqvV7ld0IZPAlMlLQDOAAbl8hHAZS2VIuK1Mvt+FB2bY1CL0bF/jogZlQ4aES8DiyXtoZQNvx3wACllbigwU9Lc/Pun826Vzh8ASesDm0XE5HyMdyLirZJDfx44Orf9MLAx6YNIRZI2Jf1v49h63l4wM7NV1eOeezEWVBVrtR5IUrrgfQAfsPKHkbWr7M/7sWIB/WK0aUdcClwUEVMk7Qec0wltQnUhLTcCXwWeJH1IiHyP/tqI+Jcy9Tvj/AWcVO3qekpxtL8Fzmrtw4qZmdVHNf/Q3w1MlnRRRLwqaSPgQeBw0sxsDCkOtTW1RpD+fT7O28BI4DhSxvjH8ox1Gemy/+3tbB9gBunS/02kc6lFP+C5/PqYQvmdwInAqQCSNsyz9/clrRER71M5OrZak4GzgCHAmbnsLuB/JF0cES/l9279Cpf2VxIRf5P0rKSREXGrpLVYNeZ1KvBNSXdHxPuSPgM8FxGrfBiRtGbu43X5lkmbHPlqZta52rwsHxELgfOAaUoxoBeR8rqPVYouPYp0H74184Hl+YGt71TRr0dI95fnA7dExKw8MJ6bt91Jmrm2mACMb3mgror2IQ3Ap+Vz2Bqo5amuc4BJkmYDrxTKfwRsmB+Umwfsn8uvAOZLmtjR6Nj8YeEJYMuIeCSXPU56LuKOfD53kp59qNZRwMl53weBvyvZfhXwODAnP9R4OZU/GH4V2BcYm/8ecyUNrqEvZmbWQb028lXSuqQH3ULS4cAREXFwW/tZ53Pkq5lZ7dRK5GtvXn52KPCzfL96KenSv5mZWY/Xawf3iJgO7FIsk7QT6TmConcjYvcu69iKvjwMrFVSfFRELOjqvpTTnd4rMzNbWa8d3MvJA+fgRvcDoLsPkt3pvTIzs5U5OMbMzKzJeOZuDefIV2s2jtC1Ruu1M3e1knom6VxJI1rZdz9Jt1Xa3sp+gyV9sdb9ulpr700V+94uaWl73h8zM+scvXZwb01EfC8i/lCHpgcD3X5w76ALqW1RHjMz62RNNbhLOl/SiYXfz1FKjzujkGj2g8IufSRdqZQ9fkfLAjhK6Xaj8utdlVLu5uXUtfVLjrmepKvztkcllf2ufF657VxgdF7YZbRS6t2AvH01SX+UNCAff3xOTXtK0pdznarT2XL9M5US6ublFfFarh7MyPtPlrRhLh+a680jrbLX0kZNx4yIu0grBpqZWYM01eBOWkr2q4Xfvwq8TAo52Y00cx4qad+8fRtSGtog0nfdDy02lgfkm4BTcmLaCNKSuEVnkdLfdiOtSHehpPVKOxYR7wHfA24qpMvdwIr0thHAvBwOAyngZTfgS6TV99amhnQ2SV8ADgZ2z33/Sd50HXBmTq1bAHw/l19DWj9+l5Km2pUI1xY58tXMrG6aanDPy7h+TNInJO0CvAbsREo1exSYQ0pSa0k0WxIRc/PrVRLTgG2BFyJiZm7/jYj4oKTO54FxSolp95ICbbaosstXA0fn18eRBtgWv8rpek8Di3O/a0lnGwFc05LwFhF/ldQP6B8R03Kda4F9JfXP5ffl8uL312tOhKuGI1/NzOqnGZ+WnwSMIq2PfhOwJfDjiLi8WEnSQFZOsFsOVLsu/UpNAYdGxKJad4yIv0h6UdIBpFn6mOLm0urUmM7WSRpxTDMz64BmHNxvAq4ENgE+R5q5/zCHtiyTtBnwfpVtLQI2lbRrRMzM99tLL8tPBU6SdFJep35IK0Ew5dLrriJdnr8+IpYXyg+TdC3wKVI2+yJqSGcjhcd8L5/3W5I2yrP31yQNzyv0HQVMi4il+Qn3fSLiflb+kFHLMdvFqXBmZp2r6Qb3iFiYB+HnIuIF4AVJ2wMPpWXkWQZ8jTRTb6ut9ySNBi7ND9u9TbrcXfRD4D9JqW+rAUtIcbTl3MOKS/g/zvfdp5Aux19TUvd/SclxGwAnRMQ7kq4i3TqYk9fEf5kUiVuu77crpbHNkvQe8Dvgu6SI2vFKwTmLgWPzLscCV0sK4I5CU1UfE0DSdNIthL6SngW+7lm/mVnX6rWpcN2FpGHAxRExvFA2Abit2jz0ns6pcGZmtZNT4bonSeOAb7LyZXAzM7MO8eBeB5L+AbigpHhJRBxSLIiI84HzS/ePiLE1HKvL09kacUwzM6ueB/c6yPeYu+Q+cyPS2ZwIZ2bWvTXV99zNzMzMM3frBpwKZ1Z/TqrrXTxzNzMzazJdMrgrRYge2cr2T0i6Ob8eK+lnNbY/VtInCr9fJWmH9ve47DEuzAEzF7Zj3+92Zl9qPPayCuUnSDq63La83bGvZmY9VFddlh8IHAn8d+kGSatHxPOkJWPbayzwGPA8QET8UwfaquR4YKOSVeSq9V3g36qtnBeLUUR82I5jVSUixterbVLs67pAqwlyZmZWH1XN3CUdneM+50m6Ps/q7s5ld0naItebIOmnShGpi5VjU0lf9xquFHX6nTzTniLpbuCuMrPEzSXdqxSJ+v3c9kp1lKJcz8nHGAZMzO2vk/cdlustk3Re7vsMSR/P5Vvl3xdI+lGlGW6uOwXoC8xWimr9iqSHlSJe/1Bos6+ka3Kb8yUdqhS1uk7u28Rc7zRJj+WfUwvnt0jSdaQPKpuX6ccJxSsHxasckr6mFDs7V9LlkvoU6pU7/3MknZ5fb53PY56kOZK2Kjlup8e+yqlwZmZ10+bgLmkQcDZwQI4DPQW4FLg2x4ZOBH5a2GVTYB/SEqwt3+EeB0zPUacX57LPAqMi4nNlDrsbKX51Z9Ia62VX4AHIq7jNAsbk9kvXfl8PmJH7fh/wjVx+CXBJROwEPNvaexARBwFvF6Ja7wf2iIghwI3AP+eq/0qKR90pvzd3R8S4wr5jJA0lLfW6O7AHKUJ1SN5/G+DnETEoIv5cpiu3AMXvyo8GblRaXnc0sHdEDCYtrduyME6l8y+aSIq+3QXYC3ihZHunx746Fc7MrH6qmbkfAEyKiFcgRYcCe7LiEvv1pMG8xa05qvRx4OOttHtnbqvStlfzQP3rkvZr9R7Qcu+3GOu6JylBDsrcLmjDJ4GpkhYAZwCDcvkI4LKWShHxWpl99wEmR8SbEbGMdH4tS8/+OSJmVDpoznpfLGkPSRuT1nB/ADgQGArMVFq3/kBS2AxUPn8AlNbh3ywiJudjvNMSE1tQl9hXMzOrj3rccy/GqKqVeq2lipWLO/2AlT+MrF1lf96PFQvoL6dzzvlS4KKImCJpP+CcTmgTWn9PWtwIfBV4kvQhIfI9+msj4l/K1O+M83fsq5lZD1LNP/R3A5MlXRQRr0raCHgQOJw0ax8DTG+jjXJRp635+3yct0kJZMcBLwIfyzPWZaTL/re3s32AGaRL/zeRzqUW/YDn8utjCuV3AicCpwJI2jDP3t+XtEZEvE96rybke/EiXWY/qoZjTwbOAoYAZ+ayu4D/kXRxRLyU37v1K1zaX0lE/E3Ss5JGRsStktYC+pRUq2vsqyNfzcw6V5uX5SNiIXAeME3SPOAi4CTgWEnzSQPTKW00Mx9Ynh/Y+k4V/XqEdH95PnBLRMzKA+O5edudpJlriwmkGNO5StGs1TgVOC2fw9ZALU91nQNMkjQbeKVQ/iNgw/yg3Dxg/1x+BSkSdmJEzMn9fYR0ifuqVvLfV5E/LDwBbBkRj+Syx0nPRdyRz+dO0rMP1ToKODnv+yDwdyXbrwIeJ8W+PgZcTisfDJViXycBB+YPDv9QQ1/MzKyDem3kq1Ke+dv5svbhwBERcXCj+9UbOfLVzKx2cuRrWUOBn+X71UtJl/7NzMx6vF47uEfEdGCXYpm6UZSppIeBtUqKj8qJbA3Xnd4rMzNbWa8d3MvpTlGm3X2Q7E7vlZmZrczBMWZmZk3GM3drOEe+mllvVM8Y3l47c1crqWeSzpU0opV991M7Es8kDZb0xVr362qtvTdt7DdY0kNK6XnzJY2uR//MzKx1nrmXERHfq1PTg0khN7+rU/uN9hZwdEQ8rRTBO1vS1IhY2uB+mZn1Kk01c5d0vqQTC7+fo5Qed0Yh0ewHhV36SLoyzzTvaFkARyndblR+vatSyt28nLq2fskx15N0dd72qKSy35WXtCZpEZ7RebGd0UqpdwPy9tUk/VHSgHz88Tk17SlJX851akpnk3SmUkLdvLwiXsvsekbef7KkDXP50FxvHmmVvZY2qj5mRDwVEU/n188DLwEDWuujmZl1vqYa3ElLyX618PtXgZdJISe7kWbOQyXtm7dvQ0pDG0T6rvuhxcbygHwTcEpOTBtBWhK36CxS+ttupBXpLpS0XmnHIuI94HvATYV0uRtYkd42ApiXw2EgBbzsBnyJtPre2tSQzibpC8DBwO657z/Jm64DzsypdQuA7+fya0jrx+9S0lS7EuEk7QasCfypwnZHvpqZ1UlTDe55GdePSfqEpF2A14CdSKlmjwJzSElqLYlmSyJibn69SmIasC3wQkTMzO2/EREflNT5PDBOKTHtXlKgzRZVdvlq4Oj8+jjSANviVzld72lgce53LelsI4BrWhLeIuKvkvoB/SNiWq5zLbCvpP65/L5cXvz+es2JcJI2zW0cGxEflqvjyFczs/ppxnvuk4BRpPXRbwK2BH4cEZcXK0kayMoJdsuBatelX6kp4NCIWFTrjhHxF0kvSjqANEsfU9xcWp3GpLPVdExJGwC/Bc5qLb7WzMzqpxkH95uAK4FNgM+RZu4/zKEtyyRtBrxfZVuLgE0l7RoRM/P99tLL8lOBkySdlNepH9JKEEy59LqrSJfnr4+I5YXywyRdC3yKlM2+iNrS2e4EvpfP+y1JG+XZ+2uShucV+o4CpkXEUklLJe0TEfez8oeMqo+Zb2NMBq6LiJsrvAercCqcmVnnarrBPSIW5kH4uYh4AXhB0vbAQ2kZeZYBXyPN1Ntq6738da5L88N2b5Mudxf9EPhPUurbasASUhxtOfew4hL+j/N99ymky/HXlNT9X1Jy3AbACRHxjqSrSLcO5iidzMukSNxyfb9d0mBglqT3SE/of5cUUTteKThnMXBs3uVY4GpJAdxRaKrqY5KecdgX2FjS2Fw2tnDrw8zMukCvTYXrLiQNAy6OiOGFsgnAbbXMfnsyp8KZmdVOToXrniSNA77JypfBzczMOsQz9zqQ9A/ABSXFSyLikDocq8vT2Tr7mJL+RnqmoKfbBHil0Z3ooGY4B2iO8/A5dB/d9Ty2jIiya4l4cLeGkzSr0qWlnqQZzqMZzgGa4zx8Dt1HTzyPpvqeu5mZmXlwNzMzazoe3K07uKLRHegkzXAezXAO0Bzn4XPoPnrcefieu5mZWZPxzN3MzKzJeHA3MzNrMh7cra4k/aOkRTmrflyZ7WtJuilvfzgH+rRs+5dcviivHdAQ7T0HSQMlvS1pbv4Z3+WdX7mfbZ3HvpLmSPpA0qiSbcdIejr/HNN1vV6ljx05h+WFv8WUruv1Kn1s6xxOk/S4pPmS7pK0ZWFbt/g75L505Dx6yt/iBEkLcj/vl7RDYVu3+Pepoojwj3/q8gP0IeW5f5qU7T4P2KGkzreA8fn14aS8e4Adcv21SOE5fwL69LBzGAg81ui/Qw3nMRDYGbgOGFUo34iUQ7ARsGF+vWFPOoe8bVkP+TvsD6ybX3+z8L+nbvF36Oh59LC/xQaF1wcBt+fX3eLfp9Z+PHO3etoN+GNELI6I94AbgYNL6hxMypUHuBk4MAfUHAzcGBHvRsQS4I+5va7WkXPoTto8j4h4JiLmAx+W7PsPwJ0R8deIeI2UOPiPXdHpEh05h+6imnO4JyLeyr/OAD6ZX3eXvwN07Dy6i2rO4Y3Cr+uxIoq7u/z7VJEHd6unzYC/FH5/NpeVrRMRHwCvAxtXuW9X6Mg5AHxK0qOSpkkaTuN05P3sSX+L1qwtaZakGZJGdmrPqlfrOXwd+H07962njpwH9KC/haQTJf0J+Alwci37NpKDY8zq5wVgi4h4VdJQ4FZJg0pmA9Z1toyI5yR9Grhb0oKI+FOjO1WJpK8Bw4DPNbovHVHhPHrM3yIiLgMuk3QkcDYpNrvb88zd6uk5YPPC75/MZWXrSFod6Ae8WuW+XaHd55Av2b0KEBGzSfflPlP3HpfXkfezJ/0tKoqI5/J/FwP3AkM6s3NVquocJI0AzgIOioh3a9m3i3TkPHrU36LgRmBkO/fteo2+6e+f5v0hXRlaTHrgpOWBlUEldU5k5YfRfpVfD2LlB1YW05gH6jpyDgNa+kx6aOc5YKPu+rco1J3Aqg/ULSE9xLVhft3l59HBc9gQWCu/3gR4mpKHp7rLOZAGuj8B25SUd4u/QyecR0/6W2xTeP0VYFZ+3S3+fWr1/BrdAf809w/wReCp/H/ys3LZuaRP8gBrA5NID6Q8Any6sO9Zeb9FwBd62jkAhwILgbnAHOAr3fxvsSvp3uGbpKsnCwv7HpfP74/AsT3tHIC9gAX5H+QFwNe78Tn8AXgx/+9mLjClu/0dOnIePexvcUnh/8P3UBj8u8u/T5V+vPysmZlZk/E9dzMzsybjwd3MzKzJeHA3MzNrMh7czczMmowHdzMzsybjwd3MzKzJeHA3MzNrMv8f+RezK204jhIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "feat_importances = pd.Series(model3.feature_importances_, index=x.columns)\n",
    "feat_importances.nlargest(20).plot(kind='barh')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cb21482-c2ed-4d1d-af5f-8f454d62af6a",
   "metadata": {},
   "source": [
    "### AdaBoost Classifier - Tune Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "8e7d21ba-6769-4e96-b7ff-297a7572b385",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "ada5 = AdaBoostClassifier(n_estimators=3000, learning_rate=0.1, random_state=10)\n",
    "model5 = ada5.fit(x_train2,y_train2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "0d4044e4-df18-4dc3-9397-8d410da1d67b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True, False, False, ...,  True, False, False])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred5 = model5.predict(x_test2)\n",
    "pred5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "cf44a924-28d1-447e-b702-27f2b38bb284",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "68.65410263374451"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_test2, pred5)*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "97004731-3865-445e-9cef-0e5fd4c4653e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       False       0.67      0.74      0.70    123286\n",
      "        True       0.71      0.63      0.67    123397\n",
      "\n",
      "    accuracy                           0.69    246683\n",
      "   macro avg       0.69      0.69      0.69    246683\n",
      "weighted avg       0.69      0.69      0.69    246683\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "#print classification report for model\n",
    "print(classification_report(y_test2, pred5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "4a583b30-d9f3-4292-99d2-e89c36d5592f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert data types\n",
    "x_test2 = x_test2.astype({\"borough\": int, \"latitude\": float, \"longitude\":float, \"on_street_name\":int, \"off_street_name\":int, \"cross_street_name\": int, \"contributing_factor_vehicle_1\":int, \"contributing_factor_vehicle_1\": int,\"vehicle_type_code_1\":int, \"vehicle_type_code_2\":int, \"hour\":int})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "90bf0d11-5eff-4d63-855b-ed58f679fbe5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert data types on train set\n",
    "\n",
    "x_train2 = x_train2.astype({\"borough\": int, \"latitude\": float, \"longitude\":float, \"on_street_name\":int, \"off_street_name\":int, \"cross_street_name\": int, \"contributing_factor_vehicle_1\":int, \"contributing_factor_vehicle_1\": int,\"vehicle_type_code_1\":int, \"vehicle_type_code_2\":int, \"hour\":int})\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3414b533-8b3c-4182-9c48-10424fbca020",
   "metadata": {},
   "source": [
    "## Logistic Regression, XGBoost, Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "800e6ef4-1352-4b77-a8c0-4419442fdf02",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10.0035812628878\n"
     ]
    }
   ],
   "source": [
    "import xgboost as xgb\n",
    "from sklearn.metrics import log_loss\n",
    " \n",
    "# importing machine learning models for prediction\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    " \n",
    "# importing voting classifier\n",
    "from sklearn.ensemble import VotingClassifier\n",
    "# initializing all the model objects with default parameters\n",
    "model_1 = LogisticRegression()\n",
    "model_2 = XGBClassifier()\n",
    "model_3 = RandomForestClassifier()\n",
    " \n",
    "# Making the final model using voting classifier\n",
    "final_model2 = VotingClassifier(\n",
    "    estimators=[('lr', model_1), ('xgb', model_2), ('rf', model_3)], voting='soft')\n",
    "\n",
    "#ensemble = VotingClassifier(estimators,voting='soft')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "2ea77471-61ff-49ff-bb3c-cc8c6af332e1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7104137698990202\n"
     ]
    }
   ],
   "source": [
    "#Import scikit-learn metrics module for accuracy calculation\n",
    "from sklearn import metrics\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Accuracy:\",metrics.accuracy_score(y_test2, pred_final))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "e912a632-fd78-40dc-be3c-73f78ebd6cff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       False       0.68      0.78      0.73    123286\n",
      "        True       0.75      0.64      0.69    123397\n",
      "\n",
      "    accuracy                           0.71    246683\n",
      "   macro avg       0.71      0.71      0.71    246683\n",
      "weighted avg       0.71      0.71      0.71    246683\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "#print classification report for model\n",
    "print(classification_report(y_test2, pred_final))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c75003d2-992c-45f6-bc74-47c009112e20",
   "metadata": {},
   "source": [
    "##### Voting Classifier Tuned (level=soft)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "94cf4a1a-658b-47a6-af93-4f94f5f9d25f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9.879111330244612\n"
     ]
    }
   ],
   "source": [
    " \n",
    "# training all the model on the train dataset\n",
    "final_model2.fit(x_train2, y_train2)\n",
    " \n",
    "# predicting the output on the test dataset\n",
    "pred_final2 = final_model2.predict(x_test2)\n",
    " \n",
    "# printing log loss between actual and predicted value\n",
    "print(log_loss(y_test2, pred_final2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "5785d557-c7e9-4ec2-8588-32b357c6341c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7139729936801482\n"
     ]
    }
   ],
   "source": [
    "#Import scikit-learn metrics module for accuracy calculation\n",
    "from sklearn import metrics\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Accuracy:\",metrics.accuracy_score(y_test2, pred_final2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "4873ccac-3d55-40a7-bb24-19402377f743",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       False       0.69      0.78      0.73    123286\n",
      "        True       0.75      0.65      0.69    123397\n",
      "\n",
      "    accuracy                           0.71    246683\n",
      "   macro avg       0.72      0.71      0.71    246683\n",
      "weighted avg       0.72      0.71      0.71    246683\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "#print classification report for model\n",
    "print(classification_report(y_test2, pred_final2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "faa77c22-65d1-4baa-ae15-06ac42dff454",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "metrics.plot_roc_curve(final_model2, x_test2, y_test2) \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "97cd0aec-dd40-4c76-9164-3c4bece88ea8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAAASbElEQVR4nO3dfZBdd13H8feHhEKRPjAkOJgHUjUMRJSCOy1SR0oBJ63SDKKQ1g4WkSpScEAZ6xOF+oQP4MhMlQZaARUKVOxsNZjRtljHaTVb+yBJLcaANKUzXUtp1WJL6tc/7gne2WyyN82ee7v7e79mdvaec3733O8vu5nP/s7vPKSqkCS16wmTLkCSNFkGgSQ1ziCQpMYZBJLUOINAkhq3ctIFHKlVq1bVhg0bJl2GJC0pN998839U1er5ti25INiwYQMzMzOTLkOSlpQk/36obR4akqTGGQSS1DiDQJIaZxBIUuMMAklqXG9BkOSKJPcm+dwhtifJ+5PsSXJ7khf2VYsk6dD6HBF8GNh8mO1nAhu7rwuAP+yxFknSIfQWBFV1A/CVwzTZAny0Bm4CTkzyzL7qefc1u3j3Nbv62r0kLVmTvKBsDXDX0PK+bt09cxsmuYDBqIH169c/pg/b/eUHH9P7JGm5WxKTxVW1raqmqmpq9ep5r5CWJD1GkwyCu4F1Q8tru3WSpDGaZBBMA6/rzh56EfBAVR10WEiS1K/e5giSfBw4HViVZB9wMfBEgKr6ALAdOAvYAzwEvL6vWiRJh9ZbEFTVOQtsL+DNfX2+JGk0S2KyWJLUH4NAkhpnEEhS4wwCSWqcQSBJjTMIJKlxBoEkNc4gkKTGGQSS1DiDQJIaZxBIUuMMAklqnEEgSY0zCCSpcQaBJDXOIJCkxhkEktQ4g0CSGmcQSFLjDAJJapxBIEmNMwgkqXEGgSQ1ziCQpMYZBJLUOINAkhpnEEhS4wwCSWqcQSBJjTMIJKlxBoEkNa7XIEiyOcmdSfYkuWie7c9Kcm2S25N8NsnaPuuRJB2styBIsgK4FDgT2ASck2TTnGa/C3y0qr4LuAT4zb7qkSTNr88RwSnAnqraW1WPAFcCW+a02QRc172+fp7tkqSe9RkEa4C7hpb3deuG3Qb8UPf6VcBxSZ4+d0dJLkgyk2Rmdna2l2IlqVWTniz+OeAlSW4BXgLcDTw6t1FVbauqqaqaWr169bhrlKRlbWWP+74bWDe0vLZb9w1V9WW6EUGSpwKvrqqv9liTJGmOPkcEO4GNSU5KcgywFZgebpBkVZIDNfwCcEWP9UiS5tFbEFTVfuBCYAdwB/DJqtqV5JIkZ3fNTgfuTPJ54JuBX++rHknS/Po8NERVbQe2z1n3zqHXVwFX9VmDJOnwJj1ZLEmaMINAkhpnEEhS4wwCSWqcQSBJjTMIJKlxBoEkNc4gkKTGGQSS1DiDQJIaZxBIUuMMAklqnEEgSY0zCCSpcQaBJDXOIJCkxhkEktQ4g0CSGmcQSFLjDAJJapxBIEmNMwgkqXEGgSQ1ziCQpMYZBJLUOINAkhpnEEhS4wwCSWqcQSBJjTMIJKlxvQZBks1J7kyyJ8lF82xfn+T6JLckuT3JWX3WI0k6WG9BkGQFcClwJrAJOCfJpjnNfhn4ZFW9ANgK/EFf9UiS5tfniOAUYE9V7a2qR4ArgS1z2hRwfPf6BODLPdYjSZpHn0GwBrhraHlft27Yu4DzkuwDtgNvmW9HSS5IMpNkZnZ2to9aJalZIwVBktOS/HWSzyfZm+QLSfYuwuefA3y4qtYCZwF/nOSgmqpqW1VNVdXU6tWrF+FjJUkHrByx3eXA24CbgUdHfM/dwLqh5bXdumFvADYDVNWNSZ4MrALuHfEzJElHadRDQw9U1Weq6t6quu/A1wLv2QlsTHJSkmMYTAZPz2nzJeBlAEmeCzwZ8NiPJI3RqCOC65P8DvBp4OEDK6vqnw71hqran+RCYAewAriiqnYluQSYqapp4GeBDyZ5G4OJ4/Orqh5jXyRJj8GoQXBq931qaF0BZxzuTVW1ncEk8PC6dw693g2cNmINkqQejBQEVfXSvguRJE3GqGcNnZDkfQdO4Uzy3iQn9F2cJKl/o04WXwH8J/Ca7utB4I/6KkqSND6jzhF8W1W9emj53Ulu7aEeSdKYjToi+FqS7z2wkOQ04Gv9lCRJGqdRRwRvAj7SzQsE+Apwfl9FSZLGZ9Szhm4Fnp/k+G75wT6LkiSNz2GDIMl5VfUnSd4+Zz0AVfW+HmuTJI3BQiOCb+q+H9d3IZKkyThsEFTVZd33d4+nHEnSuI16QdlvJzk+yROTXJtkNsl5fRcnSerfqKePfn83QfyDwBeBbwfe0VdRkqTxGTUIDhxC+gHgU1X1QE/1SJLGbNTrCP4iyb8wuIjsTUlWA//TX1mSpHEZaURQVRcBLwamqurrwH9z8IPoJUlL0ELXEZxRVdcl+aGhdcNNPt1XYX3Yfc+DvPayGyddhjRRW05ew7mnrp90GXocWejQ0EuA64BXzrOtWEJBsOXkNZMuQZq43fcMbgpgEGjYQtcRXNx9f/14yunPuaeu95dfzXNErPmMeh3BbyQ5cWj5aUl+rbeqJEljM+rpo2dW1VcPLFTV/cBZvVQkSRqrUYNgRZInHVhIcizwpMO0lyQtEaNeR/CnwLVJDjye8vXAR/opSZI0TqM+j+C3ktwGvLxb9atVtaO/siRJ4zLqiADgDmB/Vf1NkqckOa6q/rOvwiRJ4zHqWUNvBK4CLutWrQGu7qkmSdIYjTpZ/GbgNOBBgKr6V+AZfRUlSRqfUYPg4ap65MBCkpUMriyWJC1xowbB3yb5ReDYJK8APgVc019ZkqRxGTUIfh6YBf4Z+ElgO/DLfRUlSRqfBc8aSrIC2FVVzwE+2H9JkqRxWnBEUFWPAncm8Y5tkrQMjXodwdOAXUn+kcFDaQCoqrMP96Ykm4HfB1YAH6qq98zZ/nvAS7vFpwDPqKoTR6xJkrQIRg2CXznSHXeHlC4FXgHsA3Ymma6q3QfaVNXbhtq/BXjBkX6OJOnoLPSEsicDPwV8O4OJ4surav+I+z4F2FNVe7t9Xcng8Za7D9H+HODiEfctSVokC80RfASYYhACZwLvPYJ9rwHuGlre1607SJJnAScxeBrafNsvSDKTZGZ2dvYISpAkLWShQ0Obquo7AZJcDvxjT3VsBa7qJqYPUlXbgG0AU1NTXsgmSYtooRHB1w+8OIJDQgfcDawbWl7brZvPVuDjR7h/SdIiWGhE8PwkD3avw+DK4ge711VVxx/mvTuBjUlOYhAAW4Fz5zZK8hwGZyX5MFVJmoCFHl6/4rHuuKr2J7kQ2MHg9NErqmpXkkuAmaqa7ppuBa6sKg/5SNIEHMnzCI5YVW1ncDuK4XXvnLP8rj5rkCQd3qj3GpIkLVMGgSQ1ziCQpMYZBJLUOINAkhpnEEhS4wwCSWqcQSBJjTMIJKlxBoEkNc4gkKTGGQSS1DiDQJIaZxBIUuMMAklqnEEgSY0zCCSpcQaBJDXOIJCkxhkEktQ4g0CSGmcQSFLjDAJJapxBIEmNMwgkqXEGgSQ1ziCQpMYZBJLUOINAkhpnEEhS43oNgiSbk9yZZE+Siw7R5jVJdifZleRjfdYjSTrYyr52nGQFcCnwCmAfsDPJdFXtHmqzEfgF4LSquj/JM/qqR5I0vz5HBKcAe6pqb1U9AlwJbJnT5o3ApVV1P0BV3dtjPZKkefQZBGuAu4aW93Xrhj0beHaSv09yU5LN8+0oyQVJZpLMzM7O9lSuJLVp0pPFK4GNwOnAOcAHk5w4t1FVbauqqaqaWr169XgrlKRlrs8guBtYN7S8tls3bB8wXVVfr6ovAJ9nEAySpDHpMwh2AhuTnJTkGGArMD2nzdUMRgMkWcXgUNHeHmuSJM3RWxBU1X7gQmAHcAfwyaraleSSJGd3zXYA9yXZDVwPvKOq7uurJknSwXo7fRSgqrYD2+ese+fQ6wLe3n1JkiZg0pPFkqQJMwgkqXEGgSQ1ziCQpMYZBJLUOINAkhpnEEhS4wwCSWqcQSBJjTMIJKlxBoEkNc4gkKTGGQSS1DiDQJIaZxBIUuMMAklqnEEgSY0zCCSpcQaBJDXOIJCkxhkEktQ4g0CSGmcQSFLjDAJJapxBIEmNMwgkqXEGgSQ1ziCQpMYZBJLUOINAkhpnEEhS43oNgiSbk9yZZE+Si+bZfn6S2SS3dl8/0Wc9kqSDrexrx0lWAJcCrwD2ATuTTFfV7jlNP1FVF/ZVhyTp8HoLAuAUYE9V7QVIciWwBZgbBJLGaPc9D/Lay26cdBl6DDZ9y/Fc/MrvWPT99nloaA1w19Dyvm7dXK9OcnuSq5Ksm29HSS5IMpNkZnZ2to9apSZsOXkNm555/KTL0ONMnyOCUVwDfLyqHk7yk8BHgDPmNqqqbcA2gKmpqRpvidLyce6p6zn31PWTLkOPM32OCO4Ghv/CX9ut+4aquq+qHu4WPwR8d4/1SJLm0WcQ7AQ2JjkpyTHAVmB6uEGSZw4tng3c0WM9kqR59HZoqKr2J7kQ2AGsAK6oql1JLgFmqmoaeGuSs4H9wFeA8/uqR5I0v1QtrUPuU1NTNTMzM+kyJGlJSXJzVU3Nt80riyWpcQaBJDXOIJCkxhkEktS4JTdZnGQW+PfH+PZVwH8sYjlLgX1ug31uw9H0+VlVtXq+DUsuCI5GkplDzZovV/a5Dfa5DX312UNDktQ4g0CSGtdaEGybdAETYJ/bYJ/b0Eufm5ojkCQdrLURgSRpDoNAkhq3LIMgyeYkdybZk+SiebY/Kcknuu3/kGTDBMpcVCP0+e1JdndPg7s2ybMmUediWqjPQ+1enaSSLPlTDUfpc5LXdD/rXUk+Nu4aF9sIv9vrk1yf5Jbu9/usSdS5WJJckeTeJJ87xPYkeX/373F7khce9YdW1bL6YnDL638DvhU4BrgN2DSnzU8DH+hebwU+Mem6x9DnlwJP6V6/qYU+d+2OA24AbgKmJl33GH7OG4FbgKd1y8+YdN1j6PM24E3d603AFydd91H2+fuAFwKfO8T2s4DPAAFeBPzD0X7mchwRnALsqaq9VfUIcCWwZU6bLQweiwlwFfCyJBljjYttwT5X1fVV9VC3eBODJ8YtZaP8nAF+Ffgt4H/GWVxPRunzG4FLq+p+gKq6d8w1LrZR+lzAgQcxnwB8eYz1LbqquoHB81kOZQvw0Rq4CThxzkO+jthyDII1wF1Dy/u6dfO2qar9wAPA08dSXT9G6fOwNzD4i2IpW7DP3ZB5XVX95TgL69EoP+dnA89O8vdJbkqyeWzV9WOUPr8LOC/JPmA78JbxlDYxR/r/fUGTfni9xizJecAU8JJJ19KnJE8A3kd7T71byeDw0OkMRn03JPnOqvrqJIvq2TnAh6vqvUm+B/jjJM+rqv+ddGFLxXIcEdwNrBtaXtutm7dNkpUMhpP3jaW6fozSZ5K8HPgl4OyqenhMtfVloT4fBzwP+GySLzI4ljq9xCeMR/k57wOmq+rrVfUF4PMMgmGpGqXPbwA+CVBVNwJPZnBztuVqpP/vR2I5BsFOYGOSk5Icw2AyeHpOm2ngx7rXPwxcV90szBK1YJ+TvAC4jEEILPXjxrBAn6vqgapaVVUbqmoDg3mRs6tqKT/ndJTf7asZjAZIsorBoaK9Y6xxsY3S5y8BLwNI8lwGQTA71irHaxp4XXf20IuAB6rqnqPZ4bI7NFRV+5NcCOxgcMbBFVW1K8klwExVTQOXMxg+7mEwKbN1chUfvRH7/DvAU4FPdfPiX6qqsydW9FEasc/Lyoh93gF8f5LdwKPAO6pqyY52R+zzzwIfTPI2BhPH5y/lP+ySfJxBmK/q5j0uBp4IUFUfYDAPchawB3gIeP1Rf+YS/veSJC2C5XhoSJJ0BAwCSWqcQSBJjTMIJKlxBoEkNc4gkOaR5NEktyb5XJJrkpy4yPv/YneeP0n+azH3LR0pg0Ca39eq6uSqeh6Da03ePOmCpL4YBNLCbqS7qVeSb0vyV0luTvJ3SZ7Trf/mJH+e5Lbu68Xd+qu7truSXDDBPkiHtOyuLJYWU5IVDG5fcHm3ahvwU1X1r0lOBf4AOAN4P/C3VfWq7j1P7dr/eFV9JcmxwM4kf7aUr/TV8mQQSPM7NsmtDEYCdwB/neSpwIv5/9t0ADyp+34G8DqAqnqUwa3NAd6a5FXd63UMbgBnEOhxxSCQ5ve1qjo5yVMY3OfmzcCHga9W1cmj7CDJ6cDLge+pqoeSfJbBDdGkxxXnCKTD6J7q9lYGNzZ7CPhCkh+Bbzw79vld02sZPAKUJCuSnMDg9ub3dyHwHAa3wpYedwwCaQFVdQtwO4MHoPwo8IYktwG7+P/HJv4M8NIk/wzczODZuX8FrExyB/AeBrfClh53vPuoJDXOEYEkNc4gkKTGGQSS1DiDQJIaZxBIUuMMAklqnEEgSY37PwiD7CQpAnNWAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import (precision_recall_curve, PrecisionRecallDisplay)\n",
    "precision, recall, _ = precision_recall_curve(y_test2, pred_final)\n",
    "disp = PrecisionRecallDisplay(precision=precision, recall=recall)\n",
    "disp.plot()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "afd009c9-709d-4161-ad56-0dbc72dacbde",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    395287.000000\n",
       "mean        -73.634704\n",
       "std           4.527111\n",
       "min         -74.253006\n",
       "25%         -73.962540\n",
       "50%         -73.918720\n",
       "75%         -73.861860\n",
       "max           0.000000\n",
       "Name: longitude, dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.longitude.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "bed6a137-a703-4402-a288-d0f368386b5c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import plot_confusion_matrix\n",
    "plot_confusion_matrix(final_model, x_test2, y_test2)  \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b71a8241-9086-471b-8ceb-4292d2ca6663",
   "metadata": {},
   "source": [
    "# AutoML/Pycaret: Best Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "d3d570a3-b38a-4b50-9347-708a7d51844b",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = df_downsampled1.sample(frac=0.95, random_state=95)\n",
    "test_data = df_downsampled1.drop(train_data.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "9af3986b-25d4-4bcc-8bbb-ee4f2d0683fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_70852_row19_col1, #T_70852_row44_col1 {\n",
       "  background-color: lightgreen;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_70852\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_70852_level0_col0\" class=\"col_heading level0 col0\" >Description</th>\n",
       "      <th id=\"T_70852_level0_col1\" class=\"col_heading level0 col1\" >Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_70852_row0_col0\" class=\"data row0 col0\" >session_id</td>\n",
       "      <td id=\"T_70852_row0_col1\" class=\"data row0 col1\" >123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_70852_row1_col0\" class=\"data row1 col0\" >Target</td>\n",
       "      <td id=\"T_70852_row1_col1\" class=\"data row1 col1\" >serious_accident</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_70852_row2_col0\" class=\"data row2 col0\" >Target Type</td>\n",
       "      <td id=\"T_70852_row2_col1\" class=\"data row2 col1\" >Binary</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_70852_row3_col0\" class=\"data row3 col0\" >Label Encoded</td>\n",
       "      <td id=\"T_70852_row3_col1\" class=\"data row3 col1\" >False: 0, True: 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_70852_row4_col0\" class=\"data row4 col0\" >Original Data</td>\n",
       "      <td id=\"T_70852_row4_col1\" class=\"data row4 col1\" >(781160, 16)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_70852_row5_col0\" class=\"data row5 col0\" >Missing Values</td>\n",
       "      <td id=\"T_70852_row5_col1\" class=\"data row5 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_70852_row6_col0\" class=\"data row6 col0\" >Numeric Features</td>\n",
       "      <td id=\"T_70852_row6_col1\" class=\"data row6 col1\" >11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_70852_row7_col0\" class=\"data row7 col0\" >Categorical Features</td>\n",
       "      <td id=\"T_70852_row7_col1\" class=\"data row7 col1\" >4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_70852_row8_col0\" class=\"data row8 col0\" >Ordinal Features</td>\n",
       "      <td id=\"T_70852_row8_col1\" class=\"data row8 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_70852_row9_col0\" class=\"data row9 col0\" >High Cardinality Features</td>\n",
       "      <td id=\"T_70852_row9_col1\" class=\"data row9 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_70852_row10_col0\" class=\"data row10 col0\" >High Cardinality Method</td>\n",
       "      <td id=\"T_70852_row10_col1\" class=\"data row10 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_70852_row11_col0\" class=\"data row11 col0\" >Transformed Train Set</td>\n",
       "      <td id=\"T_70852_row11_col1\" class=\"data row11 col1\" >(546811, 47)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_70852_row12_col0\" class=\"data row12 col0\" >Transformed Test Set</td>\n",
       "      <td id=\"T_70852_row12_col1\" class=\"data row12 col1\" >(234349, 47)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_70852_row13_col0\" class=\"data row13 col0\" >Shuffle Train-Test</td>\n",
       "      <td id=\"T_70852_row13_col1\" class=\"data row13 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_70852_row14_col0\" class=\"data row14 col0\" >Stratify Train-Test</td>\n",
       "      <td id=\"T_70852_row14_col1\" class=\"data row14 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_70852_row15_col0\" class=\"data row15 col0\" >Fold Generator</td>\n",
       "      <td id=\"T_70852_row15_col1\" class=\"data row15 col1\" >StratifiedKFold</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_70852_row16_col0\" class=\"data row16 col0\" >Fold Number</td>\n",
       "      <td id=\"T_70852_row16_col1\" class=\"data row16 col1\" >10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_70852_row17_col0\" class=\"data row17 col0\" >CPU Jobs</td>\n",
       "      <td id=\"T_70852_row17_col1\" class=\"data row17 col1\" >-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "      <td id=\"T_70852_row18_col0\" class=\"data row18 col0\" >Use GPU</td>\n",
       "      <td id=\"T_70852_row18_col1\" class=\"data row18 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "      <td id=\"T_70852_row19_col0\" class=\"data row19 col0\" >Log Experiment</td>\n",
       "      <td id=\"T_70852_row19_col1\" class=\"data row19 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "      <td id=\"T_70852_row20_col0\" class=\"data row20 col0\" >Experiment Name</td>\n",
       "      <td id=\"T_70852_row20_col1\" class=\"data row20 col1\" >crash-severity</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "      <td id=\"T_70852_row21_col0\" class=\"data row21 col0\" >USI</td>\n",
       "      <td id=\"T_70852_row21_col1\" class=\"data row21 col1\" >07b5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "      <td id=\"T_70852_row22_col0\" class=\"data row22 col0\" >Imputation Type</td>\n",
       "      <td id=\"T_70852_row22_col1\" class=\"data row22 col1\" >simple</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row23\" class=\"row_heading level0 row23\" >23</th>\n",
       "      <td id=\"T_70852_row23_col0\" class=\"data row23 col0\" >Iterative Imputation Iteration</td>\n",
       "      <td id=\"T_70852_row23_col1\" class=\"data row23 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row24\" class=\"row_heading level0 row24\" >24</th>\n",
       "      <td id=\"T_70852_row24_col0\" class=\"data row24 col0\" >Numeric Imputer</td>\n",
       "      <td id=\"T_70852_row24_col1\" class=\"data row24 col1\" >mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row25\" class=\"row_heading level0 row25\" >25</th>\n",
       "      <td id=\"T_70852_row25_col0\" class=\"data row25 col0\" >Iterative Imputation Numeric Model</td>\n",
       "      <td id=\"T_70852_row25_col1\" class=\"data row25 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row26\" class=\"row_heading level0 row26\" >26</th>\n",
       "      <td id=\"T_70852_row26_col0\" class=\"data row26 col0\" >Categorical Imputer</td>\n",
       "      <td id=\"T_70852_row26_col1\" class=\"data row26 col1\" >constant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row27\" class=\"row_heading level0 row27\" >27</th>\n",
       "      <td id=\"T_70852_row27_col0\" class=\"data row27 col0\" >Iterative Imputation Categorical Model</td>\n",
       "      <td id=\"T_70852_row27_col1\" class=\"data row27 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row28\" class=\"row_heading level0 row28\" >28</th>\n",
       "      <td id=\"T_70852_row28_col0\" class=\"data row28 col0\" >Unknown Categoricals Handling</td>\n",
       "      <td id=\"T_70852_row28_col1\" class=\"data row28 col1\" >least_frequent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row29\" class=\"row_heading level0 row29\" >29</th>\n",
       "      <td id=\"T_70852_row29_col0\" class=\"data row29 col0\" >Normalize</td>\n",
       "      <td id=\"T_70852_row29_col1\" class=\"data row29 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row30\" class=\"row_heading level0 row30\" >30</th>\n",
       "      <td id=\"T_70852_row30_col0\" class=\"data row30 col0\" >Normalize Method</td>\n",
       "      <td id=\"T_70852_row30_col1\" class=\"data row30 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row31\" class=\"row_heading level0 row31\" >31</th>\n",
       "      <td id=\"T_70852_row31_col0\" class=\"data row31 col0\" >Transformation</td>\n",
       "      <td id=\"T_70852_row31_col1\" class=\"data row31 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row32\" class=\"row_heading level0 row32\" >32</th>\n",
       "      <td id=\"T_70852_row32_col0\" class=\"data row32 col0\" >Transformation Method</td>\n",
       "      <td id=\"T_70852_row32_col1\" class=\"data row32 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row33\" class=\"row_heading level0 row33\" >33</th>\n",
       "      <td id=\"T_70852_row33_col0\" class=\"data row33 col0\" >PCA</td>\n",
       "      <td id=\"T_70852_row33_col1\" class=\"data row33 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row34\" class=\"row_heading level0 row34\" >34</th>\n",
       "      <td id=\"T_70852_row34_col0\" class=\"data row34 col0\" >PCA Method</td>\n",
       "      <td id=\"T_70852_row34_col1\" class=\"data row34 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row35\" class=\"row_heading level0 row35\" >35</th>\n",
       "      <td id=\"T_70852_row35_col0\" class=\"data row35 col0\" >PCA Components</td>\n",
       "      <td id=\"T_70852_row35_col1\" class=\"data row35 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row36\" class=\"row_heading level0 row36\" >36</th>\n",
       "      <td id=\"T_70852_row36_col0\" class=\"data row36 col0\" >Ignore Low Variance</td>\n",
       "      <td id=\"T_70852_row36_col1\" class=\"data row36 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row37\" class=\"row_heading level0 row37\" >37</th>\n",
       "      <td id=\"T_70852_row37_col0\" class=\"data row37 col0\" >Combine Rare Levels</td>\n",
       "      <td id=\"T_70852_row37_col1\" class=\"data row37 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row38\" class=\"row_heading level0 row38\" >38</th>\n",
       "      <td id=\"T_70852_row38_col0\" class=\"data row38 col0\" >Rare Level Threshold</td>\n",
       "      <td id=\"T_70852_row38_col1\" class=\"data row38 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row39\" class=\"row_heading level0 row39\" >39</th>\n",
       "      <td id=\"T_70852_row39_col0\" class=\"data row39 col0\" >Numeric Binning</td>\n",
       "      <td id=\"T_70852_row39_col1\" class=\"data row39 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row40\" class=\"row_heading level0 row40\" >40</th>\n",
       "      <td id=\"T_70852_row40_col0\" class=\"data row40 col0\" >Remove Outliers</td>\n",
       "      <td id=\"T_70852_row40_col1\" class=\"data row40 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row41\" class=\"row_heading level0 row41\" >41</th>\n",
       "      <td id=\"T_70852_row41_col0\" class=\"data row41 col0\" >Outliers Threshold</td>\n",
       "      <td id=\"T_70852_row41_col1\" class=\"data row41 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row42\" class=\"row_heading level0 row42\" >42</th>\n",
       "      <td id=\"T_70852_row42_col0\" class=\"data row42 col0\" >Remove Multicollinearity</td>\n",
       "      <td id=\"T_70852_row42_col1\" class=\"data row42 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row43\" class=\"row_heading level0 row43\" >43</th>\n",
       "      <td id=\"T_70852_row43_col0\" class=\"data row43 col0\" >Multicollinearity Threshold</td>\n",
       "      <td id=\"T_70852_row43_col1\" class=\"data row43 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row44\" class=\"row_heading level0 row44\" >44</th>\n",
       "      <td id=\"T_70852_row44_col0\" class=\"data row44 col0\" >Remove Perfect Collinearity</td>\n",
       "      <td id=\"T_70852_row44_col1\" class=\"data row44 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row45\" class=\"row_heading level0 row45\" >45</th>\n",
       "      <td id=\"T_70852_row45_col0\" class=\"data row45 col0\" >Clustering</td>\n",
       "      <td id=\"T_70852_row45_col1\" class=\"data row45 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row46\" class=\"row_heading level0 row46\" >46</th>\n",
       "      <td id=\"T_70852_row46_col0\" class=\"data row46 col0\" >Clustering Iteration</td>\n",
       "      <td id=\"T_70852_row46_col1\" class=\"data row46 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row47\" class=\"row_heading level0 row47\" >47</th>\n",
       "      <td id=\"T_70852_row47_col0\" class=\"data row47 col0\" >Polynomial Features</td>\n",
       "      <td id=\"T_70852_row47_col1\" class=\"data row47 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row48\" class=\"row_heading level0 row48\" >48</th>\n",
       "      <td id=\"T_70852_row48_col0\" class=\"data row48 col0\" >Polynomial Degree</td>\n",
       "      <td id=\"T_70852_row48_col1\" class=\"data row48 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row49\" class=\"row_heading level0 row49\" >49</th>\n",
       "      <td id=\"T_70852_row49_col0\" class=\"data row49 col0\" >Trignometry Features</td>\n",
       "      <td id=\"T_70852_row49_col1\" class=\"data row49 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row50\" class=\"row_heading level0 row50\" >50</th>\n",
       "      <td id=\"T_70852_row50_col0\" class=\"data row50 col0\" >Polynomial Threshold</td>\n",
       "      <td id=\"T_70852_row50_col1\" class=\"data row50 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row51\" class=\"row_heading level0 row51\" >51</th>\n",
       "      <td id=\"T_70852_row51_col0\" class=\"data row51 col0\" >Group Features</td>\n",
       "      <td id=\"T_70852_row51_col1\" class=\"data row51 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row52\" class=\"row_heading level0 row52\" >52</th>\n",
       "      <td id=\"T_70852_row52_col0\" class=\"data row52 col0\" >Feature Selection</td>\n",
       "      <td id=\"T_70852_row52_col1\" class=\"data row52 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row53\" class=\"row_heading level0 row53\" >53</th>\n",
       "      <td id=\"T_70852_row53_col0\" class=\"data row53 col0\" >Feature Selection Method</td>\n",
       "      <td id=\"T_70852_row53_col1\" class=\"data row53 col1\" >classic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row54\" class=\"row_heading level0 row54\" >54</th>\n",
       "      <td id=\"T_70852_row54_col0\" class=\"data row54 col0\" >Features Selection Threshold</td>\n",
       "      <td id=\"T_70852_row54_col1\" class=\"data row54 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row55\" class=\"row_heading level0 row55\" >55</th>\n",
       "      <td id=\"T_70852_row55_col0\" class=\"data row55 col0\" >Feature Interaction</td>\n",
       "      <td id=\"T_70852_row55_col1\" class=\"data row55 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row56\" class=\"row_heading level0 row56\" >56</th>\n",
       "      <td id=\"T_70852_row56_col0\" class=\"data row56 col0\" >Feature Ratio</td>\n",
       "      <td id=\"T_70852_row56_col1\" class=\"data row56 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row57\" class=\"row_heading level0 row57\" >57</th>\n",
       "      <td id=\"T_70852_row57_col0\" class=\"data row57 col0\" >Interaction Threshold</td>\n",
       "      <td id=\"T_70852_row57_col1\" class=\"data row57 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row58\" class=\"row_heading level0 row58\" >58</th>\n",
       "      <td id=\"T_70852_row58_col0\" class=\"data row58 col0\" >Fix Imbalance</td>\n",
       "      <td id=\"T_70852_row58_col1\" class=\"data row58 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_70852_level0_row59\" class=\"row_heading level0 row59\" >59</th>\n",
       "      <td id=\"T_70852_row59_col0\" class=\"data row59 col0\" >Fix Imbalance Method</td>\n",
       "      <td id=\"T_70852_row59_col1\" class=\"data row59 col1\" >SMOTE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fe539dc7af0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pycaret\n",
    "from pycaret.classification import *\n",
    "\n",
    "clf = setup(train_data, \n",
    "            target = 'serious_accident', \n",
    "            session_id=123, \n",
    "            log_experiment=True, \n",
    "            experiment_name='crash-severity'\n",
    " )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0b87c468-82bf-4240-85a3-434919ea0524",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8a9e0e8613b44f5781ee7334320f1b9e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "IntProgress(value=0, description='Processing: ', max=79)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Initiated</th>\n",
       "      <td>. . . . . . . . . . . . . . . . . .</td>\n",
       "      <td>21:50:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Status</th>\n",
       "      <td>. . . . . . . . . . . . . . . . . .</td>\n",
       "      <td>Fitting 10 Folds</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Estimator</th>\n",
       "      <td>. . . . . . . . . . . . . . . . . .</td>\n",
       "      <td>Extra Trees Classifier</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                      \n",
       "                                                                      \n",
       "Initiated  . . . . . . . . . . . . . . . . . .                21:50:27\n",
       "Status     . . . . . . . . . . . . . . . . . .        Fitting 10 Folds\n",
       "Estimator  . . . . . . . . . . . . . . . . . .  Extra Trees Classifier"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_a7ed2 th {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_a7ed2_row0_col0, #T_a7ed2_row0_col1, #T_a7ed2_row0_col2, #T_a7ed2_row0_col3, #T_a7ed2_row0_col4, #T_a7ed2_row0_col5, #T_a7ed2_row0_col6, #T_a7ed2_row0_col7, #T_a7ed2_row0_col8, #T_a7ed2_row1_col0, #T_a7ed2_row1_col1, #T_a7ed2_row1_col2, #T_a7ed2_row1_col3, #T_a7ed2_row1_col4, #T_a7ed2_row1_col5, #T_a7ed2_row1_col6, #T_a7ed2_row1_col7, #T_a7ed2_row1_col8, #T_a7ed2_row2_col0, #T_a7ed2_row2_col1, #T_a7ed2_row2_col2, #T_a7ed2_row2_col3, #T_a7ed2_row2_col4, #T_a7ed2_row2_col5, #T_a7ed2_row2_col6, #T_a7ed2_row2_col7, #T_a7ed2_row2_col8, #T_a7ed2_row3_col0, #T_a7ed2_row3_col1, #T_a7ed2_row3_col2, #T_a7ed2_row3_col3, #T_a7ed2_row3_col4, #T_a7ed2_row3_col5, #T_a7ed2_row3_col6, #T_a7ed2_row3_col7, #T_a7ed2_row3_col8, #T_a7ed2_row4_col0, #T_a7ed2_row4_col1, #T_a7ed2_row4_col2, #T_a7ed2_row4_col3, #T_a7ed2_row4_col4, #T_a7ed2_row4_col5, #T_a7ed2_row4_col6, #T_a7ed2_row4_col7, #T_a7ed2_row4_col8, #T_a7ed2_row5_col0, #T_a7ed2_row5_col1, #T_a7ed2_row5_col2, #T_a7ed2_row5_col3, #T_a7ed2_row5_col4, #T_a7ed2_row5_col5, #T_a7ed2_row5_col6, #T_a7ed2_row5_col7, #T_a7ed2_row5_col8, #T_a7ed2_row6_col0, #T_a7ed2_row6_col1, #T_a7ed2_row6_col2, #T_a7ed2_row6_col3, #T_a7ed2_row6_col4, #T_a7ed2_row6_col5, #T_a7ed2_row6_col6, #T_a7ed2_row6_col7, #T_a7ed2_row6_col8, #T_a7ed2_row7_col0, #T_a7ed2_row7_col1, #T_a7ed2_row7_col2, #T_a7ed2_row7_col3, #T_a7ed2_row7_col4, #T_a7ed2_row7_col5, #T_a7ed2_row7_col6, #T_a7ed2_row7_col7, #T_a7ed2_row7_col8, #T_a7ed2_row8_col0, #T_a7ed2_row8_col1, #T_a7ed2_row8_col2, #T_a7ed2_row8_col3, #T_a7ed2_row8_col4, #T_a7ed2_row8_col5, #T_a7ed2_row8_col6, #T_a7ed2_row8_col7, #T_a7ed2_row8_col8, #T_a7ed2_row9_col0, #T_a7ed2_row9_col1, #T_a7ed2_row9_col2, #T_a7ed2_row9_col3, #T_a7ed2_row9_col4, #T_a7ed2_row9_col5, #T_a7ed2_row9_col6, #T_a7ed2_row9_col7, #T_a7ed2_row9_col8, #T_a7ed2_row10_col0, #T_a7ed2_row10_col1, #T_a7ed2_row10_col2, #T_a7ed2_row10_col3, #T_a7ed2_row10_col4, #T_a7ed2_row10_col5, #T_a7ed2_row10_col6, #T_a7ed2_row10_col7, #T_a7ed2_row10_col8 {\n",
       "  text-align: left;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_a7ed2\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_a7ed2_level0_col0\" class=\"col_heading level0 col0\" >Model</th>\n",
       "      <th id=\"T_a7ed2_level0_col1\" class=\"col_heading level0 col1\" >Accuracy</th>\n",
       "      <th id=\"T_a7ed2_level0_col2\" class=\"col_heading level0 col2\" >AUC</th>\n",
       "      <th id=\"T_a7ed2_level0_col3\" class=\"col_heading level0 col3\" >Recall</th>\n",
       "      <th id=\"T_a7ed2_level0_col4\" class=\"col_heading level0 col4\" >Prec.</th>\n",
       "      <th id=\"T_a7ed2_level0_col5\" class=\"col_heading level0 col5\" >F1</th>\n",
       "      <th id=\"T_a7ed2_level0_col6\" class=\"col_heading level0 col6\" >Kappa</th>\n",
       "      <th id=\"T_a7ed2_level0_col7\" class=\"col_heading level0 col7\" >MCC</th>\n",
       "      <th id=\"T_a7ed2_level0_col8\" class=\"col_heading level0 col8\" >TT (Sec)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row0\" class=\"row_heading level0 row0\" >rf</th>\n",
       "      <td id=\"T_a7ed2_row0_col0\" class=\"data row0 col0\" >Random Forest Classifier</td>\n",
       "      <td id=\"T_a7ed2_row0_col1\" class=\"data row0 col1\" >0.6985</td>\n",
       "      <td id=\"T_a7ed2_row0_col2\" class=\"data row0 col2\" >0.7733</td>\n",
       "      <td id=\"T_a7ed2_row0_col3\" class=\"data row0 col3\" >0.6388</td>\n",
       "      <td id=\"T_a7ed2_row0_col4\" class=\"data row0 col4\" >0.7254</td>\n",
       "      <td id=\"T_a7ed2_row0_col5\" class=\"data row0 col5\" >0.6793</td>\n",
       "      <td id=\"T_a7ed2_row0_col6\" class=\"data row0 col6\" >0.3970</td>\n",
       "      <td id=\"T_a7ed2_row0_col7\" class=\"data row0 col7\" >0.3999</td>\n",
       "      <td id=\"T_a7ed2_row0_col8\" class=\"data row0 col8\" >73.2700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row1\" class=\"row_heading level0 row1\" >gbc</th>\n",
       "      <td id=\"T_a7ed2_row1_col0\" class=\"data row1 col0\" >Gradient Boosting Classifier</td>\n",
       "      <td id=\"T_a7ed2_row1_col1\" class=\"data row1 col1\" >0.6956</td>\n",
       "      <td id=\"T_a7ed2_row1_col2\" class=\"data row1 col2\" >0.7682</td>\n",
       "      <td id=\"T_a7ed2_row1_col3\" class=\"data row1 col3\" >0.5878</td>\n",
       "      <td id=\"T_a7ed2_row1_col4\" class=\"data row1 col4\" >0.7494</td>\n",
       "      <td id=\"T_a7ed2_row1_col5\" class=\"data row1 col5\" >0.6588</td>\n",
       "      <td id=\"T_a7ed2_row1_col6\" class=\"data row1 col6\" >0.3912</td>\n",
       "      <td id=\"T_a7ed2_row1_col7\" class=\"data row1 col7\" >0.4007</td>\n",
       "      <td id=\"T_a7ed2_row1_col8\" class=\"data row1 col8\" >132.0030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row2\" class=\"row_heading level0 row2\" >ada</th>\n",
       "      <td id=\"T_a7ed2_row2_col0\" class=\"data row2 col0\" >Ada Boost Classifier</td>\n",
       "      <td id=\"T_a7ed2_row2_col1\" class=\"data row2 col1\" >0.6642</td>\n",
       "      <td id=\"T_a7ed2_row2_col2\" class=\"data row2 col2\" >0.7329</td>\n",
       "      <td id=\"T_a7ed2_row2_col3\" class=\"data row2 col3\" >0.6081</td>\n",
       "      <td id=\"T_a7ed2_row2_col4\" class=\"data row2 col4\" >0.6849</td>\n",
       "      <td id=\"T_a7ed2_row2_col5\" class=\"data row2 col5\" >0.6442</td>\n",
       "      <td id=\"T_a7ed2_row2_col6\" class=\"data row2 col6\" >0.3284</td>\n",
       "      <td id=\"T_a7ed2_row2_col7\" class=\"data row2 col7\" >0.3305</td>\n",
       "      <td id=\"T_a7ed2_row2_col8\" class=\"data row2 col8\" >12.4970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row3\" class=\"row_heading level0 row3\" >ridge</th>\n",
       "      <td id=\"T_a7ed2_row3_col0\" class=\"data row3 col0\" >Ridge Classifier</td>\n",
       "      <td id=\"T_a7ed2_row3_col1\" class=\"data row3 col1\" >0.6323</td>\n",
       "      <td id=\"T_a7ed2_row3_col2\" class=\"data row3 col2\" >0.0000</td>\n",
       "      <td id=\"T_a7ed2_row3_col3\" class=\"data row3 col3\" >0.5725</td>\n",
       "      <td id=\"T_a7ed2_row3_col4\" class=\"data row3 col4\" >0.6502</td>\n",
       "      <td id=\"T_a7ed2_row3_col5\" class=\"data row3 col5\" >0.6089</td>\n",
       "      <td id=\"T_a7ed2_row3_col6\" class=\"data row3 col6\" >0.2646</td>\n",
       "      <td id=\"T_a7ed2_row3_col7\" class=\"data row3 col7\" >0.2665</td>\n",
       "      <td id=\"T_a7ed2_row3_col8\" class=\"data row3 col8\" >0.4450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row4\" class=\"row_heading level0 row4\" >lda</th>\n",
       "      <td id=\"T_a7ed2_row4_col0\" class=\"data row4 col0\" >Linear Discriminant Analysis</td>\n",
       "      <td id=\"T_a7ed2_row4_col1\" class=\"data row4 col1\" >0.6323</td>\n",
       "      <td id=\"T_a7ed2_row4_col2\" class=\"data row4 col2\" >0.6706</td>\n",
       "      <td id=\"T_a7ed2_row4_col3\" class=\"data row4 col3\" >0.5725</td>\n",
       "      <td id=\"T_a7ed2_row4_col4\" class=\"data row4 col4\" >0.6502</td>\n",
       "      <td id=\"T_a7ed2_row4_col5\" class=\"data row4 col5\" >0.6089</td>\n",
       "      <td id=\"T_a7ed2_row4_col6\" class=\"data row4 col6\" >0.2645</td>\n",
       "      <td id=\"T_a7ed2_row4_col7\" class=\"data row4 col7\" >0.2664</td>\n",
       "      <td id=\"T_a7ed2_row4_col8\" class=\"data row4 col8\" >4.0560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row5\" class=\"row_heading level0 row5\" >dt</th>\n",
       "      <td id=\"T_a7ed2_row5_col0\" class=\"data row5 col0\" >Decision Tree Classifier</td>\n",
       "      <td id=\"T_a7ed2_row5_col1\" class=\"data row5 col1\" >0.6308</td>\n",
       "      <td id=\"T_a7ed2_row5_col2\" class=\"data row5 col2\" >0.6308</td>\n",
       "      <td id=\"T_a7ed2_row5_col3\" class=\"data row5 col3\" >0.6325</td>\n",
       "      <td id=\"T_a7ed2_row5_col4\" class=\"data row5 col4\" >0.6303</td>\n",
       "      <td id=\"T_a7ed2_row5_col5\" class=\"data row5 col5\" >0.6314</td>\n",
       "      <td id=\"T_a7ed2_row5_col6\" class=\"data row5 col6\" >0.2616</td>\n",
       "      <td id=\"T_a7ed2_row5_col7\" class=\"data row5 col7\" >0.2616</td>\n",
       "      <td id=\"T_a7ed2_row5_col8\" class=\"data row5 col8\" >4.7900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row6\" class=\"row_heading level0 row6\" >nb</th>\n",
       "      <td id=\"T_a7ed2_row6_col0\" class=\"data row6 col0\" >Naive Bayes</td>\n",
       "      <td id=\"T_a7ed2_row6_col1\" class=\"data row6 col1\" >0.6091</td>\n",
       "      <td id=\"T_a7ed2_row6_col2\" class=\"data row6 col2\" >0.6483</td>\n",
       "      <td id=\"T_a7ed2_row6_col3\" class=\"data row6 col3\" >0.4164</td>\n",
       "      <td id=\"T_a7ed2_row6_col4\" class=\"data row6 col4\" >0.6773</td>\n",
       "      <td id=\"T_a7ed2_row6_col5\" class=\"data row6 col5\" >0.5157</td>\n",
       "      <td id=\"T_a7ed2_row6_col6\" class=\"data row6 col6\" >0.2181</td>\n",
       "      <td id=\"T_a7ed2_row6_col7\" class=\"data row6 col7\" >0.2364</td>\n",
       "      <td id=\"T_a7ed2_row6_col8\" class=\"data row6 col8\" >0.8890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row7\" class=\"row_heading level0 row7\" >lr</th>\n",
       "      <td id=\"T_a7ed2_row7_col0\" class=\"data row7 col0\" >Logistic Regression</td>\n",
       "      <td id=\"T_a7ed2_row7_col1\" class=\"data row7 col1\" >0.6058</td>\n",
       "      <td id=\"T_a7ed2_row7_col2\" class=\"data row7 col2\" >0.6312</td>\n",
       "      <td id=\"T_a7ed2_row7_col3\" class=\"data row7 col3\" >0.5166</td>\n",
       "      <td id=\"T_a7ed2_row7_col4\" class=\"data row7 col4\" >0.6290</td>\n",
       "      <td id=\"T_a7ed2_row7_col5\" class=\"data row7 col5\" >0.5671</td>\n",
       "      <td id=\"T_a7ed2_row7_col6\" class=\"data row7 col6\" >0.2116</td>\n",
       "      <td id=\"T_a7ed2_row7_col7\" class=\"data row7 col7\" >0.2152</td>\n",
       "      <td id=\"T_a7ed2_row7_col8\" class=\"data row7 col8\" >7.8840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row8\" class=\"row_heading level0 row8\" >knn</th>\n",
       "      <td id=\"T_a7ed2_row8_col0\" class=\"data row8 col0\" >K Neighbors Classifier</td>\n",
       "      <td id=\"T_a7ed2_row8_col1\" class=\"data row8 col1\" >0.5947</td>\n",
       "      <td id=\"T_a7ed2_row8_col2\" class=\"data row8 col2\" >0.6289</td>\n",
       "      <td id=\"T_a7ed2_row8_col3\" class=\"data row8 col3\" >0.6021</td>\n",
       "      <td id=\"T_a7ed2_row8_col4\" class=\"data row8 col4\" >0.5933</td>\n",
       "      <td id=\"T_a7ed2_row8_col5\" class=\"data row8 col5\" >0.5977</td>\n",
       "      <td id=\"T_a7ed2_row8_col6\" class=\"data row8 col6\" >0.1894</td>\n",
       "      <td id=\"T_a7ed2_row8_col7\" class=\"data row8 col7\" >0.1894</td>\n",
       "      <td id=\"T_a7ed2_row8_col8\" class=\"data row8 col8\" >532.2450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row9\" class=\"row_heading level0 row9\" >svm</th>\n",
       "      <td id=\"T_a7ed2_row9_col0\" class=\"data row9 col0\" >SVM - Linear Kernel</td>\n",
       "      <td id=\"T_a7ed2_row9_col1\" class=\"data row9 col1\" >0.5279</td>\n",
       "      <td id=\"T_a7ed2_row9_col2\" class=\"data row9 col2\" >0.0000</td>\n",
       "      <td id=\"T_a7ed2_row9_col3\" class=\"data row9 col3\" >0.3979</td>\n",
       "      <td id=\"T_a7ed2_row9_col4\" class=\"data row9 col4\" >0.5918</td>\n",
       "      <td id=\"T_a7ed2_row9_col5\" class=\"data row9 col5\" >0.3255</td>\n",
       "      <td id=\"T_a7ed2_row9_col6\" class=\"data row9 col6\" >0.0557</td>\n",
       "      <td id=\"T_a7ed2_row9_col7\" class=\"data row9 col7\" >0.0779</td>\n",
       "      <td id=\"T_a7ed2_row9_col8\" class=\"data row9 col8\" >50.4320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a7ed2_level0_row10\" class=\"row_heading level0 row10\" >qda</th>\n",
       "      <td id=\"T_a7ed2_row10_col0\" class=\"data row10 col0\" >Quadratic Discriminant Analysis</td>\n",
       "      <td id=\"T_a7ed2_row10_col1\" class=\"data row10 col1\" >0.5000</td>\n",
       "      <td id=\"T_a7ed2_row10_col2\" class=\"data row10 col2\" >0.5000</td>\n",
       "      <td id=\"T_a7ed2_row10_col3\" class=\"data row10 col3\" >0.4057</td>\n",
       "      <td id=\"T_a7ed2_row10_col4\" class=\"data row10 col4\" >0.4956</td>\n",
       "      <td id=\"T_a7ed2_row10_col5\" class=\"data row10 col5\" >0.4257</td>\n",
       "      <td id=\"T_a7ed2_row10_col6\" class=\"data row10 col6\" >-0.0001</td>\n",
       "      <td id=\"T_a7ed2_row10_col7\" class=\"data row10 col7\" >-0.0012</td>\n",
       "      <td id=\"T_a7ed2_row10_col8\" class=\"data row10 col8\" >1.9140</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fe5ef687730>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best_model = compare_models()"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "HW4-Team1-AdaBoostClassifier.ipynb",
   "provenance": []
  },
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}