final_notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chicago, IL : Car Crash Analysis & Predictive Modeling\n",
    "\n",
    "\n",
    "### *Predicting severity of car crashes with Machine Learning Models*\n",
    "\n",
    "Authors: [Christos Maglaras](mailto:Christo111M@gmail.com), [Marcos Panyagua](mailto:marcosvppfernandes@gmail.com), [Jamie Dowat](mailto:jamie_dowat44@yahoo.com)\n",
    "\n",
    "Date: 3/12/2021\n",
    "\n",
    "![chicago](img/chicago_night_drive.jpg)\n",
    "\n",
    "## Stakeholder: Chicago Department of Transportation\n",
    "\n",
    "![cdot](img/cdot.png)\n",
    "\n",
    "### Business Understanding\n",
    "\n",
    "Just a week ago, the National Security Coundil released a [report](https://www.nsc.org/newsroom/motor-vehicle-deaths-2020-estimated-to-be-highest) containing some disturbing statistics from 2020. The first paragraph begins as follows:\n",
    "> *For the first time since 2007, preliminary data from the National Safety Council show that as many as 42,060 people are estimated to have died in motor vehicle crashes in 2020. That marks an 8% increase over 2019 in a year where people drove significantly less frequently because of the pandemic.*\n",
    "\n",
    "According to their data, the US hasn't seen an increase like this since **1924**.\n",
    "\n",
    "Following this trend, the **Governor's Highway Safety Association** reported that [*pedestrian* fatality rate](https://www.smartcitiesdive.com/news/ghsa-projects-highest-pedestrian-death-rate-since-1988/573203/) has reached a **30-year high**, with nighttime pedestrian fatalities having increased by 67%, and a 16% increase in daytime fatalities, highlighting the need for *safer road crossings* and increased efforts to make pedestrians and vehicles more *visible*.\n",
    "\n",
    "Narrowing our focus even further, in **Illinois**, around **1000** people were KILLED in motor vehicle crashes in **2019** alone. \n",
    "\n",
    "**Advocates for Highway and Auto Safety** have scored all US states against their [Roadmap for State Highway Safety Laws](https://saferoads.org/wp-content/uploads/2020/01/Advocates-for-Highway-and-Auto-Safety-2020-Roadmap-of-State-Highway-Safety-Laws.pdf), a set of 16 laws that cover occupant protection (i.e selt belt, helmet laws), child protection, and teen driving. \n",
    "\n",
    "![scoring](img/scoringsafety.png)\n",
    "\n",
    "When Illinois is [scored](https://saferoads.org/state/illinois/) against this Roadmap, it has been given a yellow rating (Caution), since it still lacks the following safety laws:\n",
    "\n",
    "* All-Rider Motorcycle Helmet Law\n",
    "* Booster Seat Law\n",
    "* GDL (Graduated Driver's License) – Minimum Age 16 for Learner’s Permit\n",
    "* GDL – Stronger Nighttime Restriction Provision\n",
    "* GDL- Stronger Passenger Restriction Provision\n",
    "* GDL- Age 18 for Unrestricted License\n",
    "\n",
    "Currently, the **Chicago Department of Transportation** is working with the city's new initiative, **Vision Zero**, to reduce accidents on the road. In Vision Zero's [report](https://8gq.ef1.myftpupload.com/wp-content/uploads/2016/05/17_0612-VZ-Action-Plan_FOR-WEB.pdf) and action plan, they used crash data to identify high crash corridors in the city as well as other important trends to guide education, road safety improvements, and more.\n",
    "\n",
    "![quotes](img/visionzeroquotes.png)\n",
    "\n",
    "Bearing all of this in mind, we had these current safety movements and road safety problems guide our exploration and modeling of this data. \n",
    "\n",
    "###### Even more importantly, this business understanding was the sole influencer of our target choice, *severity of crash based on injury*.\n",
    "\n",
    "******\n",
    "\n",
    "## Predictive Modeling Preview\n",
    "\n",
    "We concluded that one of the best ways a predictive model could help with this business problem is to determine how different factors of the crash relate to the severity of injuries in the crash.\n",
    "\n",
    "We compared results from *Logistic Regression, K-Nearest Neighbors, Decision Tree, Naive Bayes, and Random Forest Classifier* models, before settling on our final model produced by a **Bayesian-Optimized XGBoost Classifier**. Click [here](https://analyticsindiamag.com/implementing-bayesian-optimization-on-xgboost-a-beginners-guide/) to read more on the algorithm.\n",
    "\n",
    "We experimented with both a BINARY classification and a TERNARY, with our final model using as TERNARY classification. The targets are defined as follows, using the `MOST_SEVERE_INJURY` column in the Crashes dataset:\n",
    "\n",
    "* BINARY:\n",
    "    * Class 0: No Injury\n",
    "    * Class 1: Injury\n",
    "    \n",
    "* TERNARY\n",
    "    * Class 0: No Injury\n",
    "    * Class 1: NON-INCAPACITATING injuries\n",
    "    * Class 2: INCAPACITATING or FATAL injuries\n",
    "\n",
    "******\n",
    "\n",
    "## Data: [Chicago City Data Portal](https://data.cityofchicago.org/)\n",
    "\n",
    "![ccdp](img/chicagocitydataportal.jpg)\n",
    "\n",
    "### [Crashes](https://data.cityofchicago.org/Transportation/Traffic-Crashes-Crashes/85ca-t3if):\n",
    "\n",
    "##### Number of Rows: 482,866\n",
    "\n",
    "*Shows crash data from crash from the Chicago Police Department's **E-Crash** system*\n",
    "\n",
    "**\"All crashes are recorded as per the format specified in the Traffic Crash Report, SR1050, of the Illinois Department of Transportation.\"**\n",
    "\n",
    "| Column Name                 | Description                |\n",
    "| --------------------------- | -------------------------- |\n",
    "| crash_record_id  |  Can be used to link to the same crash in the Vehicles and People datasets. |\n",
    "| rd_no | Chicago Police Department report number|\n",
    "| crash_date | Date and time of crash as entered by the reporting officer |\n",
    "| posted_speed_limit  | Posted speed limit, as determined by reporting officer |\n",
    "| traffic_control_device | Traffic control device present at crash location, as determined by reporting officer (signals, stop sign, etc) |\n",
    "| device_condition  | Condition of traffic control device, as determined by reporting officer |\n",
    "| weather_condition | Weather condition at time of crash, as determined by reporting officer |\n",
    "| lighting_condition | Light condition at time of crash, as determined by reporting officer |\n",
    "| first_crash_type | Type of first collision in crash |\n",
    "| trafficway_type  | Trafficway type, as determined by reporting officer |\n",
    "| lane_ct | Total number of through lanes in either direction, excluding turn lanes, as determined by reporting officer (0 = intersection)|\n",
    "| alignment | Street alignment at crash location, as determined by reporting officer |\n",
    "| roadway_surface_cond        | Road surface condition, as determined by reporting officer |\n",
    "| road_defect | Road defects, as determined by reporting officer |\n",
    "| crash_type | A general severity classification for the crash. Can be either Injury and/or Tow Due to Crash or No Injury / Drive Away |\n",
    "| damage | A field observation of estimated damage. |\n",
    "| prim_contributory_cause   | The factor which was most significant in causing the crash, as determined by officer judgment |\n",
    "| sec_contributory_cause | The factor which was second most significant in causing the crash, as determined by officer judgment |\n",
    "| street_name | Street address name of crash location, as determined by reporting officer|\n",
    "| num_units | Number of units involved in the crash. A unit can be a motor vehicle, a pedestrian, a bicyclist, or another non-passenger roadway user. Each unit represents a mode of traffic with an independent trajectory. |\n",
    "| most_severe_injury | Most severe injury sustained by any person involved in the crash |\n",
    "| injuries_total | Total persons sustaining fatal, incapacitating, non-incapacitating, and possible injuries as determined by the reporting officer |\n",
    "| injuries_fatal | Total persons sustaining fatal injuries in the crash |\n",
    "| injuries_incapacitating | Total persons sustaining incapacitating/serious injuries in the crash as determined by the reporting officer. Any injury other than fatal injury, which prevents the injured person from walking, driving, or normally continuing the activities they were capable of performing before the injury occurred. Includes severe lacerations, broken limbs, skull or chest injuries, and abdominal injuries. |\n",
    "| injuries_non_incapacitating | Total persons sustaining non-incapacitating injuries in the crash as determined by the reporting officer. Any injury, other than fatal or incapacitating injury, which is evident to observers at the scene of the crash. Includes lump on head, abrasions, bruises, and minor lacerations. |\n",
    "| crash_hour | The hour of the day component of CRASH_DATE. |\n",
    "| crash_day_of_week | The day of the week component of CRASH_DATE. Sunday=1 |\n",
    "| latitude | The latitude of the crash location, as determined by reporting officer, as derived from the reported address of crash |\n",
    "| longitude | The longitude of the crash location, as determined by reporting officer, as derived from the reported address of crash |\n",
    "\n",
    "\n",
    "### [People](https://data.cityofchicago.org/Transportation/Traffic-Crashes-People/u6pd-qa9d):\n",
    "\n",
    "##### Number of Rows: 1,068,637\n",
    "\n",
    "*Information about people involved in a crash and if any injuries were sustained.*\n",
    "\n",
    "| Column Name                 | Description                |\n",
    "| --------------------------- | -------------------------- |\n",
    "| crash_record_id | This number can be used to link to the same crash in the Crashes and Vehicles datasets. This number also serves as a unique ID in the Crashes dataset. |\n",
    "| person_type | Type of roadway user involved in crash |\n",
    "| rd_no | Chicago Police Department report number. For privacy reasons, this column is blank for recent crashes. |\n",
    "| crash_date | Date and time of crash as entered by the reporting officer |\n",
    "| seat_no | Code for seating position of motor vehicle occupant: 1= driver, 2= center front, 3 = front passenger, 4 = second row left, 5 = second row center, 6 = second row right, 7 = enclosed passengers, 8 = exposed passengers, 9= unknown position, 10 = third row left, 11 = third row center, 12 = third row right |\n",
    "| city | City of residence of person involved in crash |\n",
    "| state | State of residence of person involved in crash |\n",
    "| zipcode | ZIP Code of residence of person involved in crash |\n",
    "| sex | Gender of person involved in crash, as determined by reporting officer |\n",
    "| age | Age of person involved in crash |\n",
    "| drivers_license_state | State issuing driver's license of person involved in crash |\n",
    "| drivers_license_class | Class of driver's license of person involved in crash |\n",
    "| safety_equipment | Safety equipment used by vehicle occupant in crash, if any |\n",
    "| airbag_deployed | Whether vehicle occupant airbag deployed as result of crash |\n",
    "| ejection | Whether vehicle occupant was ejected or extricated from the vehicle as a result of crash |\n",
    "| injury_classification | Severity of injury person sustained in the crash |\n",
    "| driver_action | Driver action that contributed to the crash, as determined by reporting officer |\n",
    "| driver_vision | What, if any, objects obscured the driver’s vision at time of crash |\n",
    "| physical_condition | Driver’s apparent physical condition at time of crash, as observed by the reporting officer |\n",
    "| pedpedal_action | Action of pedestrian or cyclist at the time of crash |\n",
    "| pedpedal_visibility | Visibility of pedestrian of cyclist safety equipment in use at time of crash |\n",
    "| pedpedal_location | Location of pedestrian or cyclist at the time of crash |\n",
    "| bac_result | Status of blood alcohol concentration testing for driver or other person involved in crash |\n",
    "| bac_result value | Driver’s blood alcohol concentration test result (fatal crashes may include pedestrian or cyclist results) |\n",
    "| cell_phone_use | Whether person was/was not using cellphone at the time of the crash, as determined by the reporting officer |\n",
    "\n",
    "### [Vehicles](https://data.cityofchicago.org/Transportation/Traffic-Crashes-Vehicles/68nd-jvt3):\n",
    "\n",
    "##### Number of Rows: 987,148\n",
    "\n",
    "*Information about vehicles (\"units\") involved in a traffic crash.*\n",
    "\n",
    "| Column Name                 | Description                |\n",
    "| --------------------------- | -------------------------- |\n",
    "| crash_record_id | This number can be used to link to the same crash in the Crashes and People datasets. This number also serves as a unique ID in the Crashes dataset. |\n",
    "| rd_no | Chicago Police Department report number. For privacy reasons, this column is blank for recent crashes. |\n",
    "| crash_date | Date and time of crash as entered by the reporting officer |\n",
    "| unit_type | The type of unit (i.e Driver, parked, pedestrian, bicycle, etc) |\n",
    "| num_passengers | Number of passengers in the vehicle. The driver is not included. More information on passengers is in the People dataset. |\n",
    "| make | The make (brand) of the vehicle, if relevant |\n",
    "| model | The model of the vehicle, if relevant |\n",
    "| lic_plate_state | The state issuing the license plate of the vehicle, if relevant |\n",
    "| vehicle_year | The model year of the vehicle, if relevant |\n",
    "| vehicle_defect | Indicates part of car containing defect (brakes, wheels, etc.) |\n",
    "| vehicle_type | The type of vehicle, if relevant (passenger, truck, bus, etc) |\n",
    "| vehicle_use | The normal use of the vehicle, if relevant |\n",
    "| maneuver | The action the unit was taking prior to the crash, as determined by the reporting officer |\n",
    "| towed_I | Indicator of whether the vehicle was towed |\n",
    "| occupant_cnt | The number of people in the unit, as determined by the reporting officer |\n",
    "| exceed_speed_limit_I | Indicator of whether the unit was speeding, as determined by the reporting officer |\n",
    "| first_contact_point | Indicates orientation on car that was hit (front, rear, etc) |\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Loading in our data..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "pd.set_option('display.max_rows', 500)\n",
    "pd.set_option('display.max_columns', None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "crashes = pd.read_csv('data/traffic_crashes_chicago.csv', low_memory=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "people = pd.read_csv('data/traffic_crashes_people.csv', low_memory=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "vehicles = pd.read_csv('data/traffic_crashes_vehicles.csv', low_memory=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis (EDA)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src import data_cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you might have seen above, this data is rife with repetitive columns, columns with no analytical value (like RD_NO), and columns that mostly consisted of NaN values. The following function subsets the original dataset with only the columns we'll use for our EDA."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "crashes, people, vehicles = data_cleaning.column_mask([crashes, people, vehicles])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src import eda\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['crash_record_id', 'crash_date', 'posted_speed_limit',\n",
       "       'traffic_control_device', 'weather_condition', 'lighting_condition',\n",
       "       'first_crash_type', 'trafficway_type', 'alignment',\n",
       "       'roadway_surface_cond', 'road_defect', 'crash_type', 'damage',\n",
       "       'prim_contributory_cause', 'sec_contributory_cause', 'street_name',\n",
       "       'num_units', 'most_severe_injury', 'injuries_total', 'crash_hour',\n",
       "       'crash_day_of_week', 'latitude', 'longitude'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crashes.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['crash_record_id', 'person_type', 'sex', 'age', 'safety_equipment',\n",
       "       'airbag_deployed', 'ejection', 'injury_classification', 'driver_action',\n",
       "       'driver_vision', 'physical_condition', 'pedpedal_action',\n",
       "       'pedpedal_visibility', 'pedpedal_location', 'bac_result',\n",
       "       'bac_result value', 'cell_phone_use'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "people.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['crash_record_id', 'unit_type', 'num_passengers', 'vehicle_year',\n",
       "       'vehicle_defect', 'vehicle_type', 'vehicle_use', 'maneuver',\n",
       "       'occupant_cnt'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vehicles.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We went through each variable and looked at the distribution. Here are some of our graphs!\n",
    "******"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_month(crashes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following graph definitely matches our intuition-- the are highest incidences of crashes occur during rush hour; this could be due to more traffic, eager to get home, etc..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_hour(crashes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main Takeaway: **Friday** has the highest incidences of crashes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_day(crashes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main takeaway: **Ages 22-30** have the highest number of car crash reports. \n",
    "> **IMPORTANT NOTE:** This graph's main purpose is to showcase the distribution of our features. From a car crash perspective we expected there to be more crashes for newer drivers (16-21), but this graph doesn't accomodate for demographic trends in Chicago, which may be causing some of the variance in this distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_age(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main takeaways: \n",
    "1. Pedestrians/Bicyclists have the highest number of crashes when cyclist was with **traffic** or **crossing the road**. This matches our research (see above); the National Security Council reported growing numbers of **pedestrian** and **motor vehicle fatalities**. Click on [this article](https://www.smartcitiesdive.com/news/calls-for-safer-streets-intensify-amid-45-spike-in-pedestrian-deaths/596420/) to learn more!\n",
    "2. The third most common value is **UNKNOWN/NA**. This is a pattern we saw in many columns and will seek to eliminate, as we concluded this would hurt our model's performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_pedpedal_action(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main takeaway: Most reported crashes have an estimated damage of **over 1500 dollars**. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf4AAAGtCAYAAAD3dRZ8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAABI9UlEQVR4nO3dfXzO9f///9uxUydTks1pOpOUFTJlaMMnhm1OVkoWEZ+QFH3INidzEqElOe/0XUlvSUy8mU6EhEKFibyTCbMT5mQb247teP7+8HN8rTm3Ohxe9+vl4mLH8zher+Pxeh2v7f56vl7P1+uwGWMMIiIiYgkeri5ARERE/jkKfhEREQtR8IuIiFiIgl9ERMRCFPwiIiIWouAXERGxEAW/OB04cIC7776bzz77rFj7e++9R0xMTKm9T6tWrdi+fXupze9CcnJy6Nq1K+Hh4Xz55Zclnt+zZw8DBw4kMjKSDh068NRTT7F58+ZSe//p06czduzYUpvf2bp3705SUtLfMu8LiYiI4Icffvjb5r948WKeeOIJOnbsSPv27Rk5ciQnTpwotflf6vYXExPDww8/TMeOHYv9++ijjy443Weffca8efMA+Pe//83bb79dKnUDPPPMM2RlZV3WNNu3b6dVq1bnfO7uu+8mMjKSjh070qFDB6Kiopy1y/XLy9UFyLXFw8ODSZMm0ahRI+644w5Xl3PVdu7cyZEjR/jqq69KPPfHH3/w9NNP8+qrr/Lwww8DsGHDBvr168e///1v7rrrrn+6XMubM2cOa9euZebMmVSuXBm73c6ECRPo168fn3zyyT9eT8+ePendu/dlTbNlyxbntvPkk0+Waj3ff/99qc4P4MMPP6RSpUoAZGVl0a9fP/Lz83nmmWdK/b3k2qDgl2LKlClDr169GDJkCPPnz8fHx6fY8zExMdx1113OP4ZnP27VqhURERFs3LiR48eP06dPH3766Sd27NiBl5cXs2fPpkqVKgB88skn7Nq1i4KCAnr16sVjjz0GwKpVq5g9ezZ2u50yZcowbNgwGjZsyPTp0/nll1/IyMjg7rvvJiEhoVhdX3/9NTNmzMDhcFC+fHliY2Px8/MjLi6O9PR0OnbsyKeffkqZMmWc07zzzjs8+uijztAHCA4O5vXXX6dMmTIcOHCA6Oho7rzzTg4ePMjcuXNZtGgR33zzDXl5eZw6dYphw4bRunVr9uzZw/DhwykoKMAYw2OPPUZ0dDRwegeje/fuZGZmUrlyZaZMmUJAQADp6emMHTuWQ4cOYbfbCQ8Pp1+/fhQWFjJu3Dh++uknvL29qVmzJq+++irly5cv8Xl99dVXvP322+Tl5REZGUn//v2B0wF6uXXOnj2bL7/8EofDQY0aNYiPj6dKlSr8/vvvxMXFcerUKe644w5Onjx5zm0nLS2N0aNHc/DgQYwxdOrUiT59+nDgwAF69uxJaGgoW7du5cSJEwwdOpTWrVsXm/7kyZO89dZbLF68mMqVKwPg7e3Nyy+/zFdffUVBQQFvvfVWse0gJiaGUaNGceTIETIzM6lRowZTp07l5ptv5pNPPmH+/Pl4e3vj6+vL2LFjqV27NgCffvop8fHxZGVl0bFjRwYPHnze34nzOdf89+7dy6pVq/j+++8pU6YMWVlZHD16lFGjRl3y78e3337LW2+9RUFBAVlZWXTq1IlBgwYRGxsLwNNPP83bb7+Nh4fHObefM7V9+OGH+Pn5UadOnUtepkqVKhETE8MLL7xAr169OHLkyHnX79UuD8Dbb7/NwoULKV++PEFBQXzzzTesWrWKgoICEhIS2LRpE0VFRdx7772MGDECPz+/y/6c5ByMyP9v//79pkGDBqaoqMhER0ebiRMnGmOMeffdd82wYcOMMcYMGzbMvPvuu85pzn7csmVLM2HCBGOMMf/5z39M3bp1zc6dO40xxjz33HNm9uzZztfFx8cbY4xJS0szwcHBZvfu3Wbv3r0mIiLCZGVlGWOM2b17t2nWrJnJzc0106ZNM2FhYcZut5eo+/fffzdNmzY1f/75pzHGmPXr15tmzZqZ7Oxss3HjRhMeHn7O5Y2IiDCrV6++4PqoU6eO2bRpkzHGmAMHDpju3bubU6dOGWOMWbZsmYmIiDDGGBMbG2veeustY4wxGRkZZtCgQaaoqMhMmzbNtGrVyhw5csQYY0z//v3NjBkzjDHGdO/e3XzzzTfGGGPy8vJM9+7dzX/+8x+zadMm07ZtW+NwOIwxxkyePNls2bKlRH1PPfWU6du3r7Hb7SY7O9u0bdvWrF69+orqXLx4sRk0aJBz/c6fP9/06dPHGGNMx44dzYIFC4wxxmzevNncfffdZuPGjSXqiY6ONu+//74xxpgTJ06YyMhIs2zZMud6XLVqlTHGmKSkJNOiRYsS02/fvt00adLkvJ+HMabEdvDBBx84l8fhcJg+ffqY9957zxQWFpp69eqZ9PR0Y4wxixcvNvPnzzfGnN7+xo4d61wHgYGBJjU1tcR7DRs2zDRv3tx06NCh2L9du3ZdcP5n/05MmzbNjBkzxvm+F/v9cDgc5qmnnjJ79+41xpz+/bjnnnuc20+dOnWcP59v+/n1119NcHCwycjIMMYYM3LkSNOyZctzrs+z53dGbm6us/1867c0lmft2rUmLCzMHD9+3DgcDhMbG+usc/r06WbixInO34HXX3/d+TdDrp56/FKCh4cHr732Gp06daJ58+aXNW2bNm0AuOWWW6hcuTJ169YFoFatWhw/ftz5uq5duwJQpUoVmjVrxoYNG/D09CQjI4OePXs6X2ez2fjzzz8BaNCgAV5eJTfZjRs30qRJE2655RbgdK+9UqVKJCcnY7PZzlurzWbD4XBccHm8vLxo0KABADVq1GDy5MksXbqUffv2sXXrVnJzcwFo3bo1w4YNY9u2bQQHBzNixAg8PE4PoWnWrJnzUGrdunXJysri5MmTbNq0iePHj/Pmm28Cp3u8u3btonnz5nh6etKlSxeaN29OWFgY999//znre+yxx/Dy8sLPz4+wsDDWr19PaGjoZdf57bffsn37dh599FEAHA4Hp06d4ujRo/z222906tQJgEaNGp3zFMjJkyf56aefeP/99wGoUKECUVFRrF27lvr16+Pt7U1oaCgA9957L8eOHSsxDw8Pj4t+HlB8O3j66afZvHkz//rXv0hJSeG///0v9evXx9PTk7Zt29K1a1datGhB8+bNne8Pp8cpAPj7+1O5cmWOHDlCtWrVSrzXhQ71X2j+53Ox3w+bzcacOXNYvXo1y5YtY8+ePRhjOHXqVLH5XGj7SUtLo1mzZvj7+wPwxBNPsG7duovWdsaZ3xlfX9/zrt/SWJ41a9bQtm1bbrjhBgCio6PZuHEjAKtXryY7O5v169cDYLfbufnmmy95GeTCFPxyTtWqVWPMmDEMGzbM+UcfTv9RMGd9vYPdbi823dmnBry9vc87/zOhCKdDxsvLi6KiIoKDg5k6darzuUOHDhEQEMBXX31FuXLlzjkvh8NRIuCNMRQWFl6whgYNGvDLL7/QsmXLYu0zZsygVq1aPPDAA/j4+DhDZseOHTz33HP07NmTZs2a0bhxY8aMGQNAy5YtWblyJevXr2fDhg3MnDmTRYsWARTbWTmz/hwOB8YY5s+fT9myZYHT51d9fX0pX748S5Ys4aeffmLjxo0MGjSI3r17Ow/Jn83T07PYMnt5eV1RnQ6Hgz59+tCtWzcACgoKiu2onf2Zn2vn68zy/LWtsLAQOL0tnPnMz7czVrt2bQoLC0lJSeG2225ztufn5/P888/zyiuvABTbDl577TW2bdvGo48+ykMPPURhYaGzjoSEBHbv3s369et5++23WbJkiTMkz/WZXK4Lzf98Lvb7cfLkSTp37swjjzxCUFAQjz76KF9//fU51+35tp9PP/202OvP3kYuxfbt26lZsybly5e/4Pq92uXx8vI6b50Oh4O4uDjnzlRubi75+fmXtRxyfhrVL+fVtm1bQkJC+PDDD51tN910E8nJyQCkp6fz448/XtG8Fy9eDEBqaiobNmwgODiY4OBgvv/+e/bs2QPAmjVr6NChA3l5eRecV3BwMOvWrWP//v3A6QF6hw4dKtYzOZfevXvz2WefFesNrV27lrlz5zp7LmfbtGkTgYGB9OrViwcffJBvvvmGoqIiAP7v//6P5cuXEx4eTnx8PH5+fs4jFefi5+dHgwYN+Ne//gXAiRMnePLJJ/nmm2/49ttv6dmzJw0bNmTgwIF06tTJuc7/KjExEWMMx48fZ8WKFTz88MNXVGfz5s1ZuHAhOTk5ALz55pu8/PLL3HTTTdSrV895pceOHTvYvXv3OZenfv36zhHh2dnZJCYm0rRp0wt+Bmfz8fHhf//3fxk+fDiHDx8GTu+ATJgwgVOnTjnHh5xt3bp1PP3003Tq1Imbb76Z9evXU1RURFZWFqGhoVSsWJGePXsyaNCgUr2S5ELz9/T0dO7wXK59+/aRk5PDoEGDaNWqFT/88AMFBQXOIyFn5n2h7adZs2Z8//33pKWlAf/vd+1SpKenk5CQ4BzYd771WxrLExoaypdffkl2djYACxcudE7XvHlz5s2b53ztyJEjmTJlyiW/r1yYevxyQSNGjGDLli3Ox927d2fIkCGEhYVRs2ZNmjRpckXzzc/Pp3PnztjtdkaMGMHtt98OwNixY3nppZecPYLZs2efc1Db2WrXrk18fDzPP/88RUVFlClThjlz5lChQoULTnfrrbcyZ84cpk6dyqRJk3A4HFSqVInZs2dTp04dDhw4UOz1ERERfPnll7Rr1w6Hw0HLli05fvw4OTk5PPfccwwfPpxPP/0UT09PHnnkERo3bnzBy94SEhIYN24ckZGRFBQUEBERQYcOHSgqKmLt2rVERERQrlw5brzxRsaNG3fOeZw5pJ6Xl8dTTz1FkyZNqF279mXXGRQURHp6Oo8//jg2m41q1aoxceJEAKZMmUJsbCzz58+nVq1a573aIyEhgbFjx7Jo0SIKCgqIjIwkKiqKgwcPXvBzOFu/fv0oW7as8/B6fn4+Dz74ILNmzTrn6wcMGMDkyZN588038fb25oEHHuDPP/+kUqVK9O/fn549e1KmTBk8PT2dRwwuxwcffMAXX3xRrK1+/fqMHTv2vPMPCQlxrrvLdffdd9OiRQvatWuHj48PderUoXbt2uzbt49atWrRtm1bunfvzvTp08+7/QAMHTqUp59+mvLly5/3NNEZTz/9NB4eHs4e96OPPuo8unS+9Vsay/Pwww/z+OOP88QTT1CmTBnuuusu59GL5557jkmTJtG5c2eKioq45557SvWSYquzmSs5xiUiInIVtm/fzs8//0yPHj0A+Ne//sXWrVuLneqTv4eCX0RE/nE5OTnExcXxxx9/OI8yjRs37pyndKR0KfhFREQsRIP7RERELETBLyIiYiHX/ah+h8NBbm4u3t7eF7yZi4iIyPXAGIPdbqd8+fLF7plyxnUf/Lm5uee87lhEROR6VqdOnXNe1nzdB/+Zu0nVqVOnxBfOSHHJyckEBga6ugy5jmibktKmberiCgoK2L1793nvXHrdB/+Zw/s+Pj74+vq6uJprn9aRlDZtU1LatE1dmvOd3tbgPhEREQtR8IuIiFiIgl9ERMRCFPwiIiIWouAXERGxEAW/iIiIhSj4RURELETBLyIiYiEKfhEREQtR8IuIiFiIgl9ERMRCFPwuUmgvcnUJJTRq1MjVJZzTtbiuRETc1XX/JT3XKi9vT2b2W+XqMtzCgDmtXF2CiMh1Qz1+ERERC1Hwi4iIWIiCX0RExEIU/CIiIhai4BcREbEQBb+IiIiFKPhFREQsRMEvIiJiIQp+ERERC1Hwi4iIWIiCX0RExEIU/CIiIhai4BcREbEQBb+IiIiFKPhFREQsRMEvIiJiIQp+ERERC1Hwi4iIWIiCX0RExEIU/CIiIhai4BcREbEQBb+IiIiFKPhFREQsRMEvIiJiIZcU/DNmzCA8PJzw8HAmT54MQGxsLG3atKFjx4507NiRr776CoCdO3cSFRVFWFgYw4cPp7CwEIDU1FSio6Np27Yt/fv3Jzc3F4ATJ07w7LPP0q5dO6Kjo8nMzASgoKCAoUOH0q5dOzp37syePXtKfeFFRESs5qLBv379etatW8fixYtJTExkx44dfPXVVyQnJ/Pxxx+zZMkSlixZQuvWrQEYOnQoo0aNYuXKlRhjWLBgAQBjxoyhW7duJCUlERgYyKxZswCYOnUqQUFBrFixgi5dujB+/HgA5s6dS9myZVmxYgVxcXHExsb+XetARETEMi4a/P7+/sTExODj44O3tzd33nknqamppKamEhcXR2RkJNOmTcPhcHDw4EHy8vJo0KABAFFRUSQlJWG329m0aRNhYWHF2gFWr15NZGQkABEREaxduxa73c7q1avp0KEDAI0bNyYrK4vU1NS/Yx2IiIhYxkWD/6677nIGeUpKCitWrODhhx+mSZMmTJgwgQULFrB582YWLlxIRkYG/v7+zmn9/f1JT0/n6NGj+Pn54eXlVawdKDaNl5cXfn5+ZGVlnXNeaWlppbbgIiIiVuR1qS/873//S9++fXn55Ze54447mDlzpvO57t27k5iYyJ133onNZnO2G2Ow2WzO/8/218dnT+Ph4VFimjPtVyo5OfmKp/07NGrUyNUluJUtW7a4ugS5QvrspLRpm7o6lxT8W7Zs4YUXXiAuLo7w8HB+++03UlJSnIfujTF4eXlRtWpV5+A8gMOHDxMQEEClSpXIzs6mqKgIT09PMjMzCQgIACAgIIDDhw9TtWpVCgsLyc3NpWLFilSpUoWMjAxq1apVbF5XKjAwEF9f3yueXlxLO0ruacuWLfrspFRpm7q4/Pz8C3Z2L9qFPnToEAMGDCAhIYHw8HDgdNBPmDCB48ePY7fb+fTTT2ndujU1atTA19fXuTe2ZMkSQkJC8Pb2JigoiOXLlwOQmJhISEgIAKGhoSQmJgKwfPlygoKC8Pb2JjQ0lCVLlgCwefNmfH19qV69+pWvCREREbl4j/+9994jPz+fiRMnOtu6du3Ks88+y5NPPklhYSFt2rQhIiICgISEBEaMGEFOTg716tWjR48eAMTHxxMTE8Ps2bOpVq0aU6ZMAeDFF18kJiaG8PBwKlSoQEJCAnD69MGoUaMIDw/Hx8fHeRmhiIiIXDmbMca4uoi/05lDHtfiof6Z/Va5ugS3MGBOK1eXIFdIh2WltGmburiL5Z7u3CciImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELuaTgnzFjBuHh4YSHhzN58mQA1q9fT2RkJG3atOGNN95wvnbnzp1ERUURFhbG8OHDKSwsBCA1NZXo6Gjatm1L//79yc3NBeDEiRM8++yztGvXjujoaDIzMwEoKChg6NChtGvXjs6dO7Nnz55SXXARERErumjwr1+/nnXr1rF48WISExPZsWMHy5YtIy4ujlmzZrF8+XKSk5NZs2YNAEOHDmXUqFGsXLkSYwwLFiwAYMyYMXTr1o2kpCQCAwOZNWsWAFOnTiUoKIgVK1bQpUsXxo8fD8DcuXMpW7YsK1asIC4ujtjY2L9rHYiIiFjGRYPf39+fmJgYfHx88Pb25s477yQlJYVbb72VW265BS8vLyIjI0lKSuLgwYPk5eXRoEEDAKKiokhKSsJut7Np0ybCwsKKtQOsXr2ayMhIACIiIli7di12u53Vq1fToUMHABo3bkxWVhapqal/xzoQERGxjIsG/1133eUM8pSUFFasWIHNZsPf39/5moCAANLT08nIyCjW7u/vT3p6OkePHsXPzw8vL69i7UCxaby8vPDz8yMrK+uc80pLS7v6JRYREbEwr0t94X//+1/69u3Lyy+/jKenJykpKc7njDHYbDYcDgc2m61E+5n/z/bXx2dP4+HhUWKaM+1XKjk5+Yqn/Ts0atTI1SW4lS1btri6BLlC+uyktGmbujqXFPxbtmzhhRdeIC4ujvDwcH788UfnIDyAzMxMAgICqFq1arH2w4cPExAQQKVKlcjOzqaoqAhPT0/n6+H00YLDhw9TtWpVCgsLyc3NpWLFilSpUoWMjAxq1apVbF5XKjAwEF9f3yueXlxLO0ruacuWLfrspFRpm7q4/Pz8C3Z2L9qFPnToEAMGDCAhIYHw8HAA6tevz969e9m3bx9FRUUsW7aMkJAQatSoga+vr3NvbMmSJYSEhODt7U1QUBDLly8HIDExkZCQEABCQ0NJTEwEYPny5QQFBeHt7U1oaChLliwBYPPmzfj6+lK9evUrXxMiIiJy8R7/e++9R35+PhMnTnS2de3alYkTJzJw4EDy8/MJDQ2lbdu2ACQkJDBixAhycnKoV68ePXr0ACA+Pp6YmBhmz55NtWrVmDJlCgAvvvgiMTExhIeHU6FCBRISEgDo3r07o0aNIjw8HB8fH+dlhCIiInLlbMYY4+oi/k5nDnlci4f6Z/Zb5eoS3MKAOa1cXYJcIR2WldKmberiLpZ7unOfiIiIhSj4RURELETBLyIiYiEKfhEREQtR8IuIiFiIgl9ERMRCFPwiIiIWouAXERGxEAW/iIiIhSj4RURELETBLyIiYiEKfhEREQtR8IuIiFiIgl9ERMRCFPwiIiIWouAXERGxEAW/iIiIhSj4RURELETBLyIiYiEKfhEREQtR8IuIiFiIgl9ERMRCFPwiIiIWouAXERGxEAW/iIiIhSj4RURELETBLyIiYiEKfhEREQtR8IuIiFiIgl9ERMRCFPwiIiIWouAXERGxEAW/iIiIhSj4RURELETBLyIiYiEKfhEREQtR8IuIiFiIgl9ERMRCFPwiIiIWouAXERGxEAW/iIiIhSj4RURELOSSgj8nJ4eIiAgOHDgAQGxsLG3atKFjx4507NiRr776CoCdO3cSFRVFWFgYw4cPp7CwEIDU1FSio6Np27Yt/fv3Jzc3F4ATJ07w7LPP0q5dO6Kjo8nMzASgoKCAoUOH0q5dOzp37syePXtKfcFFRESs6KLBv3XrVp588klSUlKcbcnJyXz88ccsWbKEJUuW0Lp1awCGDh3KqFGjWLlyJcYYFixYAMCYMWPo1q0bSUlJBAYGMmvWLACmTp1KUFAQK1asoEuXLowfPx6AuXPnUrZsWVasWEFcXByxsbGlvdwiIiKWdNHgX7BgAfHx8QQEBABw6tQpUlNTiYuLIzIykmnTpuFwODh48CB5eXk0aNAAgKioKJKSkrDb7WzatImwsLBi7QCrV68mMjISgIiICNauXYvdbmf16tV06NABgMaNG5OVlUVqamqpL7yIiIjVeF3sBWd64WccPnyYJk2aEB8fT4UKFejbty8LFy7krrvuwt/f3/k6f39/0tPTOXr0KH5+fnh5eRVrB8jIyHBO4+XlhZ+fH1lZWcXaz0yTlpZG9erVr36JRURELOyiwf9Xt9xyCzNnznQ+7t69O4mJidx5553YbDZnuzEGm83m/P9sf3189jQeHh4lpjnTfjWSk5OvavrS1qhRI1eX4Fa2bNni6hLkCumzk9KmberqXHbw//bbb6SkpDgP3Rtj8PLyomrVqs7BeXD6yEBAQACVKlUiOzuboqIiPD09yczMdJ42CAgI4PDhw1StWpXCwkJyc3OpWLEiVapUISMjg1q1ahWb19UIDAzE19f3quYhrqMdJfe0ZcsWfXZSqrRNXVx+fv4FO7uX3Y02xjBhwgSOHz+O3W7n008/pXXr1tSoUQNfX1/nntiSJUsICQnB29uboKAgli9fDkBiYiIhISEAhIaGkpiYCMDy5csJCgrC29ub0NBQlixZAsDmzZvx9fXVYX4REZFScNk9/rp16/Lss8/y5JNPUlhYSJs2bYiIiAAgISGBESNGkJOTQ7169ejRowcA8fHxxMTEMHv2bKpVq8aUKVMAePHFF4mJiSE8PJwKFSqQkJAAnD59MGrUKMLDw/Hx8WHy5MmltbwiIiKWZjPGGFcX8Xc6c8jjWjzUP7PfKleX4BYGzGnl6hLkCumwrJQ2bVMXd7Hc0537RERELETBLyIiYiEKfhEREQtR8IuIiFiIgl/kOlFoL3J1CSVci4OwrsX1JPJPuuzL+UTk2uTl7akrRS6BrhIRq1OPX0RExEIU/CIiIhai4BcREbEQBb+IiIiFKPhFREQsRMEvIiJiIQp+ERERC1Hwi4iIWIiCX0RExEIU/CIiIhai4BcREbEQBb+IiIiFKPhFREQsRMEvIiJiIQp+ERERC1Hwi4iIWIiCX0RExEIU/CIiIhai4BcREbEQBb+IiIiFKPhFREQsRMEvIiJiIQp+ERERC1Hwi4iIWIiCX0RExEIU/CIiIhai4BcREbEQBb+IiIiFKPhFREQsRMEvIiJiIQp+ERERC1Hwi4iIWIiCX0RExEIU/CIiIhai4BcREbEQBb+IiIiFXFLw5+TkEBERwYEDBwBYv349kZGRtGnThjfeeMP5up07dxIVFUVYWBjDhw+nsLAQgNTUVKKjo2nbti39+/cnNzcXgBMnTvDss8/Srl07oqOjyczMBKCgoIChQ4fSrl07OnfuzJ49e0p1oUVERKzqosG/detWnnzySVJSUgDIy8sjLi6OWbNmsXz5cpKTk1mzZg0AQ4cOZdSoUaxcuRJjDAsWLABgzJgxdOvWjaSkJAIDA5k1axYAU6dOJSgoiBUrVtClSxfGjx8PwNy5cylbtiwrVqwgLi6O2NjYv2PZRURELOeiwb9gwQLi4+MJCAgAYNu2bdx6663ccssteHl5ERkZSVJSEgcPHiQvL48GDRoAEBUVRVJSEna7nU2bNhEWFlasHWD16tVERkYCEBERwdq1a7Hb7axevZoOHToA0LhxY7KyskhNTS31hRcREbEar4u94Ewv/IyMjAz8/f2djwMCAkhPTy/R7u/vT3p6OkePHsXPzw8vL69i7X+dl5eXF35+fmRlZZ1zXmlpaVSvXv0qFlVEREQuGvx/5XA4sNlszsfGGGw223nbz/x/tr8+PnsaDw+PEtOcab8aycnJVzV9aWvUqJGrS3ArW7ZscXUJ1zxtU5dO25N70+d3dS47+KtWreochAeQmZlJQEBAifbDhw8TEBBApUqVyM7OpqioCE9PT+fr4fTRgsOHD1O1alUKCwvJzc2lYsWKVKlShYyMDGrVqlVsXlcjMDAQX1/fq5qHuI5CTUqTtif3tWXLFn1+F5Gfn3/Bzu5ld6Pr16/P3r172bdvH0VFRSxbtoyQkBBq1KiBr6+vc09syZIlhISE4O3tTVBQEMuXLwcgMTGRkJAQAEJDQ0lMTARg+fLlBAUF4e3tTWhoKEuWLAFg8+bN+Pr66jC/iIhIKbjsHr+vry8TJ05k4MCB5OfnExoaStu2bQFISEhgxIgR5OTkUK9ePXr06AFAfHw8MTExzJ49m2rVqjFlyhQAXnzxRWJiYggPD6dChQokJCQA0L17d0aNGkV4eDg+Pj5Mnjy5tJZXRETE0mzGGOPqIv5OZw55XIuH+mf2W+XqEtzCgDmtXF2C29A2dXHantybDvVf3MVyT3fuExERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iInFOhvcjVJZTQqFEjV5dQwrW4ni7Ey9UFiIjItcnL25OZ/Va5uoxr3oA5rVxdwmVRj19ERMRCFPwiIiIWouAXERGxEAW/iIiIhSj4RURELETBLyIiYiEKfhEREQu5quv4u3fvTlZWFl5ep2czduxYcnNzefXVV8nPz6ddu3YMHjwYgJ07dzJ8+HByc3MJCgpizJgxeHl5kZqaytChQzly5Ai33347CQkJlC9fnhMnTjBkyBD2799PpUqVmDp1Kv7+/le/xCIiIhZ2xT1+YwwpKSksWbLE+e/uu+8mLi6OWbNmsXz5cpKTk1mzZg0AQ4cOZdSoUaxcuRJjDAsWLABgzJgxdOvWjaSkJAIDA5k1axYAU6dOJSgoiBUrVtClSxfGjx9fCosrIiJibVcc/H/88QcAzzzzDB06dODjjz9m27Zt3Hrrrdxyyy14eXkRGRlJUlISBw8eJC8vjwYNGgAQFRVFUlISdrudTZs2ERYWVqwdYPXq1URGRgIQERHB2rVrsdvtV7OsIiIilnfFwX/ixAmCg4OZOXMmH3zwAfPnzyc1NbXY4fiAgADS09PJyMgo1u7v7096ejpHjx7Fz8/PeargTDtQbBovLy/8/PzIysq60nJFRESEqzjH37BhQxo2bOh8/NhjjzFt2rRiX6BgjMFms+FwOLDZbCXaz/x/tr8+PnsaD48rH4uYnJx8xdP+Ha7FL5q4lm3ZssXVJVzztE1dOm1Pl0bb1KVzp23qioN/8+bN2O12goODgdPBXKNGDTIzM52vyczMJCAggKpVqxZrP3z4MAEBAVSqVIns7GyKiorw9PR0vh5OHy04fPgwVatWpbCwkNzcXCpWrHil5RIYGIivr+8VTy+upT9AUpq0PUlpu5a2qfz8/At2dq+4C52dnc3kyZPJz88nJyeHxYsX89JLL7F371727dtHUVERy5YtIyQkhBo1auDr6+vcI1qyZAkhISF4e3sTFBTE8uXLAUhMTCQkJASA0NBQEhMTAVi+fDlBQUF4e3tfabkiIiLCVfT4W7ZsydatW+nUqRMOh4Nu3brRsGFDJk6cyMCBA8nPzyc0NJS2bdsCkJCQwIgRI8jJyaFevXr06NEDgPj4eGJiYpg9ezbVqlVjypQpALz44ovExMQQHh5OhQoVSEhIKIXFFRERsbaruo5/0KBBDBo0qFhbcHAwX3zxRYnX1q1bl4ULF5Zor1GjBnPnzi3RXrFiRebMmXM15YmIiMhf6M59IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYiIJfRETEQhT8IiIiFqLgFxERsZBrOviXLl1K+/btadOmDfPmzXN1OSIiIm7Py9UFnE96ejpvvPEGixYtwsfHh65du/LQQw9Ru3ZtV5cmIiLitq7ZHv/69etp0qQJFStWpFy5coSFhZGUlOTqskRERNzaNdvjz8jIwN/f3/k4ICCAbdu2XfZ8jDEAFBQUlFptpcXXz+bqEtxCfn6+q0twG9qmLk7b0+XRNnVx19o2dSbvzuTfX12zwe9wOLDZ/t8GZ4wp9vhS2e12AHbv3l1qtZWWhk/e4OoS3EJycrKrS3Ab2qYuTtvT5dE2dXHX6jZlt9spU6ZMifZrNvirVq3K5s2bnY8zMzMJCAi47PmUL1+eOnXq4O3tfUU7DiIiIu7EGIPdbqd8+fLnfP6aDf6mTZsyffp0srKyKFu2LF9++SXjxo277Pl4eHhQoUKFv6FCERGRa9O5evpnXLPBX6VKFQYPHkyPHj2w2+089thj3H///a4uS0RExK3ZzPnO/ouIiMh155q9nE9ERERKn4JfRETEQhT8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4RERELUfCLiIhYyDV7y175Z6SlpZGWloaHhwcBAQFUrVrV1SWJiJTgcDj49ddfqVWrFjfcoG8MvBoKfovau3cvMTExHD16lMqVK2OM4fDhw5QpU4bXXnuNunXrurpEEbGwffv2MXjwYF544QWaNm1KdHQ0R44cweFw8Prrr9OoUSNXl+i2FPwW9X//93/ExcURFBRUrH3z5s3ExcWxaNEiF1Um7mrPnj2sXLmy2BGkhx9+mPvuu8/VpYkbeuWVV+jduzehoaEsXLiQkydP8uWXX7J//35iY2OZP3++q0t0WzrHb1F5eXklQh8gKCiIgoICF1Qk7mzevHm89NJLANx3333Uq1cPgJEjR/L++++7sjRxU+np6YSHh2Oz2Vi/fj1hYWF4eXlx++23k5OT4+ry3Jp6/BYVGBjI6NGjiYyMJCAgAIDMzEwSExMJDAx0cXXibj766CMSExMpW7ZssfZevXrRuXNnnnnmGRdVJu7qzBfHGmP44YcfiI6Odj4+efKkK0tzewp+ixo/fjxz587lzTffJCMjA2MMVapUITQ0lO7du7u6PHEzXl5eFBYWlmjPy8vD29vbBRWJu7v77rt5++23KSgowMfHhwceeICCggLef/99GjRo4Ory3JrNnNmtEksqKiriyJEjeHp6UrFiRTw9PV1dkrihpUuXMnXqVIKDg/H398dms5GRkcHGjRsZPHgw4eHhri5R3Ex2djavv/46hw8fpn///tSrV4/Ro0ezZ88e3njjDSpXruzqEt2Wgt+ijhw5wiuvvMJ3332Hn58fxhhyc3MJCgpi1KhRVK9e3dUliptJT09nw4YNxY4gBQcHU6VKFVeXJiJn0eA+ixo0aBAtWrTghx9+YPXq1axZs4YffviB9u3bM2TIEFeXJ24oJyeHgwcPcuDAAQ4dOkRqaioZGRmuLkvclMPh4OOPP2b37t3A6XEkkZGRDBs2TIP7rpKC36KOHDlCx44dix3a9/T0pEOHDpw4ccKFlYk7OjOq32azaVS/lIrXX3+ddevWUa5cObZs2cKbb75JbGwstWvXZty4ca4uz61pcJ9F3XLLLbzzzjt06NABf39/4PSo/iVLlnDLLbe4uDpxNxrVL6Vt7dq1LF68GC8vLz788EPCwsJo2rQpTZs2pV27dq4uz62px29RCQkJpKWl0a1bN+rXr8/9999Pt27dSE9PZ+LEia4uT9yMRvVLafPw8MDL63Tf9Mcff6R58+bO5xwOh6vKui6ox29RFSpUYOTIkYwcOdLVpch1oF+/fnTq1Om8o/pFLlfZsmVJTU0lNzeXPXv20LRpUwB27dqFn5+fi6tzbxrVb1FTp05l0KBBZGdnM3bsWNauXYuXlxetW7dmyJAh+sWSy3b2qH6Hw0HVqlU1ql+u2A8//MCQIUPIycmhX79+9O3bl08++YSZM2fy6quvEhIS4uoS3ZaC36I6d+7M4sWLGTp0KNWqVaNPnz7OUbQ7d+5k5syZri5R3Mi2bdu4//77AVi/fj1r1qzB29ubNm3aONtFLldBQQF5eXnOb+PbunUrN954I7fddptrC3NzOsdvcbt27eKll17ihhtuoGLFijz//POkpKS4uixxM/Hx8cDp0f2vvvoq1atXp3LlyowcOZKPP/7YxdWJO/r666/x8fHhhhtu4Pjx4wDUr1+f2267jXfeecfF1bk3Bb9FZWZmsnz5cqpWrcovv/zibN+2bRu+vr6uK0zc2oIFC/joo494+umn6dmzJ/PmzVPwyxU5+6hjz549iz23fPnyf7ia64uC36KGDBnCTz/9xIkTJ3jvvfcA+OCDDxgwYADDhw93cXXibgoLC3E4HFSsWBEfHx9nu4+PDx4e+jMjl+/ss9B/PSOtM9RXR6P6LapTp0506tSpWNtjjz1Gjx499IdaLlvFihVp0aIFAOPGjWPixIls2LCB1157jbZt27q2OHF7Npvtgo/l8ij4xenMSP5vv/0WDw8PmjZtqmuw5ZLMnTsXgD/++MN550cfHx9eeOEF5w6ByOVQuP99NKpfShgxYgShoaEUFRWptyZXLTMzk3LlylG+fHlXlyJupGHDhtx3330AbN++3fmzMYYdO3bw008/ubI8t6bgF5G/VadOnWjSpAmBgYFERES4uhxxEz/++OMFn3/wwQf/oUquPwp+ERFxK2+//TbPPvusq8twWwp+i9q0adMFn2/cuPE/VImIyOV54IEHdKj/Kmhwn0XNnDmTX375hfvvv7/EpTE2m42PPvrIRZWJO0pNTb3g89WrV/+HKhErUH/16qjHb1F2u50ePXrQp08f/ud//sfV5Yibi4yMJCUlhYCAgHPuSH7zzTcuqkyuR+rxXx0Fv4Xt3buXzz//nCFDhri6FHFzOTk5dOvWjfj4eBo1auTqcuQ6EBsbe97nli5dSnJy8j9YzfVFwS8ipWLbtm189tlnjBs3ztWlyHVg8eLFF3y+c+fO/1Al1x8Fv0Wlp6dz9OhR6tatC8DKlSvZtGkTt99+O48//rhu3CMicp3SvVkt6I8//uDxxx9n165dwOk96+nTp1OzZk02b95MXFyciysUEZG/i4LfgqZPn+78jvTFixfzr3/9i2bNmlGxYkVCQkLYsGEDiYmJri1SRET+Fgp+C7rpppu46aabqFGjBr6+vhw7doxHHnmE6tWr4+/vj6enJzVq1HB1meLGdAZRSkNubi52u71YW0FBAW+99ZaLKro+KPgtqE+fPqxbt465c+fy2muv0bdvXxo3bkxBQQFTpkyhdevWuoGPXJWmTZu6ugRxc/Pnz+ehhx6iWbNmzhH8SUlJhIWFsXTpUhdX5940uM+ijh07xubNm6levTr33nsvAFu3buXYsWOEhITom7HkqjRp0oSNGze6ugxxY4888ggzZszgwIEDLF26lBtuuIFVq1YxcOBAunTpgqenp6tLdFu6c59FVaxYkUceeaRYW/369V1UjVxvtOMoV6ts2bLUrVuXunXrMmLECIKDg1m5cqXz68Plyin4RUTkmnN2j/7GG29k8uTJusy4lOgcv4iIXHPOPmpUrlw5hX4pUo/foowxrFu3jhtvvNF5aR/A7t27mTRpEu+9954LqxMRq0tJSaFHjx4lfj5DXyR25RT8FjV69GjWrl1LXl4eI0eOpFWrVkyaNImFCxfqVphy1apVq+bqEsTN6ZK9v49G9VtUq1atWLp0KVlZWcTGxpKTk8PNN99MbGwstWvXdnV5IiIAHDp0iB07dgBQr1497VSWAvX4LapChQqUL1+e8uXLs2fPHvr168fTTz/t6rJERAAoKipi1KhRLFu2jNq1a2O329m/fz8RERGMGTMGDw8NUbtSWnMWdfbAmZtvvlmhLyLXlLfeeosTJ07w3Xff8fnnn/PFF1/wzTffkJWVpdMAV0nBb1FnB79Gy4rItSYpKYlJkyZxww03ONsqVarE5MmTWbZsmQsrc3861G9RO3fu5J577gFOj/A/+2ebzcbOnTtdWZ6IWJwxhnLlypVoL1++vA7zXyUFv0Wd+Upekb9bZGSk7q0ul83Dw4MDBw5Qs2bNYu379+/Hx8fHRVVdH7TbZFFff/218+fjx48Xe+6dd975p8sRN/f777+f99/Ro0ddXZ64od69ezNgwAA2b95Mfn4+ubm5fP/99/Tr14++ffu6ujy3psv5LKpz584sXry4xM/neixyMXXr1qVGjRrn/DrejIwM57eriVyORYsWMWPGDA4dOgRArVq1ePHFF2nfvr2LK3NvOtRvUWf/gf7rH2vtC8rlqlGjBp988glVqlQp8VxoaKgLKpLrQVRUFFFRUWRlZWGz2bjppptcXdJ1QYf6pcQ3qemb1eRytWnThoMHD57zudatW//D1cj1plKlSgr9UqQev0Up3KU0DRs27LzPjRgx4h+sREQuRuf4Laphw4bcd999AGzfvt35szGGHTt28NNPP7myPHFTR48eJS0tDYCqVauqlyZyDVKP36J05yspTX/++ScjR47k119/JSAgADg9qO/ee+9lzJgx3Hbbba4tUNyOw+FgwYIFrFixgvT0dDw8PAgICCAkJITu3bvrxmNXQT1+i5o0aRKPPvqovpBHSkXXrl3p1q0bERERzpurOBwOli5dyieffMKnn37q4grF3YwcORKHw0Hnzp0JCAjAGENmZiZffPEFJ0+eJCEhwdUlui31+C2qfPnyDBgwgIoVK/LYY4/Rvn17ypcv7+qyxE0dO3aMDh06FGvz8PCgY8eOzJ4920VViTvbtGkTSUlJxdpuvfVWgoKCdDnfVdKofot6/vnnWblyJTExMSQnJ9O+fXtiY2PZvHmzq0sTN1SxYkWWLVtW4jLRL774oti91kUuVfny5dm2bVuJ9p9//lmdlKukQ/0CgN1uZ/Xq1SxbtozffvutxJ62yIWkpKQQHx/Pzp07qVKlCsYY0tPTueeeexg9ejR33HGHq0sUN7Nz505efvll8vPz8ff3x2azkZGRga+vLwkJCdx9992uLtFtKfgFgB9//JGlS5eyceNGgoODGTt2rKtLEjeUlZXlvMtatWrVqFSpkosrEneXmppKRkYGDoeDqlWrUr16dVeX5PYU/Bb266+/snTpUlasWMFtt91GVFQUYWFh+Pr6uro0cTPff/89zZo1AyA7O5uxY8fy888/c++99zJq1CgqV67s4grFHX333XckJSWRlpbmHNUfGhpKmzZtXF2aW1PwW1S7du0oKCigU6dOREVFUaNGDVeXJG7s7O93GDt2LA6Hg27duvGf//yHffv2MXXqVNcWKG7nzTffZNu2bXTo0KHYqP5ly5ZRu3btC940Si5Mo/otatSoUQQHB7u6DLlOnN1/2LJlCwsXLsTb25s6deoQGRnpwsrEXS1fvpwVK1Y4Lw89IyIigoiICAX/VVDwW1RwcDCbNm1i1qxZbN++HZvNRmBgIAMGDCAoKMjV5YmbKSgoYM+ePRhjsNlsxW6u8tc/3CKXwtfXl7S0tBLn9FNTU/Hx8XFRVdcHBb9FbdiwgZdffpn+/fsTFxeH3W7n559/ZvDgwSQkJPDQQw+5ukRxI3l5eTz77LPOnn96ejpVqlQhJydHwS9XJCYmhujoaG677bZio/pTUlJ49dVXXV2eW9M5fot66qmnGD58OPfcc0+x9uTkZF599VXmzZvnosrkenLq1CkOHz7MLbfc4upSxA3l5+ezbdu2YqP669evrx7/VdKuuEXl5OSUCH2AwMBAjh8/7oKK5HpUtmxZhb5ckW3btuHr60vjxo2pVKkSO3bsYM2aNezatcvVpbk9Bb9FnTx5ksLCwhLthYWF52wXEfknxcfHAzBv3jwmTJhA1apVqVy5MiNHjuTjjz92cXXuTcFvUc2bNy/xJRdFRUW8+uqrtGjRwjVFiYj8xYIFC/joo4/o2bMnPXv2ZN68eQr+q6TBfRY1ZMgQ+vXrR+vWrQkMDKSoqIjk5GRq167NjBkzXF2eiFhcYWEhDoeDihUrFjun7+PjowGjV0mD+yzuxx9/dF7Od//99+tSPhG5JnTv3p19+/YB0LRpUyZOnMiGDRt47bXXaNGiBS+88IKLK3RfCn4REblm/fHHH5w4cYIGDRqwZcsWsrOzdTryKulQv4iIXHNOnTrFsWPHnN/suH37drZt26ZveiwFOlEiIiLXlPT0dCIjI/nyyy8B+PbbbxkwYACpqanMmTNHN/C5Sgp+ERG5psyZM4eGDRty77338uOPP/LWW2/x+OOP06ZNGwYOHMjKlSvZtGmTq8t0Wwp+ERG5phw7dgybzcbBgwf57bff2Lt3LzVr1uTgwYOkpaVRUFDAwYMHXV2m21Lwi4jINeW5557jxx9/5LfffmPx4sVER0fTqVMn7rjjDr766iuaNGlCp06dXF2m29KofhERueakpKSwevVqatSoQevWrQH47rvv2L9/P48++ii+vr4urtB9KfhFREQsRIf6RURELETBLyIiYiEKfnELBw4c4J577qFjx4507NiRyMhIunbtyvLly11d2mXJycmhd+/e5OXlsWjRIho1auRcpjP/Xn755QvOY9u2bYwaNQo4fVOT0rx16YwZM/j6668ve7qGDRty4MCBEu2tWrVi+/btV1xPTEwMDz/8cLHP/X/+53945513rniel2vEiBEkJycDMHz4cNavX19q8z558iSTJk0iLCyMyMhIIiMjeeONN8jLy3O+Zvbs2bRo0YLY2FgWLVpEixYt6N27N2+++SaJiYmlUkdOTg59+vQp9r5y/dKd+8RtlClThiVLljgfHzx4kJ49e+Lp6UlYWJgLK7t0CQkJdOnShTJlygAQFBTEW2+9dVnz+P3330lPTwfgvvvuY9q0aaVW3w8//EDt2rVLbX6loWfPnvTu3dv5ODU1lfbt29OqVSvuvPPOv/39169fzxNPPAHA+PHjS22+hYWF9OrViwYNGpCYmEjZsmU5deoUr7/+Or179+bDDz/Ey8uLhQsXkpCQQFBQED169GDw4MF07Nix1OoA8PPzIyIigjfffJNhw4aV6rzl2qPgF7dVo0YNXnjhBd577z3CwsLYu3cvY8eOJTc3l8zMTOrWrcvUqVPx9fXlvvvuo1evXqxfv56TJ0/y/PPPk5SUxO7duwkICGDOnDmUK1eOhQsX8umnn2K32zl+/Dj/+7//S7du3SgqKmLy5MmsWrWKChUqcP/997Nnzx7mzp1LdnY248ePZ/fu3djtdoKDg3n55Zfx8ir+63Xo0CG+/fZbRowYcUnLt3nzZiZOnIjD4QCgb9++3H///UybNo3s7GxiY2Pp1KkT48aNY9myZcTExFCmTBl2797NkSNHaNWqFRUrVuTbb78lMzOTV155heDg4POup4ULF5KcnMzkyZPx9PQkNDSUhIQENm3aRFFREffeey8jRozAz8+PzZs3M27cOGw2G/fdd5+zxnP55JNP2LVrFwUFBfTq1YvHHnuMESNGcPPNNzN48GAAlixZwpdffsnMmTMvul7S0tIwxuDn5wfATz/9REJCAqdOncLDw4Pnn3+eli1bsmjRIpKSknA4HKSmplKlShUmTpxIlSpVSEtLY/To0Rw8eBBjDJ06daJPnz4cOHCA6Oho7rzzTg4ePEijRo3IyMhgyJAhTJ48mYSEBKKjowkMDKRnz56EhoaydetWTpw4wdChQ2ndujWnTp0iPj6erVu3UqFCBeeO1MSJE4stx5naYmNjnW1ly5Zl+PDhdOrUia+++oqVK1eSnp7O8OHDsdvtHD16lAMHDnD06FF27drFXXfdRe/evdm6dSuvvPIKp06dwtvbm5dffpng4GD27NnD+PHjOXbsGEVFRXTv3p3HHnuMH374gfHjx1OuXDlyc3P5/PPPadeuHQkJCfTu3ZvKlStf0jYqbsqIuIH9+/ebBg0alGjfvXu3qV+/vjHGmIkTJ5rExERjjDEFBQUmIiLCJCUlGWOMqVOnjvnwww+NMca89dZbpmHDhiYtLc0UFRWZzp07my+++MLk5OSYxx9/3GRlZRljjPn555+d7/nvf//bREdHm7y8PJOfn2+eeeYZ89RTTxljjImJiTEfffSRMcaYwsJCM2TIEPP222+XqHXu3Llm2LBhzseff/65eeCBB0yHDh2K/Vu4cKExxpgePXqYZcuWGWOM2blzpxk9erRzumeffdYYY8zGjRtNeHi4McaYYcOGmS5dupiCggKTkZFh6tSp46zrgw8+ML169broenrqqafMihUrjDHGTJ8+3UycONE4HA5jjDGvv/66iY+PN/n5+aZp06Zm/fr1xhhjli5daurUqWP2799fYplbtmxp4uPjjTHGpKWlmeDgYLN7927z66+/mmbNmhm73W6MMaZbt25m7dq1JaYfNmyYad68uenQoYNp1aqVefDBB03//v3Nhg0bjDHGHDt2zLRp08b53mlpaSYkJMQcPHjQfP7556ZBgwbmjz/+MMYY89prr5mBAwcaY4yJjo4277//vjHGmBMnTpjIyEizbNkys3//flOnTh2zadOmYsuwbdu2YuvnzOtWrVpljDEmKSnJtGjRwhhjTEJCgnnppZdMUVGRyc7ONpGRkcU+9zPGjh1rJk6cWKLdGGNeffVVM27cuPO+/5l18+6775qCggLTrFkz8+233xpjjNm+fbuJiIgw+fn5pn379iY5Odm5nO3atTM///yz2bhxo6lbt645cOBAsfft27evc/uT65d6/OLWbDab87D50KFD+f7773nnnXdISUkhIyODkydPOl975nRArVq1qFOnDlWqVAGgZs2aHD9+nPLlyzNnzhzWrFlDSkoKu3btck6/Zs0aOnbs6Lx2+IknnmDu3LkArF69mu3bt7Nw4UKA854n/eOPP6hVq1axtgsd6m/Xrh1jx45l1apVNG3alJdeeumi66Nly5Z4e3vj7+9PuXLlePjhh53LfOzYsUtaT2esXr2a7Oxs5zltu93OzTffzO7du/Hy8iI4OBiAiIgI55iDc+natSsAVapUoVmzZmzYsIEePXpQs2ZNVq9eze23305GRgbNmzc/5/RnDvWfPHmSwYMH4+Pjw0MPPQTAL7/8QmZmJgMGDHC+3maz8dtvvwHQrFkzbr/9dgAef/xxOnbsyMmTJ/npp594//33AahQoQJRUVGsXbuW+vXr4+XlRYMGDS66rr29vQkNDQXg3nvvda7fNWvWEBsbi4eHB35+fnTu3NlZz18VFhaes72goABPT8+L1gCwe/duPDw8nN9YFxgYyNKlS/n999/5888/iYuLc742Ly+PX3/9lTvvvJNq1apRo0aNYvOqWbMme/fuvaT3Ffel4Be3tn37durUqQPASy+9RFFREe3ataNFixYcOnQIc9ZtKry9vc/58xlpaWk88cQTPP744zRq1Ii2bdvy7bffApQ4bO/h8f/GxTocDt58803n+eYTJ05gs9lKzN9ms13wkPhfde3alZYtW/L999/z3XffMWPGDJKSki44jY+PT7HHf60bLr6ezl6uuLg4Z7jl5uaSn59Pampqidef633O+Ou6OvPa6OhoPv/8c2677TYef/zxc66zs5UrV47JkyfTvn17PvjgA3r16kVRURF33nknn332mfN16enpVKpUiaVLlxYLT4fDgaenJw6Ho0T9DofDGcI+Pj4XXJ4zvL29nct2du1eXl7F5n/28p/tgQce4N1338XhcJRYR5s2baJ///4XrQHA09OzxLrbvXs3xhgqVKhQbFzM4cOHqVChAr/88gvlypU75zJd6g6HuC+N6he3tXfvXmbNmsUzzzwDwLp16xgwYADt27cHYOvWrRQVFV3y/JKTk6lUqRLPPfcczZs3d4Z+UVERoaGhfPHFFxQUFFBYWMjixYud0zVv3pwPPvgAYwwFBQX079+fjz/+uMT8b7/9dvbv33/J9XTt2pWdO3cSFRXFuHHjOHHiBJmZmXh6ep63p3gpLrSezp538+bNmTdvHgUFBTgcDkaOHMmUKVO4++67McawZs0aAL755huOHz9+3vc7s65SU1PZsGGD80hBWFgYO3fuZOXKlTz66KOXVPuNN97IsGHDmDZtGunp6TRo0IB9+/Y5v7Bl586dhIWFOQc/bty40fnz/PnzadmyJX5+ftSvX5958+YBkJ2dTWJiIk2bNj3ne17u+g4NDeXzzz/H4XBw6tQpli1bds6dmrCwMMqWLcuECROcR4ny8vIYN24c5cuXd96t7mLuuOMObDYb33//PQA7duzg6aef5vbbby82IPbQoUNEREQ4r1A4lwMHDjiPkMj1Sz1+cRt5eXnO0cweHh74+vry0ksvOQ9xDh48mAEDBlCuXDn8/Pxo3Lgxf/755yXPv1mzZixcuJC2bdtis9l48MEHqVSpEvv27SMqKoq9e/fSqVMnypUrR82aNSlbtixw+hKv8ePHExkZid1up2nTpvTp06fE/B955BHeffddioqKnL2qzZs3lxih7enpyaJFixgyZAgTJkxg6tSp2Gw2nn/+eWrWrElRUREzZ87k+eefp3v37pe9Hi+0nlq1asWUKVOw2+0899xzTJo0ic6dO1NUVMQ999xDTEwM3t7ezJw5k9GjRzNlyhTuuecebr755vO+X35+Pp07d8ZutzNixAhnsPj4+BAWFsbhw4epVKnSJdffoUMHPvvsMyZNmsSUKVOYNm0akydPJj8/H2MMkydPpmbNmvz4449UqVKFoUOHkpmZSe3atRk7dixw+uqKsWPHsmjRIgoKCoiMjCQqKuqcX/zSunVrhg4dyujRoy+pvr59+zJ27FgiIyOpUKECN998s/N01Nm8vLx4//33mTVrFlFRUXh4eFBUVESrVq14//33z3lU6lx8fHyYPn06EyZMYPLkyXh7ezN9+nR8fHyYNWsW48eP591336WwsJAXX3yRRo0a8cMPP5SYT0FBAb/88kupXrkg1ybdslfkEqxbt44jR444Q/qVV17B19eXoUOHXtZ8Ro4cSXBwsLO3bWUnT57kqaeeYtSoUZd0Tv1yLVq0iJUrV1725ZJX6z//+Q9+fn6EhobicDgYOHAgzZo1o1u3bv9oHZdr0aJF/Pe//9XlfBagQ/0il+Cuu+4iMTGRyMhIwsPDOXr0KP369bvs+QwdOpTPPvvM8jdK+e6772jRogUPP/zw3xL6rnTXXXcxe/ZsOnbsSEREBAEBAXTp0sXVZV1Qbm4uy5YtY+DAga4uRf4B6vGLiIhYiHr8IiIiFqLgFxERsRAFv4iIiIUo+EVERCxEwS8iImIhCn4REREL+f8ADdAS+T8E0+oAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_damage(crashes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main takeaway: Most crashes in this dataset occurred during the day."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_lighting_condition(crashes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To clarify, this graph shows the distribution of *successful vs unsuccessful airbag deployment* (these are crashes where airbag should be deployed).\n",
    "\n",
    "Main takeaway: HUGE disparity between successfully deployed airbags and unsuccessful deployment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.airbag_deployment(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main takeaway: Significant disparity in crash incidences between **Male** and **Female**. May serve as a helpful predictor in our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.crashes_by_sex(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main Takeaway: **Location, Location, Location!!** This distribution verifies a possible contributor to crashes - **traffic congestion**. These are areas in Chicago with a lot of traffic."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.top_10_crash_sites(crashes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main takeaways: \n",
    "* Significantly more crash incidences where driver vision was **NOT** impaired.\n",
    "* Yet again, we see a high incidence of **UNKNOWN** values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.driver_vision(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main takeaway: Winning by a significant majority, the primary offense committed by drivers in our dataset is **FAILING TO YIELD**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.driver_action(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main Takeaway: This is not looking like it will serve as a helpful predictor with the HUGE presence of \"UNKNOWN\" values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.seatbelt_used(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main Takeaways: If vehicles had a defect, most had issues with the **brakes**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.vehicle_defect(vehicles)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predictive Modeling\n",
    "\n",
    "### TARGET: Severity of Crash\n",
    "\n",
    "* BINARY:\n",
    "    * Class 0: No Injury\n",
    "    * Class 1: Injury\n",
    "    \n",
    "* TERNARY\n",
    "    * Class 0: No Injury\n",
    "    * Class 1: NON-INCAPACITATING injuries\n",
    "    * Class 2: INCAPACITATING or FATAL injuries\n",
    "    \n",
    "********\n",
    "\n",
    "Now, it's time to start modeling! As stated at the beginning, the target we decided had the most real-world value was **injury**. \n",
    "\n",
    "In terms of our data, we used the `MOST_SEVERE_INJURY` column in the Crashes dataset to represent the target.\n",
    "\n",
    "However, using Injury as a target proved to have a major problem: ***severe class imbalance***. \n",
    "\n",
    "Bearing this in mind, we explored both binary and ternary classification models to see which one could best explain this data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eda.class_imbalance(people)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Modeling Data Prep - BINARY Classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Feature selection:\n",
    "\n",
    "As seen above, we have an incredible amount of repetitive features and \"UNKNOWN\" values we have to accomodate for before we start modeling. If you would like to see the full list of columns dropped, consult the `models.py` file in the `src` folder.\n",
    "\n",
    "Columns were removed if they satisfied any of these conditions:\n",
    "* Over 40% of the entries were NaN values.\n",
    "* They were a slightly modified version of another column.\n",
    "\n",
    "Columns were MODIFIED to contain less 'features' if:\n",
    "* One feature was the extreme majority\n",
    "* There were more than 50 'features' (which would make our model unnecessarily complex)\n",
    "\n",
    "Essentially, we tried to condense the dataset to be as simple as possible.\n",
    "\n",
    "##### 1st Priority Metric: RECALL\n",
    "\n",
    "*We focused first and foremost on improve RECALL, since the most damaging false predictions would be **false negatives** (predicting no injury, when there was actually injury).*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from src import models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "crash_mod = models.model_mask_binary(crashes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "pp = models.Preprocessor(target_col_name = 'injured', df=crash_mod)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_split = pp.target_split([crash_mod])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "tr_te_split = pp.split_train_test(target_split)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Modeling - Imports"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you do not have xgboost (if the import cell below breaks because of xgboost), you can download it by running the following command:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install xgboost"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If xgboost is still not working, you may need to uncomment and run the code below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !brew install libomp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.metrics import confusion_matrix, plot_confusion_matrix, classification_report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `tr_te_split` value is a **list of dictionaries**, containing X_train and y_train data for **4** different train sizes: \n",
    "\n",
    "\n",
    "|Index           |Train Size Proportion |\n",
    "|----------------|----------------------|\n",
    "| tr_te_split[0] | 0.5                  |\n",
    "| tr_te_split[1] | 0.6                  |\n",
    "| tr_te_split[2] | 0.75                 |\n",
    "| tr_te_split[3] | 0.8                  |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = tr_te_split[2]['train']['X']\n",
    "\n",
    "y_train = tr_te_split[2]['train']['y']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### First Simple Model: BINARY Classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For each of our models, we are going to score using **Stratified K-Fold Cross Validation**. Stratified was chosen over regular K-folds due to major class imbalance. *Number of folds = 5* \n",
    "\n",
    "After some hyperparameter experimentation, we found we needed to increase the number of `max_iter` to `100000` to be able to converge, and found there were *slightly* better results with a **smaller C value**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr = LogisticRegression(max_iter=100000, C=0.05)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This function, `models.kfold_validation` is a wrapper for `StratifiedKFold` from `sklearn.model_selection`. It takes in the training data, splits into 5 parts, and if `smote=True`, synthetically does a combination of over and undersampling. The proportion for **oversampling** can be adjusted by changing the `minority_size` parameter (i.e 0.7 oversamples the minority class so that its size is 70% of the majority class). The undersampling then reduces the majority class by a certain amount (which can be adjusted with the `majority_reduce` parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "rec, prec, acc, roc_auc, clf = models.kfold_validation(X_train, y_train, classifier=lr, \n",
    "                        continuous_cols=['crash_hour'], \n",
    "                        categorical_cols=tr_te_split[0]['train']['X'].drop(labels=['crash_hour'], axis=1).columns, \n",
    "                        smote=False,\n",
    "                        minority_size=0.7, \n",
    "                        majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall \n",
      " 0.22264190833559222 \n",
      "\n",
      "Precision \n",
      " 0.7689385320501001 \n",
      "\n",
      "Accuracy \n",
      " 0.8663994406136029 \n",
      "\n",
      "ROC-AUC \n",
      " 0.6050310926909876 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "score_names = ['Recall', 'Precision', 'Accuracy', 'ROC-AUC']\n",
    "\n",
    "for name, score in list(zip(score_names,[rec, prec, acc, roc_auc])):\n",
    "    print(name, '\\n', np.mean(score), '\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "fsm_prepped = pp.scale_and_ohe(tr_te_split, continuous_cols=['crash_hour'], \n",
    "                               categorical_cols=tr_te_split[0]['train']['X'].drop(labels=['crash_hour'], axis=1).columns)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = fsm_prepped[2]['test']['X']\n",
    "\n",
    "y_test = fsm_prepped[2]['test']['y']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "fsm_preds = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7fc45b5326d0>"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(clf, X_test, y_test, normalize='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### FSM Takeaways (Logistic Regression):\n",
    "* Our model does a great job at predicting our majority class, which is evident by the **high accuracy**.\n",
    "\n",
    "* **LOW RECALL VALUE** of 22% indicates the model is having trouble correctly picking up on our minority class.\n",
    "\n",
    "* Confusion Matrix shows that we really need to get that bottom right corner to be a **lighter color**.\n",
    "\n",
    "*********\n",
    "\n",
    "### Model #2: Logistic Regression, with synthetically generated data (SMOTE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr2 = LogisticRegression(max_iter=100000, C=0.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "rec, prec, acc, roc_auc, clf = models.kfold_validation(X_train, y_train, classifier=lr2, \n",
    "                        continuous_cols=['crash_hour'], \n",
    "                        categorical_cols=tr_te_split[0]['train']['X'].drop(labels=['crash_hour'], axis=1).columns, \n",
    "                        smote=True,\n",
    "                        minority_size=0.7, \n",
    "                        majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall \n",
      " 0.5701121577673329 \n",
      "\n",
      "Precision \n",
      " 0.3863556977019471 \n",
      "\n",
      "Accuracy \n",
      " 0.7886865150349943 \n",
      "\n",
      "ROC-AUC \n",
      " 0.6999443496360829 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for name, score in list(zip(score_names,[rec, prec, acc, roc_auc])):\n",
    "    print(name, '\\n', np.mean(score), '\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model #2 Takeaways:\n",
    "\n",
    "* SMOTE-ing the data *did* help the model pick up on our minority class, BUT that came at a price, as both **accuracy** and **precision** were reduced.\n",
    "\n",
    "********\n",
    "\n",
    "### Model #3: Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfc = RandomForestClassifier(n_estimators=100, class_weight={0:1, 1:6})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "rec, prec, acc, roc_auc, clf = models.kfold_validation(X_train, y_train, classifier=rfc, \n",
    "                        continuous_cols=['crash_hour'], \n",
    "                        categorical_cols=tr_te_split[0]['train']['X'].drop(labels=['crash_hour'], axis=1).columns, \n",
    "                        smote=False,\n",
    "                        minority_size=0.7, \n",
    "                        majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall \n",
      " 0.5772611937786454 \n",
      "\n",
      "Precision \n",
      " 0.3129240379524425 \n",
      "\n",
      "Accuracy \n",
      " 0.7325323847512145 \n",
      "\n",
      "ROC-AUC \n",
      " 0.6694915921915989 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for name, score in list(zip(score_names,[rec, prec, acc, roc_auc])):\n",
    "    print(name, '\\n', np.mean(score), '\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model #3 Takeaways:\n",
    "\n",
    "* Overall, RandomForest performed **worse** than logistic regression; while maintaining a similar **Recall** value, Precision and Accuracy decreased even further.\n",
    "\n",
    "********\n",
    "\n",
    "### Model #4: Random Forest with SMOTED data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfc = RandomForestClassifier(n_estimators=100, class_weight={0:1, 1:6})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "rec, prec, acc, roc_auc, clf = models.kfold_validation(X_train, y_train, classifier=rfc, \n",
    "                        continuous_cols=['crash_hour'], \n",
    "                        categorical_cols=tr_te_split[0]['train']['X'].drop(labels=['crash_hour'], axis=1).columns, \n",
    "                        smote=True,\n",
    "                        minority_size=0.7, \n",
    "                        majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall \n",
      " 0.6962122245588425 \n",
      "\n",
      "Precision \n",
      " 0.24146461460598742 \n",
      "\n",
      "Accuracy \n",
      " 0.6058443594458315 \n",
      "\n",
      "ROC-AUC \n",
      " 0.6425340701392024 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for name, score in list(zip(score_names,[rec, prec, acc, roc_auc])):\n",
    "    print(name, '\\n', np.mean(score), '\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model #4 Takeaways:\n",
    "\n",
    "* SMOTE-ing **did** help with recall, but drove down precision and accuracy *even further!!*\n",
    "\n",
    "**********\n",
    "\n",
    "### Model #5: Naive Bayes!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "gnb = GaussianNB()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "rec, prec, acc, roc_auc, clf = models.kfold_validation(X_train, y_train, classifier=gnb, \n",
    "                        continuous_cols=['crash_hour'], \n",
    "                        categorical_cols=tr_te_split[0]['train']['X'].drop(labels=['crash_hour'], axis=1).columns, \n",
    "                        smote=True,\n",
    "                        minority_size=0.7, \n",
    "                        majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall \n",
      " 0.7779410085104548 \n",
      "\n",
      "Precision \n",
      " 0.255779313552683 \n",
      "\n",
      "Accuracy \n",
      " 0.6063936545187174 \n",
      "\n",
      "ROC-AUC \n",
      " 0.6760424287715405 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for name, score in list(zip(score_names,[rec, prec, acc, roc_auc])):\n",
    "    print(name, '\\n', np.mean(score), '\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model #5 Takeaways:\n",
    "\n",
    "* Looks like Bayes gave us a bit more \"bang for our buck\" in terms of Recall vs Precision/Accuracy tradeoff. \n",
    "\n",
    "* **Highest RECALL** value so far!!\n",
    "\n",
    "*********\n",
    "\n",
    "### Model #6: XGBoost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb = XGBClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[06:09:28] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[06:11:01] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[06:13:08] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[06:14:38] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[06:16:03] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    }
   ],
   "source": [
    "rec, prec, acc, roc_auc, clf = models.kfold_validation(X_train, y_train, classifier=xgb, \n",
    "                        continuous_cols=['crash_hour'], \n",
    "                        categorical_cols=tr_te_split[0]['train']['X'].drop(labels=['crash_hour'], axis=1).columns, \n",
    "                        smote=True,\n",
    "                        minority_size=0.7, \n",
    "                        majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall \n",
      " 0.46150603928710837 \n",
      "\n",
      "Precision \n",
      " 0.4475633775847535 \n",
      "\n",
      "Accuracy \n",
      " 0.8246468834027031 \n",
      "\n",
      "ROC-AUC \n",
      " 0.6772100866778674 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for name, score in list(zip(score_names,[rec, prec, acc, roc_auc])):\n",
    "    print(name, '\\n', np.mean(score), '\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model #6 Takeaways:\n",
    "\n",
    "* Along with Random Forest, this is one of our lowest performing, with a **Low Recall**, and an **Accuracy** *still lower* than our first model!\n",
    "\n",
    "********"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Modeling Data Prep: TERNARY Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "crashes_tern = pd.read_csv('data/traffic_crashes_chicago.csv', low_memory=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "people_tern = pd.read_csv('data/traffic_crashes_people.csv', low_memory=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "vehicles_tern = pd.read_csv('data/traffic_crashes_vehicles.csv', low_memory=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "tern_df, categorical_cols, numerical_cols = models.model_mask_ternary(crashes_tern, people_tern, vehicles_tern)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "UNIT_TYPE                  0\n",
       "VEHICLE_DEFECT             0\n",
       "VEHICLE_TYPE               0\n",
       "VEHICLE_USE                0\n",
       "TRAVEL_DIRECTION           0\n",
       "MANEUVER                   0\n",
       "FIRST_CONTACT_POINT        0\n",
       "TRAFFIC_CONTROL_DEVICE     0\n",
       "DEVICE_CONDITION           0\n",
       "WEATHER_CONDITION          0\n",
       "LIGHTING_CONDITION         0\n",
       "FIRST_CRASH_TYPE           0\n",
       "TRAFFICWAY_TYPE            0\n",
       "ALIGNMENT                  0\n",
       "ROADWAY_SURFACE_COND       0\n",
       "ROAD_DEFECT                0\n",
       "REPORT_TYPE                0\n",
       "CRASH_TYPE                 0\n",
       "DAMAGE                     0\n",
       "PRIM_CONTRIBUTORY_CAUSE    0\n",
       "SEC_CONTRIBUTORY_CAUSE     0\n",
       "STREET_DIRECTION           0\n",
       "PERSON_TYPE                0\n",
       "STATE                      0\n",
       "SEX                        0\n",
       "SAFETY_EQUIPMENT           0\n",
       "AIRBAG_DEPLOYED            0\n",
       "EJECTION                   0\n",
       "DRIVER_ACTION              0\n",
       "DRIVER_VISION              0\n",
       "PHYSICAL_CONDITION         0\n",
       "BAC_RESULT                 0\n",
       "MOST_SEVERE_INJURY         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tern_df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "pp = models.Preprocessor(target_col_name = 'MOST_SEVERE_INJURY', df=tern_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_split = pp.target_split([tern_df])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following code performs a ***stratified*** train test split, which we deemed necessary with the extreme class imbalance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "tr_te_split = pp.split_train_test(target_split)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "prepped_data = pp.scale_and_ohe(tr_te_split, numerical_cols, categorical_cols[:32])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ternary Classification: Using XGBoost with Bayesian Optimization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If necessary, uncomment and run the cell below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install bayesian-optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "from bayes_opt import BayesianOptimization\n",
    "import xgboost as xgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = prepped_data[2]['train']['X'].astype('int')\n",
    "\n",
    "y_train = prepped_data[2]['train']['y'].astype('int')\n",
    "\n",
    "X_test = prepped_data[2]['test']['X'].astype('int')\n",
    "\n",
    "y_test = prepped_data[2]['test']['y'].astype('int')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [],
   "source": [
    "D_train = xgb.DMatrix(X_train, label=y_train)\n",
    "D_test  = xgb.DMatrix(X_test, label=y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bo_tune_xgb(max_depth, gamma, n_estimators ,learning_rate, scale_pos_weight, min_child_weight, colsample_bytree):\n",
    "    params = {'max_depth'       : int(max_depth),\n",
    "              'gamma'           : gamma,\n",
    "              'n_estimators'    : int(n_estimators),\n",
    "              'learning_rate'   : learning_rate,\n",
    "              'subsample'       : 0.8,\n",
    "              'eval_metric'     : 'rmse',\n",
    "              'min_child_weight': min_child_weight,\n",
    "              'scale_pos_weight': scale_pos_weight,\n",
    "              'colsample_bytree': colsample_bytree,\n",
    "              'tree_method'     : 'gpu_hist'}\n",
    "#              Disable tree_method if no gpu avail\n",
    "    cv_result = xgb.cv(params, D_train, num_boost_round=200, nfold=5)\n",
    "    return -1.0 * cv_result['test-rmse-mean'].iloc[-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_bo = BayesianOptimization(bo_tune_xgb, {'max_depth' : (3, 15),\n",
    "                        'gamma' : (0, 2),\n",
    "                        'learning_rate'    : (0,1),\n",
    "                        'n_estimators'     : (100,400),\n",
    "                        'scale_pos_weight' : (5,10),\n",
    "                        'min_child_weight' : (1,10),\n",
    "                        'colsample_bytree' : (0,1)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "# WARNING: please DO NOT run the cell below unless you want enough time to get up,\n",
    "# take a shower, go for a walk, eat dinner, call an old friend, and binge the entirety\n",
    "# of 'Days of Our Lives'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_bo.maximize(n_iter=10, init_points=12, acq='ei')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bayesian Optimization - TRIAL 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These params were produced after running the above line of code. Since runtime for 200 iterations of optimization took a loooong time, we saved the hyperparameters in here. This is our first try at some optimized hyperparameters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = {'colsample_bytree': 0.5802155452308511, \n",
    "          'gamma': 0.29764960799687223, \n",
    "          'learning_rate': 0.47464750071745865, \n",
    "          'max_depth': 11.569768971164503, \n",
    "          'min_child_weight': 5.565853600455331, \n",
    "          'n_estimators': 308.84834061214656, \n",
    "          'scale_pos_weight': 7.350150415230057}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[08:38:13] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    }
   ],
   "source": [
    "xgb_opt1 = XGBClassifier(params, use_label_encoder=False).fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Confusion Matrix: TRAINING DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7fc27167f2e0>"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(xgb_opt1, X_train, y_train, normalize='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Confusion Matrix: TEST DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7fc27167f0d0>"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAFeCAYAAADHWxsNAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAA91UlEQVR4nO3deVwVVf8H8M+9l0UWAVEuoOSSuFSoZZo8PKbmhguImj0pPdomuaFJpeIWimlGJpapuaSVgmkqGGWIZvVkoKn1U9FcSHFj32RHuDO/P8hrN5bLvSzDwOfda14vZuace76DwZdz5swZhSiKIoiIiCSilDoAIiJq3piIiIhIUkxEREQkKSYiIiKSFBMRERFJykTqAKojCAIKCgpgamoKhUIhdThERJUSRRGlpaWwsrKCUln3f9+XlZVBo9EYVVelUsHEpFH/qm/ciaigoABXrlyROgwiohrp2rUrWrZsWaefWVZWhvhzv0AjWhtVX6VSwc3NrVEno8YbGQBTU1MAQNCJ35FVUiJxNLWzfqA7Zv90Quowas1l7e9Sh1AnFhycgfd8NkkdRq0oOzwkdQh1Yt5HXnh/zjdSh1ErNvYWmLZsiPZ3Vl3SaDTQiNbo3OZtmKmyDKp7T2OPPzOCodFomIiMdX84LqukBOnF8k5EAJrENVgl50odQp3Jlvm1qKwKpQ6hzuRkNI1rqc9bCGaqLJiZpNfb50upUSciIiIqpxEFaETB4DpywERERCQDAkQIMGxFNkPLS4WJiIhIBoS//jO0jhwwERERyYAgitAYuEa1IJM1rZmIiIhkQIDhQ23y6A8xERERyYIGIpQGJiKNTO4RcYkfIiKSFHtEREQyIBoxa06USY+IiYiISAY0ogiFgZMPDJ3cIBUmIiIiGRBg+OQDTlYgIqI6I0A0ePIBH2glIqI6owFgaF4x7sURDY+z5oiISFLsERERyYAAwNC1vXmPiIiI6owABQxNRYLBqUsaTERERDIgiDD4HpEgj7kKTERERHKggQIie0RERCQVwYhEZGh5qTARERHJgCAqoBANTEQGlpcKp28TEZGk2CMiIpIBDQCFwUNz8sBEREQkAwKUMHwQSymLYS8mIiIiGRBEBWDoPR9RIYtEJIcYiYiaPQ0URm3GioqKwqhRozB8+HCEhYVVWe7HH3/E4MGDtftJSUl44YUXMGLECMyYMQMFBQV622IiIiKSAUFUQmPgJojG/YpPTU1FaGgowsPDERkZiT179iAhIaFCuYyMDLz33ns6x5YvXw5fX19ER0fDzc0NGzdu1NseExERkQwIUECA0sCtvEeUkpKC27dv62y5ublVthUbGwt3d3fY2dnB0tISnp6eiI6OrlBuyZIl8Pf31+6Xlpbi1KlT8PT0BACMHz++0nr/xHtERERNnK+vLzIyMnSO+fv7Y/bs2ZWWT0tLg4ODg3ZfrVbj3LlzOmW++OILPProo+jVq5f2WHZ2NqytrWFiUp5aHBwckJqaqjc+JiIiIhkwZomf+9O9w8PDoVKpdM7Z2NhUWU8QBCgUD9oSRVFn/8qVK4iJicFnn32GlJSUKssBqLBfGSYiIiIZEIy456P8q7yTkxPMzc1rXM/JyQmnT5/W7qenp0OtVmv3o6OjkZ6ejmeffRalpaVIS0uDr68vPv/8c+Tl5UGj0UClUlWoV2WcBlwTERFJpPwekeGbMTw8PBAXF4esrCwUFRUhJiYGAwYM0J6fM2cODh8+jIMHD2LLli1Qq9UIDw+Hqakp+vTpg0OHDgEAIiMjdepVhYmIiEgGNFAatRnD0dERAQEBmDJlCsaOHQsvLy/07NkTfn5+OH/+fLV1g4KCsHfvXowaNQqnT5/G3Llz9bbHoTkiIhkQRAU0hk7HrsWip97e3vD29tY5tnXr1grlXFxccOzYMe1+u3btsHPnToPaYiIiIpKB+1OyDSGPdRU4NEdERBJjj4iISAY0ogIaI9aakwMmIiIiGRCMmHwgl6E5JiIiIhkQYPhzRIbeU5IKExERkQyUr6ZtaI+IQ3NERFRHBCPuESl5j4iIiOqKMdO35TI0J48oiYioyWKPiIhIBu6/7M4QSiNfjNfQmIhqaaDLDbzZ5yTMVBpczmqNRccHoaDUTKfMmM5X0F19CJE+BSgqM8HKE/9GfOaDFWmdrPKx1+sAfCKfQ3aJRUNfguw8NeQuXg5Mhqm5iOt/tEDom+1RmK+qURmlUsRrQXfwiPpN7DiegX2b1fh2ZxsAQPsuxXg95BYsrDQQRQW2r3LGmZ/Kl8pfuuU6Hn60CEWF5T/YZ2NbYvOydg174TLQ1z0FL712AaamAq5fs8G693qjqNC02jJKhZf23O6D3yIj/cHPwP4vu+DHow+h5xPpeHV6PFQmAu6VqPDJRz1x5ZJ9g11XYyAasYipoa+NkEq9psuavvNcrlq1KMK7T/+A2ceGY8T+SbiVZ4O3+pzQKdPJJgfz+p5AQkYgxh58DpvO9sb6ITHa8z6ulxE26iAcrQobOnxZsrUvw5trb2HFa50wdcAjSLlhjlcWJdW4zKjJmWjXqQR/pIVg9uiuGDc1Hd0eLwAA+K+6hcNf2mPm8O5Y+8ZDWPxJIpQqEQDwyJMFePPZLpg5vDtmDu/OJFQJG9sSBASewcql/fDa5GFISbLCy9Mu6C3T1mY3AKDdQ3nIzzPF7KmDtduPRx+CiYmAwKBf8eGaJ+D/6hB8ubMb3lp8RopLlJTmr7XmDNuaeSKq6TvP5ax/21s4n6HGjVw7AMDuS4/Cu3MCAFFb5p6gxJLjA1EmtAIAxGeo0caiEKZKDdQWBRjaPhGvHh4tQfTy1HtgLi6ftUTS9fJ3q3zzRWsMHpeNv3/Pqyvz7xE5iNnbGoAK+XdN8ONBOwwenw0AUKmAlnYaAICFtYB7JeU/Ho4PlcDCWsDc92/hk6OX8Obam2hpV9Zg1ywXvfum4cqlVki6Yw0A+PZgJzwz9BZ0/m0qKWNv+QsAEY+6ZUEjKBCy/n/YsP17THrxEpRKEWVlSkx+diSuXbUDIMKpbSFyc80qtN/UNeTq2w2t3qKs6TvP5czJugApBdba/ZQCa7Q0uwcr01LtsTv5Nvjpdoe/9kQsfCoWx252RKmgQlqRFWYf80TiX4mM9HNoW4qMpAdDPenJZrCyEWBpLdSojEPbUqT/7VxGshnaOJf/e3282AXP+6di1+kLWP3ln1i/0AWCRgG7NmX4/eeWWB/ogpnDu6GoQIk3PrjZAFcrLw7qQmSkPRhWy0i3gJV1GSwsy6oto1IWwcKyDEqVgP87o8bSeR6YP+dpPNk3Fd7j/wQAaDRK2LUqxhf7ovHq9Hjs392l4S6skRBFBQQDN1EmPaJ6u0dUk3eey50SIkSx4nGhkn98paIYHz5zBE5W+Zgawx6QsZRKVPo912hqVkbxj3MKBSAIgKm5gEWbEvFBQHucPGqL7r0LsPyza7hy1hKXf7dC8NRO2jo71zrhy9/jYWIqoKxUHn9xNgSFsvJ7EoKgqFGZw988+B6XAIjY64oxz/6Jg/tcAQA52S0wZcJIdO6Sg1Whx3Fzekvcud2y7i+kkTKmhyOXHlG9JSJ97zw3xPqB7nUVVp1qZaGEncVJdLcZCAAwVaWjTLDCjiFDdcqZqjLQuXUQ2pe2w43sIGx9prJhhU+wZbAHNELV75FvFDwHStp8K4ufYWdxEu2GvAXg/vd8IZb/tKhGZWxbr8Tcz4fjbjGw+tcFcGq5DyplIdac+Dec7Ndh3KpVGLeq/HNU1sFYET0cpYItVMp85Bb3KT+uzIXSZBbe+WUB+ATEA/e/7237/AfA/e/7Uby93bdGZVpZ/Iyi0vYoLisfQbBrcRJtrArx7m4vtDS/gLvFfR80ZpaChR+54W5xv4a7QKo39ZaI9L3z3BCzfzqB9OKSugqtzti3KELUuItYEHcQN3Lt8MaTJ3E51wWLjv+kLWNlcg8Hx32FnKLhGLTPAUBcpZ91+RXgtWOxjX7WXMelJ/QXqke2rUux+fvL+Oj5YCRdN8fLgUlIUpvjgzfeq1GZMS9n48mB22D+aG+sGLoa676+ig8CH8K1i7fwxYksfDHzbVw8bQXnDiVY9/VVLB8DtHu4BHNW38ICz27IyzHBq4uSkOHcEu/Nfl/C7wSgcu2kv1ADsrUrwcYd8dgwbzuS7ljjJb8LSG7dGqGr91ZbxnlcHyz23YuXp8XD5aF8rHy7H0xMBASHxOLroy746ftIfP5VNDa+/S9cjG+N9h1zsWrtNQRNu4TM9BsSXvEDdm0sMe8jL/0Fa6F8uM3AB1qb+9Cch4cH1q9fj6ysLFhYWCAmJgYrVqyor+YkkVVsgYU/D8JHg4/AVKnBzVwbLPjfYLi1TsM7/X/C2IPP4YVH49HWKh8lZacR6ZOvrftStDdySlpIGL083c00xQdvtMfSLddhYioi+YY53n+9Pbr0LETAmpuYObx7lWUA4Jsv2qBtxxKMGLQA6w+l49DONjh/ovw+X/DUTpi+/DbMzEVoNAqsm++C5BvmSL5hjoPbHbA28ioUSiDxUgusm/eQlN+GRulujjlCV/fGouCTMDEVkHLHCmtW9UGXbtmYM+93zJ46uNIyd+7OA3AI4Z91x4y5Z7Fxx/dQmQg4/mM7HP6mIwAFVix2x2uzz0GlElFWqsT7K/oiM71x/9FW1wQooDFwOrah072lohDFykbT60ZUVBQ2b96M0tJSTJgwAX5+fgbVLykpQXx8fKPtERniS8+BmHj4J/0FGzmpe0R1ZfWvCxD41Hv6CzZija1HZKyV4f/BYt+9+gs2Yvd7RG5ubjA3N6/Tz77/e/AX8zUoVuYYVLeFYId/l7xVL3HVpXp9oLWyd54TEZHhNEb0iAwtLxWurEBEJAOiaPj7iEQu8UNERHVFA4XBa83JpUckj3RJRERNFntEREQyIBix6KlcZs0xERERyYAgGr6ygqH3lKQijyiJiJo5Q9eZu78ZS9/bE44cOQJvb2+MHj0agYGBuHfvHgAgIiIC/fv3h4+PD3x8fBAaGqq3LfaIiIhkoCHXmrv/9oQDBw7AzMwMEydORL9+/eDqWr7uX2FhIYKDgxEREYE2bdogICAAEREReP755xEfH4/AwEB4edV8pQkmIiIiGRBEI+4R/dUjSklJgUql+/JIGxsb2NhUvrbl39+eAED79gR/f38AgKWlJY4dOwZTU1MUFRUhMzNT+1nnz59HYmIiNm/ejG7dumHp0qWwtbWtNk4OzRERyUD5G1qVBm33Vzr39fXFkCFDdLbPP/+8yrYqe3tCamqqThlTU1P89NNPGDRoELKzs9G/f38AgIODA2bOnImvv/4azs7OCA4O1ntt7BERETVx4eHhlfaIqlLTtycMHDgQJ0+exNq1a7Fs2TJ88MEH2LBhg/b81KlTMWzYML3xsUdERCQD5a8KN3wDyt+G4OLiorNVl4icnJyQnp6u3f/n2xNycnJw/Phx7b63tzcuX76MvLw8fPbZZ9rjoihWSICVYSIiIpKBhpw15+Hhgbi4OGRlZaGoqAgxMTEYMGCA9rwoipg3bx6SkpIAANHR0ejduzcsLS2xbds2nD17FgCwa9euGvWIODRHRCQDolh+38fQOsZwdHREQEAApkyZon17Qs+ePeHn54c5c+agR48eWLFiBaZNmwaFQgFXV1csX74cKpUK69atw7Jly1BcXIyOHTsiJCREb3tMREREMtDQq29X9vaErVu3ar8eOnQohg4d+s9q6NOnDyIiIgxqi4mIiEgGBBg+1MYlfoiIqM6UP0dk4BI/UEAOuYiTFYiISFLsERERyYBoxOrbohy6Q2AiIiKSBY1o5GQFGeQiJiIiIhkQjJi+bWh5qTARERHJgGjEoqccmiMiojrDN7QSEZGkBNHwxCLUUyx1TR4DiERE1GSxR0REJAOcrEBERJLiZAUiIpIUJysQEZGkjFr0tJK3qjZGTERERDJgzIvu2CMiIqI6IxqRiORyj0geUyqIiKjJYo+IiEgGBBFGDM3JAxMREZEMcNYcERFJipMViIhIUk15sgITERGRDLBHREREkmrKiYjTt4mISFLsERERyYAIBURD7xHJZIkf9oiIiGTg/vRtQzdjRUVFYdSoURg+fDjCwsIqnD9y5Ai8vb0xevRoBAYG4t69ewCApKQkvPDCCxgxYgRmzJiBgoICvW0xERERyYAoPrhPVNNNFI1rKzU1FaGhoQgPD0dkZCT27NmDhIQE7fnCwkIEBwdjx44d+Pbbb1FSUoKIiAgAwPLly+Hr64vo6Gi4ublh48aNettjIiIikgFRVBi1GSM2Nhbu7u6ws7ODpaUlPD09ER0drT1vaWmJY8eOoU2bNigqKkJmZiZsbGxQWlqKU6dOwdPTEwAwfvx4nXpV4T0iIiIZMGrW3F/lU1JSoFKpdM7Z2NjAxsam0nppaWlwcHDQ7qvVapw7d06njKmpKX766SfMnz8farUa/fv3R3Z2NqytrWFiUp5aHBwckJqaqjdO9oiIiGSgNj0iX19fDBkyRGf7/PPPq2xLEAQo/jbRQRRFnf37Bg4ciJMnT+KZZ57BsmXLKi1XWb1/kkWPqMNX6bBN13/Dq1HzBFy3J0sdRa19e+d3qUOoE7/fAQ7L/FpGdDCVOoQ6IyTeljqEWhGKK+9ZNBbh4eGV9oiq4uTkhNOnT2v309PToVartfs5OTmIj49H//79AQDe3t4ICAiAvb098vLyoNFooFKpKtSrCntEREQyYOhEhb8P5Tk5OcHFxUVnqy4ReXh4IC4uDllZWSgqKkJMTAwGDBigPS+KIubNm4ekpCQAQHR0NHr37g1TU1P06dMHhw4dAgBERkbq1KuKLHpERETNngjDZ8EZOWvO0dERAQEBmDJlCkpLSzFhwgT07NkTfn5+mDNnDnr06IEVK1Zg2rRpUCgUcHV1xfLlywEAQUFBCAwMxKZNm+Ds7Iy1a9fqbY+JiIhIBhr6NRDe3t7w9vbWObZ161bt10OHDsXQoUMr1GvXrh127txpUFtMREREMmDMdGxjp283NCYiIiIZENB0Fz1lIiIikgHRiHtExq6s0NA4a46IiCTFHhERkQyU94gMvUdUT8HUMSYiIiI5MGbtOE5WICKiulKbteYaOyYiIiIZaMqTFZiIiIhkgM8RERGRpJpyIuL0bSIikhR7REREMiDC8DVMZXKLiImIiEgORBgxNMclfoiIqM404S4RExERkRzwgVYiIpISnyMiIiJJcfo2ERFRPWGPiIhIDkQYfs+HQ3NERFRXeI+IiIik1Rynb+fk5FRb0c7Oro5DISKiqjTLB1rd3d2hUCggVtK3UygU+OOPP+o1MCIi+pvm2CO6dOlSQ8ZBRETVaNbTtwVBwKefforAwEDk5+dj8+bN0Gg0DREbERE1A3oTUUhICC5fvoyzZ89CFEX8/PPPePfddxsiNiIiuk80cjNSVFQURo0aheHDhyMsLKzC+aNHj8LHxwdjxozBzJkzcffuXQBAREQE+vfvDx8fH/j4+CA0NFRvW3pnzcXFxSEiIgLjx49Hy5YtsX37dvj4+BhxWUREZDzFX5uhdQyXmpqK0NBQHDhwAGZmZpg4cSL69esHV1dXAEB+fj6WLVuG/fv3w9HRER9++CHWr1+PJUuWID4+HoGBgfDy8qpxe3p7RCYmJlAqHxQzMzODiQlnfRMRNagG7BHFxsbC3d0ddnZ2sLS0hKenJ6Kjo7XnS0tLERQUBEdHRwBAt27dkJycDAA4f/48IiIi4O3tjbfeekvbU6qO3kTUtWtXhIWFQaPR4Nq1a3j77bfRvXt3466OiIiMU4tElJKSgtu3b+tsubm5VTaVlpYGBwcH7b5arUZqaqp2v1WrVhg2bBgAoLi4GFu2bMHQoUMBAA4ODpg5cya+/vprODs7Izg4WO+l6e3aLF68GKtWrUJmZiYmTZqE/v37Y8mSJXo/mIiI6pCoMGKJn/Lyvr6+yMjI0Dnl7++P2bNnV1pNEAQoFA/aEkVRZ/++vLw8zJo1C927d8e4ceMAABs2bNCenzp1qjZhVUdvIrK2tsaqVav0fhAREdWf2izxEx4eDpVKpXPOxsamynpOTk44ffq0dj89PR1qtVqnTFpaGl599VW4u7tj0aJFAMoT0/79+/HSSy/91b5Yod3K6B2ay8zMxBtvvIF+/fqhf//+WLRoUbVdOiIialycnJzg4uKis1WXiDw8PBAXF4esrCwUFRUhJiYGAwYM0J7XaDSYPn06Ro4cicWLF2t7S5aWlti2bRvOnj0LANi1a1fd9IiWLFmCrl27Yt++fdBoNNizZw/efvttrFu3Tu+HExFRHWqglRIcHR0REBCAKVOmoLS0FBMmTEDPnj3h5+eHOXPmICUlBRcvXoRGo8Hhw4cBAG5ubli5ciXWrVuHZcuWobi4GB07dkRISIje9vQmojt37mDTpk3a/QULFsDb27sWl0hERAarxT0iY3h7e1f4Xb9161YAQI8ePapcfadPnz6IiIgwqC29Q3NqtRq3bt3S7qekpOjMpiAiovqnEI3b5KDKHtH06dMBAFlZWRg7diw8PDygVCpx8uRJdOvWrcECJCIiNM9FTz09PSs9PmjQoPqKhYiIqtLAQ3MNqcpEdH9O+D+JoogbN27UW0BERFSJ5tgjuu/LL79ESEgIioqKtMfs7e3xyy+/1GtgRETUPOhNRFu2bMGOHTuwadMmzJ07Fz/88ANSUlIaIjYiIrqvCfeI9M6as7OzQ69evfDII48gMzMTM2bMwKlTpxoiNiIiuq+BXwPRkGq0+vbdu3fRoUMHnDt3DgD4YjwioganeDBhoaabka+BaGh6h+b+85//YNq0afjkk08wduxYHDlyBA8//HBDxCYLff+Vghen/QFTMw0S/7TFuncfR1GhaYUy3dXzsTk8q0KZ8G++Q2Z6C23Z/eGu+PHIQw16DU3JyaM22PGuM0pLFOj0aDECPrgJq5ZCpWVjv7NFyJz2iLx6vsK54Fc7wt6xFP6r7tR3yE3KU4Nz8PL82zA1E3H9kgVC53dCYb6q2jJKRaHO+TbOJVgX+QdmjngMudnlPyf9huTgrbXXkHbHTFvureceQVGB/nXMmgpjnguSy3NEentEEyZMwPbt22FnZ4c9e/Zg5syZNV7eJz8/H15eXrh9+3Zt42yUbOxKMHfR71i1pC+m+Q5FSpIlXp5xsdIy1zMDKpRp91Ae8vNMMfvlZ7Qbk5DxcjJV+CDgISzdeh2fHr8Ep/Yl2L6qbaVlM25mYcuKtpUuIrl3gxrxJ63rOdqmx9a+FG+8fx0rprti6uAeSL5pjpcDb+kt09Zmt/b8kPEZWPPVJbRxKtWp9+iTedi3xQmzRrlpt+aUhAA0z6G5HTt2aLc9e/Zgx44dOHToEG7duoXw8HC9H3z27FlMmjQJiYmJdRlvo9K7bxqu/tEKSbfLf2l9G9EJg4bdxt//9e+XKdE4VyjzSI8sCBoF3tvwMz7+7AdMeukylEqZ/J/TCP32kw26PV6Idg/fAwB4vZiJYwdaVUg2xYUK7Hk7CtOCKvZ2zsZa4/SPLTF6SkaFc1S93gNyceWcFZISy3v43+5SY7BPFnR+HiopY2/5CwAR9up78PDMxuLJFR+Yf+TJAjzukYeN38VjzVd/wO2pvIa4JGogVQ7NXblypVYfvHfvXgQFBWH+/Pm1+pzGzMGxCOlpFtr9jPQWsLIug4VlmXbo7X4Z5w4Vy6hUIv7vtAN2fPIoTExELAs5gcICExz8qrMUlyN76XdM0abtg7+kHZzvoTBPhcJ8pc7w3EcLHsJT4x9Hp0e/16mfmWKCTW+3w8qwP3FoZ5sGi7upcHC+h/SkB0Nn6clmsLLRwNJa0A7PVVZGpSyCpbWArDQzrJjWpdLPzstR4YeDrXH8UCs81icfQduuYuYIN2SkmFVanuRFIYqGvuHCMIMHD8YXX3wBFxcXg+uWlJQgPj6+HqKqG47WkTBVZeL23Vf/OqLBE+3+i7NJOyCILWpc5j7bFifhYB2NhIyghruIJuSHHbHISc3DuMDyVUE0ZQIW/ysEwf97A2YW5b+w4r76DbcvJuO5oNHISsrBuomfIvh/b0JTpsG2mV9iiN+/4dq3I45s+RmFOUXwmT9cykuSlbr8eXii3SScS94MjVD5qwoetn8fOcV9kVU4qP4uyAhubm4wNzev08+8/3tw5ukTSC8pNqiug3kLbOzjXi9x1SW9kxUagzX+UchJL5A6jAqeGX4L/Z9JwoqFXwIAHBwLsX6HKRb9J7JCGfOHgSXPf6lT5hnPW7ieYIPEP20BAP8elIRRYzOwZO6XUlxOjXx7PFLqEKqU9Ugr/C/BDk+02wEASL1tCmu7bujnOkVbZntMF5QUKfGhbypMcRNlJebY8mIQ/FfdRl5KBxz7+AqOAchON4GgUcDW5AgCPrhVRYvSGtHhKalD0DF4XAaeHpWN5X53AQDqdiXYcEiFBf/aWG2Z7cetdMoAQPQN4J2Rm5CbbQormzJ4TU7Dng3OuD8LbPn2P/FTVDaORfzWMBenh52TDRYcmFq/jYgwYomfeomkzumdrEBV++1XNbo9lo22LvkAgFFjE3HiZ6dKy5irkiuU6fhwLv776iUolSLMzDTwevYa/vd9u4a9iCbkyYF5uPSbJe5cK+/9fPtFG/xr+F2dMusPXcWWHy7j9fBXsGLXNZi1ELDp6GU89lQBws5cxKajl7Hp6GWMnpyJAWOyG20SaozO/M8W3Z/IR9uO5X+1j34hDXExrfSWuVvUp9rPLcpXwXtKGv49MhsA0PmxAnTrVYDTP9rWw1U0Ys1xsgLpdzfHHOtWPYGF75zCJ7u+R8eHc7HtYze4dsvG+h0/6JTp1HqdThkACN/eDXl5Ztjw+TF8/PkP+OO8PQ5HdZDykmTNrk0Z3gy9iRWvdcLUAd1x/VILvBaUhCtnLTBjKFeMr293M02xdl4nLNmUgC3fn0fH7kXY8s5D6NKjABsOxVdZ5s7d/1b7uYKgwPKpXTDhtRR8EhOPN96/jlX+nbVTu5uNJpyI9A7NCYKA7du34+rVq1i6dCnCwsIwderUGr2HvDk4fcIRp0846hxLuGyG2S8/o1PmUtrrWPJf3SG3khITfPjuEw0SZ3Px1JA8PDXkss4xm1ZF2HT0coWyTg/dw8GEis8QAcDkt7iMlTFO/WCHUz/Y6Ry7et4Es0a5VVnm3diKU+VHdOj7j8+wQsC4R+s0Vrlpys8R6U1EISEhyMrKwvnz5T+wP//8M9LT07FkyZIaNXDs2LHaRUhEROVkklgMpXdoLi4uDqtXr4a5uTmsra2xfft2rrxNRER1Rm+PyMTEBErlg3xlZmYGExNZTLYjImo6mvDq23ozSteuXREWFgaNRoNr167hs88+Q/fu3RsiNiIi+ktTvkekd2hu8eLFuHDhAjIzMzFp0iQUFBRg0aJFDREbERHdZ+jK28a8WlwientE1tbWWLVqVUPEQkREVWnOQ3PvvPNOpcdrOmuOiIhqr1kPzdnZ2Wk3Kysr/Prrrw0RFxER/V1zfqDV399fZ9/Pzw8zZsyot4CIiEh6UVFR2LRpE8rKyvDiiy/ihRde0Dl/9OhRrF+/HqIowsXFBe+++y5sbW2RlJSEefPmITMzE506dcKaNWtgZWVVbVsGL/FjbW2NtLQ0Q6sREVFtiA+G52q6GdsjSk1NRWhoKMLDwxEZGYk9e/YgISFBez4/Px/Lli3Dli1b8PXXX6Nbt25Yv349AGD58uXw9fVFdHQ03NzcsHHjxqqa0dLbI1qxYgUUivKZF6Io4sKFC3xVOBGRFBpoqC02Nhbu7u6ws7MDAHh6eiI6Olo7QlZaWoqgoCA4OpYvb9atWzdERUWhtLQUp06dwoYNGwAA48ePx3//+1/Mmzev2vb0JqJWrXRXzx0zZgzGjBlj8IUREVEt1GLWXEpKSoX1QW1sbGBjU/n7ntLS0uDg4KDdV6vVOHfunHa/VatWGDZsGACguLgYW7ZsweTJk5GdnQ1ra2vtogcODg5ITU3VG6beRHTz5k2EhITo/SAiIqo/tZk15+vri4yMDJ1z/v7+mD17dqX1BEHQjoQB5aNhf9+/Ly8vD7NmzUL37t0xbtw4pKamVihXWb1/0puILl26VGUQRETU+IWHh1faI6qKk5MTTp8+rd1PT0+HWq3WKZOWloZXX30V7u7u2kUO7O3tkZeXB41GA5VKVWm9yuhNRA4ODhg9ejR69eqlM/OBzxERETWgWgzNOTk5GfSqcA8PD6xfvx5ZWVmwsLBATEwMVqxYoT2v0Wgwffp0jBw5EjNnztQeNzU1RZ8+fXDo0CF4e3sjMjISAwYM0NtelYno3r17MDMzwxNPPIEnnuA7c4iImgtHR0cEBARgypQpKC0txYQJE9CzZ0/4+flhzpw5SElJwcWLF6HRaHD48GEAgJubG1auXImgoCAEBgZi06ZNcHZ2xtq1a/W2V2Uiev755xEREVHhOSIiImp4Db2ygre3N7y9vXWObd26FQDQo0cPXLp0qdJ67dq1w86dOw1qq8pEJIoyeSSXiKg5aI5rzZWUlODixYtVJqTHHnus3oIiIqJ/aI6J6NatW5g9e3aliUihUOD777+v18CIiOiBprzoaZWJyNXVFZGRkQ0YChERVUsmicVQBq81R0REVJeq7BH16dOnIeMgIqJqNMuhOT6wSkTUiDTHyQpERNSIMBEREZGUmuXQHBERNSLsERERkaSacCLi9G0iIpIUe0RERDLAe0RERCQ9mSQWQzERERHJAHtEREQkrSY8WYGJiIhIDpiIiIhIagqpA6gnnL5NRESSYo+IiEgOODRHRERS4qw5IiKSFntEREQkKSYiIiKSkgJGDM3VSyR1j4mIiEgOmnCPiNO3iYhIUkxEREQycH/WnKGbsaKiojBq1CgMHz4cYWFhVZabP38+Dhw4oN2PiIhA//794ePjAx8fH4SGhupti0NzRERy0IBDc6mpqQgNDcWBAwdgZmaGiRMnol+/fnB1ddUpExQUhLi4OLi7u2uPx8fHIzAwEF5eXjVuTxaJSLAyh1AsSB1GrQktLaQOodY82z4udQh1YvUp+V9LxmtPSh1CnUl/Wd7XIlib13sbtXmOKCUlBSqVSuecjY0NbGxsKq0XGxsLd3d32NnZAQA8PT0RHR0Nf39/bZmoqCgMGTJEW+a+8+fPIzExEZs3b0a3bt2wdOlS2NraVhsnh+aIiORANHID4OvriyFDhuhsn3/+eZVNpaWlwcHBQbuvVquRmpqqU2bq1Kl47rnnKtR1cHDAzJkz8fXXX8PZ2RnBwcF6L00WPSIiomavFkNz4eHhlfaIqiIIAhSKB5O/RVHU2a/Ohg0btF9PnToVw4YN01uHiYiISAZqMzTn5OQEc/OaDx86OTnh9OnT2v309HSo1Wq99fLy8rB//3689NJLAMoT2D8TYGU4NEdERDo8PDwQFxeHrKwsFBUVISYmBgMGDNBbz9LSEtu2bcPZs2cBALt27WKPiIioSWmgB1QdHR0REBCAKVOmoLS0FBMmTEDPnj3h5+eHOXPmoEePHpXWU6lUWLduHZYtW4bi4mJ07NgRISEhettjIiIikgGFKEIhGpaJDC3/d97e3vD29tY5tnXr1grlVq9erbPfp08fREREGNQWExERkRw04SV+mIiIiGSA7yMiIiJpsUdERESSMmbtOJkkIk7fJiIiSbFHREQkBxyaIyIiKXGyAhERSU8micVQTERERDLAHhEREUlLBGDoSglMREREVFeaco+I07eJiEhS7BEREckBp28TEZGUFEL5ZmgdOWAiIiKSA/aIiIhISgoYMVmhXiKpe0xERERyIIpGTN+WR5eIiYiISAY4fZuIiKiesEdERCQHnKxARERSaspDc0xERERywMkKREQkqSb8qnAmIiIiOeA9IiIiklJTfqCV07eJiKiCqKgojBo1CsOHD0dYWFiV5ebPn48DBw5o95OSkvDCCy9gxIgRmDFjBgoKCvS2xURERCQHgmjcZoTU1FSEhoYiPDwckZGR2LNnDxISEiqUmT59Og4fPqxzfPny5fD19UV0dDTc3NywceNGve0xERERyYFo5GaE2NhYuLu7w87ODpaWlvD09ER0dLROmaioKAwZMgQjR47UHistLcWpU6fg6ekJABg/fnyFepXhPSIiIhmozXNEKSkpUKlUOudsbGxgY2NTab20tDQ4ODho99VqNc6dO6dTZurUqQCAM2fOaI9lZ2fD2toaJiblqcXBwQGpqal642QiIiKSg1o8R+Tr64uMjAydU/7+/pg9e3al1QRBgELxYKqDKIo6+1U3V7FcTeoxERERyUBtekTh4eGV9oiq4uTkhNOnT2v309PToVar9bZnb2+PvLw8aDQaqFSqGtfjPSIiIjmoxT0iJycnuLi46GzVJSIPDw/ExcUhKysLRUVFiImJwYABA/SGaGpqij59+uDQoUMAgMjIyBrVYyIiIiIdjo6OCAgIwJQpUzB27Fh4eXmhZ8+e8PPzw/nz56utGxQUhL1792LUqFE4ffo05s6dq7c9Ds0REcmAAiIUBt4jUtRiaQVvb294e3vrHNu6dWuFcqtXr9bZb9euHXbu3GlQW0xERERyIPy1GVpHBpiIiIhkQCEa0SPi6ttERFRnuOgpERFJiu8jor/r2zcJL798FqamAq5ft8O6dU+hsNC02jJKxTDtudGjr2LEiGswM9MgIaEV1q17CqWlKrRvfxdz5pxCixZlAIDt23vht9+cG/Ta5OCpIbl4eWEyTM1FXL/YAqFvPoTCfFWNyiiVIl4LSsIj6jex45cM7PvEAd/ubAMAaGlXhpnv3EH7rsUwbyFi94dqfL/fHgDw7LQ0eE7MgqZMgbtZJvhwvguSb5g3+LXLSf8uN+A/9CRMVRokpLZG8NeDUFBiVklJEcvHHkNCmj12xj6uPfr9/M+Qmmul3d/5Sy98d75r/QfeSDXlN7TW6/Ttjz/+GKNHj8bo0aMREhJSn001GFvbYrzxxkm8805/+PmNRkqKFV5++azeMm1tdgMAPDxuYcyYq1i4cBCmTx8JMzMNxo69DACYNes0YmIehr//CISG9sOiRbFQKmVyt7GB2NqX4c3QW1jh1xFTn+6OlJtmeGVRco3LjJqciXYPl+CPtBDMHtUF4/wy0O3xQgDAm+tuISPZFLOGd0Pg8w9jxooktHG+hyeezoPnpCzM9e6CGcO64fghW7wZeqvBr11O7CyLEDT2B8zbMxzPfjwJt7NtMHvoiQrlOrbJhmubdzDk0Ws6xzu0zsHdInP4fvKcdmvOSaipq7dEFBsbi+PHjyMiIgKRkZG4cOECjhw5Ul/NNZjevVNw5Yo9kpJaAgC++cYVzzxzA38fjK2sjL3lLwBEDBmSiAMHuiE/3xyiqMDHH/fBsWMdAQBKpQhr63sAAAuLUty7x8e8/qn3wDxc/j8LJF0v741883kbDB6fDZ3vfzVl/j3yLmL22ANQIf+uCX48aIfBz2ajpV0Zej+dh11rnQAAGclmeN2rC/JyTJCVZoL1gS7aXtfVs5ZQu9xryMuWnX91voWLd9S4lWUHANh3+lGM7JGAf960+M9T8cgsGIyjFx/WOd7zoRQIggLbXo7ElzP2wm/gaSgVzfyPMhEPhudqvEkddM3U29Ccg4MDAgMDYWZW3hXv3LkzkpKS6qu5BtOmTSHS0y21+xkZlrCyKoWlZZl2eK6yMiplESwty+DikocrV4qxYsWPaN26CPHxDvj008cBABs29MHq1ccwbtxl2NqWYPXqf0EQmIz+zqHdPWQkPRjeSU82hZWNAEtrQZsoqivj0LYU6UkPhlEzkk3R6ZFitO14D1lpphj/Wjr6Ds6FqZmIfZ844M61Vrhx2UJb3tRMwCuLkvHzN3b1f7Ey5mhbgJRca+1+Wq41rFvcg5V5qc7wXMihp7HN798AYnTqmygFnLzmgvVH+8FEJeBD30PILzHD7hM9G+oSGh2FCBiai+UyNFdviahLly7arxMTE/Hdd99h9+7d9dVcg1EqRVT23kONRlGjMiqVgCeeSEVw8NO4d0+JN988iZdeOoft23th4cJYrF3bD7/+2g7du2dg2bKfceWKPTIyrCp8VnOlVFR+/1WjqVkZhULUOacAIGgAlakI5w73UJivxBs+XdC2YwnWRCTgzjVzJJwv/6PC1r4MS7YmoiBPhR3vOtXthTUxCkXlf41rhJq9MzTit0cf7JQCYXG9MLHf+WadiJryZAWFKNZvpFevXsW0adMwe/ZsjBs3zqC6JSUliI+Pr6fIjNPK4mfYWZzE9ay3AACmqnR0Vy/E+eRtNSrj2mY5cor6IaNgBADAxvw3ONkcwK2cl/Gw/TpcSF2v/RzXNsHIyPdETnG/BrzCxq3W3//WK5FeMBx3i/sCAJxa7oNKWYj0/BF4zOl1nE3aDkEs7wF1tF+H/JLHkFEwDC1MbuDh1mtwt7gv7tz9L7g6VvVq8u/0d+3tNqG47CGk5Xtp6xeVtkdxWQcAgF2Lk2hjdRQJmYsb5gKM5ObmBnPzup3Ecv/34HtBPyI7q8iguq3sLbBg+aB6iasu1eusuTNnzmDOnDlYtGgRRo8ebfTnhCyIQU6mYf8A9cXWthibNsXj47d3ISmpJV566SySWqkRGnqw2jLOXn2waOpBjBljiaef/hbLFhfg3j0V5sw5hQulKnzxxW/47LO72PnBp/jjjzZwds7D2rXXsGLeVSQnp0h4xbqEs39I2r5t61JsPnYFH/1nOZKum+PlhclIUpvhg4DVNSoz5pUsPDlwG8wf7Y0VQ1ZjXdRVfLDABedPJOHjaAsc37MUX+9oA7s2pdhw+ApCXslCVlocNh65go/nOyPmyyQAjWPiTcZr/5I6hCq1sirCnhkXsXz/QdzKsoP/kJO4luyC5Qd/qlB2m99AxF5NwZ9p97Aztvz87KEn0LHNIczbMxymKgEf/fdbhP3SBRG/VazfGLS2Nsd7k9zrtxEjHmiVS4+o3hJRcnIyZs2ahdDQUPzrX433B8ZQd++2QGhoPyxe/AtMTAQkJ1tjzZp+6NIlC6+//iv8/UdUWubO3TcAfI9vvnGFtfU9rF8fA6VSREJCK2zb9gQKC02xYkV/TJ/+G0xNNRAEBT76qC+Sk1tKfcmNyt1MU3wQ8BCWbkmEiZmI5EQzvP96e3TpWYiAD25h5rBuVZYByicutO1wDyMGLcD6Q+k4tKs1zp8ov5ex/NWO8F91B15TMqBQAmGhTrhy1hJz3ruFFhYCxr6agbGvlr/TpbREide9ulQZZ3OXXWCB5QcHIeQ/R2Cq0uB2tg3ejhiMR9qmYemYn+D7yXPV1t/605OYP+o49sz8CiZKDY5e7IyI3x5pmOAbKw7NGe6dd97B/v370b59e+2xiRMnYtKkSTX+jPtd0sbUIzLWqm0+WDT1oP6CjZzUPaK6svpUIAL7rtZfsBFrzD0iQ2zzG4ipWxtnT6em7veI6nVobukx5Bg4NGdnb4EFKwY336G5JUuWYMmSJfX18UREzQsXPSUiIik19GsgGhITERGRHDThe0RMREREcsBEREREkuI9IiIiklJTfjEeHw8nIiJJsUdERCQH91ffNrSODDARERHJAScrEBGRtIxIRDLpEjERERHJAWfNERGRlDhrjoiIpGXwa8KNGcp7ICoqCqNGjcLw4cMRFhZW4fwff/yB8ePHw9PTE4sXL0ZZWRkAICIiAv3794ePjw98fHwQGhqqty32iIiISEdqaipCQ0Nx4MABmJmZYeLEiejXrx9cXV21ZebNm4d33nkHjz/+OBYtWoS9e/fC19cX8fHxCAwMhJeXV43bY4+IiEgOBNG4zQixsbFwd3eHnZ0dLC0t4enpiejoaO35O3fuoLi4GI8//jgAYPz48drz58+fR0REBLy9vfHWW2/h7t27ettjIiIikoNaDM2lpKTg9u3bOltubm6VTaWlpcHBwUG7r1arkZqaWuV5BwcH7XkHBwfMnDkTX3/9NZydnREcHKz30jg0R0QkC8ZP3/b19UVGRobOGX9/f8yePbvSWoIgQKFQPPgUUdTZr+78hg0btMenTp2KYcOG6Y2SiYiISA5qsbJCeHg4VCqVzikbG5sqqzk5OeH06dPa/fT0dKjVap3z6enp2v2MjAyo1Wrk5eVh//79eOmll8qbF8UK7VaGQ3NERHJQi3tETk5OcHFx0dmqS0QeHh6Ii4tDVlYWioqKEBMTgwEDBmjPt2vXDubm5jhz5gwA4ODBgxgwYAAsLS2xbds2nD17FgCwa9cu9oiIiJoMUSjfDK1jBEdHRwQEBGDKlCkoLS3FhAkT0LNnT/j5+WHOnDno0aMH1qxZgyVLliA/Px+PPfYYpkyZApVKhXXr1mHZsmUoLi5Gx44dERISorc9JiIiIqrA29sb3t7eOse2bt2q/bp79+7Yt29fhXp9+vRBRESEQW0xERERyQEXPSUiIkkZ81yQkc8RNTQmIiIiWeDq20REJCUOzRERkaT4hlYiIpKUIJRvhtaRAT7QSkREkmKPiIhIDniPiIiIJMVEREREkhKNeI6IiYiIiOqKKAoQDVw7ztDyUmEiIiKSAwFGrKxQL5HUOSYiIiI5aML3iDh9m4iIJMUeERGRHDThB1qZiIiI5KAJD80xERERyYAoChAN7OFw1hwREdUd9oiIiEhSnL5NRESSEoXyzdA6MsDp20REJCn2iIiIZEAURIgGDs0ZWl4qTERERLJgxNCcTG4SMREREcmBET0igyc3SKRRJyLxr6mHNq1aSBxJ3bBrbSF1CLUmOttKHUKdaSXzaxGtzaUOoc60lvm1tLIqj1+sx+nSNg7WBk/HtnGwrqdo6pZCrM/vXC3l5eXhypUrUodBRFQjXbt2RcuWLev0M8vKyhAfHw+NRmNUfZVKBTc3N5iYNN5+R6NORIIgoKCgAKamplAoFFKHQ0RUKVEUUVpaCisrKyiVdT8ZuaysrFaJqDEnIaCRJyIiImr6+BwRERFJiomIiIgkxURERESSYiIiIiJJMREREZGkmIiIiEhSTERERCQpJiIiIpIUExEREUmKiYiIiCTFRNQAoqKiMGrUKAwfPhxhYWFSh9Ps5efnw8vLC7dv35Y6lGbv448/xujRozF69GiEhIRIHQ5JhImonqWmpiI0NBTh4eGIjIzEnj17kJCQIHVYzdbZs2cxadIkJCYmSh1KsxcbG4vjx48jIiICkZGRuHDhAo4cOSJ1WCQBJqJ6FhsbC3d3d9jZ2cHS0hKenp6Ijo6WOqxma+/evQgKCoJarZY6lGbPwcEBgYGBMDMzg6mpKTp37oykpCSpwyIJNO61wZuAtLQ0ODg4aPfVajXOnTsnYUTN28qVK6UOgf7SpUsX7deJiYn47rvvsHv3bgkjIqmwR1TPBEHQeZeSKIp8txLR31y9ehWvvPIK5s+fj44dO0odDkmAiaieOTk5IT09Xbufnp7OYSGiv5w5cwYvvfQS3nzzTYwbN07qcEgiTET1zMPDA3FxccjKykJRURFiYmIwYMAAqcMiklxycjJmzZqFNWvWYPTo0VKHQxLiPaJ65ujoiICAAEyZMgWlpaWYMGECevbsKXVYRJL79NNPUVJSgtWrV2uPTZw4EZMmTZIwKpICXxVORESS4tAcERFJiomIiIgkxURERESSYiIiIiJJMREREZGkmIioXty+fRuPPPIIfHx8tNuYMWOwb9++Wn/2tGnTcODAAQCAj48PcnNzqyybl5eHKVOmGNxGdHQ0Jk+eXOH4yZMn4eXlpbd+t27dkJWVZVCbgYGB+PTTTw2qQ9QU8DkiqjctWrTAwYMHtfupqanw8vKCm5sbunfvXidt/P3zK3P37l2cP3++TtoiovrBREQNxtHRER06dEBiYiIuXryIffv2oaioCNbW1ti5cye++uor7N69G4IgwM7ODkuXLkXnzp2RmpqKwMBApKWloW3btsjMzNR+Zrdu3RAXFwd7e3ts3rwZERERMDExQYcOHbB69WosXLgQxcXF8PHxwYEDB5CYmIiVK1ciJycHGo0GkydPxoQJEwAAH374IaKiomBnZ4cOHTrovZ7r168jODgYBQUFSE9PR/fu3bFu3TqYm5sDANatW4fz589DEATMnTsXzzzzDABUeZ1EzRUTETWY33//HTdv3kSvXr0QFxeHhIQEHDt2DNbW1vj1118RGRmJsLAwWFhY4Pjx4/D398d3332H4OBg9OrVC3PnzsWNGzcwduzYCp/9/fff48CBA9i7dy9sbW3x7rvvYteuXXj33Xfh7e2NgwcPoqysDHPmzEFISAgee+wx5OXl4fnnn4erqysyMjIQExODyMhItGjRArNmzdJ7PXv37sXYsWPh4+OD0tJSjB8/Hj/++CM8PT0BAC4uLggODsaVK1cwefJkfPfdd0hISKjyOomaKyYiqjf3eyIAoNFo0KpVK7z//vtwdnYGUN6bsba2BgD8+OOPuHHjBiZOnKitn5ubi5ycHMTGxmLBggUAgA4dOqBfv34V2oqLi8OIESNga2sLAFi4cCEA6LyFNTExETdv3sSiRYt0Yrx48SL+/PNPDBs2TBvPs88+i507d1Z7ffPmzcMvv/yCrVu3IjExEWlpaSgsLNSev79UTdeuXdG5c2f8/vvvOHPmTJXXSdRcMRFRvfnnPaJ/srS01H4tCAJ8fHwwb9487X5aWhpsbW2hUCjw95WoTEwq/m+rUql0Xq+Rm5tbYRKDRqNBy5YtdWLKyMhAy5YtERISotOGSqXSe31vvPEGNBoNRo4ciUGDBiE5OVnnM5TKB3OBBEGAiYlJtddJ1Fxx1hw1Cv3798e3336LtLQ0AMDu3bvx4osvAgCefvpp7NmzBwCQlJSEkydPVqjv4eGBI0eOID8/HwCwfv16fPbZZzAxMYFGo4EoiujUqZNOckxOToaXlxfi4+MxYMAAREdHIzc3F4Ig6J0EAQDHjx/HrFmzMGrUKADlryHXaDTa8xEREQCACxcuaIckq7tOouaKPSJqFPr37w8/Pz+88sorUCgUsLa2xscffwyFQoGgoCAsXLgQI0eOhJOTU6Uz7gYOHIiEhATtcJirqytWrFgBCwsL9OzZE6NHj0ZYWBg2btyIlStXYtu2bSgrK8Prr7+OJ598EgBw+fJlPPvss7CxsUH37t2RnZ1dbcwBAQGYNWsWLC0tYW1tjb59++LmzZva87du3cLYsWOhUCiwdu1a2NnZVXudRM0VV98mIiJJcWiOiIgkxURERESSYiIiIiJJMREREZGkmIiIiEhSTERERCQpJiIiIpIUExEREUnq/wEnZZETHIGyGwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(xgb_opt1, X_test, y_test, normalize='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metric Evaluation using `sklearn.metrics.classification_report`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.64      0.77      0.70      5790\n",
      "           1       0.97      0.87      0.92     10995\n",
      "           2       0.69      0.67      0.68      4901\n",
      "\n",
      "    accuracy                           0.80     21686\n",
      "   macro avg       0.77      0.77      0.77     21686\n",
      "weighted avg       0.82      0.80      0.81     21686\n",
      "\n"
     ]
    }
   ],
   "source": [
    "y_pred = xgb_opt1.predict(X_test)\n",
    "\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metric Evaluation using `StratifiedKFold`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[08:49:20] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[08:52:58] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[08:56:46] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:00:15] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:03:44] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    }
   ],
   "source": [
    "metrics = models.kfold_xgboost(X_train, y_train, classifier=xgb_opt1, \n",
    "                                         continuous_cols=numerical_cols, categorical_cols=categorical_cols, smote=False,\n",
    "                                         minority_size=0.7, majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.63      0.76      0.69      3474\n",
      "           1       0.97      0.88      0.92      6597\n",
      "           2       0.69      0.67      0.68      2941\n",
      "\n",
      "    accuracy                           0.80     13012\n",
      "   macro avg       0.76      0.77      0.76     13012\n",
      "weighted avg       0.82      0.80      0.80     13012\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.63      0.77      0.69      3474\n",
      "           1       0.97      0.88      0.92      6597\n",
      "           2       0.69      0.67      0.68      2941\n",
      "\n",
      "    accuracy                           0.80     13012\n",
      "   macro avg       0.76      0.77      0.76     13012\n",
      "weighted avg       0.82      0.80      0.80     13012\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.63      0.76      0.69      3474\n",
      "           1       0.97      0.87      0.92      6597\n",
      "           2       0.68      0.66      0.67      2941\n",
      "\n",
      "    accuracy                           0.79     13012\n",
      "   macro avg       0.76      0.76      0.76     13012\n",
      "weighted avg       0.81      0.79      0.80     13012\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.64      0.76      0.69      3474\n",
      "           1       0.97      0.87      0.92      6597\n",
      "           2       0.69      0.69      0.69      2940\n",
      "\n",
      "    accuracy                           0.80     13011\n",
      "   macro avg       0.77      0.77      0.77     13011\n",
      "weighted avg       0.82      0.80      0.81     13011\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.63      0.78      0.70      3474\n",
      "           1       0.97      0.87      0.92      6596\n",
      "           2       0.69      0.66      0.68      2941\n",
      "\n",
      "    accuracy                           0.80     13011\n",
      "   macro avg       0.76      0.77      0.76     13011\n",
      "weighted avg       0.82      0.80      0.80     13011\n",
      " \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for report in metrics:\n",
    "    print(report, '\\n\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Trial 1 Takeaways:\n",
    "\n",
    "* Best **RECALL** so far, with an overall average of **80%**!\n",
    "\n",
    "* Our **confusion matrix** looks *much better*, with a clearly defined diagonal line (lighter colors) of correctly predicted values.\n",
    "\n",
    "********\n",
    "\n",
    "## Bayesian Optimization - TRIAL 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = {'colsample_bytree': 0.9404827047850027, \n",
    "          'gamma': 0.29360102121743203, \n",
    "          'learning_rate': 0.2368194786682022, \n",
    "          'max_depth': 12.503988212911377, \n",
    "          'min_child_weight': 4.061256592155039, \n",
    "          'n_estimators': 155.7883908687575, \n",
    "          'scale_pos_weight': 5.564329182083066}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "# If these params are not type cast into int, it raises an error during model fitting\n",
    "params['max_depth']= int(params['max_depth'])\n",
    "params['n_estimators']= int(params['n_estimators'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[09:20:35] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:541: \n",
      "Parameters: { scale_pos_weight } might not be used.\n",
      "\n",
      "  This may not be accurate due to some parameters are only used in language bindings but\n",
      "  passed down to XGBoost core.  Or some parameters are not used but slip through this\n",
      "  verification. Please open an issue if you find above cases.\n",
      "\n",
      "\n",
      "[09:20:41] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    }
   ],
   "source": [
    "xgb_opt2 = XGBClassifier(**params).fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Confusion Matrix: TRAINING Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_p2 = xgb_opt2.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_p2 = np.array(train_p2.astype(int))\n",
    "y_train  = np.array(y_train.astype(int))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "cf = confusion_matrix(train_p2, y_train)\n",
    "sns.heatmap(cf/np.sum(cf), annot=True, fmt='.2%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Confusion Matrix: TEST Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_p2 = xgb_opt2.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_p2 = np.array(test_p2.astype(int))\n",
    "y_test  = np.array(y_test.astype(int))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cf = confusion_matrix(test_p2, y_test)\n",
    "sns.heatmap(cf/np.sum(cf), annot=True, fmt='.2%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metric Evaluation using `sklearn.metrics.classification_report`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.76      0.84      0.80      5790\n",
      "           1       0.97      0.92      0.95     10995\n",
      "           2       0.80      0.80      0.80      4901\n",
      "\n",
      "    accuracy                           0.87     21686\n",
      "   macro avg       0.85      0.85      0.85     21686\n",
      "weighted avg       0.88      0.87      0.87     21686\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, test_p2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metric Evaluation using `StratifiedKFold`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[09:38:16] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:541: \n",
      "Parameters: { scale_pos_weight } might not be used.\n",
      "\n",
      "  This may not be accurate due to some parameters are only used in language bindings but\n",
      "  passed down to XGBoost core.  Or some parameters are not used but slip through this\n",
      "  verification. Please open an issue if you find above cases.\n",
      "\n",
      "\n",
      "[09:38:21] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:48:23] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:541: \n",
      "Parameters: { scale_pos_weight } might not be used.\n",
      "\n",
      "  This may not be accurate due to some parameters are only used in language bindings but\n",
      "  passed down to XGBoost core.  Or some parameters are not used but slip through this\n",
      "  verification. Please open an issue if you find above cases.\n",
      "\n",
      "\n",
      "[09:48:27] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[10:38:52] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:541: \n",
      "Parameters: { scale_pos_weight } might not be used.\n",
      "\n",
      "  This may not be accurate due to some parameters are only used in language bindings but\n",
      "  passed down to XGBoost core.  Or some parameters are not used but slip through this\n",
      "  verification. Please open an issue if you find above cases.\n",
      "\n",
      "\n",
      "[10:38:58] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[10:58:40] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:541: \n",
      "Parameters: { scale_pos_weight } might not be used.\n",
      "\n",
      "  This may not be accurate due to some parameters are only used in language bindings but\n",
      "  passed down to XGBoost core.  Or some parameters are not used but slip through this\n",
      "  verification. Please open an issue if you find above cases.\n",
      "\n",
      "\n",
      "[10:58:44] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[11:08:20] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:541: \n",
      "Parameters: { scale_pos_weight } might not be used.\n",
      "\n",
      "  This may not be accurate due to some parameters are only used in language bindings but\n",
      "  passed down to XGBoost core.  Or some parameters are not used but slip through this\n",
      "  verification. Please open an issue if you find above cases.\n",
      "\n",
      "\n",
      "[11:08:25] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    }
   ],
   "source": [
    "metrics = models.kfold_xgboost(X_train, pd.DataFrame(y_train), classifier=xgb_opt2, \n",
    "                     continuous_cols=numerical_cols, categorical_cols=categorical_cols, smote=False,\n",
    "                     minority_size=0.7, majority_reduce=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.73      0.82      0.77      3474\n",
      "           1       0.97      0.91      0.94      6597\n",
      "           2       0.77      0.76      0.76      2941\n",
      "\n",
      "    accuracy                           0.85     13012\n",
      "   macro avg       0.82      0.83      0.83     13012\n",
      "weighted avg       0.86      0.85      0.86     13012\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.75      0.83      0.79      3474\n",
      "           1       0.97      0.92      0.95      6597\n",
      "           2       0.79      0.78      0.78      2941\n",
      "\n",
      "    accuracy                           0.86     13012\n",
      "   macro avg       0.84      0.84      0.84     13012\n",
      "weighted avg       0.87      0.86      0.87     13012\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.74      0.83      0.78      3474\n",
      "           1       0.97      0.91      0.94      6597\n",
      "           2       0.78      0.78      0.78      2941\n",
      "\n",
      "    accuracy                           0.86     13012\n",
      "   macro avg       0.83      0.84      0.84     13012\n",
      "weighted avg       0.87      0.86      0.86     13012\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.75      0.82      0.78      3474\n",
      "           1       0.97      0.91      0.94      6597\n",
      "           2       0.77      0.79      0.78      2940\n",
      "\n",
      "    accuracy                           0.86     13011\n",
      "   macro avg       0.83      0.84      0.83     13011\n",
      "weighted avg       0.87      0.86      0.86     13011\n",
      " \n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.74      0.83      0.78      3474\n",
      "           1       0.97      0.91      0.94      6596\n",
      "           2       0.78      0.77      0.78      2941\n",
      "\n",
      "    accuracy                           0.86     13011\n",
      "   macro avg       0.83      0.84      0.83     13011\n",
      "weighted avg       0.87      0.86      0.86     13011\n",
      " \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for report in metrics:\n",
    "    print(report, '\\n\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Ternary Classification Model Takeaways:\n",
    "* Our **confusion matrix** looks *even better* than Trial 1, with a clearly defined diagonal line of correctly predicted values.\n",
    "\n",
    "* Recall is at it's **HIGHEST**, averaging at about **85%**, with high precision AND accuracy values to boot!\n",
    "\n",
    "### Let's check out its **features!**\n",
    "\n",
    "*Scroll all the way down to see features with the highest **gain** metric*."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Features Importances TRIAL 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[0.30892483141249993, 'x0_DRIVER'],\n",
       " [0.37696212748, 'x0_DRIVERLESS'],\n",
       " [0.412579572385, 'x13_CURVE ON HILLCREST'],\n",
       " [0.441790849, 'x23_OH'],\n",
       " [0.510764718, 'x23_NM'],\n",
       " [0.5337560175, 'x23_MN'],\n",
       " [0.5703626875000001, 'x23_MS'],\n",
       " [0.6053069589594445, 'x8_UNKNOWN'],\n",
       " [0.6078768252000001, 'x23_CT'],\n",
       " [0.6325415375, 'x23_OR'],\n",
       " [0.63833692675, 'x23_DC'],\n",
       " [0.6433590055, 'x23_NB'],\n",
       " [0.643772602, 'x23_RI'],\n",
       " [0.65645053, 'x23_MO'],\n",
       " [0.6657915415, 'x23_NH'],\n",
       " [0.6822504156666667, 'x23_CA'],\n",
       " [0.7042528690000001, 'x23_NV'],\n",
       " [0.7158821595, 'x20_VISION OBSCURED (SIGNS, TREE LIMBS, BUILDINGS, ETC.)'],\n",
       " [0.725632568, 'x23_KS'],\n",
       " [0.7413493095, 'x23_ME'],\n",
       " [0.758955479, 'x23_SC'],\n",
       " [0.7762215933333333, 'x23_IA'],\n",
       " [0.84448278, 'x23_OK'],\n",
       " [0.874778926, 'x23_IL'],\n",
       " [0.891543648181818, 'x20_DISREGARDING OTHER TRAFFIC SIGNS'],\n",
       " [0.9028219122500001, 'x21_E'],\n",
       " [0.9288267896666668, 'x23_IN'],\n",
       " [0.9458862196363633, 'x20_DISTRACTION - FROM INSIDE VEHICLE'],\n",
       " [0.9619256544999999, 'x20_FOLLOWING TOO CLOSELY'],\n",
       " [0.985537648, 'x23_DE'],\n",
       " [1.0034742301666666, 'x20_MOTORCYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [1.00415233575, 'x10_UNKNOWN'],\n",
       " [1.0250859984, 'x23_AK'],\n",
       " [1.0250922205416668, 'x12_DIVIDED - W/MEDIAN BARRIER'],\n",
       " [1.0694840167272728, 'x12_PARKING LOT'],\n",
       " [1.0762299297777775, 'x20_DISTRACTION - FROM OUTSIDE VEHICLE'],\n",
       " [1.094390991888889, 'x20_DRIVING SKILLS/KNOWLEDGE/EXPERIENCE'],\n",
       " [1.1658213725, 'x23_NJ'],\n",
       " [1.19339859, 'x23_MD'],\n",
       " [1.20315599, 'x23_AL'],\n",
       " [1.2654472785000002, 'x20_EXCEEDING AUTHORIZED SPEED LIMIT'],\n",
       " [1.2955818183333334, 'x20_ROAD ENGINEERING/SURFACE/MARKING DEFECTS'],\n",
       " [1.3031183866666665, 'x20_TEXTING'],\n",
       " [1.3752822871428572, 'x19_NOT APPLICABLE'],\n",
       " [1.3809783593999998, 'x20_DISREGARDING TRAFFIC SIGNALS'],\n",
       " [1.4018699065, 'x21_N'],\n",
       " [1.420889447333333, 'x20_FAILING TO YIELD RIGHT-OF-WAY'],\n",
       " [1.43657875, 'x23_KY'],\n",
       " [1.4447223530000002, 'x19_TEXTING'],\n",
       " [1.459107122, 'x20_DISREGARDING YIELD SIGN'],\n",
       " [1.4648101355217393, 'x5_PARKED IN TRAFFIC LANE'],\n",
       " [1.4757969486, 'x23_FL'],\n",
       " [1.4804444733333335, 'x20_ROAD CONSTRUCTION/MAINTENANCE'],\n",
       " [1.4849299510000002, 'x20_EVASIVE ACTION DUE TO ANIMAL, OBJECT, NONMOTORIST'],\n",
       " [1.5133712305, 'x23_MA'],\n",
       " [1.5135983235000001, 'x19_DISREGARDING ROAD MARKINGS'],\n",
       " [1.5554514896999998, 'x15_WORN SURFACE'],\n",
       " [1.5652076579999998, 'x13_CURVE, LEVEL'],\n",
       " [1.5905473728333333, 'x23_LA'],\n",
       " [1.594432406344, 'x14_SAND, MUD, DIRT'],\n",
       " [1.6155598150000001, 'x23_MT'],\n",
       " [1.6289112166666666, 'x20_IMPROPER BACKING'],\n",
       " [1.685266986, 'x19_HAD BEEN DRINKING (USE WHEN ARREST IS NOT MADE)'],\n",
       " [1.6873345383333334, 'x23_ID'],\n",
       " [1.6926985769999998, 'x20_EXCEEDING SAFE SPEED FOR CONDITIONS'],\n",
       " [1.6932263739, 'x19_IMPROPER TURNING/NO SIGNAL'],\n",
       " [1.6951764800000002, 'x23_MI'],\n",
       " [1.7164374534166662, 'x12_T-INTERSECTION'],\n",
       " [1.744301466833333, 'x20_IMPROPER LANE USAGE'],\n",
       " [1.7495716437036366, 'x7_PEDESTRIAN CROSSING SIGN'],\n",
       " [1.7668686333333332, 'x23_AR'],\n",
       " [1.7788280463750004, 'x20_IMPROPER TURNING/NO SIGNAL'],\n",
       " [1.8098926542263156, 'x12_NOT DIVIDED'],\n",
       " [1.8202064291428572, 'x19_FOLLOWING TOO CLOSELY'],\n",
       " [1.8502032473853995, 'x7_TRAFFIC SIGNAL'],\n",
       " [1.853817968625, 'x19_DISREGARDING YIELD SIGN'],\n",
       " [1.862005113636364, 'x20_ANIMAL'],\n",
       " [1.8666587154666667, 'x19_ANIMAL'],\n",
       " [1.881114976142857, 'x22_DRIVER'],\n",
       " [1.8913902218799998, 'x7_OTHER RAILROAD CROSSING'],\n",
       " [1.8924575324127273, 'x0_NON-CONTACT VEHICLE'],\n",
       " [1.8928105955999999, 'x20_TURNING RIGHT ON RED'],\n",
       " [1.8939290735000003, 'x14_ICE'],\n",
       " [1.894229889, 'x20_WEATHER'],\n",
       " [1.8945633014285714, 'x19_MOTORCYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [1.905134182,\n",
       "  'x19_OPERATING VEHICLE IN ERRATIC, RECKLESS, CARELESS, NEGLIGENT OR AGGRESSIVE MANNER'],\n",
       " [1.922159285333333, 'x19_DRIVING SKILLS/KNOWLEDGE/EXPERIENCE'],\n",
       " [1.9401953588125003, 'x11_REAR END'],\n",
       " [1.965979406285714, 'x20_DISREGARDING ROAD MARKINGS'],\n",
       " [1.9781861983333335, 'x20_PASSING STOPPED SCHOOL BUS'],\n",
       " [1.98731240023, 'x5_DRIVERLESS'],\n",
       " [2.012021448111111, 'x20_OBSTRUCTED CROSSWALKS'],\n",
       " [2.0191918513563016, 'x3_STATE OWNED'],\n",
       " [2.0249505644, 'x19_DRIVING ON WRONG SIDE/WRONG WAY'],\n",
       " [2.0428901007857143, 'x11_FIXED OBJECT'],\n",
       " [2.0501829132189653, 'x9_UNKNOWN'],\n",
       " [2.05707654375,\n",
       "  'x20_DISTRACTION - OTHER ELECTRONIC DEVICE (NAVIGATION DEVICE, DVD PLAYER, ETC.)'],\n",
       " [2.05779409, 'x23_NC'],\n",
       " [2.061235863333333, 'x19_IMPROPER OVERTAKING/PASSING'],\n",
       " [2.0894002051428573, 'x2_UNKNOWN/NA'],\n",
       " [2.100752235, 'x23_GA'],\n",
       " [2.117375102052778, 'x5_OTHER'],\n",
       " [2.120000860909091, 'x20_IMPROPER OVERTAKING/PASSING'],\n",
       " [2.131699385, 'x20_RELATED TO BUS STOP'],\n",
       " [2.134489696333333, 'x23_AZ'],\n",
       " [2.13657093, 'x18_$501 - $1,500'],\n",
       " [2.145473876666667, 'x21_S'],\n",
       " [2.169967197549019, 'x5_STRAIGHT AHEAD'],\n",
       " [2.1707698572000003, 'x19_ROAD ENGINEERING/SURFACE/MARKING DEFECTS'],\n",
       " [2.17612028525, 'x15_OTHER'],\n",
       " [2.17888782, 'x6_ROOF'],\n",
       " [2.1895005032, 'x19_IMPROPER BACKING'],\n",
       " [2.1913842784615385, 'x11_PARKED MOTOR VEHICLE'],\n",
       " [2.20298812, 'x7_NO CONTROLS'],\n",
       " [2.2047593375, 'x19_RELATED TO BUS STOP'],\n",
       " [2.2060191497222226, 'x16_AMENDED'],\n",
       " [2.216958505714286, 'x19_OBSTRUCTED CROSSWALKS'],\n",
       " [2.21890349,\n",
       "  'x20_OPERATING VEHICLE IN ERRATIC, RECKLESS, CARELESS, NEGLIGENT OR AGGRESSIVE MANNER'],\n",
       " [2.2217345004, 'x19_FAILING TO YIELD RIGHT-OF-WAY'],\n",
       " [2.22532558625, 'x1_TIRES'],\n",
       " [2.24984952824793, 'x5_SLOW/STOP - LEFT TURN'],\n",
       " [2.271206381766666, 'x5_ENTER FROM DRIVE/ALLEY'],\n",
       " [2.278451629285714, 'x19_BICYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [2.2801747963063836, 'x5_U-TURN'],\n",
       " [2.289776094318181, 'x3_PERSONAL'],\n",
       " [2.293841189375, 'x19_PHYSICAL CONDITION OF DRIVER'],\n",
       " [2.3182908887924527, 'x3_RIDESHARE SERVICE'],\n",
       " [2.319060828, 'x3_CAMPER/RV - SINGLE UNIT'],\n",
       " [2.3204845183986107, 'x5_TURNING RIGHT'],\n",
       " [2.3469971800000002, 'x19_PASSING STOPPED SCHOOL BUS'],\n",
       " [2.351235901, 'x20_DISREGARDING STOP SIGN'],\n",
       " [2.3619904519999997, 'x14_OTHER'],\n",
       " [2.3651893533333332,\n",
       "  'x20_UNDER THE INFLUENCE OF ALCOHOL/DRUGS (USE WHEN ARREST IS EFFECTED)'],\n",
       " [2.37700468475, 'x10_DAYLIGHT'],\n",
       " [2.378432095944444, 'x5_UNKNOWN/NA'],\n",
       " [2.4007059146153846, 'x5_PARKED'],\n",
       " [2.4088459532499997, 'x15_RUT, HOLES'],\n",
       " [2.414978662333334, 'x5_CHANGING LANES'],\n",
       " [2.431232042222222, 'x11_OVERTURNED'],\n",
       " [2.4409199266000003, 'x19_DISREGARDING STOP SIGN'],\n",
       " [2.469490975, 'x23_HI'],\n",
       " [2.4868909462857154, 'x12_UNKNOWN'],\n",
       " [2.5029616102142853, 'x5_NEGOTIATING A CURVE'],\n",
       " [2.503413298306482, 'x3_SCHOOL BUS'],\n",
       " [2.51192713, 'x23_NY'],\n",
       " [2.51490283, 'x23_PA'],\n",
       " [2.532485245, 'x20_HAD BEEN DRINKING (USE WHEN ARREST IS NOT MADE)'],\n",
       " [2.5415941745151516, 'x4_NW'],\n",
       " [2.542241184360656, 'x2_MOTORCYCLE (OVER 150CC)'],\n",
       " [2.54747252125, 'x18_$500 OR LESS'],\n",
       " [2.6357341215555556, 'x12_ONE-WAY'],\n",
       " [2.640974289972222, 'x7_SCHOOL ZONE'],\n",
       " [2.6599807215185174, 'x8_FUNCTIONING PROPERLY'],\n",
       " [2.671976458833333, 'x7_OTHER'],\n",
       " [2.6794869057142856,\n",
       "  'x19_VISION OBSCURED (SIGNS, TREE LIMBS, BUILDINGS, ETC.)'],\n",
       " [2.689943055165854, 'x10_DUSK'],\n",
       " [2.7010104165442854, 'x12_NOT REPORTED'],\n",
       " [2.7116580079545454, 'x5_TURNING ON RED'],\n",
       " [2.721825143161403, 'x5_MERGING'],\n",
       " [2.74235036492, 'x5_SLOW/STOP - RIGHT TURN'],\n",
       " [2.746277683930232, 'x8_WORN REFLECTIVE MATERIAL'],\n",
       " [2.7632839075, 'x10_DARKNESS'],\n",
       " [2.7658967814285718, 'x16_NOT ON SCENE (DESK REPORT)'],\n",
       " [2.7746481484, 'x19_DISREGARDING TRAFFIC SIGNALS'],\n",
       " [2.781916095256918, 'x3_COMMERCIAL - SINGLE UNIT'],\n",
       " [2.8010420340000004, 'x20_EQUIPMENT - VEHICLE CONDITION'],\n",
       " [2.826096963989247, 'x5_ENTERING TRAFFIC LANE FROM PARKING'],\n",
       " [2.8263781455, 'x12_DIVIDED - W/MEDIAN (NOT RAISED)'],\n",
       " [2.838871007751135, 'x2_BUS UP TO 15 PASS.'],\n",
       " [2.849297265, 'x22_NON-CONTACT VEHICLE'],\n",
       " [2.8538361400000003, 'x20_UNABLE TO DETERMINE'],\n",
       " [2.8593124835592594, 'x11_ANIMAL'],\n",
       " [2.87437103, 'x20_BICYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [2.9349862516666665, 'x20_FAILING TO REDUCE SPEED TO AVOID CRASH'],\n",
       " [2.945406107326786, 'x5_LEAVING TRAFFIC LANE TO PARK'],\n",
       " [2.9632375872087913, 'x1_OTHER'],\n",
       " [2.965245644851852, 'x5_DRIVING WRONG WAY'],\n",
       " [2.9706896346666665, 'x20_DRIVING ON WRONG SIDE/WRONG WAY'],\n",
       " [2.9714301020817313, 'x5_SLOW/STOP - LOAD/UNLOAD'],\n",
       " [2.9762300349807704, 'x7_YIELD'],\n",
       " [2.9828298699999998, 'x7_OTHER WARNING SIGN'],\n",
       " [2.9839294201208344, 'x11_OTHER NONCOLLISION'],\n",
       " [3.000481254691071, 'x11_HEAD ON'],\n",
       " [3.0051284760000003, 'x20_PHYSICAL CONDITION OF DRIVER'],\n",
       " [3.0174455120625003, 'x12_RAMP'],\n",
       " [3.035811207727271, 'x11_TRAIN'],\n",
       " [3.0421033665714283, 'x19_EXCEEDING SAFE SPEED FOR CONDITIONS'],\n",
       " [3.0539455357241376, 'x6_OTHER'],\n",
       " [3.057873301727143, 'x3_DRIVER EDUCATION'],\n",
       " [3.077744794607143, 'x5_PASSING/OVERTAKING'],\n",
       " [3.0795790747777776, 'x19_WEATHER'],\n",
       " [3.128294952918368, 'x13_STRAIGHT AND LEVEL'],\n",
       " [3.146677337, 'x11_SIDESWIPE OPPOSITE DIRECTION'],\n",
       " [3.16326375096, 'x3_COMMERCIAL - MULTI-UNIT'],\n",
       " [3.18767528728, 'x7_DELINEATORS'],\n",
       " [3.2233505209166675, 'x12_FIVE POINT, OR MORE'],\n",
       " [3.2410020057936504, 'x6_UNDER CARRIAGE'],\n",
       " [3.2436176423999994,\n",
       "  'x19_DISTRACTION - OTHER ELECTRONIC DEVICE (NAVIGATION DEVICE, DVD PLAYER, ETC.)'],\n",
       " [3.272464139583334, 'x19_DISTRACTION - FROM OUTSIDE VEHICLE'],\n",
       " [3.3004398233750005, 'x9_OTHER'],\n",
       " [3.3138742588874996, 'x9_RAIN'],\n",
       " [3.3332808561999996, 'x11_ANGLE'],\n",
       " [3.3371913235555564, 'x17_INJURY AND / OR TOW DUE TO CRASH'],\n",
       " [3.33757646375, 'x19_ROAD CONSTRUCTION/MAINTENANCE'],\n",
       " [3.3701589536363628, 'x13_STRAIGHT ON GRADE'],\n",
       " [3.3816553330845056, 'x4_S'],\n",
       " [3.38361536, 'x19_EXCEEDING AUTHORIZED SPEED LIMIT'],\n",
       " [3.4105208778571425, 'x5_DIVERGING'],\n",
       " [3.4272010117000002, 'x19_EVASIVE ACTION DUE TO ANIMAL, OBJECT, NONMOTORIST'],\n",
       " [3.4346473215352753, 'x7_POLICE/FLAGMAN'],\n",
       " [3.438598256, 'x19_IMPROPER LANE USAGE'],\n",
       " [3.4407183812105266, 'x15_SHOULDER DEFECT'],\n",
       " [3.47277264, 'x12_ROUNDABOUT'],\n",
       " [3.4759873848888887, 'x11_REAR TO FRONT'],\n",
       " [3.5143358738095247, 'x6_SIDE-RIGHT'],\n",
       " [3.516642566129032, 'x13_STRAIGHT ON HILLCREST'],\n",
       " [3.5205764678, 'x12_TRAFFIC ROUTE'],\n",
       " [3.570411948461538, 'x8_OTHER'],\n",
       " [3.5770612071071435, 'x7_RR CROSSING SIGN'],\n",
       " [3.5841340174444443, 'x14_WET'],\n",
       " [3.591736343454546, 'x7_FLASHING CONTROL SIGNAL'],\n",
       " [3.593304028125, 'x6_REAR-RIGHT'],\n",
       " [3.6004078711923073, 'x12_DRIVEWAY'],\n",
       " [3.616481571593068, 'x3_CONSTRUCTION/MAINTENANCE'],\n",
       " [3.628144359934483, 'x6_NONE'],\n",
       " [3.65747456, 'x19_CELL PHONE USE OTHER THAN TEXTING'],\n",
       " [3.658953604400001, 'x12_L-INTERSECTION'],\n",
       " [3.670341535714286, 'x11_TURNING'],\n",
       " [3.678551870137255, 'x5_SKIDDING/CONTROL LOSS'],\n",
       " [3.6929269328571435, 'x14_UNKNOWN'],\n",
       " [3.7013959866666664, 'x19_DISREGARDING OTHER TRAFFIC SIGNS'],\n",
       " [3.7271704508749997, 'x17_NO INJURY / DRIVE AWAY'],\n",
       " [3.7413444392666673, 'x12_ALLEY'],\n",
       " [3.752306412642857, 'x11_PEDALCYCLIST'],\n",
       " [3.7591316065652176, 'x8_MISSING'],\n",
       " [3.7749622925000006, 'x20_CELL PHONE USE OTHER THAN TEXTING'],\n",
       " [3.8740653380000003, 'x11_REAR TO SIDE'],\n",
       " [3.879517615999999, 'x6_REAR'],\n",
       " [3.886677642771086, 'x5_BACKING'],\n",
       " [3.8897159217812507, 'x10_DAWN'],\n",
       " [3.9013050745522952, 'x1_UNKNOWN'],\n",
       " [3.91014957, 'x19_FAILING TO REDUCE SPEED TO AVOID CRASH'],\n",
       " [3.911942962342104, 'x1_SUSPENSION'],\n",
       " [3.9264811381818188, 'x4_UNKNOWN'],\n",
       " [3.927368686666666, 'x11_REAR TO REAR'],\n",
       " [3.9466132349999996, 'x19_DISTRACTION - FROM INSIDE VEHICLE'],\n",
       " [3.9728669570910133, 'x2_3-WHEELED MOTORCYCLE (2 REAR WHEELS)'],\n",
       " [4.000578531428572, 'x12_FOUR WAY'],\n",
       " [4.00119810029375, 'x6_FRONT-LEFT'],\n",
       " [4.0210633901914905, 'x7_NO PASSING'],\n",
       " [4.0349925190307685, 'x3_NOT IN USE'],\n",
       " [4.042806472571429, 'x4_SE'],\n",
       " [4.0452363175, 'x8_NOT FUNCTIONING'],\n",
       " [4.058905576666666, 'x4_E'],\n",
       " [4.06350994,\n",
       "  'x19_UNDER THE INFLUENCE OF ALCOHOL/DRUGS (USE WHEN ARREST IS EFFECTED)'],\n",
       " [4.071469997181818, 'x3_UNKNOWN/NA'],\n",
       " [4.086161061317073, 'x4_NE'],\n",
       " [4.086374974, 'x14_SNOW OR SLUSH'],\n",
       " [4.109297772, 'x3_AGRICULTURE'],\n",
       " [4.1453288564355555, 'x2_ALL-TERRAIN VEHICLE (ATV)'],\n",
       " [4.191035285890586, 'x2_TRACTOR W/O SEMI-TRAILER'],\n",
       " [4.251449019374999, 'x7_UNKNOWN'],\n",
       " [4.259354922222222, 'x3_POLICE'],\n",
       " [4.265053935625001, 'x6_TOTAL (ALL AREAS)'],\n",
       " [4.305084049000001, 'x19_EQUIPMENT - VEHICLE CONDITION'],\n",
       " [4.310978030999999, 'x6_UNKNOWN'],\n",
       " [4.330003445454545, 'x18_OVER $1,500'],\n",
       " [4.369034831538461, 'x15_UNKNOWN'],\n",
       " [4.376094540833333, 'x5_TURNING LEFT'],\n",
       " [4.403042420526316, 'x11_OTHER OBJECT'],\n",
       " [4.410187573333334, 'x7_STOP SIGN/FLASHER'],\n",
       " [4.412300782674419, 'x2_BUS OVER 15 PASS.'],\n",
       " [4.41695846, 'x6_FRONT'],\n",
       " [4.4219322812686555, 'x1_WHEELS'],\n",
       " [4.437753458347826, 'x8_FUNCTIONING IMPROPERLY'],\n",
       " [4.447279959190769, 'x8_NO CONTROLS'],\n",
       " [4.460784528153846, 'x12_UNKNOWN INTERSECTION TYPE'],\n",
       " [4.47362882, 'x19_UNABLE TO DETERMINE'],\n",
       " [4.4778579781111105, 'x3_OTHER TRANSIT'],\n",
       " [4.496009171666667, 'x6_FRONT-RIGHT'],\n",
       " [4.6052710134999995, 'x15_NO DEFECTS'],\n",
       " [4.623241698682666, 'x2_TRUCK - SINGLE UNIT'],\n",
       " [4.643680143999999, 'x12_Y-INTERSECTION'],\n",
       " [4.645488877320327, 'x3_TOW TRUCK'],\n",
       " [4.647890677222223, 'x9_SNOW'],\n",
       " [4.658395864, 'x9_CLEAR'],\n",
       " [4.7346955699999995, 'x19_TURNING RIGHT ON RED'],\n",
       " [4.746624615263159, 'x2_FARM EQUIPMENT'],\n",
       " [4.7531806058620685, 'x1_WINDOWS'],\n",
       " [4.7726055029375, 'x10_DARKNESS, LIGHTED ROAD'],\n",
       " [4.807781627142857, 'x11_SIDESWIPE SAME DIRECTION'],\n",
       " [4.864821369076925, 'x3_AMBULANCE'],\n",
       " [4.889823654545454, 'x2_MOTOR DRIVEN CYCLE'],\n",
       " [4.91053724, 'x20_NOT APPLICABLE'],\n",
       " [5.05116558, 'x21_W'],\n",
       " [5.077941689999999, 'x7_BICYCLE CROSSING SIGN'],\n",
       " [5.099598640661841, 'x3_OTHER'],\n",
       " [5.102152017290323, 'x3_FIRE'],\n",
       " [5.16720752888, 'x12_CENTER TURN LANE'],\n",
       " [5.272518074619048, 'x2_MOPED OR MOTORIZED BICYCLE'],\n",
       " [5.2838588556, 'x7_OTHER REG. SIGN'],\n",
       " [5.310910474548333, 'x7_LANE USE MARKING'],\n",
       " [5.3395556, 'x13_CURVE ON GRADE'],\n",
       " [5.340123116466793, 'x4_W'],\n",
       " [5.35935545, 'x14_DRY'],\n",
       " [5.420692579285714, 'x6_REAR-LEFT'],\n",
       " [5.600444555, 'x16_ON SCENE'],\n",
       " [5.6170702, 'x23_CO'],\n",
       " [5.735460398571429, 'x9_CLOUDY/OVERCAST'],\n",
       " [5.7578743532162155, 'x2_SPORT UTILITY VEHICLE (SUV)'],\n",
       " [5.8396075216000005, 'x12_OTHER'],\n",
       " [5.994296453909476, 'x2_PASSENGER'],\n",
       " [6.064100342433334, 'x3_TAXI/FOR HIRE'],\n",
       " [6.721780764999999, 'x11_PEDESTRIAN'],\n",
       " [6.776406587999999, 'x15_DEBRIS ON ROADWAY'],\n",
       " [6.929598028464585, 'x5_DISABLED'],\n",
       " [7.138131133846154, 'x3_MILITARY'],\n",
       " [7.518232015239999, 'x5_AVOIDING VEHICLES/OBJECTS'],\n",
       " [7.646255087283335, 'x3_CTA'],\n",
       " [8.199650908265305, 'x1_NONE'],\n",
       " [8.275792375639623, 'x4_N'],\n",
       " [8.470531427707316, 'x7_RAILROAD CROSSING GATE'],\n",
       " [8.746307563155781, 'x2_PICKUP'],\n",
       " [8.853465766888787, 'x1_LIGHTS'],\n",
       " [8.861354970999995, 'x5_SLOW/STOP IN TRAFFIC'],\n",
       " [9.385229698704546, 'x2_TRACTOR W/ SEMI-TRAILER'],\n",
       " [9.450639506499998, 'x2_OTHER'],\n",
       " [9.662038195971792, 'x3_LAWN CARE/LANDSCAPING'],\n",
       " [9.857386657972969, 'x5_STARTING IN TRAFFIC'],\n",
       " [10.563383409565215, 'x6_SIDE-LEFT'],\n",
       " [11.994906177705884, 'x1_STEERING'],\n",
       " [12.27188630156129, 'x4_SW'],\n",
       " [12.678178422454545, 'x1_FUEL SYSTEM'],\n",
       " [13.104317637800003, 'x2_OTHER VEHICLE WITH TRAILER'],\n",
       " [13.456966521182716, 'x1_CARGO'],\n",
       " [13.567932414832, 'x0_PARKED'],\n",
       " [13.87950683994521, 'x2_VAN/MINI-VAN'],\n",
       " [14.239722848076921, 'x3_MASS TRANSIT'],\n",
       " [14.66843751647059, 'x1_SIGNALS'],\n",
       " [15.495660230641025, 'x2_RECREATIONAL OFF-HIGHWAY VEHICLE (ROV)'],\n",
       " [15.958429388636365, 'x2_SINGLE UNIT TRUCK WITH TRAILER'],\n",
       " [20.21625447885587, 'x1_ENGINE/MOTOR'],\n",
       " [23.46065583081553, 'x1_BRAKES'],\n",
       " [1091.9994696917283, 'x0_DISABLED VEHICLE']]"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "models.show_feature_importances(xgb_opt1, X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Features Importances TRIAL 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[0.15461746425, 'x23_NH'],\n",
       " [0.22121609, 'x23_NM'],\n",
       " [0.2503278226, 'x23_MS'],\n",
       " [0.2834506485, 'x23_NJ'],\n",
       " [0.2836215734, 'x23_MI'],\n",
       " [0.2854856936, 'x23_NY'],\n",
       " [0.3111125635714286, 'x23_HI'],\n",
       " [0.3154645544083333, 'x23_LA'],\n",
       " [0.3236448228, 'x23_MN'],\n",
       " [0.341232276311111, 'x23_ID'],\n",
       " [0.3413169984210526, 'x23_MT'],\n",
       " [0.34180631226666675, 'x23_GA'],\n",
       " [0.3475196125, 'x21_S'],\n",
       " [0.34898385940714277, 'x23_DC'],\n",
       " [0.35582444976923083, 'x23_AK'],\n",
       " [0.3751244628571428, 'x21_W'],\n",
       " [0.38135370975, 'x23_NC'],\n",
       " [0.40180358275, 'x23_NB'],\n",
       " [0.40780508275, 'x23_CA'],\n",
       " [0.4391285765, 'x23_IA'],\n",
       " [0.439166106, 'x23_MD'],\n",
       " [0.4636766552, 'x23_CO'],\n",
       " [0.4688251338461539, 'x21_E'],\n",
       " [0.4691060221944443, 'x20_ROAD CONSTRUCTION/MAINTENANCE'],\n",
       " [0.48555309179999995, 'x23_AZ'],\n",
       " [0.48764141055555554, 'x23_IN'],\n",
       " [0.5087341412, 'x23_KS'],\n",
       " [0.5170276155, 'x17_NO INJURY / DRIVE AWAY'],\n",
       " [0.55960960275, 'x22_DRIVER'],\n",
       " [0.5614726344500001, 'x23_ME'],\n",
       " [0.5879501588, 'x23_AL'],\n",
       " [0.589485638, 'x23_CT'],\n",
       " [0.589487329, 'x23_FL'],\n",
       " [0.5948460378903225, 'x20_IMPROPER BACKING'],\n",
       " [0.604896179, 'x20_WEATHER'],\n",
       " [0.6085600303392859,\n",
       "  'x20_UNDER THE INFLUENCE OF ALCOHOL/DRUGS (USE WHEN ARREST IS EFFECTED)'],\n",
       " [0.6282240496666668, 'x23_AR'],\n",
       " [0.6335290873684211, 'x20_PHYSICAL CONDITION OF DRIVER'],\n",
       " [0.6896593677499998,\n",
       "  'x20_OPERATING VEHICLE IN ERRATIC, RECKLESS, CARELESS, NEGLIGENT OR AGGRESSIVE MANNER'],\n",
       " [0.6964320040359998, 'x20_BICYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [0.7074741704, 'x20_RELATED TO BUS STOP'],\n",
       " [0.7234962016190476, 'x20_EQUIPMENT - VEHICLE CONDITION'],\n",
       " [0.7565394025555557, 'x20_TURNING RIGHT ON RED'],\n",
       " [0.759217007892857, 'x20_DISREGARDING OTHER TRAFFIC SIGNS'],\n",
       " [0.7631638494166667, 'x23_IL'],\n",
       " [0.7642301242857144, 'x20_IMPROPER TURNING/NO SIGNAL'],\n",
       " [0.7822228267777778, 'x23_MA'],\n",
       " [0.788145446090909, 'x22_NON-CONTACT VEHICLE'],\n",
       " [0.791457788368421, 'x19_EXCEEDING SAFE SPEED FOR CONDITIONS'],\n",
       " [0.7952656845, 'x20_OBSTRUCTED CROSSWALKS'],\n",
       " [0.7970766361333332, 'x20_ROAD ENGINEERING/SURFACE/MARKING DEFECTS'],\n",
       " [0.8056129135333335, 'x23_KY'],\n",
       " [0.8160389137099999, 'x19_DRIVING SKILLS/KNOWLEDGE/EXPERIENCE'],\n",
       " [0.821343898857143, 'x19_HAD BEEN DRINKING (USE WHEN ARREST IS NOT MADE)'],\n",
       " [0.8248695466944445, 'x19_PHYSICAL CONDITION OF DRIVER'],\n",
       " [0.8424674389266668,\n",
       "  'x19_VISION OBSCURED (SIGNS, TREE LIMBS, BUILDINGS, ETC.)'],\n",
       " [0.8880332916666666, 'x23_NV'],\n",
       " [0.8925726632666667, 'x20_PASSING STOPPED SCHOOL BUS'],\n",
       " [0.921304472888889, 'x19_MOTORCYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [0.9213092141111113, 'x19_ROAD ENGINEERING/SURFACE/MARKING DEFECTS'],\n",
       " [0.9298646664516129, 'x19_IMPROPER BACKING'],\n",
       " [0.934389850625, 'x12_NOT DIVIDED'],\n",
       " [0.9382225555814813, 'x20_IMPROPER LANE USAGE'],\n",
       " [0.942322852590909, 'x20_CELL PHONE USE OTHER THAN TEXTING'],\n",
       " [0.9503743334705883, 'x19_TURNING RIGHT ON RED'],\n",
       " [0.952406524857143, 'x23_DE'],\n",
       " [0.9541837175833335, 'x20_EXCEEDING SAFE SPEED FOR CONDITIONS'],\n",
       " [0.9803511399599999, 'x12_L-INTERSECTION'],\n",
       " [1.005516106235294, 'x20_FAILING TO REDUCE SPEED TO AVOID CRASH'],\n",
       " [1.0107108878260864, 'x19_TEXTING'],\n",
       " [1.0242187265000002, 'x12_FOUR WAY'],\n",
       " [1.0443655721133331, 'x18_$501 - $1,500'],\n",
       " [1.0748585631875, 'x12_ROUNDABOUT'],\n",
       " [1.0793031758181815, 'x15_RUT, HOLES'],\n",
       " [1.082319803225, 'x11_FIXED OBJECT'],\n",
       " [1.0912814694962407, 'x15_WORN SURFACE'],\n",
       " [1.1165614300869562, 'x19_DISREGARDING YIELD SIGN'],\n",
       " [1.121666541722222, 'x20_FOLLOWING TOO CLOSELY'],\n",
       " [1.122144341222222, 'x20_DRIVING SKILLS/KNOWLEDGE/EXPERIENCE'],\n",
       " [1.1244718360000001, 'x21_N'],\n",
       " [1.125409080502703, 'x19_NOT APPLICABLE'],\n",
       " [1.13432851228, 'x6_UNDER CARRIAGE'],\n",
       " [1.1350147830833333, 'x20_HAD BEEN DRINKING (USE WHEN ARREST IS NOT MADE)'],\n",
       " [1.1373259667647062, 'x20_IMPROPER OVERTAKING/PASSING'],\n",
       " [1.1572940026543777, 'x3_UNKNOWN/NA'],\n",
       " [1.1619393855416664, 'x7_POLICE/FLAGMAN'],\n",
       " [1.1690903724243955, 'x3_MASS TRANSIT'],\n",
       " [1.1701073687307693, 'x20_DISREGARDING TRAFFIC SIGNALS'],\n",
       " [1.1744359975517242, 'x19_UNABLE TO DETERMINE'],\n",
       " [1.1783855415714282, 'x20_NOT APPLICABLE'],\n",
       " [1.1818767701, 'x20_DISTRACTION - FROM OUTSIDE VEHICLE'],\n",
       " [1.1822049942500001, 'x19_OBSTRUCTED CROSSWALKS'],\n",
       " [1.1881551177066667, 'x14_OTHER'],\n",
       " [1.1900795800384614, 'x19_EXCEEDING AUTHORIZED SPEED LIMIT'],\n",
       " [1.231020512826087, 'x20_DISREGARDING YIELD SIGN'],\n",
       " [1.2382357341249999, 'x20_MOTORCYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [1.2486773312994144, 'x2_TRUCK - SINGLE UNIT'],\n",
       " [1.25214632076, 'x15_SHOULDER DEFECT'],\n",
       " [1.2543223712307694, 'x12_PARKING LOT'],\n",
       " [1.2654181297049178, 'x10_DAYLIGHT'],\n",
       " [1.2713253415, 'x20_EXCEEDING AUTHORIZED SPEED LIMIT'],\n",
       " [1.2740325137894737, 'x15_DEBRIS ON ROADWAY'],\n",
       " [1.274088386142857, 'x19_DISREGARDING OTHER TRAFFIC SIGNS'],\n",
       " [1.2839260976500002, 'x20_DRIVING ON WRONG SIDE/WRONG WAY'],\n",
       " [1.289856193714286, 'x12_NOT REPORTED'],\n",
       " [1.2947716240877194, 'x19_BICYCLE ADVANCING LEGALLY ON RED LIGHT'],\n",
       " [1.2948571784090908, 'x5_U-TURN'],\n",
       " [1.3195967440094827, 'x7_DELINEATORS'],\n",
       " [1.3405111713555284, 'x5_AVOIDING VEHICLES/OBJECTS'],\n",
       " [1.3410930624516135, 'x7_UNKNOWN'],\n",
       " [1.359902877125,\n",
       "  'x19_OPERATING VEHICLE IN ERRATIC, RECKLESS, CARELESS, NEGLIGENT OR AGGRESSIVE MANNER'],\n",
       " [1.3639183353737354, 'x2_OTHER'],\n",
       " [1.3653475111666669, 'x20_EVASIVE ACTION DUE TO ANIMAL, OBJECT, NONMOTORIST'],\n",
       " [1.3673335156250006, 'x19_DISREGARDING TRAFFIC SIGNALS'],\n",
       " [1.3681180307313947, 'x12_T-INTERSECTION'],\n",
       " [1.3751626534444445, 'x19_IMPROPER TURNING/NO SIGNAL'],\n",
       " [1.3803394854137934,\n",
       "  'x20_DISTRACTION - OTHER ELECTRONIC DEVICE (NAVIGATION DEVICE, DVD PLAYER, ETC.)'],\n",
       " [1.3860683545527954, 'x5_SLOW/STOP - LOAD/UNLOAD'],\n",
       " [1.395139243069548, 'x5_STARTING IN TRAFFIC'],\n",
       " [1.4060531843970359, 'x3_LAWN CARE/LANDSCAPING'],\n",
       " [1.4080763212142855, 'x13_CURVE ON HILLCREST'],\n",
       " [1.4250429415755392, 'x5_SLOW/STOP IN TRAFFIC'],\n",
       " [1.4333497638608976, 'x3_FIRE'],\n",
       " [1.4348348633691141, 'x3_DRIVER EDUCATION'],\n",
       " [1.4395207225655429, 'x2_VAN/MINI-VAN'],\n",
       " [1.439546165310345, 'x20_ANIMAL'],\n",
       " [1.440246295125, 'x20_FAILING TO YIELD RIGHT-OF-WAY'],\n",
       " [1.4454979160771433, 'x11_REAR TO REAR'],\n",
       " [1.4493217251066666, 'x7_RAILROAD CROSSING GATE'],\n",
       " [1.452130006515837, 'x3_AMBULANCE'],\n",
       " [1.460447630772686, 'x3_TAXI/FOR HIRE'],\n",
       " [1.4615470205185193, 'x7_NO PASSING'],\n",
       " [1.462231569479167, 'x10_DAWN'],\n",
       " [1.4647991103529407, 'x11_REAR TO FRONT'],\n",
       " [1.4657268672976178, 'x5_UNKNOWN/NA'],\n",
       " [1.4709765544545454, 'x19_ROAD CONSTRUCTION/MAINTENANCE'],\n",
       " [1.473253512195517, 'x1_SUSPENSION'],\n",
       " [1.4858709654, 'x19_IMPROPER OVERTAKING/PASSING'],\n",
       " [1.4899094963207542, 'x19_ANIMAL'],\n",
       " [1.4927892837916672, 'x16_AMENDED'],\n",
       " [1.499147474925926, 'x19_DISREGARDING ROAD MARKINGS'],\n",
       " [1.5023393963246572, 'x12_DRIVEWAY'],\n",
       " [1.5136752002857141, 'x12_CENTER TURN LANE'],\n",
       " [1.5173944674721458, 'x5_PARKED IN TRAFFIC LANE'],\n",
       " [1.5204002931871676, 'x5_BACKING'],\n",
       " [1.5250979355217917, 'x4_N'],\n",
       " [1.5315197781892402, 'x2_MOTORCYCLE (OVER 150CC)'],\n",
       " [1.53216345988, 'x19_CELL PHONE USE OTHER THAN TEXTING'],\n",
       " [1.5325265782222228, 'x3_TOW TRUCK'],\n",
       " [1.5449806719717534, 'x5_ENTER FROM DRIVE/ALLEY'],\n",
       " [1.5551560531931812, 'x6_REAR-RIGHT'],\n",
       " [1.5656913072370373, 'x11_REAR END'],\n",
       " [1.5688557753939394, 'x5_SLOW/STOP - LEFT TURN'],\n",
       " [1.5726696018947368, 'x8_MISSING'],\n",
       " [1.5738946094999997, 'x18_OVER $1,500'],\n",
       " [1.5866311414077312, 'x3_OTHER'],\n",
       " [1.58779085, 'x20_UNABLE TO DETERMINE'],\n",
       " [1.5878991513428569, 'x20_DISTRACTION - FROM INSIDE VEHICLE'],\n",
       " [1.590063996434783, 'x19_FAILING TO REDUCE SPEED TO AVOID CRASH'],\n",
       " [1.5972259772412816, 'x8_FUNCTIONING IMPROPERLY'],\n",
       " [1.6019065180861327, 'x2_ALL-TERRAIN VEHICLE (ATV)'],\n",
       " [1.60193373125, 'x13_CURVE ON GRADE'],\n",
       " [1.6031470643621784, 'x4_NW'],\n",
       " [1.614963779485714, 'x16_ON SCENE'],\n",
       " [1.6199160834929593, 'x3_SCHOOL BUS'],\n",
       " [1.6230209662568627, 'x6_FRONT-LEFT'],\n",
       " [1.6243455620819673, 'x3_COMMERCIAL - MULTI-UNIT'],\n",
       " [1.6368833290033553, 'x4_S'],\n",
       " [1.6491606762285715, 'x14_SNOW OR SLUSH'],\n",
       " [1.660106949544366, 'x2_UNKNOWN/NA'],\n",
       " [1.6674230233399996, 'x2_OTHER VEHICLE WITH TRAILER'],\n",
       " [1.6712252220000006, 'x14_UNKNOWN'],\n",
       " [1.6732167390172414, 'x4_UNKNOWN'],\n",
       " [1.6805881644909082, 'x14_SAND, MUD, DIRT'],\n",
       " [1.6854303048857142, 'x19_RELATED TO BUS STOP'],\n",
       " [1.6863341149779456, 'x2_MOPED OR MOTORIZED BICYCLE'],\n",
       " [1.6869975709812497, 'x5_PARKED'],\n",
       " [1.688596955153846,\n",
       "  'x20_VISION OBSCURED (SIGNS, TREE LIMBS, BUILDINGS, ETC.)'],\n",
       " [1.6938436491261544, 'x8_WORN REFLECTIVE MATERIAL'],\n",
       " [1.6961833968647058, 'x14_WET'],\n",
       " [1.7001751987272729, 'x6_ROOF'],\n",
       " [1.7059364353783786,\n",
       "  'x19_DISTRACTION - OTHER ELECTRONIC DEVICE (NAVIGATION DEVICE, DVD PLAYER, ETC.)'],\n",
       " [1.7124313693764706, 'x7_SCHOOL ZONE'],\n",
       " [1.7173426088182737, 'x3_CTA'],\n",
       " [1.718044439480519, 'x11_OVERTURNED'],\n",
       " [1.7265015563904764, 'x3_PERSONAL'],\n",
       " [1.7380019667142856, 'x12_Y-INTERSECTION'],\n",
       " [1.7458327409999999, 'x15_NO DEFECTS'],\n",
       " [1.7542368238891894, 'x1_LIGHTS'],\n",
       " [1.7549643833407857, 'x5_CHANGING LANES'],\n",
       " [1.7604533873132526, 'x10_UNKNOWN'],\n",
       " [1.766883915464286, 'x15_OTHER'],\n",
       " [1.7712266661470584, 'x14_ICE'],\n",
       " [1.7741610920000002, 'x20_DISREGARDING STOP SIGN'],\n",
       " [1.7765330037352935, 'x12_UNKNOWN'],\n",
       " [1.7767562730933335, 'x3_RIDESHARE SERVICE'],\n",
       " [1.7768251724806632, 'x7_RR CROSSING SIGN'],\n",
       " [1.7854702259062503, 'x14_DRY'],\n",
       " [1.7954709927726402, 'x6_REAR'],\n",
       " [1.7993162268974352, 'x12_DIVIDED - W/MEDIAN BARRIER'],\n",
       " [1.8025147306984526, 'x6_NONE'],\n",
       " [1.803210193870968, 'x12_OTHER'],\n",
       " [1.8055610092199976, 'x4_W'],\n",
       " [1.8071569680909088, 'x19_PASSING STOPPED SCHOOL BUS'],\n",
       " [1.8075071748235294, 'x19_WEATHER'],\n",
       " [1.8079100968000001,\n",
       "  'x19_UNDER THE INFLUENCE OF ALCOHOL/DRUGS (USE WHEN ARREST IS EFFECTED)'],\n",
       " [1.80866586715, 'x19_DRIVING ON WRONG SIDE/WRONG WAY'],\n",
       " [1.820512021180299, 'x1_FUEL SYSTEM'],\n",
       " [1.8207032199729734, 'x5_DRIVERLESS'],\n",
       " [1.8219542343800004, 'x19_DISTRACTION - FROM INSIDE VEHICLE'],\n",
       " [1.8255221463571425, 'x12_TRAFFIC ROUTE'],\n",
       " [1.8336494677065562, 'x2_SINGLE UNIT TRUCK WITH TRAILER'],\n",
       " [1.8343300543571428, 'x13_STRAIGHT ON HILLCREST'],\n",
       " [1.8369874207083334, 'x5_NEGOTIATING A CURVE'],\n",
       " [1.8399549713333327, 'x11_TURNING'],\n",
       " [1.8423043515116277, 'x6_REAR-LEFT'],\n",
       " [1.8441251756666663, 'x7_NO CONTROLS'],\n",
       " [1.8499931884309266, 'x4_E'],\n",
       " [1.854661709709845, 'x7_LANE USE MARKING'],\n",
       " [1.8578318187222225, 'x16_NOT ON SCENE (DESK REPORT)'],\n",
       " [1.8603652168910894, 'x8_FUNCTIONING PROPERLY'],\n",
       " [1.866192469809836, 'x6_SIDE-RIGHT'],\n",
       " [1.8682699305000001, 'x23_MO'],\n",
       " [1.8775980619066144, 'x2_BUS UP TO 15 PASS.'],\n",
       " [1.8820590628855292, 'x5_STRAIGHT AHEAD'],\n",
       " [1.8833385077706024, 'x7_OTHER'],\n",
       " [1.8848321444545457, 'x12_ALLEY'],\n",
       " [1.8880800899250008, 'x18_$500 OR LESS'],\n",
       " [1.8945070823658539, 'x10_DARKNESS, LIGHTED ROAD'],\n",
       " [1.8980052846114863, 'x1_NONE'],\n",
       " [1.9047518577272724, 'x19_EQUIPMENT - VEHICLE CONDITION'],\n",
       " [1.9100973924464288, 'x15_UNKNOWN'],\n",
       " [1.9125960032758618, 'x19_DISREGARDING STOP SIGN'],\n",
       " [1.920447217976608, 'x6_OTHER'],\n",
       " [1.9256878526638506, 'x4_NE'],\n",
       " [1.9276282665882354, 'x11_OTHER OBJECT'],\n",
       " [1.9355211739614173, 'x2_PICKUP'],\n",
       " [1.9473023645848373, 'x2_3-WHEELED MOTORCYCLE (2 REAR WHEELS)'],\n",
       " [1.9590885248222223, 'x10_DARKNESS'],\n",
       " [1.9667436925000004, 'x19_IMPROPER LANE USAGE'],\n",
       " [1.9711850223742569, 'x2_FARM EQUIPMENT'],\n",
       " [1.9714220554537014, 'x3_MILITARY'],\n",
       " [1.9912059548554224, 'x6_FRONT'],\n",
       " [1.9948726796666676, 'x6_UNKNOWN'],\n",
       " [1.996349776299412, 'x5_LEAVING TRAFFIC LANE TO PARK'],\n",
       " [2.014426982206896, 'x7_STOP SIGN/FLASHER'],\n",
       " [2.0211665886666697, 'x1_TIRES'],\n",
       " [2.0504806909848488, 'x5_TURNING ON RED'],\n",
       " [2.0580126552500007, 'x3_STATE OWNED'],\n",
       " [2.0670484523846153, 'x11_SIDESWIPE OPPOSITE DIRECTION'],\n",
       " [2.0693591616896554, 'x12_UNKNOWN INTERSECTION TYPE'],\n",
       " [2.074227147686567, 'x7_FLASHING CONTROL SIGNAL'],\n",
       " [2.076611516666667, 'x20_TEXTING'],\n",
       " [2.0934134606315786, 'x7_OTHER RAILROAD CROSSING'],\n",
       " [2.1029152467368415, 'x11_TRAIN'],\n",
       " [2.1040310509242426, 'x17_INJURY AND / OR TOW DUE TO CRASH'],\n",
       " [2.1083047771142858, 'x13_STRAIGHT ON GRADE'],\n",
       " [2.120987644020834, 'x13_CURVE, LEVEL'],\n",
       " [2.1248973681492602, 'x7_BICYCLE CROSSING SIGN'],\n",
       " [2.140169582333334, 'x11_ANIMAL'],\n",
       " [2.1462567401736106, 'x6_FRONT-RIGHT'],\n",
       " [2.1535006995365857, 'x19_FOLLOWING TOO CLOSELY'],\n",
       " [2.1601142553356145, 'x9_CLEAR'],\n",
       " [2.176648988086956, 'x12_ONE-WAY'],\n",
       " [2.181339669363637, 'x12_FIVE POINT, OR MORE'],\n",
       " [2.20379133683495, 'x3_AGRICULTURE'],\n",
       " [2.212525836564705, 'x11_ANGLE'],\n",
       " [2.212720745784092, 'x5_DIVERGING'],\n",
       " [2.220200081720258, 'x5_OTHER'],\n",
       " [2.2360918254728994, 'x1_STEERING'],\n",
       " [2.2560346165853664, 'x3_CAMPER/RV - SINGLE UNIT'],\n",
       " [2.282376545794117, 'x7_PEDESTRIAN CROSSING SIGN'],\n",
       " [2.282983630271604, 'x10_DUSK'],\n",
       " [2.2844633259561404, 'x3_NOT IN USE'],\n",
       " [2.290306631339621, 'x2_RECREATIONAL OFF-HIGHWAY VEHICLE (ROV)'],\n",
       " [2.3197951913466666, 'x12_DIVIDED - W/MEDIAN (NOT RAISED)'],\n",
       " [2.324435682104, 'x11_REAR TO SIDE'],\n",
       " [2.3279693671081083, 'x13_STRAIGHT AND LEVEL'],\n",
       " [2.350113947341881, 'x8_OTHER'],\n",
       " [2.3541250592698404, 'x8_NO CONTROLS'],\n",
       " [2.3744739590000004, 'x5_TURNING LEFT'],\n",
       " [2.395232390326923, 'x5_SKIDDING/CONTROL LOSS'],\n",
       " [2.3958695655781255, 'x11_HEAD ON'],\n",
       " [2.4039831737642867, 'x7_YIELD'],\n",
       " [2.4139903200434785, 'x19_EVASIVE ACTION DUE TO ANIMAL, OBJECT, NONMOTORIST'],\n",
       " [2.414126332010871, 'x5_DRIVING WRONG WAY'],\n",
       " [2.4154114250975605, 'x7_OTHER WARNING SIGN'],\n",
       " [2.419990567136587, 'x1_SIGNALS'],\n",
       " [2.4430288100908233, 'x9_CLOUDY/OVERCAST'],\n",
       " [2.496124404022223, 'x2_SPORT UTILITY VEHICLE (SUV)'],\n",
       " [2.4993171945159225, 'x2_TRACTOR W/O SEMI-TRAILER'],\n",
       " [2.5249686916300815, 'x1_OTHER'],\n",
       " [2.564765368571429, 'x20_DISREGARDING ROAD MARKINGS'],\n",
       " [2.6293920657920804, 'x3_POLICE'],\n",
       " [2.661027077175573, 'x5_ENTERING TRAFFIC LANE FROM PARKING'],\n",
       " [2.6793395131785713, 'x4_SE'],\n",
       " [2.707008275156076, 'x2_MOTOR DRIVEN CYCLE'],\n",
       " [2.7322167200243586, 'x6_SIDE-LEFT'],\n",
       " [2.7530977494036466, 'x1_UNKNOWN'],\n",
       " [2.7910518666190476, 'x3_COMMERCIAL - SINGLE UNIT'],\n",
       " [2.7915599245789466, 'x12_RAMP'],\n",
       " [2.797327621851063, 'x11_SIDESWIPE SAME DIRECTION'],\n",
       " [2.801637975599999, 'x8_UNKNOWN'],\n",
       " [2.840123306166666, 'x11_OTHER NONCOLLISION'],\n",
       " [2.893940880129413, 'x9_UNKNOWN'],\n",
       " [2.930702243427873, 'x3_OTHER TRANSIT'],\n",
       " [2.949067178313868, 'x2_BUS OVER 15 PASS.'],\n",
       " [2.9533824648593745, 'x11_PEDALCYCLIST'],\n",
       " [3.0338538739736842, 'x11_PEDESTRIAN'],\n",
       " [3.0422698078999972, 'x5_SLOW/STOP - RIGHT TURN'],\n",
       " [3.0999349480583933, 'x2_TRACTOR W/ SEMI-TRAILER'],\n",
       " [3.1282969950099964, 'x7_TRAFFIC SIGNAL'],\n",
       " [3.1558599891250005, 'x5_PASSING/OVERTAKING'],\n",
       " [3.2373313017719285, 'x7_OTHER REG. SIGN'],\n",
       " [3.7599160791142836, 'x5_DISABLED'],\n",
       " [3.892170659090909, 'x9_RAIN'],\n",
       " [3.897779264119208, 'x4_SW'],\n",
       " [3.951969676188406, 'x1_WINDOWS'],\n",
       " [4.025639975056095, 'x0_DRIVERLESS'],\n",
       " [4.245074750731708, 'x11_PARKED MOTOR VEHICLE'],\n",
       " [4.5839038124239995, 'x5_TURNING RIGHT'],\n",
       " [4.7183142083729335, 'x0_NON-CONTACT VEHICLE'],\n",
       " [4.784998560063638, 'x9_SNOW'],\n",
       " [4.877268641530436, 'x8_NOT FUNCTIONING'],\n",
       " [5.236058486363637, 'x19_FAILING TO YIELD RIGHT-OF-WAY'],\n",
       " [5.407252042050507, 'x1_CARGO'],\n",
       " [5.4896170662884645, 'x9_OTHER'],\n",
       " [5.743054246909011, 'x6_TOTAL (ALL AREAS)'],\n",
       " [6.213408361608576, 'x2_PASSENGER'],\n",
       " [6.661194422624999, 'x1_BRAKES'],\n",
       " [6.9467828724177245, 'x5_MERGING'],\n",
       " [7.003850229715686, 'x3_CONSTRUCTION/MAINTENANCE'],\n",
       " [8.156844538455172, 'x0_PARKED'],\n",
       " [8.403192918830682, 'x1_WHEELS'],\n",
       " [8.716943740641431, 'x1_ENGINE/MOTOR'],\n",
       " [11.037010628775658, 'x0_DRIVER'],\n",
       " [181.23680767714285, 'x19_DISTRACTION - FROM OUTSIDE VEHICLE'],\n",
       " [810.6248598214955, 'x0_DISABLED VEHICLE']]"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "models.show_feature_importances(xgb_opt2, X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Top 5 Gain Metrics:\n",
    "\n",
    "*[Gain](https://datascience.stackexchange.com/questions/12318/how-to-interpret-the-output-of-xgboost-importance) is defined as the relative \"contribution\" of a given feature to the model.*\n",
    "\n",
    "\n",
    "###### TRIAL 1\n",
    "\n",
    "* `UNIT_TYPE` - DISABLED VEHICLE (1092)\n",
    "* `VEHICLE_DEFECT` - BRAKES (23.5)\n",
    "* `VEHICLE_DEFECT` - ENGINE/MOTOR (8.71)\n",
    "* `VEHICLE_TYPE` - SINGLE UNIT TRUCK WITH TRAILER (15.96)\n",
    "* `VEHICLE_TYPE` - RECREATIONAL OFF-HIGHWAY VEHICLE (15.5)\n",
    "\n",
    "\n",
    "###### TRIAL 2\n",
    "\n",
    "* `UNIT_TYPE` - DISABLED VEHICLE (810)\n",
    "* `DRIVER_ACTION` - DISTRACTION - FROM OUTSIDE VEHICLE (181)\n",
    "* `PERSON_TYPE` - DRIVER (11)\n",
    "* `VEHICLE_DEFECT` - ENGINE/MOTOR (8.71)\n",
    "* `VEHICLE_DEFECT` - WHEELS (8.4)\n",
    "\n",
    "\n",
    "*NOTE: This is NOT the same as Logistic Regression coefficient values.*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Next Steps\n",
    "\n",
    "* Find a way to automate optimization of feature selection AND engineering (trying different combinations of different versions of features)\n",
    "\n",
    "* Properly implement **pipelines** for the modeling process.\n",
    "\n",
    "* Subset the data and create different types of models. There were columns that were removed because they contained too many NaN values in the context of the entire dataset, but still has thousands of entries. Possibilities for subset models could include:\n",
    "    * Predicting crash injury for **Pedestrians only**\n",
    "    * Predicting crash injury for certain **age groups**\n",
    "    * Predicting crash injury based on **cell phone use**\n",
    "\n",
    "\n",
    "## Sources\n",
    "\n",
    "##### Source Code\n",
    "* [Changing font size for Confusion Matrix](https://stackoverflow.com/questions/59839782/confusion-matrix-font-size)\n",
    "*[SMOTE-ing Data Tutorial](https://machinelearningmastery.com/smote-oversampling-for-imbalanced-classification/)\n",
    "*[Bayesian Optimization: A Beginner's Guide](https://analyticsindiamag.com/implementing-bayesian-optimization-on-xgboost-a-beginners-guide/)\n",
    "* Source code for this notebook can be found in the `src` folder.\n",
    "\n",
    "##### Business Understanding\n",
    "\n",
    "* [Speeding Increases During Pandemic, Prompting Safety Groups To Take Action](http://www.ghsa.org/about/news/Forbes/Speed-Pilot21)\n",
    "* [Calls for safer streets intensify amid 45% spike in pedestrian deaths](https://www.smartcitiesdive.com/news/calls-for-safer-streets-intensify-amid-45-spike-in-pedestrian-deaths/596420/)\n",
    "* [Chicago unveils its West Side Vision Zero Traffic Safety Plan](https://chi.streetsblog.org/2019/09/12/the-city-of-chicago-unveils-its-west-side-vision-zero-traffic-safety-plan/)\n",
    "* [Vision Zero Chicago Webpage](https://visionzerochicago.org/what-is-vision-zero/)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}