1_Abstract_and_cleaning_accidents_data.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project abstract"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Project aims**\n",
    "\n",
    "Each year about 1.25 million people die in road traffic accidents, and an additional 20-50 million are injured or disabled. If the locations of traffic accidents could be predicted, this could have a huge beneficial impact in potentially helping to reduce the number of accidents each year. For example, routing software could avoid the most dangerous areas - particularly in the context of the coming advent of driverless cars. It could also be useful in an insurance context, in order to predict risk, as well as for governments and local road authorities looking to create more efficient systems of road maintenance and improvements. The aim of this project is to predict where traffic accidents are likely to occur. \n",
    "\n",
    "\n",
    "**Methodology**\n",
    "\n",
    "For this problem, we didn’t want to just look at traditional structured data and machine learning models. Instead, we wanted to find out if satellite imagery could be combined with other datasets in order to increase our ability to predict where traffic accidents are likely to occur. Our methodology for this project was to make four models:\n",
    "\n",
    "*Model 1* uses a combination of accident, population density and traffic data from the UK, where we focused on accidents in London.  Different machine learning models were built to see if the level of accident severity could be predicted.  \n",
    "\n",
    "*Model 2* uses satellite images of London that were scraped using Google Maps Static API and fed into a Convolutional Neural Network (CNN) in order to predict where traffic accidents are likely to occur.\n",
    "\n",
    "*Model 3* then makes use of Keras functional API to combine the top features from model 1 and the image features extracted from a CNN (similar to model 2) to create a mixed-input or mixed data model. Both data types are fed into separate deep learning models and their outputs are combined for the final layers in order to predict whether a given area is likely to have traffic accidents or not. \n",
    "\n",
    "*Model 4* uses the same model architecture as model 3, but applies it to the task of distinguishing between areas with no traffic accidents and areas with serious or fatal traffic accidents, in order to predict the locations of the worst traffic accidents.\n",
    "\n",
    "\n",
    "**Results**\n",
    "\n",
    "Our findings showed that being able to predict accident severity was quite difficult due to the imbalance between slight, serious, and fatal accidents (58% accuracy).  Using satellite images with a CNN model yielded strong results in predicting traffic accident locations (77% accuracy). The mixed data models had an even higher accuracy (80% for no accidents vs. any accidents, and 82% for no accidents vs. serious or fatal accidents), producing better results from combining image features with structured data. There are also several promising options to improve the model further, beyond this initial proof of concept.\n",
    "\n",
    "Overall, we were able to demonstrate that combining satellite images with structured data can increase the predictive capacity of a model to predict the location of traffic accidents.\n",
    "\n",
    "\n",
    "**Notebook structure**\n",
    "\n",
    "The following notebooks show the steps we took in order to reach these findings:\n",
    "1. Cleaning the accident data\n",
    "2. Cleaning the traffic and population density data, merging this with the accident data, and downloading satellite images from the Google Maps Static API\n",
    "3. Exploring the geographical distribution of the data, and comparing the characteristics of accidents of different levels of severity\n",
    "4. Exploring the changes in accident numbers over time, and investigating various other features of the accident dataset\n",
    "5. Model 1 - predicting traffic accident severity with structured data\n",
    "6. Model 2 - predicting traffic accident location from satellite imagery\n",
    "7. Model 3 - predicting traffic accident location using mixed data (satellite imagery and structured data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xXYQlI6ZZlLl"
   },
   "source": [
    "# Cleaning the traffic accident dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook, the dataset on traffic accidents in England and Wales from 2013-2017 will be cleaned and pre-processed.\n",
    "\n",
    "Source: https://data.gov.uk/dataset/cb7ae6f0-4be6-4935-9277-47e5ce24a11f/road-safety-data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "JJn0DkgjZlLo"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import urllib, os\n",
    "import glob\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.preprocessing import MinMaxScaler, LabelEncoder, StandardScaler\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, auc, roc_curve, classification_report\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "WJRllLqhZlL5"
   },
   "source": [
    "## Importing the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df = pd.concat([pd.read_csv(f) for f in glob.glob('data/Accidents/Accidents*.csv')], ignore_index = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "L0gPGqqSZlMj",
    "outputId": "aeb058d4-6f43-4b19-da20-3cb9c123fb66"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "691641"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(accidents_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Accident_Index</th>\n",
       "      <th>Location_Easting_OSGR</th>\n",
       "      <th>Location_Northing_OSGR</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Police_Force</th>\n",
       "      <th>Accident_Severity</th>\n",
       "      <th>Number_of_Vehicles</th>\n",
       "      <th>Number_of_Casualties</th>\n",
       "      <th>Date</th>\n",
       "      <th>...</th>\n",
       "      <th>Pedestrian_Crossing-Human_Control</th>\n",
       "      <th>Pedestrian_Crossing-Physical_Facilities</th>\n",
       "      <th>Light_Conditions</th>\n",
       "      <th>Weather_Conditions</th>\n",
       "      <th>Road_Surface_Conditions</th>\n",
       "      <th>Special_Conditions_at_Site</th>\n",
       "      <th>Carriageway_Hazards</th>\n",
       "      <th>Urban_or_Rural_Area</th>\n",
       "      <th>Did_Police_Officer_Attend_Scene_of_Accident</th>\n",
       "      <th>LSOA_of_Accident_Location</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>201501BS70001</td>\n",
       "      <td>525130.0</td>\n",
       "      <td>180050.0</td>\n",
       "      <td>-0.198465</td>\n",
       "      <td>51.505538</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>12/01/2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>E01002825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>201501BS70002</td>\n",
       "      <td>526530.0</td>\n",
       "      <td>178560.0</td>\n",
       "      <td>-0.178838</td>\n",
       "      <td>51.491836</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>12/01/2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>E01002820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>201501BS70004</td>\n",
       "      <td>524610.0</td>\n",
       "      <td>181080.0</td>\n",
       "      <td>-0.205590</td>\n",
       "      <td>51.514910</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>12/01/2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>E01002833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>201501BS70005</td>\n",
       "      <td>524420.0</td>\n",
       "      <td>181080.0</td>\n",
       "      <td>-0.208327</td>\n",
       "      <td>51.514952</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>13/01/2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>E01002874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>201501BS70008</td>\n",
       "      <td>524630.0</td>\n",
       "      <td>179040.0</td>\n",
       "      <td>-0.206022</td>\n",
       "      <td>51.496572</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>09/01/2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>E01002814</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  Accident_Index  Location_Easting_OSGR  Location_Northing_OSGR  Longitude  \\\n",
       "0  201501BS70001               525130.0                180050.0  -0.198465   \n",
       "1  201501BS70002               526530.0                178560.0  -0.178838   \n",
       "2  201501BS70004               524610.0                181080.0  -0.205590   \n",
       "3  201501BS70005               524420.0                181080.0  -0.208327   \n",
       "4  201501BS70008               524630.0                179040.0  -0.206022   \n",
       "\n",
       "    Latitude  Police_Force  Accident_Severity  Number_of_Vehicles  \\\n",
       "0  51.505538             1                  3                   1   \n",
       "1  51.491836             1                  3                   1   \n",
       "2  51.514910             1                  3                   1   \n",
       "3  51.514952             1                  3                   1   \n",
       "4  51.496572             1                  2                   2   \n",
       "\n",
       "   Number_of_Casualties        Date            ...              \\\n",
       "0                     1  12/01/2015            ...               \n",
       "1                     1  12/01/2015            ...               \n",
       "2                     1  12/01/2015            ...               \n",
       "3                     1  13/01/2015            ...               \n",
       "4                     1  09/01/2015            ...               \n",
       "\n",
       "   Pedestrian_Crossing-Human_Control Pedestrian_Crossing-Physical_Facilities  \\\n",
       "0                                  0                                       0   \n",
       "1                                  0                                       0   \n",
       "2                                  0                                       1   \n",
       "3                                  0                                       0   \n",
       "4                                  0                                       5   \n",
       "\n",
       "   Light_Conditions Weather_Conditions  Road_Surface_Conditions  \\\n",
       "0                 4                  1                        1   \n",
       "1                 1                  1                        1   \n",
       "2                 4                  2                        2   \n",
       "3                 1                  1                        2   \n",
       "4                 1                  2                        2   \n",
       "\n",
       "   Special_Conditions_at_Site  Carriageway_Hazards  Urban_or_Rural_Area  \\\n",
       "0                           0                    0                    1   \n",
       "1                           0                    0                    1   \n",
       "2                           0                    0                    1   \n",
       "3                           0                    0                    1   \n",
       "4                           0                    0                    1   \n",
       "\n",
       "   Did_Police_Officer_Attend_Scene_of_Accident  LSOA_of_Accident_Location  \n",
       "0                                            1                  E01002825  \n",
       "1                                            1                  E01002820  \n",
       "2                                            1                  E01002833  \n",
       "3                                            2                  E01002874  \n",
       "4                                            2                  E01002814  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 691641 entries, 0 to 691640\n",
      "Data columns (total 32 columns):\n",
      "Accident_Index                                 691641 non-null object\n",
      "Location_Easting_OSGR                          691588 non-null float64\n",
      "Location_Northing_OSGR                         691588 non-null float64\n",
      "Longitude                                      691578 non-null float64\n",
      "Latitude                                       691578 non-null float64\n",
      "Police_Force                                   691641 non-null int64\n",
      "Accident_Severity                              691641 non-null int64\n",
      "Number_of_Vehicles                             691641 non-null int64\n",
      "Number_of_Casualties                           691641 non-null int64\n",
      "Date                                           691641 non-null object\n",
      "Day_of_Week                                    691641 non-null int64\n",
      "Time                                           691610 non-null object\n",
      "Local_Authority_(District)                     691641 non-null int64\n",
      "Local_Authority_(Highway)                      691641 non-null object\n",
      "1st_Road_Class                                 691641 non-null int64\n",
      "1st_Road_Number                                691641 non-null int64\n",
      "Road_Type                                      691641 non-null int64\n",
      "Speed_limit                                    691604 non-null float64\n",
      "Junction_Detail                                691641 non-null int64\n",
      "Junction_Control                               691641 non-null int64\n",
      "2nd_Road_Class                                 691641 non-null int64\n",
      "2nd_Road_Number                                691641 non-null int64\n",
      "Pedestrian_Crossing-Human_Control              691641 non-null int64\n",
      "Pedestrian_Crossing-Physical_Facilities        691641 non-null int64\n",
      "Light_Conditions                               691641 non-null int64\n",
      "Weather_Conditions                             691641 non-null int64\n",
      "Road_Surface_Conditions                        691641 non-null int64\n",
      "Special_Conditions_at_Site                     691641 non-null int64\n",
      "Carriageway_Hazards                            691641 non-null int64\n",
      "Urban_or_Rural_Area                            691641 non-null int64\n",
      "Did_Police_Officer_Attend_Scene_of_Accident    691641 non-null int64\n",
      "LSOA_of_Accident_Location                      648221 non-null object\n",
      "dtypes: float64(5), int64(22), object(5)\n",
      "memory usage: 168.9+ MB\n"
     ]
    }
   ],
   "source": [
    "accidents_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cleaning the dataset\n",
    "Below will drop initial columns based on domain knowledge and reading through the data dictionary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(691641, 24)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# First make initial list of column to drop that won't be useful\n",
    "# Accident Severity should be object\n",
    "# Date to date_time\n",
    "# Day_of_Week to object\n",
    "# Time to date_time\n",
    "# 1st/2nd_Road_Class is object but are numbers\n",
    "# Note Road_Type is object but are numbers\n",
    "# Junction_Detail is float but should be objects\n",
    "# Junction_Control is object but are numbers\n",
    "# Light/Weather/Road_Surface_Conditions/Special_Conditions_at_Site/Carriageway_Hazards are objects but numbers\n",
    "# Urban_or_Rural_Area is int should be strings\n",
    "# Also note nulls are usually marked by -1\n",
    "drop_list = ['Police_Force', 'Local_Authority_(District)', 'Local_Authority_(Highway)', \n",
    "             '1st_Road_Number', '2nd_Road_Number', 'Pedestrian_Crossing-Human_Control', \n",
    "             'Pedestrian_Crossing-Physical_Facilities','Did_Police_Officer_Attend_Scene_of_Accident']\n",
    "accidents_df = accidents_df.drop(labels=drop_list, axis=1)\n",
    "accidents_df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dealing with nulls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Accident_Index                0.000000\n",
       "Location_Easting_OSGR         0.000077\n",
       "Location_Northing_OSGR        0.000077\n",
       "Longitude                     0.000091\n",
       "Latitude                      0.000091\n",
       "Accident_Severity             0.000000\n",
       "Number_of_Vehicles            0.000000\n",
       "Number_of_Casualties          0.000000\n",
       "Date                          0.000000\n",
       "Day_of_Week                   0.000000\n",
       "Time                          0.000045\n",
       "1st_Road_Class                0.000000\n",
       "Road_Type                     0.000000\n",
       "Speed_limit                   0.000053\n",
       "Junction_Detail               0.000000\n",
       "Junction_Control              0.000000\n",
       "2nd_Road_Class                0.000000\n",
       "Light_Conditions              0.000000\n",
       "Weather_Conditions            0.000000\n",
       "Road_Surface_Conditions       0.000000\n",
       "Special_Conditions_at_Site    0.000000\n",
       "Carriageway_Hazards           0.000000\n",
       "Urban_or_Rural_Area           0.000000\n",
       "LSOA_of_Accident_Location     0.062778\n",
       "dtype: float64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.isna().sum()/len(accidents_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For Speed_limit replace NaNs with median value\n",
    "accidents_df.Speed_limit.fillna(accidents_df.Speed_limit.median(), inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correcting dtypes and labels\n",
    "A lot of the data uses numerical labeling, so majority of the cleaning will be checking these datatypes and assinging the correct labels according to the data dictionary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3    0.836837\n",
       "2    0.151230\n",
       "1    0.011932\n",
       "Name: Accident_Severity, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Accident_Severity.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change values to listed meanings\n",
    "accidents_df.Accident_Severity = accidents_df.Accident_Severity.replace([1,2,3], ['Fatal', 'Serious', 'Slight'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change Date to datetime\n",
    "accidents_df['Date'] = pd.to_datetime(accidents_df['Date'], format='%d/%m/%Y')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Take just hour from Time string and convert to datetime\n",
    "# Can possibly label the times into groups later\n",
    "accidents_df['Hour'] = pd.to_datetime(accidents_df['Time'], format='%H:%M').dt.hour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df = accidents_df.drop('Time', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df.Hour.fillna(accidents_df.Hour.median(), inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df.Hour = accidents_df.Hour.astype('int')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def two_hour_groupings(hour):\n",
    "    '''Creates two hour bins from Hour column'''\n",
    "    if hour >= 0 and hour < 2:\n",
    "        return '12am-2am'\n",
    "    if hour >= 2 and hour < 4:\n",
    "        return '2am-4am'\n",
    "    if hour >= 4 and hour < 6:\n",
    "        return '4am-6am'\n",
    "    if hour >= 6 and hour < 8:\n",
    "        return '6am-8am'\n",
    "    if hour >= 8 and hour < 10:\n",
    "        return '8am-10am'\n",
    "    if hour >= 10 and hour < 12:\n",
    "        return '10am-12pm'\n",
    "    if hour >= 12 and hour < 14:\n",
    "        return '12pm-2pm'\n",
    "    if hour >= 14 and hour < 16:\n",
    "        return '2pm-4pm'\n",
    "    if hour >= 16 and hour < 18:\n",
    "        return '4pm-6pm'\n",
    "    if hour >= 18 and hour < 20:\n",
    "        return '6pm-8pm'\n",
    "    if hour >= 20 and hour < 22:\n",
    "        return '8pm-10pm'\n",
    "    if hour >= 22 and hour < 24:\n",
    "        return '10pm-12am'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# apply function (probably drop Hour later before modeling)\n",
    "accidents_df['Two_Hour_Groupings'] = accidents_df['Hour'].apply(two_hour_groupings)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I kept the below function for possibly looking at during EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use function to group house into Time_of_Day\n",
    "def Hour_to_groups(hour):\n",
    "    if hour >= 5 and hour < 10:\n",
    "        return 'Morning'\n",
    "    elif hour >= 10 and hour < 15:\n",
    "        return 'Office hours'\n",
    "    elif hour >= 15 and hour < 19:\n",
    "        return 'Rush hour'\n",
    "    elif hour >= 19 and hour < 23:\n",
    "        return 'Evening'\n",
    "    else:\n",
    "        return 'Night'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# apply function (probably drop Hour later before modeling)\n",
    "accidents_df['Time_of_Day'] = accidents_df['Hour'].apply(Hour_to_groups)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6    0.163672\n",
       "5    0.152089\n",
       "3    0.151223\n",
       "4    0.150698\n",
       "2    0.141935\n",
       "7    0.130423\n",
       "1    0.109960\n",
       "Name: Day_of_Week, dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Day_of_Week.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change Day_of_Week from int to object\n",
    "accidents_df.Day_of_Week = accidents_df.Day_of_Week.replace([1,2,3,4,5,6,7], \n",
    "                                 ['Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3    0.454639\n",
       "6    0.306250\n",
       "4    0.123197\n",
       "5    0.077224\n",
       "1    0.035877\n",
       "2    0.002814\n",
       "Name: 1st_Road_Class, dtype: float64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['1st_Road_Class'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "A               0.454639\n",
       "Unclassified    0.306250\n",
       "B               0.123197\n",
       "C               0.077224\n",
       "Motorway        0.038691\n",
       "Name: 1st_Road_Class, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['1st_Road_Class'] = accidents_df['1st_Road_Class'].replace([1,2,3,4,5,6], \n",
    "                                                                        ['Motorway', 'A(M)', 'A', 'B', \n",
    "                                                                         'C', 'Unclassified'])\n",
    "# Combining A(M) (motorway sections of A roads) into the motorway category\n",
    "accidents_df['1st_Road_Class'] = accidents_df['1st_Road_Class'].replace('A(M)', 'Motorway')\n",
    "accidents_df['1st_Road_Class'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 6    0.745468\n",
       " 3    0.146459\n",
       " 1    0.067383\n",
       " 2    0.021767\n",
       " 7    0.010601\n",
       " 9    0.008321\n",
       "-1    0.000001\n",
       "Name: Road_Type, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Road_Type.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Single carriageway    0.753790\n",
       "Dual carriageway      0.146459\n",
       "Roundabout            0.067383\n",
       "One way street        0.021767\n",
       "Slip road             0.010601\n",
       "Name: Road_Type, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Road_Type to proper labels\n",
    "# Rather than dropping added unknown(9) and missing data(-1) to the mode ('Single carriageway')\n",
    "accidents_df.Road_Type = accidents_df.Road_Type.replace([1,2,3,6,7,9,-1], \n",
    "                               ['Roundabout', 'One way street', 'Dual carriageway', 'Single carriageway', \n",
    "                                'Slip road', 'Single carriageway', 'Single carriageway'])\n",
    "accidents_df.Road_Type.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 0    0.401522\n",
       " 3    0.316150\n",
       " 6    0.097717\n",
       " 1    0.087626\n",
       " 8    0.034965\n",
       " 9    0.025311\n",
       " 5    0.014574\n",
       " 2    0.012800\n",
       " 7    0.008289\n",
       "-1    0.001047\n",
       "Name: Junction_Detail, dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Junction_Detail.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Not at junction or within 20 meters    0.402569\n",
       "T or staggered junction                0.316150\n",
       "Crossroads                             0.097717\n",
       "Roundabout                             0.087626\n",
       "Private drive or entrance              0.034965\n",
       "Other junction                         0.025311\n",
       "Slip road                              0.014574\n",
       "Mini-roundabout                        0.012800\n",
       "More than 4 arms (not roundabout)      0.008289\n",
       "Name: Junction_Detail, dtype: float64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Junction_Detail to proper labels\n",
    "# Rather than dropping added missing data(-1) to the mode (Not at junction or within 20 meters)\n",
    "accidents_df.Junction_Detail = accidents_df.Junction_Detail.replace([0,1,2,3,5,6,7,8,9,-1], \n",
    "                                                                    ['Not at junction or within 20 meters', \n",
    "                                                                     'Roundabout', 'Mini-roundabout', \n",
    "                                                                     'T or staggered junction', 'Slip road', \n",
    "                                                                     'Crossroads', \n",
    "                                                                     'More than 4 arms (not roundabout)', \n",
    "                                                                     'Private drive or entrance', 'Other junction', \n",
    "                                                                     'Not at junction or within 20 meters'])\n",
    "accidents_df.Junction_Detail.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 4    0.482139\n",
       "-1    0.405216\n",
       " 2    0.105125\n",
       " 3    0.005081\n",
       " 1    0.001878\n",
       " 0    0.000561\n",
       "Name: Junction_Control, dtype: float64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Junction_Control.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Give way or uncontrolled               0.482139\n",
       "Data missing or out of range           0.405216\n",
       "Auto traffic signal                    0.105125\n",
       "Stop sign                              0.005081\n",
       "Authorised person                      0.001878\n",
       "Not at junction or within 20 metres    0.000561\n",
       "Name: Junction_Control, dtype: float64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Junction_Control to proper labels\n",
    "accidents_df.Junction_Control = accidents_df.Junction_Control.replace([0,1,2,3,4,-1], \n",
    "                                                                      ['Not at junction or within 20 metres', \n",
    "                                                                       'Authorised person', 'Auto traffic signal', \n",
    "                                                                       'Stop sign', 'Give way or uncontrolled', \n",
    "                                                                       'Data missing or out of range'])\n",
    "accidents_df.Junction_Control.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop Junction_Control because majority of the feature data is missing\n",
    "accidents_df = accidents_df.drop('Junction_Control', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 6    0.406821\n",
       "-1    0.406115\n",
       " 3    0.098274\n",
       " 5    0.042593\n",
       " 4    0.039237\n",
       " 1    0.006246\n",
       " 2    0.000714\n",
       "Name: 2nd_Road_Class, dtype: float64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df['2nd_Road_Class'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unclassified                           0.406821\n",
       "Not at junction or within 20 meters    0.406115\n",
       "A                                      0.098274\n",
       "C                                      0.042593\n",
       "B                                      0.039237\n",
       "Motorway                               0.006960\n",
       "Name: 2nd_Road_Class, dtype: float64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change 2nd_Road_Class to proper labels\n",
    "accidents_df['2nd_Road_Class'] = accidents_df['2nd_Road_Class'].replace([1,2,3,4,5,6,-1], \n",
    "                                                                        ['Motorway', 'A(M)', 'A', 'B', 'C', \n",
    "                                                                         'Unclassified', \n",
    "                                                                         'Not at junction or within 20 meters'])\n",
    "# Combining A(M) (motorway sections of A roads) into the motorway category\n",
    "accidents_df['2nd_Road_Class'] = accidents_df['2nd_Road_Class'].replace('A(M)', 'Motorway')\n",
    "accidents_df['2nd_Road_Class'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    0.728686\n",
       " 4    0.198545\n",
       " 6    0.051253\n",
       " 7    0.015485\n",
       " 5    0.006010\n",
       "-1    0.000020\n",
       "Name: Light_Conditions, dtype: float64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Light_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Daylight                       0.728706\n",
       "Darkness - lights lit          0.198545\n",
       "Darkness - no lighting         0.051253\n",
       "Darkness - lighting unknown    0.015485\n",
       "Darkness - lights unlit        0.006010\n",
       "Name: Light_Conditions, dtype: float64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Light_Conditions to proper labels\n",
    "# Add missing data(-1) to mode (Daylight)\n",
    "accidents_df.Light_Conditions = accidents_df.Light_Conditions.replace([1,4,5,6,7,-1], \n",
    "                                                      ['Daylight', \n",
    "                                                       'Darkness - lights lit', \n",
    "                                                       'Darkness - lights unlit', \n",
    "                                                       'Darkness - no lighting', \n",
    "                                                       'Darkness - lighting unknown', \n",
    "                                                       'Daylight'])\n",
    "\n",
    "accidents_df.Light_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    0.812651\n",
       " 2    0.110719\n",
       " 9    0.022883\n",
       " 8    0.016841\n",
       " 5    0.013507\n",
       " 4    0.012032\n",
       " 3    0.005047\n",
       " 7    0.004920\n",
       " 6    0.001378\n",
       "-1    0.000020\n",
       "Name: Weather_Conditions, dtype: float64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Weather_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Fine no high winds       0.835555\n",
       "Raining no high winds    0.110719\n",
       "Other                    0.016841\n",
       "Raining + high winds     0.013507\n",
       "Fine + high winds        0.012032\n",
       "Snowing no high winds    0.005047\n",
       "Fog or mist              0.004920\n",
       "Snowing + high winds     0.001378\n",
       "Name: Weather_Conditions, dtype: float64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Weather_Conditions to proper labels\n",
    "# Add unknown(9) and missing data(-1) to mode (Fine no high winds)\n",
    "accidents_df.Weather_Conditions = accidents_df.Weather_Conditions.replace([1,2,3,4,5,6,7,8,9,-1], \n",
    "                                                                ['Fine no high winds', \n",
    "                                                                 'Raining no high winds', \n",
    "                                                                 'Snowing no high winds', \n",
    "                                                                 'Fine + high winds', \n",
    "                                                                 'Raining + high winds', \n",
    "                                                                 'Snowing + high winds', \n",
    "                                                                 'Fog or mist', 'Other', 'Fine no high winds', \n",
    "                                                                 'Fine no high winds'])\n",
    "accidents_df.Weather_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    0.711846\n",
       " 2    0.262678\n",
       " 4    0.014380\n",
       "-1    0.005018\n",
       " 3    0.004780\n",
       " 5    0.001297\n",
       "Name: Road_Surface_Conditions, dtype: float64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Road_Surface_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dry                     0.716865\n",
       "Wet or damp             0.262678\n",
       "Frost or ice            0.014380\n",
       "Snow                    0.004780\n",
       "Flood over 3cm. deep    0.001297\n",
       "Name: Road_Surface_Conditions, dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Road_Surface_Conditions to proper labels\n",
    "# Add missing data(-1) to mode (Dry)\n",
    "accidents_df.Road_Surface_Conditions = accidents_df.Road_Surface_Conditions.replace([1,2,3,4,5,-1], \n",
    "                                                                                    ['Dry', \n",
    "                                                                                     'Wet or damp', \n",
    "                                                                                     'Snow', \n",
    "                                                                                     'Frost or ice', \n",
    "                                                                                     'Flood over 3cm. deep', \n",
    "                                                                                     'Dry'])\n",
    "accidents_df.Road_Surface_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 0    0.974209\n",
       " 4    0.010630\n",
       "-1    0.004074\n",
       " 7    0.002822\n",
       " 6    0.002568\n",
       " 5    0.002251\n",
       " 1    0.001822\n",
       " 3    0.001229\n",
       " 2    0.000395\n",
       "Name: Special_Conditions_at_Site, dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Special_Conditions_at_Site.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None                                          0.978284\n",
       "Roadworks                                     0.010630\n",
       "Mud                                           0.002822\n",
       "Oil or diesel                                 0.002568\n",
       "Road surface defective                        0.002251\n",
       "Auto traffic singal - out                     0.001822\n",
       "Road sign or marking defective or obscured    0.001229\n",
       "Auto signal part defective                    0.000395\n",
       "Name: Special_Conditions_at_Site, dtype: float64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Special_Conditions_at_Site to proper labels\n",
    "# Add missing data (-1) to mode (None)\n",
    "accidents_df.Special_Conditions_at_Site = \\\n",
    "accidents_df.Special_Conditions_at_Site.replace([0,1,2,3,4,5,6,7,-1],  \n",
    "                                                ['None', 'Auto traffic singal - out', \n",
    "                                                 'Auto signal part defective', \n",
    "                                                 'Road sign or marking defective or obscured', \n",
    "                                                 'Roadworks', 'Road surface defective', \n",
    "                                                 'Oil or diesel', 'Mud', 'None'])\n",
    "\n",
    "accidents_df.Special_Conditions_at_Site.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 0    0.979288\n",
       " 2    0.007258\n",
       " 7    0.004437\n",
       "-1    0.003928\n",
       " 6    0.002380\n",
       " 3    0.001413\n",
       " 1    0.001295\n",
       "Name: Carriageway_Hazards, dtype: float64"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Carriageway_Hazards.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None                                               0.983217\n",
       "Other object on road                               0.007258\n",
       "Any animal in carriageway (except ridden horse)    0.004437\n",
       "Pedestrian in carriageway - not injured            0.002380\n",
       "Previous accident                                  0.001413\n",
       "Vehicle load on road                               0.001295\n",
       "Name: Carriageway_Hazards, dtype: float64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Carriageway_Hazards to proper labels\n",
    "# Add missing data(-1) to mode (None)\n",
    "accidents_df.Carriageway_Hazards = \\\n",
    "accidents_df.Carriageway_Hazards.replace([0,1,2,3,6,7,-1],  \n",
    "                                                ['None', 'Vehicle load on road', \n",
    "                                                 'Other object on road', \n",
    "                                                 'Previous accident',  \n",
    "                                                 'Pedestrian in carriageway - not injured', \n",
    "                                                 'Any animal in carriageway (except ridden horse)', \n",
    "                                                 'None'])\n",
    "\n",
    "accidents_df.Carriageway_Hazards.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    0.656437\n",
       "2    0.343538\n",
       "3    0.000025\n",
       "Name: Urban_or_Rural_Area, dtype: float64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Urban_or_Rural_Area.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Urban    0.656462\n",
       "Rural    0.343538\n",
       "Name: Urban_or_Rural_Area, dtype: float64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Change Urban_or_Rural_Area to proper labels\n",
    "# Add unallocated (3) to mode (Urban)\n",
    "accidents_df.Urban_or_Rural_Area = accidents_df.Urban_or_Rural_Area.replace([1,2,3], ['Urban', 'Rural', 'Urban'])\n",
    "accidents_df.Urban_or_Rural_Area.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Rest of the rows with missing values are small enough to drop\n",
    "# Also are from features that wouldn't make sense to impute with another value\n",
    "accidents_df = accidents_df.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{dtype('int64'): Index(['Number_of_Vehicles', 'Number_of_Casualties', 'Hour'], dtype='object'),\n",
       " dtype('float64'): Index(['Location_Easting_OSGR', 'Location_Northing_OSGR', 'Longitude',\n",
       "        'Latitude', 'Speed_limit'],\n",
       "       dtype='object'),\n",
       " dtype('<M8[ns]'): Index(['Date'], dtype='object'),\n",
       " dtype('O'): Index(['Accident_Index', 'Accident_Severity', 'Day_of_Week', '1st_Road_Class',\n",
       "        'Road_Type', 'Junction_Detail', '2nd_Road_Class', 'Light_Conditions',\n",
       "        'Weather_Conditions', 'Road_Surface_Conditions',\n",
       "        'Special_Conditions_at_Site', 'Carriageway_Hazards',\n",
       "        'Urban_or_Rural_Area', 'LSOA_of_Accident_Location',\n",
       "        'Two_Hour_Groupings', 'Time_of_Day'],\n",
       "       dtype='object')}"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.columns.groupby(accidents_df.dtypes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Further Feature Selection \n",
    "Now will plot the categorical features to see if there are any that are heavily imbalanced that can be dropped."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x864 with 13 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(15,12))\n",
    "ax1 = plt.subplot(441)\n",
    "sns.countplot(accidents_df['Accident_Severity'])\n",
    "ax2 = plt.subplot(442)\n",
    "sns.countplot(accidents_df['Day_of_Week'])\n",
    "ax3 = plt.subplot(443)\n",
    "sns.countplot(accidents_df['1st_Road_Class'])\n",
    "ax4 = plt.subplot(444)\n",
    "sns.countplot(accidents_df['Road_Type'])\n",
    "ax5 = plt.subplot(445)\n",
    "sns.countplot(accidents_df['Junction_Detail'])\n",
    "ax6 = plt.subplot(446)\n",
    "sns.countplot(accidents_df['2nd_Road_Class'])\n",
    "ax7 = plt.subplot(447)\n",
    "sns.countplot(accidents_df['Light_Conditions'])\n",
    "ax8 = plt.subplot(448)\n",
    "sns.countplot(accidents_df['Weather_Conditions'])\n",
    "ax9 = plt.subplot(449)\n",
    "sns.countplot(accidents_df['Road_Surface_Conditions'])\n",
    "ax10 = plt.subplot(4,4,10)\n",
    "sns.countplot(accidents_df['Special_Conditions_at_Site'])\n",
    "ax11 = plt.subplot(4,4,11)\n",
    "sns.countplot(accidents_df['Carriageway_Hazards'])\n",
    "ax12 = plt.subplot(4,4,12)\n",
    "sns.countplot(accidents_df['Urban_or_Rural_Area'])\n",
    "ax13 = plt.subplot(4,4,13)\n",
    "sns.countplot(accidents_df['Time_of_Day'])\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Drop `Special_Conditions_at_Site` and `Carriageway_Hazards` due to one dominant label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df = accidents_df.drop(columns=['Special_Conditions_at_Site', 'Carriageway_Hazards'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also possibly change `Light_Conditions`, `Weather_Conditions`, and `Road_Surface_Conditions` to binary (might have to drop one since they are possibly collinear to each other)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Daylight                       0.727956\n",
       "Darkness - lights lit          0.200164\n",
       "Darkness - no lighting         0.049813\n",
       "Darkness - lighting unknown    0.016103\n",
       "Darkness - lights unlit        0.005964\n",
       "Name: Light_Conditions, dtype: float64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Light_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The below functions will be used to combine minority classes for features that are imbalanced. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "def was_daylight(condition):\n",
    "    if condition == 'Daylight':\n",
    "        return 'Yes'\n",
    "    else:\n",
    "        return 'No'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df['Was_Daylight'] = accidents_df['Light_Conditions'].apply(was_daylight)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Fine no high winds       0.840448\n",
       "Raining no high winds    0.107624\n",
       "Other                    0.016843\n",
       "Raining + high winds     0.012683\n",
       "Fine + high winds        0.011700\n",
       "Fog or mist              0.004983\n",
       "Snowing no high winds    0.004566\n",
       "Snowing + high winds     0.001152\n",
       "Name: Weather_Conditions, dtype: float64"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Weather_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "def was_bad_weather(condition):\n",
    "    if condition == 'Fine no high winds':\n",
    "        return 'No'\n",
    "    else:\n",
    "        return 'Yes'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df['Was_Bad_Weather'] = accidents_df['Weather_Conditions'].apply(was_bad_weather)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dry                     0.724585\n",
       "Wet or damp             0.256415\n",
       "Frost or ice            0.013635\n",
       "Snow                    0.004179\n",
       "Flood over 3cm. deep    0.001186\n",
       "Name: Road_Surface_Conditions, dtype: float64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.Road_Surface_Conditions.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "def was_road_dry(condition):\n",
    "    if condition == 'Dry':\n",
    "        return 'Yes'\n",
    "    else:\n",
    "        return 'No'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df['Was_Road_Dry'] = accidents_df['Road_Surface_Conditions'].apply(was_road_dry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Location_Easting_OSGR</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>457684.471736</td>\n",
       "      <td>88611.341073</td>\n",
       "      <td>90185.000000</td>\n",
       "      <td>398206.750000</td>\n",
       "      <td>459530.500000</td>\n",
       "      <td>528820.000000</td>\n",
       "      <td>655391.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Location_Northing_OSGR</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>263312.280412</td>\n",
       "      <td>122443.236168</td>\n",
       "      <td>10290.000000</td>\n",
       "      <td>174840.000000</td>\n",
       "      <td>217436.000000</td>\n",
       "      <td>366841.250000</td>\n",
       "      <td>656404.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Longitude</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>-1.164207</td>\n",
       "      <td>1.282775</td>\n",
       "      <td>-6.317417</td>\n",
       "      <td>-2.027918</td>\n",
       "      <td>-1.121212</td>\n",
       "      <td>-0.142525</td>\n",
       "      <td>1.759641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Latitude</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>52.257879</td>\n",
       "      <td>1.105293</td>\n",
       "      <td>49.912941</td>\n",
       "      <td>51.460000</td>\n",
       "      <td>51.842163</td>\n",
       "      <td>53.193894</td>\n",
       "      <td>55.800956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Number_of_Vehicles</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>1.843288</td>\n",
       "      <td>0.712747</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>67.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Number_of_Casualties</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>1.327304</td>\n",
       "      <td>0.804570</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>93.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Speed_limit</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>37.858172</td>\n",
       "      <td>13.849555</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>70.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hour</th>\n",
       "      <td>648208.0</td>\n",
       "      <td>13.549365</td>\n",
       "      <td>5.101049</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>23.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           count           mean            std           min  \\\n",
       "Location_Easting_OSGR   648208.0  457684.471736   88611.341073  90185.000000   \n",
       "Location_Northing_OSGR  648208.0  263312.280412  122443.236168  10290.000000   \n",
       "Longitude               648208.0      -1.164207       1.282775     -6.317417   \n",
       "Latitude                648208.0      52.257879       1.105293     49.912941   \n",
       "Number_of_Vehicles      648208.0       1.843288       0.712747      1.000000   \n",
       "Number_of_Casualties    648208.0       1.327304       0.804570      1.000000   \n",
       "Speed_limit             648208.0      37.858172      13.849555      0.000000   \n",
       "Hour                    648208.0      13.549365       5.101049      0.000000   \n",
       "\n",
       "                                  25%            50%            75%  \\\n",
       "Location_Easting_OSGR   398206.750000  459530.500000  528820.000000   \n",
       "Location_Northing_OSGR  174840.000000  217436.000000  366841.250000   \n",
       "Longitude                   -2.027918      -1.121212      -0.142525   \n",
       "Latitude                    51.460000      51.842163      53.193894   \n",
       "Number_of_Vehicles           1.000000       2.000000       2.000000   \n",
       "Number_of_Casualties         1.000000       1.000000       1.000000   \n",
       "Speed_limit                 30.000000      30.000000      40.000000   \n",
       "Hour                        10.000000      14.000000      17.000000   \n",
       "\n",
       "                                  max  \n",
       "Location_Easting_OSGR   655391.000000  \n",
       "Location_Northing_OSGR  656404.000000  \n",
       "Longitude                    1.759641  \n",
       "Latitude                    55.800956  \n",
       "Number_of_Vehicles          67.000000  \n",
       "Number_of_Casualties        93.000000  \n",
       "Speed_limit                 70.000000  \n",
       "Hour                        23.000000  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df.describe().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Accident_Index</th>\n",
       "      <th>Location_Easting_OSGR</th>\n",
       "      <th>Location_Northing_OSGR</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Accident_Severity</th>\n",
       "      <th>Number_of_Vehicles</th>\n",
       "      <th>Number_of_Casualties</th>\n",
       "      <th>Date</th>\n",
       "      <th>Day_of_Week</th>\n",
       "      <th>...</th>\n",
       "      <th>Weather_Conditions</th>\n",
       "      <th>Road_Surface_Conditions</th>\n",
       "      <th>Urban_or_Rural_Area</th>\n",
       "      <th>LSOA_of_Accident_Location</th>\n",
       "      <th>Hour</th>\n",
       "      <th>Two_Hour_Groupings</th>\n",
       "      <th>Time_of_Day</th>\n",
       "      <th>Was_Daylight</th>\n",
       "      <th>Was_Bad_Weather</th>\n",
       "      <th>Was_Road_Dry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1947</th>\n",
       "      <td>201501CW11286</td>\n",
       "      <td>527880.0</td>\n",
       "      <td>180840.0</td>\n",
       "      <td>-0.158577</td>\n",
       "      <td>51.512023</td>\n",
       "      <td>Slight</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2015-08-12</td>\n",
       "      <td>Wednesday</td>\n",
       "      <td>...</td>\n",
       "      <td>Fine no high winds</td>\n",
       "      <td>Dry</td>\n",
       "      <td>Urban</td>\n",
       "      <td>E01004689</td>\n",
       "      <td>18</td>\n",
       "      <td>6pm-8pm</td>\n",
       "      <td>Rush hour</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22799</th>\n",
       "      <td>201501YE80704</td>\n",
       "      <td>529330.0</td>\n",
       "      <td>192460.0</td>\n",
       "      <td>-0.133423</td>\n",
       "      <td>51.616118</td>\n",
       "      <td>Slight</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2015-09-23</td>\n",
       "      <td>Wednesday</td>\n",
       "      <td>...</td>\n",
       "      <td>Fine no high winds</td>\n",
       "      <td>Dry</td>\n",
       "      <td>Urban</td>\n",
       "      <td>E01001534</td>\n",
       "      <td>15</td>\n",
       "      <td>2pm-4pm</td>\n",
       "      <td>Rush hour</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121311</th>\n",
       "      <td>2015521507211</td>\n",
       "      <td>361824.0</td>\n",
       "      <td>173552.0</td>\n",
       "      <td>-2.550877</td>\n",
       "      <td>51.459648</td>\n",
       "      <td>Slight</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2015-08-13</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>...</td>\n",
       "      <td>Fine no high winds</td>\n",
       "      <td>Dry</td>\n",
       "      <td>Urban</td>\n",
       "      <td>E01014681</td>\n",
       "      <td>21</td>\n",
       "      <td>8pm-10pm</td>\n",
       "      <td>Evening</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Accident_Index  Location_Easting_OSGR  Location_Northing_OSGR  \\\n",
       "1947    201501CW11286               527880.0                180840.0   \n",
       "22799   201501YE80704               529330.0                192460.0   \n",
       "121311  2015521507211               361824.0                173552.0   \n",
       "\n",
       "        Longitude   Latitude Accident_Severity  Number_of_Vehicles  \\\n",
       "1947    -0.158577  51.512023            Slight                   2   \n",
       "22799   -0.133423  51.616118            Slight                   1   \n",
       "121311  -2.550877  51.459648            Slight                   2   \n",
       "\n",
       "        Number_of_Casualties       Date Day_of_Week     ...       \\\n",
       "1947                       1 2015-08-12   Wednesday     ...        \n",
       "22799                      1 2015-09-23   Wednesday     ...        \n",
       "121311                     1 2015-08-13    Thursday     ...        \n",
       "\n",
       "        Weather_Conditions Road_Surface_Conditions  Urban_or_Rural_Area  \\\n",
       "1947    Fine no high winds                     Dry                Urban   \n",
       "22799   Fine no high winds                     Dry                Urban   \n",
       "121311  Fine no high winds                     Dry                Urban   \n",
       "\n",
       "       LSOA_of_Accident_Location Hour Two_Hour_Groupings Time_of_Day  \\\n",
       "1947                   E01004689   18            6pm-8pm   Rush hour   \n",
       "22799                  E01001534   15            2pm-4pm   Rush hour   \n",
       "121311                 E01014681   21           8pm-10pm     Evening   \n",
       "\n",
       "       Was_Daylight Was_Bad_Weather Was_Road_Dry  \n",
       "1947            Yes              No          Yes  \n",
       "22799           Yes              No          Yes  \n",
       "121311           No              No          Yes  \n",
       "\n",
       "[3 rows x 26 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents_df[accidents_df['Speed_limit']<=10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping the three accidents with 0 or 10 mph, as this may be data entry error\n",
    "accidents_df = accidents_df.drop(accidents_df[accidents_df.Speed_limit < 20].index)\n",
    "\n",
    "# Converting to integers\n",
    "accidents_df.Speed_limit = accidents_df.Speed_limit.astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x864 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "accidents_df.hist(figsize=(15,12))\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like need to log transform `Number_of_Vehicles` and `Number_of_Casualties`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Log transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "accidents_df['log_Number_of_Casualties'] = np.log(accidents_df.Number_of_Casualties)\n",
    "accidents_df['log_Number_of_Vehicles'] = np.log(accidents_df.Number_of_Vehicles)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "accidents_df[['log_Number_of_Casualties', 'log_Number_of_Vehicles']].hist()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Checking for collinearity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set the style of the visualization\n",
    "sns.set(style=\"white\")\n",
    "\n",
    "# Create a covariance matrix\n",
    "corr = accidents_df.corr()\n",
    "\n",
    "# Set up the matplotlib figure\n",
    "f, ax = plt.subplots(figsize=(11,9))\n",
    "\n",
    "# Generate a custom diverging colormap\n",
    "cmap = sns.diverging_palette(220, 10, as_cmap=True)\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, cmap=cmap, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5}, annot=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obivous colliearity with latitude and longitude with northing and easting but will leave those and LSOA for now for adding the other datasets later. Further cleaning will be done after cleaning and merging the population and traffic datasets with this accidents dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create csv of cleaned data to use later\n",
    "#accidents_df.to_csv('data/Accidents/Cleaned_accidents.csv')"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Traffic_accident_data.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "learn-env",
   "language": "python",
   "name": "learn-env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}