{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Weather and Motor Vehicle Collisions Frequency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import datetime\n",
    "from datetime import date\n",
    "from dateutil.rrule import rrule, DAILY\n",
    "from __future__ import division\n",
    "import geoplotlib as glp\n",
    "from geoplotlib.utils import BoundingBox, DataAccessObject\n",
    "\n",
    "pd.set_option('display.max_columns', None)\n",
    "%matplotlib inline  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/masve/anaconda/lib/python2.7/site-packages/IPython/core/interactiveshell.py:2902: DtypeWarning: Columns (11) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n",
      "/Users/masve/anaconda/lib/python2.7/site-packages/IPython/core/interactiveshell.py:2902: DtypeWarning: Columns (30,31) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n"
     ]
    }
   ],
   "source": [
    "# Read and filter weather data\n",
    "weather = pd.read_csv('datasets/weather_data_nyc_kjfk_clean2.csv')\n",
    "incidents = pd.read_csv(\"datasets/NYPD_Motor_Vehicle_Collisions_weather4.csv\")\n",
    "weather['date'] = weather.Year.astype('str') +'/'+ weather.Month.astype('str') \\\n",
    "                  +'/'+ weather.Day.astype('str') +'/'+ weather.Hour.astype('str')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*** Frequency of measured weather conditions from 7/1/2012 to 3/1/2016, on hourly basis ***"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Blowing Snow': 3,\n",
       " 'Clear': 2142,\n",
       " 'Fog': 212,\n",
       " 'Haze': 107,\n",
       " 'Heavy Rain': 96,\n",
       " 'Heavy Snow': 11,\n",
       " 'Heavy Thunderstorms and Rain': 18,\n",
       " 'Ice Pellets': 2,\n",
       " 'Light Drizzle': 291,\n",
       " 'Light Freezing Drizzle': 8,\n",
       " 'Light Freezing Rain': 41,\n",
       " 'Light Ice Pellets': 11,\n",
       " 'Light Rain': 1759,\n",
       " 'Light Rain Showers': 1,\n",
       " 'Light Snow': 569,\n",
       " 'Light Thunderstorms and Rain': 49,\n",
       " 'Mist': 15,\n",
       " 'Mostly Cloudy': 10366,\n",
       " 'Overcast': 5823,\n",
       " 'Partly Cloudy': 4102,\n",
       " 'Patches of Fog': 7,\n",
       " 'Rain': 295,\n",
       " 'Scattered Clouds': 6060,\n",
       " 'Shallow Fog': 9,\n",
       " 'Snow': 33,\n",
       " 'Squalls': 2,\n",
       " 'Thunderstorm': 27,\n",
       " 'Thunderstorms and Rain': 15,\n",
       " 'Thunderstorms with Small Hail': 1,\n",
       " 'Unknown': 7}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initialize condition dictionary\n",
    "conditions = list(weather.Conditions.unique())\n",
    "condic = {}\n",
    "for cond in conditions:\n",
    "    condic[cond] = 0;\n",
    "\n",
    "# Fill condic with every occurrence of incident in given weather condition\n",
    "for d in weather.date.unique():\n",
    "    condi = weather[weather.date == d]['Conditions'].iloc[0]\n",
    "    condic[condi] += 1\n",
    "\n",
    "condic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Blowing Snow': 884,\n",
       " 'Clear': 1043732,\n",
       " 'Fog': 116518,\n",
       " 'Haze': 112200,\n",
       " 'Heavy Rain': 82246,\n",
       " 'Heavy Snow': 5644,\n",
       " 'Heavy Thunderstorms and Rain': 21522,\n",
       " 'Ice Pellets': 1496,\n",
       " 'Light Drizzle': 229636,\n",
       " 'Light Freezing Drizzle': 4760,\n",
       " 'Light Freezing Rain': 30770,\n",
       " 'Light Ice Pellets': 13090,\n",
       " 'Light Rain': 1548632,\n",
       " 'Light Rain Showers': 1326,\n",
       " 'Light Snow': 479026,\n",
       " 'Light Thunderstorms and Rain': 52258,\n",
       " 'Mist': 9826,\n",
       " 'Mostly Cloudy': 8756700,\n",
       " 'Overcast': 3677100,\n",
       " 'Partly Cloudy': 2912236,\n",
       " 'Patches of Fog': 2346,\n",
       " 'Rain': 267410,\n",
       " 'Scattered Clouds': 5272788,\n",
       " 'Shallow Fog': 3366,\n",
       " 'Snow': 45628,\n",
       " 'Squalls': 1564,\n",
       " 'Thunderstorm': 29104,\n",
       " 'Thunderstorms and Rain': 8670,\n",
       " 'Thunderstorms with Small Hail': 1190,\n",
       " 'Unknown': 6834}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get frequency of collision per hour of specific weather condition\n",
    "conditionCount = {}\n",
    "for c in incidents.Conditions.unique():\n",
    "    if (pd.notnull(c)):\n",
    "        mask = ((incidents.Conditions == c))\n",
    "        filtered_incidents = incidents[mask]\n",
    "        conditionCount[c] = filtered_incidents.size\n",
    "    \n",
    "conditionCount"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Blowing Snow': 34.882029379408529,\n",
       " 'Clear': 57.681994068578057,\n",
       " 'Fog': 65.062072577957025,\n",
       " 'Haze': 124.13087378006628,\n",
       " 'Heavy Rain': 101.41782339998706,\n",
       " 'Heavy Snow': 60.738638569739337,\n",
       " 'Heavy Thunderstorms and Rain': 141.54054228952307,\n",
       " 'Ice Pellets': 88.546689963113963,\n",
       " 'Light Drizzle': 93.41523649029547,\n",
       " 'Light Freezing Drizzle': 70.434867016113373,\n",
       " 'Light Freezing Rain': 88.841191149244054,\n",
       " 'Light Ice Pellets': 140.86973403222675,\n",
       " 'Light Rain': 104.22049290505748,\n",
       " 'Light Rain Showers': 156.96913220733839,\n",
       " 'Light Snow': 99.659235891541186,\n",
       " 'Light Thunderstorms and Rain': 126.24885201605393,\n",
       " 'Mist': 77.545434543454348,\n",
       " 'Mostly Cloudy': 100.0,\n",
       " 'Overcast': 74.753130372588132,\n",
       " 'Partly Cloudy': 84.043018728121126,\n",
       " 'Patches of Fog': 39.673516931525086,\n",
       " 'Rain': 107.30658190445166,\n",
       " 'Scattered Clouds': 103.00028332477977,\n",
       " 'Shallow Fog': 44.273344981556981,\n",
       " 'Snow': 163.67721478030157,\n",
       " 'Squalls': 92.571539506891867,\n",
       " 'Thunderstorm': 127.60263738792179,\n",
       " 'Thunderstorms and Rain': 68.422442244224428,\n",
       " 'Thunderstorms with Small Hail': 140.86973403222675,\n",
       " 'Unknown': 115.57067975705134}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate ratios\n",
    "ratios = {}\n",
    "\n",
    "for k,v in conditionCount.iteritems():\n",
    "    conditionCountValue = conditionCount[k]\n",
    "    weatherConditionCountValue = condic[k]\n",
    "    ratio =  conditionCountValue / weatherConditionCountValue\n",
    "    ratios[k] = ratio\n",
    "    #print \"%s: %s\" % (k, ratio)\n",
    "\n",
    "# Normalize on Mostly Cloudy (Most common weather condition)\n",
    "reference = ratios[\"Mostly Cloudy\"]\n",
    "for k in ratios:\n",
    "    ratios[k] = (ratios[k]/reference)*100\n",
    "\n",
    "ratios"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10d437090>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAngAAAHaCAYAAACTqzRCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x104065b90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot ratios\n",
    "df = pd.DataFrame(pd.Series(ratios, name=\"Collision Frequency (Normalized)\").sort_values())\n",
    "df.plot(kind='barh', figsize=(8,8))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Export to json for d3 viz\n",
    "from collections import OrderedDict\n",
    "import json\n",
    "with open('datasets/freq_weather2.json', 'w') as fp:\n",
    "    json.dump(OrderedDict(sorted(ratios.items(), key=lambda x: x[1], reverse=True)), fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}