Haversine performance

Heversine.ipynb

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from numpy import sin, cos, pi, sqrt, arcsin, random"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def get_distance(lat, lon, pcode_lat, pcode_lon):\n",
      "    \"\"\"\n",
      "    Find the distance between `(lat,lon)` and the reference point\n",
      "    `(pcode_lat,pcode_lon)`.\n",
      "    \"\"\"\n",
      "    RAD_FACTOR = pi / 180.0  # degrees to radians for trig functions\n",
      "    lat_in_rad = lat * RAD_FACTOR\n",
      "    lon_in_rad = lon * RAD_FACTOR\n",
      "    pcode_lat_in_rad = pcode_lat * RAD_FACTOR\n",
      "    pcode_lon_in_rad = pcode_lon * RAD_FACTOR\n",
      "    delta_lon = lon_in_rad - pcode_lon_in_rad\n",
      "    delta_lat = lat_in_rad - pcode_lat_in_rad\n",
      "    # Next two lines is the Haversine formula\n",
      "    inverse_angle = (sin(delta_lat / 2) ** 2 + cos(pcode_lat_in_rad) *\n",
      "                     cos(lat_in_rad) * sin(delta_lon / 2) ** 2)\n",
      "    haversine_angle = 2 * asin(sqrt(inverse_angle))\n",
      "    EARTH_RADIUS = 6367  # kilometers\n",
      "    return haversine_angle * EARTH_RADIUS\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "reference = (52.3905927,4.8412508)\n",
      "points = random.randn(400000, 2) * 0.01\n",
      "points[:, 0] = points[:, 0] + reference[0]\n",
      "points[:, 1] = points[:, 1] + reference[1]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 35
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def iterate_distance():\n",
      "    d = []\n",
      "    for p in points:\n",
      "        d.append(get_distance(p[0], p[1], reference[0], reference[1]))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 27
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%timeit iterate_distance()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "1 loops, best of 3: 3.62 s per loop\n"
       ]
      }
     ],
     "prompt_number": 36
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%timeit [get_distance(p[0], p[1], reference[0], reference[1]) for p in points]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "1 loops, best of 3: 3.58 s per loop\n"
       ]
      }
     ],
     "prompt_number": 37
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%timeit get_distance(points[:, 0], points[:, 1], reference[0], reference[1])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 22.7 ms per loop\n"
       ]
      }
     ],
     "prompt_number": 38
    }
   ],
   "metadata": {}
  }
 ]
}