global_tdrive_stats.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Global Statistics for T-Drive data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import tdrive\n",
    "import numpy as np\n",
    "from collections import Counter\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading 10336 taxis...\n",
      "  1000\n",
      "  2000\n",
      "  3000\n",
      "  4000\n",
      "  5000\n",
      "  6000\n",
      "  7000\n",
      "  8000\n",
      "  9000\n",
      "  10000\n",
      "Done.\n"
     ]
    }
   ],
   "source": [
    "dir_path = r'C:\\dev\\stanford\\project\\data'\n",
    "taxis = tdrive.load_taxis(dir_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Min and Max for lat/lon and time stamps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min/Max\n",
      "   time stamps : [1201987844.000000, 1202521159.000000]\n",
      "   latitude    : [0.000000, 255.300000]\n",
      "   longitude   : [0.000000, 96.067670]\n",
      "   duration of t-drive data : 148.143056 hours\n"
     ]
    }
   ],
   "source": [
    "min_ts, min_lat, min_lon = [float('inf')]*3\n",
    "max_ts, max_lat, max_lon = [float('-inf')]*3\n",
    "\n",
    "\n",
    "for taxi in taxis:\n",
    "    min_ts = min(min_ts, min(taxi.ts))\n",
    "    max_ts = max(max_ts, max(taxi.ts))\n",
    "    \n",
    "    min_lat = min(min_lat, min(taxi.lat))\n",
    "    max_lat = max(max_lat, max(taxi.lat))\n",
    "    \n",
    "    min_lon = min(min_lon, min(taxi.lon))\n",
    "    max_lon = max(max_lon, max(taxi.lon))\n",
    "\n",
    "\n",
    "print('Min/Max')\n",
    "print('   timestamps  : [%f, %f]' % (min_ts, max_ts))\n",
    "print('   latitude    : [%f, %f]' % (min_lat, max_lat))\n",
    "print('   longitude   : [%f, %f]' % (min_lon, max_lon))\n",
    "print('   duration    : %f hours' % ((max_ts - min_ts) / 3600))\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Histograms of all of the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZsAAAEWCAYAAACwtjr+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxcbMzEr3/wHfWQda6whjUgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x77963400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCpKkjqEgSeoYCpKkjqEgSeoYCpKkzn8AUiivCPbMrboAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x75a6b8d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEWCAYAAABxMXBSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVpI4IiKilSSOiIho5f8Dr9nJpCx0ylIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x77607898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ts_counts = Counter()\n",
    "lat_counts = Counter()\n",
    "lon_counts = Counter()\n",
    "\n",
    "for taxi in taxis:\n",
    "    ts_counts.update([int(round(x / 3600.0)) for x in taxi.ts])\n",
    "    lat_counts.update([int(round(x)) for x in taxi.lat])\n",
    "    lon_counts.update([int(round(x)) for x in taxi.lon])\n",
    "    \n",
    "X = np.array(ts_counts.keys())\n",
    "X = X - min(X)\n",
    "Y = ts_counts.values()\n",
    "plt.bar(X, Y)\n",
    "plt.title('Histogram of Timestamps')\n",
    "plt.xlabel('hours')\n",
    "plt.ylabel('num data points')\n",
    "plt.show()\n",
    "\n",
    "X = lat_counts.keys()\n",
    "Y = lat_counts.values()\n",
    "plt.bar(X, Y)\n",
    "plt.title('Histogram of Latitude')\n",
    "plt.xlabel('latitude')\n",
    "plt.ylabel('num data points')\n",
    "plt.show()\n",
    "\n",
    "X = lon_counts.keys()\n",
    "Y = lon_counts.values()\n",
    "plt.bar(X, Y)\n",
    "plt.title('Histogram of Longitudes')\n",
    "plt.xlabel('longitude')\n",
    "plt.ylabel('num data points')\n",
    "plt.show()"
   ]
  }
 ],
 "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": "2.7.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}