{
 "metadata": {
  "name": "",
  "signature": "sha256:feac633e1109801a7af35fcee14e424dd0ed0b5221807af52f96e0176106e770"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from __future__ import print_function, division\n",
      "\n",
      "import pandas as pd\n",
      "import thinkstats2\n",
      "import thinkplot\n",
      "\n",
      "import math\n",
      "import gzip\n",
      "\n",
      "%matplotlib inline\n",
      "formats = ['png', 'pdf']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def ReadFile(filename='soc-Slashdot0902.txt.gz', n=None):\n",
      "    \"\"\"Reads a compressed data file.\n",
      "\n",
      "    Args:\n",
      "        filename: string name of the file to read\n",
      "    \"\"\"\n",
      "    if filename.endswith('gz'):\n",
      "        fp = gzip.open(filename)\n",
      "    else:\n",
      "        fp = open(filename)\n",
      "\n",
      "    srcs = {}\n",
      "    for i, line in enumerate(fp):\n",
      "        if i == n:\n",
      "            break\n",
      "\n",
      "        if line.startswith('#'):\n",
      "            continue\n",
      "\n",
      "        src, dest = line.split()\n",
      "        srcs.setdefault(src, []).append(dest)\n",
      "\n",
      "    fp.close()\n",
      "\n",
      "    return srcs\n",
      "\n",
      "srcs = ReadFile()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def Summarize(srcs):\n",
      "    \"\"\"Computes the number of edges for each source.\"\"\"\n",
      "    lens = [len(t) for t in srcs.itervalues()]\n",
      "    mu, sigma2 = thinkstats2.MeanVar(lens)\n",
      "    print(mu, math.sqrt(sigma2))\n",
      "    return lens\n",
      "\n",
      "lens = Summarize(srcs)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "12.0914317767 37.6969124125\n"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pmf = thinkstats2.Pmf(lens, 'actual')\n",
      "cdf = pmf.MakeCdf()\n",
      "thinkplot.PrePlot(2)\n",
      "thinkplot.Plot(cdf.xs, cdf.ps, label=cdf.label)\n",
      "thinkplot.Config(xlabel='number of friends/foes', ylabel='CDF', xscale='log', \n",
      "                 xlim=[1, 3000],\n",
      "                 loc='lower right')\n",
      "thinkplot.Save('social1', formats=formats)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Writing social1.png\n",
        "Writing"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " social1.pdf\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "<matplotlib.figure.Figure at 0x7f96849628d0>"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def BiasPmf(pmf, label, invert=False):\n",
      "    \"\"\"Returns the Pmf with oversampling proportional to value.\n",
      "\n",
      "    If pmf is the distribution of true values, the result is the\n",
      "    distribution that would be seen if values are oversampled in\n",
      "    proportion to their values; for example, if you ask students\n",
      "    how big their classes are, large classes are oversampled in\n",
      "    proportion to their size.\n",
      "\n",
      "    If invert=True, computes in inverse operation; for example,\n",
      "    unbiasing a sample collected from students.\n",
      "\n",
      "    Args:\n",
      "      pmf: Pmf object.\n",
      "      invert: boolean\n",
      "\n",
      "     Returns:\n",
      "       Pmf object\n",
      "    \"\"\"\n",
      "    new_pmf = pmf.Copy()\n",
      "    new_pmf.label = label\n",
      "\n",
      "    for x, p in pmf.Items():\n",
      "        if invert:\n",
      "            new_pmf.Mult(x, 1.0/x)\n",
      "        else:\n",
      "            new_pmf.Mult(x, x)\n",
      "        \n",
      "    new_pmf.Normalize()\n",
      "    return new_pmf\n",
      "\n",
      "\n",
      "biased_pmf = BiasPmf(pmf, 'biased')\n",
      "cdf2 = biased_pmf.MakeCdf()\n",
      "\n",
      "pmf.Mean(), biased_pmf.Mean()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "(12.091431776749411, 129.61739823546282)"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "thinkplot.PrePlot(2)\n",
      "thinkplot.Plot(cdf.xs, cdf.ps, label=cdf.label)\n",
      "thinkplot.Plot(cdf2.xs, cdf2.ps, label=cdf2.label)\n",
      "thinkplot.Config(xlabel='number of friends/foes', ylabel='CDF', xscale='log', \n",
      "                 xlim=[1, 3000],\n",
      "                 loc='lower right')\n",
      "thinkplot.Save('social2', formats=formats)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Writing social2.png\n",
        "Writing"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " social2.pdf\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "<matplotlib.figure.Figure at 0x7f96813506d0>"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    }
   ],
   "metadata": {}
  }
 ]
}