jmMAGALLANES/SEGREGATION-1D.ipynb

## SEGREGATION-1D.ipynb
{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "<font color=\"red\">SEGREGATION in 1D</font>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Annetta BURGER & Jos\u00e9 Manuel MAGALLANES**  \n",
      "PhD Program in Computational Social Science  \n",
      "George Mason University  \n",
      "SPRING 2014"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "LIBRARIES TO IMPORT"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import random  \n",
      "import copy\n",
      "import matplotlib.pyplot as plot\n",
      "from random import choice\n",
      "from collections import Counter\n",
      "import numpy as np\n",
      "%matplotlib inline"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "THE *CLASS* 'AGENT'"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Create an Agent Class with a set of functions.\n",
      "class Agent:\n",
      "    def __init__(self,code):\n",
      "        self.code=code\n",
      "        self.group = self.set_group()               # the agent's type\n",
      "        self.tolerance = -1                         # sets the agent's tolerance level\n",
      "        self.mood = UNHAPPY                           # agent starts happy\n",
      "        \n",
      "    def stats(self):                                # display the agent's stats\n",
      "        return self.code, self.group, self.tolerance, self.mood\n",
      "        \n",
      "    def set_group(self):\n",
      "        if self.code%2==0: return 'red'\n",
      "        else: return 'blue'\n",
      "        \n",
      "    def get_neighbors(self,list_of_agents):\n",
      "        leftIndex=(self.code-1)%len(list_of_agents)\n",
      "        rigthIndex=(self.code+1)%len(list_of_agents)\n",
      "        return list_of_agents[leftIndex],list_of_agents[rigthIndex]\n",
      "    \n",
      "    def switch_place_with(self,otherAgent):\n",
      "        temp_self=copy.deepcopy(self)\n",
      "        self=otherAgent    \n",
      "        otherAgent=temp_self    \n",
      "        self.mood=otherAgent.mood=HAPPY\n",
      "        del temp_self\n",
      "        \n",
      "    def get_mood(self,leftNeighbor,rigthNeighbor):\n",
      "        if self.tolerance==0 and (leftNeighbor.group!=self.group and rigthNeighbor.group!=self.group):\n",
      "            return HAPPY\n",
      "        if self.tolerance==1 and (\n",
      "            (leftNeighbor.group==self.group and rigthNeighbor.group!=self.group) or \n",
      "            (leftNeighbor.group!=self.group and rigthNeighbor.group==self.group)):\n",
      "            return HAPPY\n",
      "        if self.tolerance==2 and (leftNeighbor.group==self.group and rigthNeighbor.group==self.group):\n",
      "            return HAPPY\n",
      "        return UNHAPPY"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "SETTING FUNCTIONS"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def set_neighborhood(number_of_agents):\n",
      "    list_of_agents=[Agent(i) for i in range(number_of_agents)]\n",
      "    random.shuffle(list_of_agents)\n",
      "    for i in range(len(list_of_agents)):\n",
      "        list_of_agents[i].code=i\n",
      "    return list_of_agents\n",
      "\n",
      "def change_tolerance(list_of_agents,intolerance_value):\n",
      "    for agent in list_of_agents:\n",
      "        agent.tolerance=intolerance_value\n",
      "        \n",
      "def set_tolerance(list_of_agents, population,proportion_of_indifferent_agents):\n",
      "    \n",
      "    ZERO =random.sample(list_of_agents,int(population*proportion_of_indifferent_agents))\n",
      "    change_tolerance(ZERO,0)\n",
      "    ForONEandTWO=[agent for agent in list_of_agents if not agent in ZERO]\n",
      "    ONE=random.sample(ForONEandTWO,int(population/3))\n",
      "    change_tolerance(ONE,1)\n",
      "    TWO=[agent for agent in ForONEandTWO if not agent in ONE]\n",
      "    change_tolerance(TWO,2)\n",
      "    print len(ZERO),len(ONE),len(TWO)\n",
      "\n",
      "    \n",
      "def set_mood(list_of_agents):\n",
      "    \n",
      "    for agent in list_of_agents:\n",
      "        left,rigth=agent.get_neighbors(list_of_agents)\n",
      "        agent.mood=agent.get_mood(left,rigth)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "COMPUTING FUNCTIONS"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def get_average_neighbor_the_same(list_of_agents):\n",
      "    neighboorhoods = 0.0\n",
      "    for agent in list_of_agents:\n",
      "        leftNeighbor,rigthNeighbor=agent.get_neighbors(list_of_agents)\n",
      "        if (agent.group==leftNeighbor.group and agent.group==rigthNeighbor.group):\n",
      "            neighboorhoods += 1\n",
      "    return neighboorhoods/POPULATION\n",
      "\n",
      "def get_UnhappyAgents(list_of_agents,color):\n",
      "    list_of_unhappyAgents=[a for a in list_of_agents if a.mood==UNHAPPY and a.group==color]\n",
      "    return list_of_unhappyAgents\n",
      "    \n",
      "def switchUnhappy(list_of_agents):\n",
      "    go=True\n",
      "    timeTaking=0\n",
      "    while go:\n",
      "        \n",
      "        unhappyBlue_all=get_UnhappyAgents(list_of_agents,'blue')\n",
      "        unhappyRed_all=get_UnhappyAgents(list_of_agents,'red')\n",
      "        \n",
      "        if len(unhappyBlue_all)*len(unhappyRed_all)!=0:\n",
      "            allPairs_RedBlue=((blue,red) for blue in unhappyBlue_all for red in unhappyRed_all)\n",
      "            \n",
      "            for unhappyBlue,unhappyRed in allPairs_RedBlue:\n",
      "                timeTaking+=1\n",
      "                leftB,rigthB=unhappyBlue.get_neighbors(list_of_agents)\n",
      "                leftR,rigthR=unhappyRed.get_neighbors(list_of_agents)\n",
      "                if (unhappyBlue.get_mood(leftR,rigthR)==HAPPY and unhappyRed.get_mood(leftB,rigthB)==HAPPY):\n",
      "                    unhappyBlue.switch_place_with(unhappyRed)\n",
      "        else:\n",
      "            go=False\n",
      "            return timeTaking,get_average_neighbor_the_same(list_of_agents)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "MAIN FUNCTION"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def main_segregate(population, proportion):\n",
      "    hood = set_neighborhood(population)\n",
      "    set_tolerance(hood, population,proportion)\n",
      "    set_mood(hood)\n",
      "    return switchUnhappy(hood)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "EXPERIMENT FUNCTION"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def Experiments(population, initial_proportion,final_proportion):\n",
      "    timeTakenHistory=[]\n",
      "    historyOfMeans=[]\n",
      "    proportions=np.arange(initial_proportion,final_proportion,0.05)\n",
      "    for p in proportions:\n",
      "        time,average=main_segregate(population, p)\n",
      "        timeTakenHistory.append(time)\n",
      "        historyOfMeans.append(average)\n",
      "    return timeTakenHistory,historyOfMeans"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "RUNNING EXPERIMENTS"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "UNHAPPY = 'UNhappy'\n",
      "HAPPY = 'HAPPY'\n",
      "POPULATION=100\n",
      "INITIAL_PROPORTION=0.05\n",
      "FINAL_PROPORTION=0.65\n",
      "\n",
      "# to experiment and retrieve a list of run results:\n",
      "result=Experiments(POPULATION, INITIAL_PROPORTION, FINAL_PROPORTION)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "5 33 62\n",
        "10 33 57\n",
        "15"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 33 52\n",
        "20 33 47\n",
        "25"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 33 42\n",
        "30 33 37\n",
        "35"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 33 32\n",
        "40 33 27\n",
        "45"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 33 22\n",
        "50 33 17\n",
        "55"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 33 12\n",
        "60 33 7\n"
       ]
      }
     ],
     "prompt_number": 22
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "PLOTTING RESULTS"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plot.figure(figsize=(14,6))\n",
      "plot.plot(result[1])\n",
      "plot.xticks(range(0,12),np.arange(INITIAL_PROPORTION,FINAL_PROPORTION,0.05))\n",
      "plot.xlabel(\"Value of p\")\n",
      "plot.ylabel(\"Proportion of same type neighbors\")\n",
      "plot.title(\"Proportion of same type neighbors at different p values\",fontsize=14)\n",
      "plot.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA0MAAAGKCAYAAADOuJq6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10bf48410>"
       ]
      }
     ],
     "prompt_number": 24
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}