wordshift_demo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline\n",
    "import matplotlib.pylab as plt\n",
    "\n",
    "from wordshifts import WordShifts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1183891\n"
     ]
    }
   ],
   "source": [
    "# Load demo data\n",
    "sampleText = ''\n",
    "with open('wordshifts/data/moby-dick.txt','r') as fl:\n",
    "    for line in fl:\n",
    "        sampleText += line.strip()\n",
    "print len(sampleText)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'comparison': 5.907102040816327, 'difference': 0.334428773489595, 'reference': 5.572673267326732, 'word_shift': {'all': {'case': '+d', 'shift': -2.808192640266889, 'value_difference': 0.6473267326732675, 'frequency_difference': -0.014507981410385937}, 'shot': {'case': '-d', 'shift': 18.193714183053633, 'value_difference': -3.0726732673267323, 'frequency_difference': -0.019801980198019802}, 'money': {'case': '+u', 'shift': 4.216331699583332, 'value_difference': 1.7273267326732675, 'frequency_difference': 0.00816326530612245}, 'magnetic': {'case': '+u', 'shift': 0.8388677822568925, 'value_difference': 0.6873267326732675, 'frequency_difference': 0.004081632653061225}, 'soon': {'case': '+d', 'shift': -1.335216610898637, 'value_difference': 0.7673267326732676, 'frequency_difference': -0.005819357445948676}, 'valuable': {'case': '+u', 'shift': 1.8884796457952633, 'value_difference': 1.5473267326732678, 'frequency_difference': 0.004081632653061225}, 'not': {'case': '-u', 'shift': -11.65176775500695, 'value_difference': -1.7126732673267324, 'frequency_difference': 0.022752071125479896}, 'sleep': {'case': '+u', 'shift': 3.8745976044778168, 'value_difference': 1.5873267326732678, 'frequency_difference': 0.00816326530612245}, 'hell': {'case': '-d', 'shift': 9.925816035099604, 'value_difference': -3.352673267326732, 'frequency_difference': -0.009900990099009901}, 'sung': {'case': '+u', 'shift': 0.985325251587828, 'value_difference': 0.8073267326732676, 'frequency_difference': 0.004081632653061225}, 'touches': {'case': '+u', 'shift': 0.985325251587828, 'value_difference': 0.8073267326732676, 'frequency_difference': 0.004081632653061225}, 'considerable': {'case': '+u', 'shift': 0.8388677822568925, 'value_difference': 0.6873267326732675, 'frequency_difference': 0.004081632653061225}, 'better': {'case': '+u', 'shift': 1.7420221764643276, 'value_difference': 1.4273267326732677, 'frequency_difference': 0.004081632653061225}, 'tears': {'case': '-u', 'shift': -3.7989520798893945, 'value_difference': -3.1126732673267323, 'frequency_difference': 0.004081632653061225}, 'pride': {'case': '+d', 'shift': -2.863836237166239, 'value_difference': 0.9673267326732677, 'frequency_difference': -0.009900990099009901}, 'thousands': {'case': '+u', 'shift': 1.384820625851914, 'value_difference': 0.5673267326732674, 'frequency_difference': 0.00816326530612245}, 'hats': {'case': '+u', 'shift': 1.0219396189205623, 'value_difference': 0.8373267326732678, 'frequency_difference': 0.004081632653061225}, 'sinking': {'case': '-d', 'shift': 6.195500789022067, 'value_difference': -2.0926732673267323, 'frequency_difference': -0.009900990099009901}, 'lonely': {'case': '-d', 'shift': 8.031052735504666, 'value_difference': -2.7126732673267324, 'frequency_difference': -0.009900990099009901}, 'seized': {'case': '-d', 'shift': 6.313923495246751, 'value_difference': -2.1326732673267323, 'frequency_difference': -0.009900990099009901}, 'very': {'case': '+u', 'shift': 0.668000734704135, 'value_difference': 0.5473267326732678, 'frequency_difference': 0.004081632653061225}, 'myself': {'case': '+u', 'shift': 0.8876869387005377, 'value_difference': 0.7273267326732675, 'frequency_difference': 0.004081632653061225}, 'sight': {'case': '+d', 'shift': -0.31557294240488004, 'value_difference': 0.6073267326732674, 'frequency_difference': -0.0017377247928874514}, 'touching': {'case': '+u', 'shift': 1.2660354011387882, 'value_difference': 1.037326732673268, 'frequency_difference': 0.004081632653061225}, 'altogether': {'case': '+u', 'shift': 1.0097348298096511, 'value_difference': 0.8273267326732681, 'frequency_difference': 0.004081632653061225}, 'cook': {'case': '+u', 'shift': 1.302649768471521, 'value_difference': 1.0673267326732674, 'frequency_difference': 0.004081632653061225}, 'world': {'case': '+u', 'shift': 1.1561922991405857, 'value_difference': 0.9473267326732673, 'frequency_difference': 0.004081632653061225}, 'bringing': {'case': '+u', 'shift': 0.5703624218168447, 'value_difference': 0.46732673267326774, 'frequency_difference': 0.004081632653061225}, 'day': {'case': '+d', 'shift': -1.9756659404811112, 'value_difference': 0.6673267326732679, 'frequency_difference': -0.009900990099009901}, 'charm': {'case': '+u', 'shift': 1.8152509111297948, 'value_difference': 1.4873267326732673, 'frequency_difference': 0.004081632653061225}, 'heaven': {'case': '+d', 'shift': -6.771785542580802, 'value_difference': 2.287326732673268, 'frequency_difference': -0.009900990099009901}, 'magic': {'case': '+u', 'shift': 2.995852788492204, 'value_difference': 1.2273267326732675, 'frequency_difference': 0.00816326530612245}, 'like': {'case': '+d', 'shift': -3.7224618752585967, 'value_difference': 1.6473267326732675, 'frequency_difference': -0.007557082238836129}, 'university': {'case': '+u', 'shift': 1.4246976595806347, 'value_difference': 1.167326732673268, 'frequency_difference': 0.004081632653061225}, 'brother': {'case': '+u', 'shift': 2.0105275369043754, 'value_difference': 1.6473267326732675, 'frequency_difference': 0.004081632653061225}, 'pistol': {'case': '-u', 'shift': -2.41981091035642, 'value_difference': -1.9826732673267324, 'frequency_difference': 0.004081632653061225}, 'she': {'case': '+d', 'shift': -1.7980318811440845, 'value_difference': 0.6073267326732674, 'frequency_difference': -0.009900990099009901}, 'moral': {'case': '+u', 'shift': 1.180601877362409, 'value_difference': 0.9673267326732677, 'frequency_difference': 0.004081632653061225}, 'river': {'case': '+u', 'shift': 1.3514689249151663, 'value_difference': 1.1073267326732674, 'frequency_difference': 0.004081632653061225}, 'soft': {'case': '+d', 'shift': -2.6862021778292147, 'value_difference': 0.9073267326732681, 'frequency_difference': -0.009900990099009901}, 'mean': {'case': '-u', 'shift': -2.309967808358218, 'value_difference': -1.8926732673267321, 'frequency_difference': 0.004081632653061225}, 'enjoy': {'case': '+u', 'shift': 2.5475382577844727, 'value_difference': 2.087326732673268, 'frequency_difference': 0.004081632653061225}, 'poor': {'case': '-u', 'shift': -7.939638254884306, 'value_difference': -3.2526732673267325, 'frequency_difference': 0.00816326530612245}, 'right': {'case': '+u', 'shift': 1.180601877362409, 'value_difference': 0.9673267326732677, 'frequency_difference': 0.004081632653061225}, 'old': {'case': '-u', 'shift': -9.719120675154398, 'value_difference': -1.5926732673267323, 'frequency_difference': 0.02040816326530612}, 'picture': {'case': '+u', 'shift': 1.302649768471521, 'value_difference': 1.0673267326732674, 'frequency_difference': 0.004081632653061225}, 'house': {'case': '+u', 'shift': 0.9365060951441829, 'value_difference': 0.7673267326732676, 'frequency_difference': 0.004081632653061225}, 'we': {'case': '+u', 'shift': 0.985325251587828, 'value_difference': 0.8073267326732676, 'frequency_difference': 0.004081632653061225}, 'rivers': {'case': '+u', 'shift': 1.546745550689747, 'value_difference': 1.2673267326732676, 'frequency_difference': 0.004081632653061225}, 'see': {'case': '+u', 'shift': 2.379088000154667, 'value_difference': 0.4873267326732673, 'frequency_difference': 0.0163265306122449}, 'sea': {'case': '+u', 'shift': 2.627110467782365, 'value_difference': 1.367326732673268, 'frequency_difference': 0.006425540513234998}, 'our': {'case': '+u', 'shift': 0.6191815782604898, 'value_difference': 0.5073267326732678, 'frequency_difference': 0.004081632653061225}, 'sailed': {'case': '+d', 'shift': -2.8342305606100684, 'value_difference': 0.957326732673268, 'frequency_difference': -0.009900990099009901}, 'pale': {'case': '-u', 'shift': -4.327020678054566, 'value_difference': -1.7726732673267325, 'frequency_difference': 0.00816326530612245}, 'strike': {'case': '-u', 'shift': -1.9682337132527024, 'value_difference': -1.6126732673267323, 'frequency_difference': 0.004081632653061225}, 'living': {'case': '+d', 'shift': -3.278315708952631, 'value_difference': 1.1073267326732674, 'frequency_difference': -0.009900990099009901}, 'driving': {'case': '+u', 'shift': 0.7412294693696022, 'value_difference': 0.6073267326732674, 'frequency_difference': 0.004081632653061225}, 'greeks': {'case': '+u', 'shift': 0.9975300406987391, 'value_difference': 0.8173267326732674, 'frequency_difference': 0.004081632653061225}, 'sail': {'case': '+d', 'shift': -2.2052609831616192, 'value_difference': 1.2673267326732676, 'frequency_difference': -0.005819357445948676}, 'please': {'case': '+u', 'shift': 0.9609156733660059, 'value_difference': 0.787326732673268, 'frequency_difference': 0.004081632653061225}, 'find': {'case': '+u', 'shift': 1.043086530746399, 'value_difference': 0.4273267326732677, 'frequency_difference': 0.00816326530612245}, 'content': {'case': '+u', 'shift': 0.5459528435950215, 'value_difference': 0.4473267326732673, 'frequency_difference': 0.004081632653061225}, 'indian': {'case': '+d', 'shift': -1.1960095113365594, 'value_difference': 0.6873267326732675, 'frequency_difference': -0.005819357445948676}, 'landscape': {'case': '+u', 'shift': 1.3514689249151663, 'value_difference': 1.1073267326732674, 'frequency_difference': 0.004081632653061225}, 'receiving': {'case': '+u', 'shift': 1.180601877362409, 'value_difference': 0.9673267326732677, 'frequency_difference': 0.004081632653061225}, 'bird': {'case': '+d', 'shift': -6.319786005455894, 'value_difference': 1.0673267326732674, 'frequency_difference': -0.019801980198019802}, 'poets': {'case': '+u', 'shift': 1.0829635644751183, 'value_difference': 0.8873267326732677, 'frequency_difference': 0.004081632653061225}, 'full': {'case': '+d', 'shift': -1.1264059615555209, 'value_difference': 0.6473267326732675, 'frequency_difference': -0.005819357445948676}, 'desires': {'case': '+u', 'shift': 1.4124928704697237, 'value_difference': 1.1573267326732681, 'frequency_difference': 0.004081632653061225}, 'funeral': {'case': '-u', 'shift': -4.238324487882201, 'value_difference': -3.472673267326732, 'frequency_difference': 0.004081632653061225}, 'never': {'case': '-u', 'shift': -5.449861276258404, 'value_difference': -2.2326732673267324, 'frequency_difference': 0.00816326530612245}, 'disappeared': {'case': '-d', 'shift': 6.313923495246751, 'value_difference': -2.1326732673267323, 'frequency_difference': -0.009900990099009901}, 'water': {'case': '+u', 'shift': 5.503514012547958, 'value_difference': 1.1273267326732679, 'frequency_difference': 0.0163265306122449}, 'sink': {'case': '-d', 'shift': 10.555449631561533, 'value_difference': -1.7826732673267323, 'frequency_difference': -0.019801980198019802}, 'alone': {'case': '-d', 'shift': 6.669191613920801, 'value_difference': -2.2526732673267325, 'frequency_difference': -0.009900990099009901}, 'sadly': {'case': '-u', 'shift': -4.018638283885799, 'value_difference': -3.2926732673267325, 'frequency_difference': 0.004081632653061225}, 'november': {'case': '+u', 'shift': 0.6435911564823118, 'value_difference': 0.5273267326732674, 'frequency_difference': 0.004081632653061225}, 'strong': {'case': '+u', 'shift': 1.8152509111297948, 'value_difference': 1.4873267326732673, 'frequency_difference': 0.004081632653061225}, 'care': {'case': '+u', 'shift': 2.605299536943042, 'value_difference': 1.0673267326732674, 'frequency_difference': 0.00816326530612245}, 'boy': {'case': '+u', 'shift': 0.8144582040350705, 'value_difference': 0.6673267326732679, 'frequency_difference': 0.004081632653061225}, 'great': {'case': '+d', 'shift': -17.676947068854908, 'value_difference': 2.3073267326732676, 'frequency_difference': -0.025621337643968477}, 'fixed': {'case': '+d', 'shift': -0.2324353690553061, 'value_difference': 0.4473267326732673, 'frequency_difference': -0.0017377247928874514}, 'last': {'case': '-d', 'shift': 21.703012794246487, 'value_difference': -1.832673267326732, 'frequency_difference': -0.039603960396039604}, 'healthy': {'case': '+u', 'shift': 5.973821331554556, 'value_difference': 2.4473267326732673, 'frequency_difference': 0.00816326530612245}, 'refugees': {'case': '-u', 'shift': -3.066664733234717, 'value_difference': -2.5126732673267322, 'frequency_difference': 0.004081632653061225}, 'against': {'case': '-d', 'shift': 3.7806442980077, 'value_difference': -2.1726732673267324, 'frequency_difference': -0.005819357445948676}, 'leaving': {'case': '-d', 'shift': 5.899444023460357, 'value_difference': -1.9926732673267322, 'frequency_difference': -0.009900990099009901}, 'american': {'case': '+u', 'shift': 1.4246976595806347, 'value_difference': 1.167326732673268, 'frequency_difference': 0.004081632653061225}, 'nothing': {'case': '-u', 'shift': -4.08292489583634, 'value_difference': -1.6726732673267324, 'frequency_difference': 0.00816326530612245}, 'cherish': {'case': '+u', 'shift': 2.3522616320098915, 'value_difference': 1.9273267326732677, 'frequency_difference': 0.004081632653061225}, 'distinction': {'case': '+u', 'shift': 0.985325251587828, 'value_difference': 0.8073267326732676, 'frequency_difference': 0.004081632653061225}, 'upward': {'case': '+d', 'shift': -1.8572432342564276, 'value_difference': 0.6273267326732679, 'frequency_difference': -0.009900990099009901}, 'top': {'case': '+d', 'shift': -3.5151611214019987, 'value_difference': 1.1873267326732675, 'frequency_difference': -0.009900990099009901}, 'first': {'case': '+u', 'shift': 3.0446719449358506, 'value_difference': 1.247326732673268, 'frequency_difference': 0.00816326530612245}, 'own': {'case': '+u', 'shift': 0.7168198911477802, 'value_difference': 0.5873267326732678, 'frequency_difference': 0.004081632653061225}, 'family': {'case': '+u', 'shift': 2.6207669924499397, 'value_difference': 2.1473267326732675, 'frequency_difference': 0.004081632653061225}, 'gone': {'case': '-u', 'shift': -5.254584650483824, 'value_difference': -2.1526732673267324, 'frequency_difference': 0.00816326530612245}, 'voyage': {'case': '+u', 'shift': 1.6199742853552157, 'value_difference': 1.327326732673268, 'frequency_difference': 0.004081632653061225}, 'down': {'case': '-d', 'shift': 1.987687675701816, 'value_difference': -1.9126732673267322, 'frequency_difference': -0.003475449585774903}, 'lies': {'case': '-d', 'shift': 5.137913518737951, 'value_difference': -2.952673267326732, 'frequency_difference': -0.005819357445948676}, 'maintained': {'case': '+d', 'shift': -2.3309340591551626, 'value_difference': 0.787326732673268, 'frequency_difference': -0.009900990099009901}, 'rising': {'case': '+d', 'shift': -2.863836237166239, 'value_difference': 0.9673267326732677, 'frequency_difference': -0.009900990099009901}, 'reached': {'case': '+d', 'shift': -3.003950231164752, 'value_difference': 0.5073267326732678, 'frequency_difference': -0.019801980198019802}, 'open': {'case': '+d', 'shift': -1.5611864686947166, 'value_difference': 0.5273267326732674, 'frequency_difference': -0.009900990099009901}, 'boat': {'case': '+d', 'shift': -1.501975115582376, 'value_difference': 0.5073267326732678, 'frequency_difference': -0.009900990099009901}, 'late': {'case': '-u', 'shift': -2.5784731687982667, 'value_difference': -2.1126732673267323, 'frequency_difference': 0.004081632653061225}, 'devil': {'case': '-u', 'shift': -4.067457440329442, 'value_difference': -3.332673267326732, 'frequency_difference': 0.004081632653061225}, 'eye': {'case': '+u', 'shift': 1.384820625851914, 'value_difference': 0.5673267326732674, 'frequency_difference': 0.00816326530612245}, 'caught': {'case': '-d', 'shift': 6.01786672968504, 'value_difference': -2.0326732673267323, 'frequency_difference': -0.009900990099009901}, 'flying': {'case': '+d', 'shift': -1.9756659404811112, 'value_difference': 0.6673267326732679, 'frequency_difference': -0.009900990099009901}, 'visit': {'case': '+u', 'shift': 1.180601877362409, 'value_difference': 0.9673267326732677, 'frequency_difference': 0.004081632653061225}, 'breath': {'case': '+d', 'shift': -2.3901454122675037, 'value_difference': 0.8073267326732676, 'frequency_difference': -0.009900990099009901}, 'paint': {'case': '+u', 'shift': 0.8876869387005377, 'value_difference': 0.7273267326732675, 'frequency_difference': 0.004081632653061225}, 'interest': {'case': '+u', 'shift': 0.9609156733660059, 'value_difference': 0.787326732673268, 'frequency_difference': 0.004081632653061225}, 'poet': {'case': '+u', 'shift': 1.1561922991405857, 'value_difference': 0.9473267326732673, 'frequency_difference': 0.004081632653061225}, 'survive': {'case': '+d', 'shift': -1.2651297031330084, 'value_difference': 0.4273267326732677, 'frequency_difference': -0.009900990099009901}, 'philosophical': {'case': '+u', 'shift': 0.7412294693696022, 'value_difference': 0.6073267326732674, 'frequency_difference': 0.004081632653061225}, 'valley': {'case': '+u', 'shift': 0.8388677822568925, 'value_difference': 0.6873267326732675, 'frequency_difference': 0.004081632653061225}, 'bubble': {'case': '+d', 'shift': -3.714486468512855, 'value_difference': 0.6273267326732679, 'frequency_difference': -0.019801980198019802}, 'more': {'case': '+u', 'shift': 4.072291020175352, 'value_difference': 0.6673267326732679, 'frequency_difference': 0.02040816326530612}, 'mountains': {'case': '+u', 'shift': 1.6932030200206825, 'value_difference': 1.3873267326732677, 'frequency_difference': 0.004081632653061225}, 'eyes': {'case': '+u', 'shift': 0.8144582040350705, 'value_difference': 0.6673267326732679, 'frequency_difference': 0.004081632653061225}, 'buy': {'case': '+u', 'shift': 0.8632773604787155, 'value_difference': 0.707326732673268, 'frequency_difference': 0.004081632653061225}, 'savage': {'case': '-d', 'shift': 13.397594580953944, 'value_difference': -2.2626732673267322, 'frequency_difference': -0.019801980198019802}, 'boys': {'case': '+u', 'shift': 0.7168198911477802, 'value_difference': 0.5873267326732678, 'frequency_difference': 0.004081632653061225}, 'flag': {'case': '+d', 'shift': -2.034877293593452, 'value_difference': 0.6873267326732675, 'frequency_difference': -0.009900990099009901}, 'virtue': {'case': '+u', 'shift': 1.0951683535860295, 'value_difference': 0.8973267326732675, 'frequency_difference': 0.004081632653061225}, 'authors': {'case': '+u', 'shift': 0.7900486258132473, 'value_difference': 0.6473267326732675, 'frequency_difference': 0.004081632653061225}, 'rocking': {'case': '+d', 'shift': -2.626990824716871, 'value_difference': 0.8873267326732677, 'frequency_difference': -0.009900990099009901}, 'herself': {'case': '+d', 'shift': -1.5611864686947166, 'value_difference': 0.5273267326732674, 'frequency_difference': -0.009900990099009901}, 'white': {'case': '+d', 'shift': -2.0940886467057953, 'value_difference': 0.707326732673268, 'frequency_difference': -0.009900990099009901}, 'especially': {'case': '+u', 'shift': 0.7168198911477802, 'value_difference': 0.5873267326732678, 'frequency_difference': 0.004081632653061225}, 'me': {'case': '+d', 'shift': -2.276254785363684, 'value_difference': 1.0073267326732678, 'frequency_difference': -0.007557082238836129}, 'loved': {'case': '+u', 'shift': 2.9136819311118107, 'value_difference': 2.3873267326732677, 'frequency_difference': 0.004081632653061225}, 'kind': {'case': '+u', 'shift': 2.034937115126199, 'value_difference': 1.667326732673268, 'frequency_difference': 0.004081632653061225}, 'romantic': {'case': '+u', 'shift': 2.6207669924499397, 'value_difference': 2.1473267326732675, 'frequency_difference': 0.004081632653061225}, 'glory': {'case': '+u', 'shift': 4.704523264019783, 'value_difference': 1.9273267326732677, 'frequency_difference': 0.00816326530612245}, 'straight': {'case': '+u', 'shift': 0.5947720000386667, 'value_difference': 0.4873267326732673, 'frequency_difference': 0.004081632653061225}, 'soul': {'case': '+u', 'shift': 0.9365060951441829, 'value_difference': 0.7673267326732676, 'frequency_difference': 0.004081632653061225}, 'nations': {'case': '+u', 'shift': 1.043086530746399, 'value_difference': 0.4273267326732677, 'frequency_difference': 0.00816326530612245}, 'will': {'case': '+u', 'shift': 2.183811374380086, 'value_difference': 0.4473267326732673, 'frequency_difference': 0.0163265306122449}, 'sadness': {'case': '-u', 'shift': -4.5068298483222495, 'value_difference': -3.6926732673267324, 'frequency_difference': 0.004081632653061225}, 'lead': {'case': '+u', 'shift': 0.8632773604787155, 'value_difference': 0.707326732673268, 'frequency_difference': 0.004081632653061225}, 'loves': {'case': '+u', 'shift': 2.9136819311118107, 'value_difference': 2.3873267326732677, 'frequency_difference': 0.004081632653061225}, 'coffin': {'case': '-d', 'shift': 4.546283345599123, 'value_difference': -2.6126732673267323, 'frequency_difference': -0.005819357445948676}, 'growing': {'case': '+u', 'shift': 1.1561922991405857, 'value_difference': 0.9473267326732673, 'frequency_difference': 0.004081632653061225}, 'making': {'case': '+u', 'shift': 1.628916408070141, 'value_difference': 0.6673267326732679, 'frequency_difference': 0.00816326530612245}, 'closing': {'case': '-d', 'shift': 4.774428314325862, 'value_difference': -1.6126732673267323, 'frequency_difference': -0.009900990099009901}, 'my': {'case': '+d', 'shift': -6.1320038213037815, 'value_difference': 0.5873267326732678, 'frequency_difference': -0.034916144675692054}, 'damned': {'case': '-d', 'shift': 6.846825673257826, 'value_difference': -2.3126732673267325, 'frequency_difference': -0.009900990099009901}, 'heart': {'case': '+d', 'shift': -4.877022242985861, 'value_difference': 1.6473267326732675, 'frequency_difference': -0.009900990099009901}, 'picked': {'case': '+d', 'shift': -1.501975115582376, 'value_difference': 0.5073267326732678, 'frequency_difference': -0.009900990099009901}, 'stream': {'case': '+u', 'shift': 1.0341444080314732, 'value_difference': 0.8473267326732676, 'frequency_difference': 0.004081632653061225}, 'give': {'case': '+d', 'shift': -1.6832343598038302, 'value_difference': 0.9673267326732677, 'frequency_difference': -0.005819357445948676}, 'mind': {'case': '+u', 'shift': 0.5703624218168447, 'value_difference': 0.46732673267326774, 'frequency_difference': 0.004081632653061225}, 'high': {'case': '+u', 'shift': 3.9079493054145624, 'value_difference': 1.0673267326732674, 'frequency_difference': 0.012244897959183673}, 'professor': {'case': '+d', 'shift': -0.24282756572400305, 'value_difference': 0.46732673267326774, 'frequency_difference': -0.0017377247928874514}, 'dropped': {'case': '-d', 'shift': 5.011273726775229, 'value_difference': -1.6926732673267324, 'frequency_difference': -0.009900990099009901}, 'sense': {'case': '+u', 'shift': 0.8876869387005377, 'value_difference': 0.7273267326732675, 'frequency_difference': 0.004081632653061225}, 'floating': {'case': '+d', 'shift': -4.898713530759693, 'value_difference': 0.8273267326732681, 'frequency_difference': -0.019801980198019802}, 'attract': {'case': '+u', 'shift': 1.4002880813588117, 'value_difference': 1.1473267326732675, 'frequency_difference': 0.004081632653061225}, 'trip': {'case': '+u', 'shift': 1.7420221764643276, 'value_difference': 1.4273267326732677, 'frequency_difference': 0.004081632653061225}, 'huge': {'case': '+u', 'shift': 0.692410312925957, 'value_difference': 0.5673267326732674, 'frequency_difference': 0.004081632653061225}, 'court': {'case': '-u', 'shift': -2.1879199172491055, 'value_difference': -1.7926732673267325, 'frequency_difference': 0.004081632653061225}, 'no': {'case': '-u', 'shift': -7.662190771729331, 'value_difference': -2.0926732673267323, 'frequency_difference': 0.012244897959183673}, 'make': {'case': '+u', 'shift': 0.5215432653731995, 'value_difference': 0.4273267326732677, 'frequency_difference': 0.004081632653061225}, 'book': {'case': '+u', 'shift': 2.034937115126199, 'value_difference': 1.667326732673268, 'frequency_difference': 0.004081632653061225}, 'you': {'case': '+u', 'shift': 12.216873060526055, 'value_difference': 0.6673267326732679, 'frequency_difference': 0.061224489795918366}, 'surf': {'case': '+d', 'shift': -2.804624884053896, 'value_difference': 0.9473267326732673, 'frequency_difference': -0.009900990099009901}, 'noble': {'case': '+u', 'shift': 1.7420221764643276, 'value_difference': 1.4273267326732677, 'frequency_difference': 0.004081632653061225}, 'begin': {'case': '+u', 'shift': 1.726554720957431, 'value_difference': 0.707326732673268, 'frequency_difference': 0.00816326530612245}, 'established': {'case': '+u', 'shift': 0.7412294693696022, 'value_difference': 0.6073267326732674, 'frequency_difference': 0.004081632653061225}, 'act': {'case': '+d', 'shift': -1.2651297031330084, 'value_difference': 0.4273267326732677, 'frequency_difference': -0.009900990099009901}, 'prepared': {'case': '+d', 'shift': -2.0312521087090234, 'value_difference': 1.167326732673268, 'frequency_difference': -0.005819357445948676}, 'most': {'case': '+u', 'shift': 1.5800972516264946, 'value_difference': 0.6473267326732675, 'frequency_difference': 0.00816326530612245}, 'meaning': {'case': '+u', 'shift': 0.5215432653731995, 'value_difference': 0.4273267326732677, 'frequency_difference': 0.004081632653061225}, 'fell': {'case': '-d', 'shift': 6.96524837948251, 'value_difference': -2.352673267326732, 'frequency_difference': -0.009900990099009901}, 'stab': {'case': '-d', 'shift': 6.284317818690578, 'value_difference': -2.122673267326732, 'frequency_difference': -0.009900990099009901}, 'hearts': {'case': '+u', 'shift': 2.059346693348021, 'value_difference': 1.6873267326732675, 'frequency_difference': 0.004081632653061225}, 'dust': {'case': '-u', 'shift': -2.407606121245508, 'value_difference': -1.9726732673267322, 'frequency_difference': 0.004081632653061225}, 'whale': {'case': '+d', 'shift': -2.567779471604531, 'value_difference': 0.8673267326732681, 'frequency_difference': -0.009900990099009901}, 'friends': {'case': '+u', 'shift': 2.8648627746681656, 'value_difference': 2.3473267326732676, 'frequency_difference': 0.004081632653061225}, 'grow': {'case': '+u', 'shift': 2.7029378498303327, 'value_difference': 1.1073267326732674, 'frequency_difference': 0.00816326530612245}, 'pool': {'case': '+u', 'shift': 1.4735168160242798, 'value_difference': 1.207326732673268, 'frequency_difference': 0.004081632653061225}, 'land': {'case': '+u', 'shift': 0.7412294693696022, 'value_difference': 0.6073267326732674, 'frequency_difference': 0.004081632653061225}, 'natural': {'case': '+d', 'shift': -2.9230475902785797, 'value_difference': 0.9873267326732673, 'frequency_difference': -0.009900990099009901}, 'faster': {'case': '+d', 'shift': -4.780290824535007, 'value_difference': 0.8073267326732676, 'frequency_difference': -0.019801980198019802}, 'surprising': {'case': '+u', 'shift': 1.4246976595806347, 'value_difference': 1.167326732673268, 'frequency_difference': 0.004081632653061225}, 'conscious': {'case': '+u', 'shift': 0.692410312925957, 'value_difference': 0.5673267326732674, 'frequency_difference': 0.004081632653061225}, 'home': {'case': '+d', 'shift': -4.640176830536493, 'value_difference': 1.5673267326732674, 'frequency_difference': -0.009900990099009901}, 'up': {'case': '+d', 'shift': -2.9615965859803306, 'value_difference': 0.5673267326732674, 'frequency_difference': -0.017458072337846027}, 'clear': {'case': '+d', 'shift': -1.2656130611175984, 'value_difference': 0.7273267326732675, 'frequency_difference': -0.005819357445948676}, 'together': {'case': '+d', 'shift': -2.1356574333805804, 'value_difference': 1.2273267326732675, 'frequency_difference': -0.005819357445948676}, 'thought': {'case': '+u', 'shift': 0.985325251587828, 'value_difference': 0.8073267326732676, 'frequency_difference': 0.004081632653061225}, 'chief': {'case': '+u', 'shift': 0.7412294693696022, 'value_difference': 0.6073267326732674, 'frequency_difference': 0.004081632653061225}, 'without': {'case': '-u', 'shift': -7.442504567732926, 'value_difference': -2.0326732673267323, 'frequency_difference': 0.012244897959183673}, 'green': {'case': '+u', 'shift': 1.4735168160242798, 'value_difference': 1.207326732673268, 'frequency_difference': 0.004081632653061225}, 'principle': {'case': '+u', 'shift': 0.5215432653731995, 'value_difference': 0.4273267326732677, 'frequency_difference': 0.004081632653061225}, 'velocity': {'case': '+d', 'shift': -1.501975115582376, 'value_difference': 0.5073267326732678, 'frequency_difference': -0.009900990099009901}, 'travel': {'case': '+u', 'shift': 1.9617083804607303, 'value_difference': 1.6073267326732674, 'frequency_difference': 0.004081632653061225}, 'wine': {'case': '+u', 'shift': 1.0341444080314732, 'value_difference': 0.8473267326732676, 'frequency_difference': 0.004081632653061225}}}\n",
      "[-17.676947068854908, -11.65176775500695, -9.719120675154398, -7.939638254884306, -7.662190771729331, -7.442504567732926, -6.771785542580802, -6.319786005455894, -6.1320038213037815, -5.449861276258404, -5.254584650483824, -4.898713530759693, -4.877022242985861, -4.780290824535007, -4.640176830536493, -4.5068298483222495, -4.327020678054566, -4.238324487882201, -4.08292489583634, -4.067457440329442, -4.018638283885799, -3.7989520798893945, -3.7224618752585967, -3.714486468512855, -3.5151611214019987, -3.278315708952631, -3.066664733234717, -3.003950231164752, -2.9615965859803306, -2.9230475902785797, -2.863836237166239, -2.863836237166239, -2.8342305606100684, -2.808192640266889, -2.804624884053896, -2.6862021778292147, -2.626990824716871, -2.5784731687982667, -2.567779471604531, -2.41981091035642, -2.407606121245508, -2.3901454122675037, -2.3309340591551626, -2.309967808358218, -2.276254785363684, -2.2052609831616192, -2.1879199172491055, -2.1356574333805804, -2.0940886467057953, -2.034877293593452, -2.0312521087090234, -1.9756659404811112, -1.9756659404811112, -1.9682337132527024, -1.8572432342564276, -1.7980318811440845, -1.6832343598038302, -1.5611864686947166, -1.5611864686947166, -1.501975115582376, -1.501975115582376, -1.501975115582376, -1.335216610898637, -1.2656130611175984, -1.2651297031330084, -1.2651297031330084, -1.1960095113365594, -1.1264059615555209, -0.31557294240488004, -0.24282756572400305, -0.2324353690553061, 0.5215432653731995, 0.5215432653731995, 0.5215432653731995, 0.5459528435950215, 0.5703624218168447, 0.5703624218168447, 0.5947720000386667, 0.6191815782604898, 0.6435911564823118, 0.668000734704135, 0.692410312925957, 0.692410312925957, 0.7168198911477802, 0.7168198911477802, 0.7168198911477802, 0.7412294693696022, 0.7412294693696022, 0.7412294693696022, 0.7412294693696022, 0.7412294693696022, 0.7900486258132473, 0.8144582040350705, 0.8144582040350705, 0.8388677822568925, 0.8388677822568925, 0.8388677822568925, 0.8632773604787155, 0.8632773604787155, 0.8876869387005377, 0.8876869387005377, 0.8876869387005377, 0.9365060951441829, 0.9365060951441829, 0.9609156733660059, 0.9609156733660059, 0.985325251587828, 0.985325251587828, 0.985325251587828, 0.985325251587828, 0.985325251587828, 0.9975300406987391, 1.0097348298096511, 1.0219396189205623, 1.0341444080314732, 1.0341444080314732, 1.043086530746399, 1.043086530746399, 1.0829635644751183, 1.0951683535860295, 1.1561922991405857, 1.1561922991405857, 1.1561922991405857, 1.180601877362409, 1.180601877362409, 1.180601877362409, 1.180601877362409, 1.2660354011387882, 1.302649768471521, 1.302649768471521, 1.3514689249151663, 1.3514689249151663, 1.384820625851914, 1.384820625851914, 1.4002880813588117, 1.4124928704697237, 1.4246976595806347, 1.4246976595806347, 1.4246976595806347, 1.4735168160242798, 1.4735168160242798, 1.546745550689747, 1.5800972516264946, 1.6199742853552157, 1.628916408070141, 1.6932030200206825, 1.726554720957431, 1.7420221764643276, 1.7420221764643276, 1.7420221764643276, 1.8152509111297948, 1.8152509111297948, 1.8884796457952633, 1.9617083804607303, 1.987687675701816, 2.0105275369043754, 2.034937115126199, 2.034937115126199, 2.059346693348021, 2.183811374380086, 2.3522616320098915, 2.379088000154667, 2.5475382577844727, 2.605299536943042, 2.6207669924499397, 2.6207669924499397, 2.627110467782365, 2.7029378498303327, 2.8648627746681656, 2.9136819311118107, 2.9136819311118107, 2.995852788492204, 3.0446719449358506, 3.7806442980077, 3.8745976044778168, 3.9079493054145624, 4.072291020175352, 4.216331699583332, 4.546283345599123, 4.704523264019783, 4.774428314325862, 5.011273726775229, 5.137913518737951, 5.503514012547958, 5.899444023460357, 5.973821331554556, 6.01786672968504, 6.195500789022067, 6.284317818690578, 6.313923495246751, 6.313923495246751, 6.669191613920801, 6.846825673257826, 6.96524837948251, 8.031052735504666, 9.925816035099604, 10.555449631561533, 12.216873060526055, 13.397594580953944, 18.193714183053633, 21.703012794246487]\n",
      "100.0\n"
     ]
    }
   ],
   "source": [
    "wShift = WordShifts()\n",
    "wsData = wShift.computeWordShifts(sampleText[:10000], sampleText[-5000:])\n",
    "\n",
    "print wsData['arousal']\n",
    "\n",
    "vals = list()\n",
    "for w,d in wsData['arousal']['word_shift'].items():\n",
    "    vals.append(d['shift'])\n",
    "print sorted(vals)\n",
    "print sum(vals)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVwAAAJ4CAYAAAApw1uRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3XtYlXW6+P/3B5AFIqAiImAI5CiIHBWRSUjHyW+lptRM\n489xq1BqtKffVM5M4+z2jDX5dWZ202/apVk2ldcWy5oyJ5tr9pSHwNEURATFZR6hFDey5bBAFsfn\n9weykqMLWCdY9+u6uGI961nPcz9d0z0fPutzf26laRpCCCGsz8XeAQghhLOQhCuEEDYiCVcIIWxE\nEq4QQtiIJFwhhLARSbhCCGEjknCFEMJGHC7hKqVGKaV2KaVqlVIXlVL/j71jEkIIS3CzdwDd2AwY\nAX8gAfhUKVWgadpp+4YlhBADoxyp0kwpNRyoBKZomnb+5rFtwGVN035l1+CEEGKAHG1KYRLQ1J5s\nbzoBRNkpHiGEsBhHS7gjgJpOx2oAbzvEIoQQFuVoc7i1gE+nY76A4dYDSinHmQcRQjgdTdNUfz7n\naCPcrwA3pdSdtxyLBU51PlHTtEH785vf/MbuMUj89o9D4h+cPwPhUAlX07QbwEfA80qp4UqpWcBC\n4L/sG5kQQgycQyXcm/4VGA6UA9uBxzRZEiaEGAIcbQ4XTdMqgTR7x2FNs2fPtncIAyLx25fEP3g5\n1DpccymltMEYtxBi8FNKoQ2RL82EEGLIcrgphcHorbfeIiMjw6xzDxw4gE6nIzk5GYCLF9dbMTLR\nnbCw9fYOQTgppx3hlpeXD/ga7dMa5iZbIYRzc8oRbk1NDbt27SIjI4Nhw4b1eN6JEyc4fPgwSikC\nAgKYPXs227dvZ/z48ZSVlbF06VJ8fX3ZuHEj69ato7CwkKNHj9LS0kJwcDDz589HKUV2djaFhYV4\neXnh4+NDUFCQDZ9WCOEonC7h6vV6DAYD06dPJy8vj6CgICZMmNDlvGvXrpGTk8MjjzyCp6cnRqMR\no9HI9evXSUtLIzg4uMP5FRUVnDp1ioyMDFxcXPj0008pKirC39+f4uJiHnvsMVpaWnjjjTck4Qrh\npJwu4YaHh+Pu7k5BQQHJyck0NjZ2e97FixeZMmUKnp6eAHh4eGA0Ghk5cmSXZAtw4cIFysrK2Lp1\nKwDNzc2MGDGCGzduEBERgZubG25ubkyaNMl6DyeEcGhDKuE2NTVRXFzc4Zimabi7uzNlyhQA3N3d\ngbalHbe+NldvUxCxsbHMnTu3w7Evv/yyT9cXQgxdQ+pLs2HDhhEbG9vhJy4uzpRs+yIsLIzi4mLq\n6+sBTP/sSXh4OMXFxdTV1ZnOr66uZsKECej1epqbm2loaOCrr77q+4MJIYaEITXCtSR/f39SUlJ4\n5513cHFxYdy4cdx99909nj9mzBi+973vsX37djRNw9XVlfvvv5/g4GCioqLYsmULXl5e3U5HCCGc\ng9NWmhUUFBAXF2ehiPpP1uHanqzDFQMxkEozpx3hurq62jsEQP7jF8KZOO0IVwgh+kNGuMKi2gs5\nOhsqZcnyV4WwlyG1SkF8yxKly0IIy5IR7hBkbuny4cOHKSgoACA+Pp6ZM2d2OUfKkoWwHBnhDjF6\nvZ4zZ86YSpdLSkq6Pa+srIwTJ06watUqHnnkEfLz87l69WqHc65cuWIqS166dClXrlyxxSMIMWTJ\nCHeIMbd0ubS01FRyDBAZGUlpaSnjxo3r9hwpSxZi4CThDiK2KF0WQliPTCkMIpYsXQ4JCTGVHDc2\nNqLX67vsmiZlyUJYloxwnVRgYCBxcXGm3c0SEhIICAjoco6UJQthOVL4METZonRZ1uEKZySFD6IL\nW5QuS+ISom9khCuEEH0gI1zhsBxx2kFG5sJeZJWCE7Nk52IhxO3JCNdJWbJzcX5+Pp6enqbS4H37\n9uHl5UVSUpKtHkeIQUESrhOydOfi+Ph43n//fWbOnImmaZw8eZJVq1bZ+rGEcHiScJ2QpTsXjxw5\nkuHDh3P16lVqa2sJDAw0fUYI8S1JuEOMvToXx8fHU1BQQG1tLfHx8f0JXYghTxLuENNe/msJYWFh\n7Ny5k+TkZDw9PXvtXBwREcH+/ftpbW3loYcessj9hRhqJOGKHvWlc7GrqyuhoaF4enqaRs5CiI4k\n4Toxc5Z0tW+Sc6vMzMxur3X58mV++MMfWiw+IYYaSbhOzFLlv9euXePdd98lMjKS0aNHd3hPigyE\n+JaU9gohRB9Iaa8QfWDvcmMZ9TsvKe0VQggbkYQrhJMZyHRcUVGRBSNxPjKlIISdVVVVkZWVRUhI\nCF9//TU+Pj4sWbIENzc3CgsLOXr0KC0tLQQHBzN//nz27t2Lr68viYmJABw4cACdTkdycnK351dX\nV3fZ/8LX17dfsba2tlry0Z2OjHCFcADXr19nxowZPP7443h4eHD69GkqKio4deoUGRkZrFmzBqUU\nRUVFTJ06lVOnTpk+W1xcTFRUVI/nt18/MTGRzMzMfidbMXAywhXCAYwaNcrUUy4wMJCqqirq6+sp\nKysz9Z1rbm5mxIgRxMTEcOPGDWpra6mrq8PT0xMfHx+OHj3a7fkhISFd9r8Q9iEJVwgHcOuaaKWU\n6U/32NhY5s6d2+X8KVOmcOrUKWpra4mKijId7+78qqqqXrfgNGf/DWEZMqUghIMKCwvj9OnT1NXV\nAVBfX091dTUAUVFRnDp1itOnT5uSYm/n96Z9/41bf+Li4jokW6PR2OEzPe0wJ3onI1whHJS/vz9z\n5sxh+/btaJqGq6sr999/P76+vvj7+9PQ0ICPjw8jRozo9XwvL68Bx1JaWorBYEApRW5uLoGBgYwf\nP37A13U2UmkmhDBLTU0N7733Hunp6b1OUQx1A6k0k4QrhDBbRUUFY8aMsXcYdiUJVwghbET2UhA2\nVVVVxbvvvtvtNo09MRqNFBUVmRbrg/32NJC9DIS9yCoFJ2eJVunmMBqN5OXl2eReQjgqGeE6MXNa\npTc1NfHBBx9gMBhobW0lNTWV4OBgWltb+eSTT7qUogIcPnyYgoICABISEkhKSmLv3r1UVlby+uuv\nEx4ezj333GOz5xTCUUjCdVLmtko/d+4c3t7eLF26FICGhgbq6+u5fv06P/jBD1i4cCF/+ctfOH36\nNNHR0ZSVlXHixAlWrVpFa2srb775JhMmTGDu3LmUl5ezZs0aWz+qEA5DphScVHh4OImJibi6upKc\nnExgYGC3540dO5YLFy7w+eefU1paik6nA7ovRYW29ZoRERG4ubnh7u5OZGQkpaWltnkoIRycjHCH\nIEu2Svfz82PNmjWcPXuWffv2ER4eTkxMTI+lqEKInknCHYIs2SrdYDDg6elJdHQ0Op2O48ePExMT\n0+P5ISEh7N69m1mzZtHa2operyctLQ2dTifloMLpScIVvSovL+ezzz5DKYWrqyvz58/v9fzAwEDi\n4uJMO1YlJCQwbtw4AO644w5ee+01Jk6cKF+aCackhQ9OrqCggLi4OLvcW9bhisFICh9Ev1mqVXp/\nSOITzkZGuEII0QcywhWD3vr1Q/NeQtxK1uEKampqaGpq6vH9HTt22DAaIYYuSbiC48ePU1ZW1uP7\nzc3NvX6+pKSEyMhI0tPTmTx5MsuWLWPv3r3MmjWLyZMnd9hDISsri6SkJBISEsjMzBxQy24hBhtJ\nuMIizp8/z89//nPOnDmDXq/n3Xff5eDBg/zHf/wHGzZsANrKiXfu3MmhQ4fIz8/HxcWFrKwsO0cu\nhO3IHK4AGPBIMywszFTFFhUVZWpkGB0dTUlJCQB79+4lPz+fxMRENE3DaDSayoOFcAaScJ3UuXPn\n2L17N9A2OvXz82PkyJEopVi+fHmfd/Vv32MBwMXFxfTaxcXFNCWhaRorVqwwjXiFcDaScJ3UxIkT\nWbt2LQDZ2dmEhoYSEhLS4Zzq6mp8fX1Nr2tra00NCzvrbYTc/t7cuXNZvHgxTz75JP7+/lRWVmIw\nGLrcV4ihShKu6DFZ5uTkcOXKFVxcXNi8eTMJCQnMnDmz23PbN8Hp/PutryMjI3nhhReYN28era2t\nuLu7s2nTJkm4wmlI4YPocYQLcPnyZRYtWkR2djbDhw+3WgyyDlcMFtJEUgxIfX09Op0OF5fuF62c\nOXOGyZMn2zgqIRyTJFwhhLARKe0Vg55MKQhnIIUPTkxKeoWwLUm4TmygJb2/+c1vePnll02vn332\nWV555RUAXnrpJaKjo4mJiTGdU1JSQnR0tOn8P/7xjzz//PMDeQQhBhWZUhD9lpGRwYMPPshPf/pT\nNE3jvffeIzc3l/z8fLZt20Zubi4tLS0kJSUxe/ZsU2GFEM5KRrhObiBfPk6YMIExY8Zw4sQJ/vGP\nf5CQkMCoUaM4ePAgaWlpeHh44OXlxYMPPkhOTo4FoxZicJIRrpOxdEnvo48+yttvv83Vq1fJyMjo\n9Vw3NzdaWlpMr41GY98fQIhBTBKuk7F0Se/ixYv593//d5qbm3n33XcBSElJIT09nV/+8pe0tLSw\na9cusrKyCAgI4Nq1a1RWVjJ8+HD27NnDfffdZ6UnFcLxSMJ1YpYo6R02bBhz5sxh1KhRpvnZ+Ph4\nVq5cSWJiIkopVq9ebWqt/utf/5rExETGjx9PZGSkdR5MCAclhQ9OzBIlva2trUybNo2//OUv3Hnn\nnf2ORdbhisFCKs1Evwy0pPf06dMsWLCAhx56iD/84Q/WClMIhyIJVwghbGQgCVeWhQkhhI3Il2bC\n6Vy8uN5m9woLs929hOOTEa4QQ1RRUZG9QxCdyAhXCBuqqqoiKyuLkJAQvv76a3x8fFiyZAlubm4c\nPnyYgoICoG1pXU9L8czV2tpqiZCFBckIVwgbu379OjNmzODxxx/Hw8OD06dPU1ZWxokTJ1i1ahWP\nPPII+fn5XL161d6hCguTEa4QNjZq1ChTe/jAwECqqqq4ceMGERERuLm1/ScZGRlJaWkp48aNs2eo\nwsIk4QphY66urqbflVKmP/1vt9SxqamJ4uLiDsc0TcPd3Z0pU6ZYPlBhcZJwhXAAISEh7N69m5SU\nFFpbW9Hr9aSlpXU4Z9iwYcTGxt72WkajEQ8PD9PrxsZG3N3dLR6z6DtJuEI4gMDAQGJjY9m6dSsA\nCQkJ/Z5OKC0txWAwoJQiNzeXwMBAxo8fb8lwRT9JpZlwOs6wDrempob33nuP9PR0hg0bZpcYhiop\n7RVCdFFRUdHn/Y3F7UnCFUIIG5E26WLIsMWf+1JuK+xFCh9EB+Xl5Xa7t/zVIoY6GeEKk5qaGnbt\n2kVGRkaPX7S0l6YGBgZSVlbG2LFjSUtL67U0tafjVVVVbN++nfHjx1NWVsbSpUtt8JRC2I8kXAGA\nXq/HYDAwffp08vLyCAoKYsKECd2eW1FRwaJFixg/fjy7d+8mNzeX0NBQU2lqa2srb775JqGhoWia\n1u3x9iVP169fJy0tjeDg4JuvbfbIQticJFwBQHh4OO7u7hQUFJCcnExjY2OP5/r6+prWdcbExHD0\n6FFcXFy6lKaWlJQA9FqyOnLkSFOyFWKok4TrBMwpCW2vRGpvBNmXyqT2z3Seg735bW6vc7OyRlQ4\nE/nSzAm0l4Te+hMXF9fv+vvq6mq++eYboG3P1TvuuIOQkBDOnDlDc3MzjY2N6PV6QkJCejwuhDOS\nEa7oszFjxpCbm8vu3bsZO3YsiYmJuLm5ERcX121pak/HhXA2knBFB+YszXJxcemysQrAzJkzu900\nu6fjI0eOJDMzs3+BCjEIScIVHdy6daA9SFGCGMqktFcIIfpgUJX2KqX+FVgJRAM7NE3LuOW9ucCr\nwB3AESBd07RSW8co7Mva5b0yihb2Yo9VCpeB3wJ/vvWgUsoP+BD4N2A0cAzYafPoRL9t3Lixz58x\nGo3k5uZaIRohHI/NE66maR9rmvZXoHNN0YPASU3TPtI0rRFYD8QqpSbZOkbRlbX2WDAajeTl5Vnl\n2kI4Gkf60iwKONH+QtO0G0qpczePf2W3qIRZeyx0tnPnTmpqamhubiYpKYmEhASampr44IMPMBgM\ntLa2kpqail6v5/r167z++uuEh4dzzz33WPlphLAfR0q4I4DOw6gawNsOsYib+rLHwq0WLVqEh4cH\nzc3NbN26lcjISC5duoS3t7dpk5qGhgaCg4MpLy9nzZo11n4UIezOkRJuLeDT6ZgvYOju5PXr15t+\nnz17NrNnz7ZWXE6tL3ss3OrLL79Er9cDbSPk69evM3bsWP7xj3/w+eefM2nSJEJCQqivr7dm+EIM\n2IEDBzhw4IBFruVICfcUsKL9hVLKC7jz5vEubk24on+stcfCpUuXuHjxIo8++ihubm5s27aN5uZm\n/Pz8WLNmDWfPnmXfvn2Eh4cTExNj4acSwrI6D+iee+65fl/LHsvCXIFhgCvgppTSAc3ALuAPSqk0\n4G/Ab4ACTdNk/tZKzG273VcNDQ14enri5uZGRUWFad8Fg8GAp6cn0dHR6HQ6jh8/TmJiotmjZiEG\nO3uMcJ+lLZm2Vy78GHhO07TnlVIPAZuA7bStw11ih/jEAE2cOJG8vDw2b96Mn5+faSvH8vJyPvvs\nM5RSuLq6Mn/+fDw9Pbnjjjt47bXXmDhxonxpJoY0qTQTZikoKCAuLs7eYQhhdwOpNJPtGYVZ7L3H\nghBDgYxwhRCiDwbVXgpC2Ju1F7j09folJSUsWLCAoqIisz/j7e2NwdDtiknhwGRKQQgH0L7szlrn\nt6upqaGpqanXc06ePNmva4vbk4QrhJW9+OKLvPrqqwA89dRTzJ07F4D9+/ezbNkyAJqbm1m9ejVT\np07l3nvvpaGhAYC0tDQSExOJjo7mzTff7Pb6WVlZpvLpzMzMXjeRP378OGVlZT2+f/nyZZYvX86N\nGzf69ayid5JwhbCylJQUcnJyADh27Bh1dXW0tLSQk5PD3XffDcDZs2d54oknOHnyJL6+vnz44YcA\nvP322+Tm5pKbm8vLL79MZWVlh2vr9Xp27tzJoUOHyM/Px8XFhaysrH7F+fHHH/PXv/6VzMxMtmzZ\nYopZWI7M4QphZdOmTePYsWMYDAZ0Oh3Tpk0jNzeXnJwcXnnlFaCthDo6Otp0/qVLlwD405/+xMcf\nfwzAN998w9mzZ5kxY4bp2nv37iU/P5/ExEQ0TcNoNBIQENBrPD2NgO+55x68vLzYtm0bTz/9NLW1\ntQN9dNGJJFwhrMzNzY3Q0FDeeecd7rrrLmJiYti/fz/nz58nIiKCkpISdDqd6XxXV1eMRiNffPEF\n+/bt48iRI+h0OubMmYPRaOxwbU3TWLFiBRs2bOjx/ufOnWP37t0AnD9/Hj8/P0aOHIlSiuXLlzNm\nzBgAvLy8gG/nh0eMGGHRfw9CEq4QNpGSksKLL77I22+/zdSpU3nqqaeYPn266f3uRp3V1dWMGjUK\nnU6HXq/nyy+/7HL+3LlzWbx4MU8++ST+/v5UVlZiMBg6tKKfOHEia9euBSA7O5vQ0FBpVW8nMocr\nhA2kpKRw9epVkpOTGTt2LJ6enqSmppre727VwX333UdTUxNRUVH86le/Ijk5ucv5kZGRvPDCC8yb\nN4/Y2FjmzZvH1atXe4xD1q/blxQ+CKfjaOtwbcmcEe62bdtYsWJFj+87Oyl8EKIPHDkhWltiYmKH\n+eLumLMFp+gfGeEKIUQfyAhXiD6wZht2acEueiNfmgnhJPqyV4OwDhnhCmFFhw4dws3NjRkzZvD3\nv/+d8vJyli9fzsWLFykoKCAtLY3Dhw9TUFAAQHx8PDNnzgSgsLCQo0eP0tLSQnBwMPPnz+/3HgoA\nra2tFnkm0X8ywhXCikJCQigtLQWgrKyMxsZGWltbKS0tJSQkhLKyMk6cOMGqVat45JFHyM/P5+rV\nq1RUVHDq1CkyMjJYs2YNSikZoQ4BMsIVwoqCgoK4cuUKDQ0NuLm5ERgYyJUrVygtLeW+++4zVZu5\nubX9pxgZGdkhQW/duhVo29xGKr8GP0m4QliRi4sLI0eOpKCggDvuuIOAgAAuXrzI9evXGTNmDOfP\nn+/xs7GxsaadxbpjTtdl4VhkSkEIKwsJCeHw4cNMmDCBkJAQjh07RmBgoOk9vV5Pc3MzjY2N6PV6\nQkJCCAsL4/Tp09TV1QFQX19PdXV1h+u2d12+9ScuLq5Lsu28/4J0SbYfGeEKYWUTJkzg4MGDjB8/\nnmHDhuHm5maq9AoMDCQuLs40dZCQkMC4ceMAmDNnDtu3b0fTNFxdXbn//vvx9fXt8/1LS0sxGAwo\npcjNzSUwMNDUSVnYlhQ+CKfjjOtwa2pqeO+990hPT2fYsGH2DmdQG0jhgyRcIZxERUWFaStG0X+S\ncIUQwkaktFcMWdb4899R/+wXQ5+sUhBWdeTIETZt2sSuXbt6PW/jxo0d/inEUCQjXNFv5eXljB07\nttdz8vLyWL58Od7e3jaKSgjHJQlX9EtNTQ27du0iIyOjx2+99+zZQ2VlJVlZWcTFxTF8+HCL7g0g\nxGAjUwqiz/R6PWfOnGH69Onk5eVRUlLS7XkLFizA29ubFStWMHHiRNkbQDg9GeGKPgsPD8fd3Z2C\nggKSk5PNqly6cOGC7A0gnJ4kXNGBOfX57S1Y2qcDzG3Jcru9AYQY6iThig7a6/MtLSwsjJ07dzJz\n5ky8vLyor6+nsbGxX6WqQgxWknCFTfj7+1tsbwAhBiupNBP9VlBQQFxcnL3DEMKmBlJpJqsURL+5\nurraOwQhBhUZ4YpBwZIlvlLaKwZCRrhiyDAajeTm5to7DCGsQhKuGJDy8nKLXs9oNJKXl2fRawrh\nKGSVgug3c8p7q6qqyMrKIiQkhK+//hofHx+WLFmCm5tbh/bgCQkJJCUlsXfvXiorK3n99dcJDw/n\nnnvuseUjCWFVknBFv+j1egwGg6m8NygoiAkTJnR77vXr1/nBD37AwoUL+ctf/sLp06cZM2aMqT14\na2srb775JhMmTGDu3LmUl5ezZs0aGz+RENYnCVf0S1/Ke0eNGkVAQADQ1sOrqqqKGzdudNsefNKk\nSTaJXwh7kIQrurB0ee+ty8eUUrS2tpquKYQzkYQrurBWee+tQkJC2L17NykpKbS2tqLX60lLS0On\n00kbbzFkScIVdhEYGEhsbGy37cHvuOMOXnvtNSZOnChfmokhRQofxIBIea9wNlL4IOxGynuFMJ+M\ncIUQog+kTboQfbB+vX2vV1JSwoIFC7ptMTRr1iwOHjxIdXU1O3bsIDMz0zJBCocgUwpC2EFPzTMP\nHjwIQGVlJZs3b+739Wtqamhqaur1nJMnT/b7+qJ/JOEKYWEvvvgir776KgBPPfWUqa3Q/v37WbZs\nGdDW02316tVMnTqVe++9l4aGBgBTO/l169Zx4cIFEhISeOaZZwDIysoiKSmJhIQEMjMze13HfPz4\nccrKynp8//LlyyxfvpwbN24M/IGF2SThCmFhKSkp5OTkAHDs2DHq6upoaWkhJyeHu+++G4CzZ8/y\nxBNPcPLkSXx9ffnwww+Bb0e+v/vd77jzzjvJz8/n97//PXq9np07d3Lo0CHy8/NxcXEhKyurX/F9\n/PHH/PWvfyUzM5MtW7aYYhXWJ3O4QljYtGnTOHbsGAaDAZ1Ox7Rp08jNzSUnJ4dXXnkFaCuNjo6O\nNp1/6dIloOfqu71795Kfn09iYiKapmE0Gk3l0j3p6Vr33HMPXl5ebNu2jaeffpra2tp+PqnoK0m4\nQliYm5sboaGhvPPOO9x1113ExMSwf/9+zp8/T0REBCUlJeh0OtP5rq6uGI3GXq+paRorVqxgw4YN\nPZ5z7tw5du/eDcD58+fx8/Nj5MiRKKVYvnw5Y8aMAcDLywv4djQt7eptRxKuEFaQkpLCiy++yNtv\nv83UqVN56qmnmD59uun92y1r9Pb2xmAwmF7PnTuXxYsX8+STT+Lv709lZSUGg4GQkBDTORMnTmTt\n2rUAZGdnExoa2uF9YX8yhyuEFaSkpHD16lWSk5MZO3Ysnp6epKammt7vaZVC+/HRo0ebRsfPPPMM\nkZGR/Pa3v2XevHnExsYyb948rl692uP9ZZ26Y5LCB+F07L0O1xbMGeFu27aNFStW2DCqoUEKH4To\nA0dMkJaWmJjYYZ64O71tqSmsQ0a4QgjRBzLCFYNSVVUV7777brflq0ajkaKiIhITE03HLNUqXdqk\nC3uRL82cmKU77lqSdO8VQ5GMcJ2UOR13AU6cOMHhw4dRShEQEMDixYvZuXMnNTU1NDc3m0pNoeuI\n9dChQzQ1NXH33XfzxRdfUFRUhJeXFz4+PgQFBREZGUlrayuffPJJl46+0r1XDEWScJ2QuR13r127\nRk5ODo888gienp6mxfmLFi3Cw8OD5uZmtm7dSmRkJJ6enj3e78qVK+j1ejIzM2lpaeH1118nKCgI\n6L6jb3R0tHTvFUOSJFwnZG7H3YsXLzJlyhRTMvXw8ADgyy+/RK/XA20j5evXrxMcHNzj/UpLS5k8\neTKurq64urp26MzbXUdfIYYqSbhDjKU77nZ26dIlLl68yKOPPoqbmxvbtm2jubkZABcXlw4L7tuP\n97TIH3ru6CvEUCRfmg0x7R13b/2Ji4szJdu+CAsLo7i4mPr6egDq6+tpaGjA09MTNzc3Kioq+Oab\nb0znjxgxgrq6Ourr62lububs2bNAW1PIr776iubmZhobG03HeyPde8VQJCNc0SN/f39SUlJ45513\ncHFxYdy4cSxYsIC8vDw2b96Mn58f48ePN53v4uLC3XffzdatW/Hx8TFtlhIUFMTkyZPZsmULI0aM\nICAg4LaL8j09PaV7rxhypPDBidmy425jYyPu7u40NTXxzjvvsHDhQlNbdHPJOlzhCKTwQfSLLTvu\n7tmzh2vXrtHc3ExcXFyfky1IohSDn4xwhRCiD2SEK0QfyNSEsBdZpSCElb311lv2DkE4CEm4QlhZ\nRkaG1a5dVFRktWsLy5MpBSEspKmpiQ8++ACDwUBrayupqalERUWxceNG1q1bR1VVFVlZWYSEhHTZ\nO6K/pFBkcJGEK4SFnDt3Dm9vb5YuXQpAQ0NDl3N62jtCOAeZUhDCQsaOHcuFCxf4/PPPKS0t7ba4\nQ/aOcG63ytyfAAAgAElEQVQywhXCQvz8/FizZg1nz55l3759hIeHd2gcCbffO8KcvTDE4CUJVwgL\nMRgMeHp6Eh0djU6n4/jx432+RvteGLdjNBpNu7fBt5V8wrFJwhXCQsrLy/nss89QSuHq6sr8+fOt\ndq/S0lIMBgNKKXJzcwkMDOywr4VwTFJpJsQgVVNTw3vvvUd6enqvXTuEZQ2k0kwSrhCDWEVFhWlX\nNmEbknCFEMJGZC8FMWgcOXLE1EctKCjI9HtaWhpvvfXWbauyLLEPguyBIOxFEq7oUXl5OWPHjrXo\nNfPy8li+fDne3t5s2rTJ9DtYtwRWCEcgCVd0ayBt1A8fPkxBQQEA8fHxzJw5E2jbE7eyspKsrCwq\nKioAyMrKIi4ujpkzZ7Jx40Yef/xxtm/fbtHyVyEchfyvWHQxkDbqZWVlnDhxglWrVtHa2sqbb75J\naGioqT3P+fPnWbFiBZ6enrz88sum39tpmiblr2LIkoQruhhIG/XS0lIiIiJMI9LIyEhKS0vN7vCg\nlJLyVzFkScJ1MtZuo24J0jpdDFWyeY2TsXYb9ZCQEPR6vaklul6vJyQkxNKPIcSgJCNc0W/dtVFf\ntGgRsbGxbN26FYCEhIR+NYwUYiiSwgfRI1u2UTeXrMMV9iaFD8IqbNlG3VySLMVgJiNcIYToAxnh\nCtEH69db/holJSUsWLBgwE0dq6ur2bFjB5mZmQO6jnBMskpBCAtpX0Y3EJWVlWzevLnfn6+pqaGp\nqanH90+ePNnva4uBk4QrhIU0NTWxbNkypkyZwsMPP4zRaATgpZdeIjo6mpiYGF5++WXT+WlpaSQm\nJhIdHc2bb74JwLp167hw4QIJCQk888wz3LhxgwULFhAfH09MTAwffPBBrzEcP36csrKybt+7fPky\ny5cv58aNGxZ6YtFXMocrnI61phTCwsI4dOgQM2fO5JFHHiEqKorZs2eTnp7OkSNHaGlpISkpiays\nLGJjY6mqqmLkyJEYjUYSExPJzs6mpqaGhQsXUlhYCMBHH33Ef//3f/P6668DbW182jf76c4XX3xB\nWFhYl7XPH3/8MWVlZbi5uWEwGEhMTCQlJWXg/yKc0EDmcGWEK4SFhISEmDbqWbZsGTk5ORw8eJC0\ntDQ8PDzw8vLiwQcfJCcnB4A//elPpo17vvnmG86ePdvlmtHR0Xz22WesW7eOgwcP9pps23U3GLnn\nnnvIzMzE3d2dp59+mvj4+AE+regP+dJMCAvpPIfr4uLSPhrqcu4XX3zBvn37OHLkCDqdjjlz5pim\nIG71ne98h/z8fP72t7/x7LPP8v3vf59nn322wznnzp1j9+7dAJw/fx4/Pz9GjhyJUorly5czZswY\nvLy8OsQ4YsQIizyz6BtJuEJYSElJCUeOHCEpKYkdO3Ywa9YsZs2aRXp6OuvWraOlpYVdu3axfft2\nSkpKGDVqFDqdDr1ez5dffgmAt7c3BoPBdM2ysjJGjx7N0qVL8fX15c9//nOX+06cOJG1a9cCkJ2d\nTWhoqJRTOyhJuEJYSEREBJs2bSI9PZ2oqCgyMzPx8PBg5cqVJCYmopRi9erVxMbGEhkZyZYtW4iK\nimLy5MkkJycDMHr0aL773e8SExPDfffdx9y5c/n5z3+Oi4sL7u7uvPbaa73GIN9tODb50kw4HWt8\naeYobjfC3bZtGytWrLBxVEOLFD4I0QeOmiwtITExEZ1O1+P7tt5qU3QkI1whhOiDQTPCVUq5A5uB\n7wOjgPPArzRN+/vN9+cCrwJ3AEeAdE3TSm0Zo3BssluYGMxsvQ7XDSgFUjRN8wX+HXhfKRWilPID\nPgT+DRgNHAN22jg+0UdvvfWWTT4jxFBg9ykFpdQJYD0wBlihadqsm8eHAxVAnKZpX3X6jEwpDIA1\n2p/bioxwhb0NmimFzpRSAcB3gFPA48CJ9vc0TbuhlDoHRAFfdX8F0VfmtD+vqqoiKyuL8ePH8/XX\nXxMUFERcXBwHDhzgxo0bPPjggwQFBQGwceNGfvazn/HBBx9gMBhobW0lNTWVqKgompqauj2+ceNG\n1q1bZ7pPdy3Rv/jiC4qKivDy8sLHx4egoCDT0ikhBiu7JVyllBuwHXhH07SvlFIjgPJOp9UAt69l\nFGYxt/05wPXr13n44Yfx9/fnjTfe4OTJk2RkZHDmzBlycnL40Y9+ZDr33LlzeHt7s3TpUgAaGhp6\nPd75Pp1bovv5+aHX68nMzKSlpYXXX3/dlOCFGMzsknBVW33hdqABeOLm4VrAp9OpvoCBbqy/ZW3P\n7NmzmT17tqXDHHLMbX8OMGrUKPz9/QEYO3YsYWFhpt87ty0PCAjgH//4B59//jmTJk0yrQEdO3Zs\nt8c736dzS/S6ujomT56Mq6srrq6uTJo0ySLPL0R/HDhwgAMHDljkWvYa4f6Ztjnb+zVNa7l57BRg\nWpGtlPIC7rx5vIv1Q3kxZT9Yuv1551blbm5upt87ty0fPXo0a9as4ezZs+zbt4/w8HBSU1Px8/Pr\n9nhv92ltbbXIvrJCWErnAd1zzz3X72vZPOEqpbYAEcD3NU27dYi1C/iDUioN+BvwG6Cg8xdmonvt\n7c/toba2Fg8PD6Kjo9HpdBw/fhxo20rQ09Ozy/HbueOOO9izZw+zZs2itbWVs2fPMm3aNGs+ghA2\nYet1uCHAasAI/M/NkYwGrNE07V2l1EPAJtqmG44AS2wZn+if//mf/+Gzzz5DKYWrqyvz588H2lZD\ndHf8doKCgpg8eTJbtmxhxIgRBAQE9Fo9JcRgYfdlYf0hy8IGxhHbn3fW2NiIu7s7TU1NvPPOOyxc\nuJBx48bZOywhBu+yMGEfjtj+vLM9e/Zw7do1mpubiYuLk2QrhgQZ4QqnI8UTYiBkhCuEAzhy5Ihp\nfXNaWpq9wxEOSBKuEBaSl5fH8uXLzeo71htN0/q9NK6oqIjo6OgB3V9YjyRcISxgz549VFZWkpWV\nRVxcHMOHD+fo0aO0tLQQHBzM/PnzUUqxc+dOampqaG5uJikpiYSEBKqqqti+fTvjx4+nrKzM1E6n\nPzqvkRaORRKuEBawYMECzp8/z4oVK6irq+Ozzz4jIyMDFxcXPv30U4qKioiJiWHRokV4eHjQ3NzM\n1q1biYyMBNpKnNPS0ggODrbzkwhrkoQrhIVduHCBsrIytm7dCkBzc7OpS+6XX36JXq8H2jYSun79\nOl5eXowcOVKSrROQhCuEFcTGxjJ37twOxy5dusTFixd59NFHcXNzY9u2bTQ3NwP0uHMbmFe2LQYH\nSbhCWFhYWBg7d+5k5syZeHl5UV9fT2NjIw0NDXh6euLm5kZFRQXffPONWdczp2zbaDTi4eFhet1e\nOCIciyRcISzM39+fOXPmsH37djRNw9XVlfvvv5+JEyeSl5fH5s2b8fPzY/z48Ra7Z2lpKQaDAaUU\nubm5BAYGWvT6wjKk8EGIIaKmpob33nuP9PT0XqcoxMAMpPBBEq4QQ0hFRQVjxoyxdxhDmiRcIYSw\nESntFU5nIPvPy971wl5s3SZdOKmSkpIBlZw+99xzvPTSSxaMSAjbk4QrzFJTU0NTU1OP7+/YseO2\n15DWOcLZScIVZjl+/DhlZWU9vt++gL83TU1NLFu2jClTpvDwww9jNBoByMrKMu0rkJmZSfv8/IYN\nG5g8eTKpqamcOXPGMg8ihB1JwhU2c+bMGX7yk59QXFyMt7c3mzdvRq/Xs3PnTg4dOkR+fj4uLi5k\nZWWRn5/P+++/T2FhIZ9++im5ubn2Dl+IAZMvzYTZBroyJCQkhJkzZwKwbNky/vM//xOdTsexY8dI\nTExE0zSMRiMBAQH87//+L2lpaeh0OnQ6HQ888IAlHkEIu5KEK3p07tw5du/eDcD58+fx8/Nj5MiR\nKKVYvnx5n9d7dp7DdXFp+wNr5cqVbNiwocN7L7/88gAiF8IxScIVPZo4cSJr164FIDs7m9DQUEJC\nQjqcU11d3WHv1traWtPOWJ2VlJRw5MgRkpKS2LFjB7NmzeJ73/seixcv5sknn8Tf35/KykoMBgOp\nqamkp6ezbt06Ghsb+eSTT3jssces97BC2IAkXGGWnqYTcnJyuHLlCi4uLmzevJmEhATTtEFnERER\nbNq0ifT0dKKiosjMzMTDw4MXXniBefPm0drairu7O5s2bWLGjBk8/PDDxMTEEBAQwIwZM6z5eELY\nhFmVZkqpGZqmHe3hvWWapm23eGS9xyOVZjbW0wgX4PLlyyxatIjs7GyGDx9uh+iEsB2rl/YqpcqB\nVE3T9J2O/wuwRdM0r/7cvL8k4dpefX09Op3ONO/a2ZkzZ5g8ebKNoxLC9myRcJ8B/hX4rqZp39w8\nthx4DfiRpml7+nPz/pKEKwZioG3SpUW6cxtIwjVrHa6mab8H3gc+V0qNUUqtALYAD9s62QrhiDZu\n3AjAW2+9ZedIhCMzu/BB07SfAYeBI8Bm4Aeapn1qrcCEGIwyMjKsdu2ioiKrXVvYRo+rFJRSD3Zz\n+G/AXOBdwKP9HE3TPrJOeEIMLhs3bmTdunUAFBYWdtsqvb+kBfrg19uysL/08l7GzR8ADXC1WERC\nDAEVFRWcOnWq21bpwnn1mHA1TZN9FoTop95apQvnJYUPQlhJd63SO5MW6M7F7ISrlEqibf52LJ2+\nbNM07f+1cFxCDGo9tUq/tQwapAW6szEr4Sqlfgb8ATgHXKFt3radLIgVopOeWqV3TrjmkBboQ4e5\nhQ9fA7/XNO1V64d0e1L4IJyNtEB3HLaoNKsG4jVNu9Cfm1iaJFzhjKQFumOwRcLdAhRqmra5Pzex\nNEm4YiDluVKaKwbCFm3SvwaeU0rdBRQCHboJapom7VTFgBw4cACdTkdycjJvvfWWVSu2hLAXcxPu\no0At8N2bP7fSAEm4ooPy8nLGjh3br89KshVDlVkJV9O0MGsHIoaOmpoadu3aRUZGRq9f8GRnZ1NY\nWIiXlxc+Pj4EBQUB1i2PFcKepJpMWJRer+fMmTNMnz6dvLw8SkpKuj2vrKyM4uJiHnvsMZYuXcqV\nK1e6nHNreeyaNWtQSskGLmJQ60vhwyTgB0AI0GHVtaZp8jegACA8PBx3d3cKCgpITk6msbGx2/NK\nSkqIiIjAzc0NNzc3Jk2a1OUcKY8VQ425hQ/zgQ+B48A0IBe4E9ABOVaLTjgUc8pQ2yug2v/sH2hF\nlDnlsUIMFuZOKTwPPKdpWjLQAPwLEAp8DhywSmTC4bSXod76ExcX16+a/wkTJqDX62lubqahoYGv\nvvqqyzlhYWGcPn2auro6oK3NT3V19YCfQwh7MXdKYTKw8+bvTcBwTdOMSqnngU+RVQqijwIDA4mK\nimLLli14eXkRHBzc5RxLlscK4QjMLXwoA+ZqmlaslDoF/JumaR8rpeKBbE3TvK0daKd4pPDBwRUU\nFBAXF2fvMISwOFsUPhwBZgHFtI1o/6iUigXSaGu7I0QHrq6yJ70QnZk7wg0HRmiaVqiUGg78EbgL\n+Ap4WtO0UuuG2SUeGeEKIezCqnspKKXcgHnAEU3T/rc/N7E0SbiDh7e3NwaDweLXXb/ePp8Vwqpt\n0jVNawY+Amw6Tytsq6amhqamph7f37FjR7+uK1VhQnzL3GVhJ4CJ1gxE2Nfx48cpKyvr8f3m5uZe\nP3/jxg0WLFhAfHw8MTExfPDBB0DbOt12WVlZJCUlkZCQQGZmpum97o6XlJQQGRnJsmXLmDJlCg8/\n/DBGo9ECTyqE/ZibcNfT9kXZYqXUHUqp0bf+WDE+MUj8/e9/Jzg4mOPHj1NYWMi9994LfDvC1ev1\n7Ny5k0OHDpGfn4+LiwtZWVk9Hgc4c+YMP/nJTyguLsbb25vNmx1id1Ah+s3cVQqf3vznR3RsqaOQ\nNulDxkDmxaOjo/nZz37GunXrmD9/PrNmzerw/t69e8nPzycxMRFN0zAajQQEBFBdXc2xY8e6HE9J\nSSEkJISZM2cCsGzZMl555RWefvrpAT2jEPZkbsKdY9UohF2cO3eO3bt3A3D+/Hn8/PwYOXIkSimW\nL1/ep+4C3/nOd8jPz+dvf/sbzz77LN///vd59tlnTe9rmsaKFSvYsGFDh8+9+uqrrFy5ssvx7ja9\nkflgMdiZuz3jF9YORNjexIkTWbt2LdC2VWJoaCghISEdzqmuru5Q2VVbW9vtBjJlZWWMHj2apUuX\n4uvry5///Gfg21Hz3LlzWbx4MU8++ST+/v5UVlZiMBh6PA5tzROPHDlCUlISO3bs6DJqFmKwMXu3\nMDG09TSdkJOTw5UrV3BxcWHz5s0kJCSY/sy/VVFRET//+c9xcXHB3d2dLVu2AN+OSiMjI3nhhReY\nN28era2tuLu7s2nTJmbMmNHt8YCAACZPnsymTZtIT08nKiqKzMxM6/0LEMIGzCp8cDSyDtfyehrh\nAly+fJlFixaRnZ3N8OHDbRJPSUkJCxYs6HH/W1mHK+zF6k0kHY0kXMurr69Hp9Ph4tL9wpUzZ84w\nefJkm8VTUlLCwoULKSwstNk9hTCHJFwhhLARW2xeI5xQdXU1O3bsMM2d3u7PfEvo6R6dY2nXn+kB\nmVIQ9tJj4YNSar9Sap85P7YMWJhvoOW6lZWVXYoNbLE0q7t7dBeLEINNb5VmJ4FTN3/0tLXWCQa+\nufkTdPPYaSvHKPppoOW669at48KFCyQkJPDMM8+YPrN69WqmTp3KvffeS0NDAwAvvfQS0dHRxMTE\n8PLLLwNto9Xo6GjT9f74xz/y/PPPm17/9re/JSIigtTUVJYuXcpLL73U7T2MRmO3sQgx2PQ4paBp\n2hPtvyul/j9gG/DTWydPlVJ/oq3aTAxBv/vd7zh16hT5+flAWwI9e/YsO3fu5I033uBHP/oRH374\nIREREWzbto3c3FxaWlpISkpi9uzZpiKK7uTl5bFr1y6KiopoaGggISGB6dOnA3S5x0cffdQlFiEG\nI3P3UlgOvNrNN1WbaetvJhyUpb9cDA8PN41ap02bxqVLlzh48CBpaWl4eHjg5eXFgw8+SE5O771F\n//nPf7Jo0SKGDRvGiBEjWLhwYa/3EGIoMPdLMwVE07bh+K2iuzlX2JEly3W7o9PpTL+7urpiNBrb\nv7Xtcq6bmxstLS2m1+bu9tXdPYQYCsxNuG8BbyqlvgN8efPYTOAXwNvWCEz0jyXLdbvbPLy7xDpr\n1izS09NZt24dLS0t7Nq1i6ysLAICArh27RqVlZUMHz6cPXv2cN999wFw11138dhjj/HLX/6SpqYm\n9uzZw5o1a3q8h7U2MhfClsxNuL8AyoGfAv/35rEy4He0tdsRDmig5bqjR4/mrrvuIiYmhvvuu4/H\nH3+82znZ+Ph4VqxYQWJiIkopVq9eTUxMDAC//vWvSUxMZPz48URGRpo+M336dB544AFiY2MJCAgg\nJibG9H8C3d1j9OjRfPe73zXF8vvf/75f/06EsCdzWuy4ABFAiaZpdUopHwBN02psEF9PMUnhgxkc\nrVy3s7q6Ory8vKivryc1NZWtW7f2udOvrMMVtmbtnmYKaACmaJp2rj83sTRJuOZxtHLdzn784x9T\nXFxMQ0MDK1eu5Be/+IXdYhHCXFYv7VVKFQGrNU1ziJboknCFEPZii4R7H/As8K/ACXtnO0m4ot3F\ni+v7/JmwsL5/Roh2Vu3ae9P7wAzgGGBUStXc+tOfG4uhY+PGjYPimkLYm7mrFH5i1SiEwyovL2fs\n2LH2DkOIIcHcFjvbrB2IcDw1NTXs2rWLjIwMhg0bZtZnDh8+TEFBAdC2XGzmzJlUVVWRlZVFSEgI\nX3/9NT4+PixZsgQ3NzcKCws5evQoLS0tBAcHM3/+/A7Lwvbv34+np6dp2dq+ffvw8vIiKSnJ8g8s\nhJWZO6WAUkqnlMpQSr2olPoPpdRKpZTu9p8Ug5Fer+fMmTNMnz6dvLy8bps6dnblyhVOnDjBqlWr\neOSRR8jPz+fq1asAXL9+nRkzZvD444/j4eHB6dOnqaio4NSpU2RkZLBmzRqUUl22ZYyPjzdtQq5p\nGidPnjSt8RVisDFrhKuUmgL8HfAB2v+LWAU8p5S6V9M02TFsiAkPD8fd3Z2CggKSk5NpbGy87We+\n/vprIiIicHNr+59VZGQkpaWlTJo0iVGjRhEQEABAYGAgVVVV1NfXU1ZWxtatW4G2XcI6V7yNHDmS\n4cOHc/XqVWprawkMDMTT09PCTyuEbZg7h/sycBz4l/aCh5sFENuBPwH/xzrhCWtoamqiuLi4wzFN\n03B3d2fKlCkAuLu7A99WfbW/7i9XV1fT70opWltbAYiNjWXu3Lm9fjY+Pp6CggJqa2uJj48fUBxC\n2JO5Uwp3Ab+6tbrs5u//Bkjv6kFm2LBhxMbGdviJi4szJdv+CgkJQa/X09zcTGNjI3q9vtsqt3Zh\nYWGcPn2auro6oK1Qo7q6ust5ERERnDt3jitXrnDnnXcOKEYh7MncEa4RGNnNcd+b7wlBYGAgcXFx\npimChIQExo0bR1VVVbfn+/v7M2fOHLZv346mabi6unL//fd32FgH2kbHoaGheHp62qTjhBDWYm7h\nwzYgkbZ52/bdwpKB14GjmqalWy3C7uORwgcbKSgo6PP+BpamaRpvvPEGP/zhDxk9erRdYxHCFk0k\nf0pbx4ccoH2DUxfgr8CT/bmxGBxunXu1h2vXrvHuu+8SGRkpyVYMen1qk66Umgi077F32l6b2cgI\nVwhhL1bbS0EpNQf4p6Zpt18TZEOScMVA9Gf/Bbj9HgwHDhxAp9ORnJzc52u/9dZbZGRkAG1lzevW\nretPiMIGrLmXwl6gSim1Vyn1rFLqu0op+/6NKcQQ1J5sLaFz8YhwHLebw/0O8D1gNpAJPA/UKaUO\nAvuA/cAxGW4KZ5WdnU1hYSFeXl74+PgQFBQE0G3J8t69e/H19SUxMRHoOCK25Ki2fY2zcDy9jnA1\nTTuvadpWTdN+rGlaMDCFtnY71cBa4Ajwv9YPUwjHU1ZWRnFxMY899hhLly7lypUrAD2WLE+dOpVT\np06ZPl9cXMzUqVPtFb6wA3NXKQCgaZpeKXUduE5b0l0CdO0+KIQTKCkpMZUyu7m5MWnSJAAuXLjQ\nbclyTEwMN27coLa2lrq6Ojw9PfH29rbnIwgbu23CVUr50TalMIe26YVw2vbF/QJ4GDhoxfiEGJR6\nKlmeMmUKp06dora2lqioKLOvZ045tnB8vU4pKKUKgW9oW2tbRdt63NGapt2ladqvNE37b03T6mwQ\npxAOZ8KECaZS5oaGBr766iug95LlqKgoTp06xenTp/uUKM0pxzYaOxZ9mrPhkLCt241wJwKVwEXg\nAnBe07QbVo9KiEEgMDCQqKgotmzZgpeXF8HBwUDvJcv+/v40NDTg4+PTZWe0gSotLcVgMKCUIjc3\nl8DAQMaPH2/Re4iBud063GFAEm3TCXNoK+ctp2064QBwQNO0C9YPs0tcsjBC9Ju11uE6gpqaGt57\n7z3S09PN3jRe9I3Vm0jeciMdbUl3Nm0JeAbwP5qmhfbn5v0lCVeInlVUVDBmzBh7hzFk2aKJZLvW\nW340QAF39OfGQgjrkGTruHqdw1VKudE2ir11SsETKKGt6OHPN/8phM0899xzeHt78/TTTwOwfn3f\nPt/X84WwlNuNcKto2yEsEygDngDCNU0L0zQtQ9O0/9I07RtrBylsr6amhqamph7f37Fjhw2jEWJo\nuF3CfRqI0DRtvKZp/6Jp2luapl2yQVzCzo4fP05ZWVmP7zc3N9/2Gr/97W+JiIggNTWVpUuX8tJL\nLwHw0ksvER0dTUxMDC+//LLp/J6OA2zYsIHJkyeTmprKmTNn+vlUQthXr1MKmqa9YatAxNCSl5fH\nrl27KCoqoqGhgYSEBKZPn05+fj7btm0jNzeXlpYWkpKSmD17Ni0tLd0ej42NJT8/n/fff5/CwkIa\nGxtN1xJisOlTaa9wLgNZCfLPf/6TRYsWMWzYMIYNG8bChQsBOHjwIGlpaXh4eADw0EMPkZ2djaZp\nHY4/+OCD5OTkEBsbS05ODmlpaeh0OnQ6HQ888MDAH04IO5CEK0zOnTvH7t27ATh//jx+fn6MHDkS\npRTLly+3yLffN5fUmF5rmmbqUyZL/cRQ19dlYWIImzhxImvXrmXt2rUsWbKEVatWsXbtWp5++mlT\nsu3cVbe2trbba91111188sknNDQ0UFtby549ewCYNWsWu3fvxmg0UldXx65du0hJSenxOEBqaiof\nf/wxDQ0NGAwGPvnkEyv+WxDCemSEK7rV02gzJyeHK1eu4OLiwubNm0lISGDmzJldzps+fToPPPAA\nsbGxBAQEEBMTg6+vL/Hx8axYsYLExESUUqxevZrY2FgAVq5c2e3x+Ph4fvSjHxETE0NAQAAzZsyw\n3oMLYUV9qjRzFFJpZn3Z2dmEhoYSEhLS5b3Lly+zaNEisrOzGT58eI/XqKurw8vLi/r6elJTU9m6\ndatVOgDLOlxhSzYr7XUUknCtr76+Hp1Oh4tL97NOZ86cYfLkyb1e48c//jHFxcU0NDSwcuVKfvGL\nX1gjVCFsShKuEELYyEASrs3ncJVS/wV8n7YS4avAf2ia9ueb780FXqVtf4YjQLqmaaW2jlEMLjKl\nIAYLe6xS2AiEaZo2EngAeEEpFX+zs8SHwL8Bo2nrKrHTDvEJB/fcc8+ZqtaEGExsnnA1TSvWNK19\na3pF265jdwIPAic1TftI07RGYD0Qq5SaZOsYRe9knwUh+scu63CVUpuUUnXAaeAK8DcgCjjRfs7N\nzhLnbh4XDsQe+yzIXgpiKLDLOlxN0/5VKfUTvt3MvJG27r/lnU6tAaSt6RDTn30WZC8FMRTYrfDh\n5jKDQ0qpf6Ft+8dawKfTab6AobvPr7/lm4/Zs2cze/Zsq8QpumfLfRZaW1tlLwVhNwcOHODAgQMW\nuZYjVJq50dZ6/SSwsv2gUsqLtrndU919aL181WxT9txnofNxIWyp84Duueee6/e1bDqHq5TyV0r9\nSN28UsIAACAASURBVCnlpZRyUUr9H2AJ8DnwMRCllEq72TvtN0CBpmlf2TJG0T177rOQkpLC7t27\nZS8FMejZeoSr0TZ98Bptyb4E+KmmaZ8CKKUeAjYB22lbh7vExvEJM9hjnwXZS0EMBVJpJvpsMO2z\nIISlSWmvsCnZZ0E4M0m4wqlJaa+wpYEkXNmAXAghbEQSrjCblPQKMTCScIXZBlrSW1JSwpQpU1i9\nejVTp07l3nvvpaGhAYCsrCySkpJISEggMzOT1tZWXnzxRV599VUAnnrqKebOnQvA/v37WbZsmYWe\nSgjbkYQrbOrcuXM88cQTnDx5El9fXz788EP0ej07d+7k0KFD5Ofn4+Liwo4dO0hJSSEnJweAY8eO\nUVdXR0tLCzk5Odx99912fhIh+s4RKs3EIDLQLyvDwsKIjo4GYNq0aVy6dInKykqOHTtGYmIimqZh\nNBoJCAhgyZIlHDt2DIPBgE6nY9q0aeTm5pKTk8Mrr7xiiccRwqYk4YpeWbqkV6fTmX53dXXFaGzb\nqXPlypVs2LChy/mhoaG888473HXXXcTExLB//37Onz9PRETEAJ5KCPuQhCt61V7SCz0XPFRXV+Pr\n62t6XVtby4gRI7q9Xncj5O9973ssXryYJ598En9/fyorKzEYDISEhJCSksKLL77I22+/zdSpU3nq\nqadkpzAxaMkcrjBbbyW9b7zxBo2NjWzevJmTJ0/2eA2lui5fjIyM5IUXXmDevHnExsYyb948rl69\nCkBKSgpXr14lOTmZsWPH4unpSWpqqmUeSAgbk8IHYTZLlPQKMdhJpZmwCUuU9Aox2EnCFUIIGxlU\nbdKFsLeLF9f3+TNhYX3/jBCdyZdmQljAgQMHOHz4cJ8/ZzQayc3NNb2uqqritddes2RowoFIwhXC\njoxGI3l5eVa9R1FRkVWvL8wnUwpC9FN2djaFhYV4eXnh4+NDUFAQAIWFhRw9epSWlhaCg4OZP38+\nSil27txJTU0Nzc3Npn0j9u7dS2VlJa+//jrh4eEkJibS2trKJ598wtdff42Pjw9LlizBza3//6m2\ntrZa6pHFAEnCFaIfysrKKC4u5rHHHqOlpYU33niDoKAgKioqOHXqFBkZGbi4uPDpp59SVFRETEwM\nixYtwsPDg+bmZrZu3cr/396dx0V13osf/zwwgBRliRqCbNEY17BoxOgrxo1qk5g0JWkaa281IbWG\n9JXb/mKXa5fbpG2ut/21vemiYE1N02gSs6k18XqbmhL154qIoIC7oHFyDUUEZJfz++OcmQ7DDMwM\nwyzwfb9evGDOOXPmOWeG7zzzzPk+34kTJ5Kdnc2VK1dYsWIFoA8p1NbW8sUvfpEHH3yQt99+m4qK\nCms6tAhuEnCF8EBVVRUTJkzAZDJhMpkYN24cAOfOncNsNrN+/XpAn0HNknV34MABKisrAX2qy9ra\nWqKiorrtOy4ujvj4eAASEhKoq6vzxSEJH5CAK4SXZWRkWKeStLhw4QLnz5/na1/7GiaTiVdeecXp\ndJahoaHWv5VSDocE2tvbKS8v77JM0zTCw8OZNGmSF45C9Af50kwID6SmplJZWUlHRwetra2cOnUK\n0GdDq6io4Pr164CeLHLt2jVaW1uJjIzEZDJRU1PDpUuXAH0yn7a2NrcfPywsjIyMjC4/mZmZXYKt\nZWIgC08eR3iX9HCF8EBCQgKTJ0+moKCAqKgoEhMTARg5ciTz5s1j48aNaJpGaGgo999/P2PHjqWo\nqIi1a9cyfPhwkpKSAIiMjCQ5OZn8/HzGjh1LVlaW19pYXV1NQ0MDSikOHz5MQkKC9XGFf0immRAD\nWH19PW+88QZPPPEEYWFh/m7OgCCpvUIIp2pqatyet1g4JwFXCDe4m9orab3ClpRJFyJAHDx4kDVr\n1rBlyxa37rdhw4ZuyzxNFxaBS740E8KLioqKWLp0KcOGDeuyXNM0h5OvW+Tm5vZLe8rKyiRpIoBI\nwBXCA47SdN977z2uXr3Kpk2byMzMpKioiKSkJMxmM0uWLGHnzp3d7mOxevVqVq1a5TRd2FOS1htY\nJOAK4QFHaboPPPAAZ8+eZdmyZbS2tvLXv/6VnJwc6yVjju4TGRlp3aezdGExcEjAFcIDjtJ0LYHV\nIjY2tsuy3u5TXV3tMF1YDBwScIVwk6tpurbXvbp6H1euvpG03uAlAVcINzlL0+3rfVJTU9m6dSv3\n3HMPN27c4NSpUw5LwlvSenvS0tLCkCFDrLfb2toIDw934ehEf5KAK4SbnKXp9vU+t9xyi8N0YU9I\nWm9gksQHIQYoSevtH5JpJoRwSNJ6vU8CrhBueO65/t1eDGyS2itEgJg1a5a/myACmARcIbxo7969\nDpc7+0RWX19Pe3u70/0dP37cK+0SgUECrhAeyMnJISsri7S0NF566SXrcsscCpaaZ8uWLSMtLc3p\npWNHjx7FbDY7XPfxxx+zdOlSmpqavH8Awi/ksjAhPPDyyy8TGxtLS0sLWVlZPPLII8TFxXWZoObM\nmTO8+uqrHlVx2Lp1K2azmby8PAoKCsjKyuKee+7x5iEIP5CAK4QHXnzxRbZu3QrApUuXOH36NNOn\nT++yTWpqqkvB1tFww4IFC4iKiuKVV17h2WefpbGx0TsNF34lAVcIN3300Ud8+OGHHDx4kIiICObN\nm9etYCPgsAQ66D3fbdu2AXD27FmGDx9ObGwsSimWLl3KiBEjrPe19JgtpdZFcJOAK4Sbrl27Rlxc\nHBEREVRWVnLgwAHrOtveqrMvysaOHcvKlSsB2L17N7feeispKSn922gREORLMyHcdO+999Le3s7k\nyZP5/ve/z8yZM63rbMdwe5pw3EKuJx9cpIcrBp2+JjKEh4ezY8cOh+vq6+sBffy2tLS01331FpQl\nIA8sEnCF8KOsrCwiIiKcrpcZvgYWSe0VQgg39CW1V3q4YtBxp0x6byXSCwsLiYiI6DKO6wlv7UcE\nNvnSTIgBoqyszN9NEL2QHq4QbnJWWbe0tJRDhw5x48YNEhMTWbRoEUophxV+e9qPp6RCb+CTgCuE\nG5xV1q2pqeHEiRPk5uYSEhLC+++/T1lZGenp6Q6r9dbV1UmF3kFIAq4QbrBMSmNfWffcuXOYzWbW\nr18PQEdHhzU7zFG13osXL0qF3kFIAq4QXpKRkUF2dnaXZa5W63VGKvQOLPKlmRBuSE1NpbKyko6O\nDlpbWzl16hQAo0ePpqKiguvXrwPQ3NzMtWvXnFbrdbYfe5YKvbY/mZmZXYKt/TwObW1t/XHowguk\nhyuEGxISEhxW1h05ciTz5s1j48aNaJpGaGgo999/v9Nqvc724wmp0Bs8JPFBDDrevA43UEiFXt+R\nIpJCCKnQ6yMScIUQwkcktVcElLq6Ol5//XXy8vJ8+riuDhUEyzCBGHjkKgXRoytXrvi7CUIMGNLD\nFU7V19ezZcsWcnNznX4R097ezltvvUVDQwOdnZ3Mnj2bxMREOjs72b59OxcvXiQ6OprFixdjMukv\nN0eprnV1dWzatImkpCQuXrzIqFGjyMzMpLCwkKamJh5++GFGjRrlNH1WiGAgAVc4VFlZSUNDA9Om\nTaOoqIhRo0aRmprabbszZ84wbNgwlixZAkBrayvNzc3U1tbyxS9+kQcffJC3336biooK0tLSABym\nugLU1tbypS99iZEjR/KHP/yB48ePk5uby8mTJ9mzZw/Z2dlO02eFCAYScIVDY8aMITw8nJKSEmbO\nnOn0Yvqbb76Zv/71r/ztb39j3LhxpKSk0NzcTFxcHPHx8YB+zWldXZ31Po5SXaOiooiLi2PkyJHW\n/Y4ePdr6d11dXY/ps0IEAwm4g5Ar6aKWSgOWj+vOKg8MHz6cFStWcPr0aT788EPGjBlDeno6oaGh\n1m2UUtaZrHpKdbW/j2UIwvb+jtJnhQgWEnAHIUu6qDc0NDQQGRlJWloaERERHD16tMeP+M5SXV0x\nZswYXn/9dWbMmEFUVBTNzc20tbURExPjjUMRot9JwBV9cuXKFT744AOUUoSGhrJo0aIet3eW6uqK\nESNGMH/+/G7psxJwRbCQxAfRo5KSEjIzM/3dDJfIdbjCFyTxQfQb23HVQCeBVAQ66eEKIYQbpIcr\nhBsG4mxhIjhIaq8QbqirqyM/P9/hug0bNgD6hOCHDx+2Lre/LQYvCbhCeElubi6gB9iioiLrcvvb\nrurLsJmUTA9MMqQghJs6Ozt59913MZvN3HzzzeTk5GAymVi9ejWrVq1i165d1NbWsm7dOsaMGWPN\nprPcXrBggcM5Ia5du8bGjRtJSkrCbDazZMkSjy95k5LpgUkCrhBuqqmp4aGHHiIpKYlt27Zx+PBh\nZs6caV2fnZ3NlStXWLFiBaAPQ9jedlZSPSUlhdraWnJycvpUckcELgm4QrgpJibGmrCRnp7OoUOH\nugTc3jibEyIlJYXY2FgJtgOYBFwh3GQ/HaQn00M6mhOirq6ux2kwpVx68JOAK4Sb6urquHTpEklJ\nSdahAFsRERFdZlezvz169Gg2b97cbU6Inrg6/0VLSwtDhgyx3m5ra3M68ZDwPQm4QrhpxIgRHD58\nmG3btnHzzTczbdq0LusjIyNJTk4mPz+fsWPHsmDBgm63HZVUj4qK6nPbpGR6YJNMMyEGGCmZ3r+k\naq8Qogspmd5/JOAKgcwWJnyjLwFXMs2ETx08eJA1a9awZcsWp9u4kwq7evVqbzVNiH4nAVd4jSsl\n1YuKili6dCk5OTlOt/E0FVaIQCdXKQivcKWk+nvvvcfVq1fZtGkTmZmZVFVVdSuXDrBr1y6uXr3a\nJRXWUWl1IYKNBFzRZ66WVH/ggQc4e/Ysy5YtIzIykszMzG7l0iMjI7ulxoLj0uqRkZG+PEwh+kwC\nrugzV0uq23NULt1ZWqs72woRqCTgih55s6S6rZ7KpfdlWyECmQRc0SNvllS31VO5dPtU2L6UVhci\nkEjAFX7RU7l0+9TY+fPne1xaXYhAIokPwmuCqaS6EJ6SxAcREIKppLoQ/iA9XCGEcIOUSRd+UVdX\nx+uvv05eXp5f92EhcymIQCdDCsIpV1J1XSGfRoTQSQ9XOORKqi44rmDb2NjYrfpseXk5JSUlAEyZ\nMoUZM2Z029fVq1d58803efDBB6mpqXFY1XbTpk2kpKRw8eJFoqOjWbx4MSaTvIxFcJAeruimsrKS\nkydPWlN1q6qqnG5bU1PD9OnT+cY3vkF4eLh1lq/a2lqysrLIy8ujqamJY8eOsXz5cp588kmKi4v5\n5JNPuuznH//4B2+++SY5OTmEh4dbq9quWLECpRRlZWXW/U6fPp2nn36aIUOGUFFR0X8nQggvk66B\n6MadVF1HFWwnTpzYpfpsdXU1EyZMsPZEJ06cSHV1NbfccgsA169f54033uCxxx5jxIgRHDp0yGlV\n27i4OOLj4wFISEigrq6u386DEN4mAXeQ8XaqrrMKtu6UdhkyZAgxMTFUVVVZqxQ4q2pre+mZUorO\nzk6XH0cIf5MhhUHGkqpr+5OZmelxqW1LBVuAsrIykpOTu22TkpJCZWUlHR0dtLW1UVlZ2aXSbWho\nKI899hilpaWUlZUxevRoKioquH79OgDNzc1cu3bNo/YJEUikhyv6xL6CbVZWFo2NjV22SUhIIDMz\n0zpEMHXqVOtwgkVYWBhf/vKX2bhxI7Nnz+63qrZC+JMkPgingi1VV67DFb4giQ+iXwRbqq4EUhHo\npIcrhBBuCMoerlLqdqAUeEvTtKXGsmzg90AycBB4QtO0an+1UQxMMvQg/MWfVyn8HjhkuaGUGgG8\nA/wAuAk4Amz2T9OE8MyGDRuA7qXe6+rqyM/P7/P+3SkhLwKPXwKuUmoxcBXYZbM4Bziuadq7mqa1\nAc8BGUqpcX5oohAeyc3NBfqv1Ls392vJ3hO+4/OAq5SKBp4HngVsx0EmA8csNzRNawLOGMuF8Lt9\n+/Zx6JD+oWznzp38+c9/BuD8+fNs2bIFgNWrVwNdS71/8MEHgD7vxPbt21m7di0bN2601mXbv38/\n+fn55Ofnc+DAAaB7j3jfvn0UFhaya9cuamtru+zXU5I04nv+6OH+BFivadplu+VDAfur2+uBYT5p\nlRC9SElJobpa/0rBbDbT1tZGZ2cn1dXVXRI5ALKzs4mLi2PFihUsWLAAcDwPhNls7nWeCQulFJ/9\n7Ge56aabuuxXBA+ffmmmlMoEPgs4urizEYi2WxYDNDja13PPPWf9e+7cucydO9crbRTCmVGjRnH5\n8mVaW1sxmUwkJCRw+fJlqqurue+++3q9v6N5IJqamhzOMzFunOORNLk6x/cKCwspLCz0yr58fZXC\nHCAVqFZ60v1QIEQpNQkoAB63bKiUigJuA0442pFtwBXCF0JCQoiNjaWkpITk5GTi4+M5f/48tbW1\n1jkgeuJsHghHQTQkJKTLclfKwrsyT4Zwn32H7vnnn/d4X74eUliHHkQzgQz0IPs+sBDYCkxWSuUo\npSKAHwMlmqad8nEbhXAqJSWF/fv3k5qaSkpKCkeOHCEhIaHbdval3nva38mTJ7vMM5GamsrQoUO5\nfv06zc3NdHR0cPr06V736+o8GS0tLV1uu9JO4R0+7eFqmtYCWJ9tpVQj0KJpWq1x+xFgDbAR/Trc\nxb5snxC9SU1NZe/evSQlJREWFobJZOo2fgvdS71nZWU53F9CQgIZGRld5pmwDDvMmTOH9evXEx0d\nbe1B2+/Xk3Hc6upqGhoaUEpx+PBhEhISpPS8j0immRCDUH19PW+88QZPPPGEW1Npir5lmknAFWKQ\nqqmpcWnsWXQlAVcIgytpu5KyK/qiLwFXJiAXQceSPgv/TDQQIhhIwBX9ylul1m1Z0meFCDYScEW/\nsZRab29v73G70tJSXnrpJdatW8d7772Hpmn87W9/6zJJS2FhIfv37wekVyuClwRc0S9cLbVeU1Pj\nsCT6HXfcwYkT/8x5KS8v54477vBV84XoF1LxQfQLV0utnzt3zmFJ9PT0dJqammhsbOT69etERkYy\nbJhMqyGCmwRc4TZvl1p3VBIdYNKkSZw4cYLGxkYmT5ZJ40Twk4Ar3GZJIfWG0aNHs3nzZmbMmEFU\nVBTNzc20tbURExPD5MmT2b59O01NTTz++ONeeTwh/EkCrvCrkSNHOiyJHhMTw8iRI2ltbSU6Opqh\nQ4f6u6lC9JkkPoh+FWyl1oXojSQ+iIAVbKXWhehP0sMVQgg3BGWZdCH8xdW56x1tN2zYMBoaGpg1\naxZ79+4F4Le//S0FBQXceeedvPrqq15rpxh4JOAK4QbLZW6WYAuQn5/Prl27GDVqlNv7q6+vJzIy\nsscpEo8fPy5JHwOEjOEK4QFLEkZeXh7nzp3jvvvu4ze/+Q0AmzZt4q677mLq1Knk5eX1WIfs6NGj\nmM1mp+s//vhjli5dSlNTk3cPQPiFBFwhPGDp6ebn55OYmEhhYSHf/OY3qaysZPPmzezbt4/i4mJC\nQkLYtGmTR4+xdetW/vKXv5CXl0dBQQF79uzx5iEIP5AhBSE8YNtr1TTNenvXrl0UFxeTlZWFpmm0\ntLRYS+a4si9bCxYsICoqildeeYVnn32WxsZG7x2A8AsJuEJ4kaZpLFu2jBdeeMHpNmfOnGHbtm0A\nnD17luHDhxMbG4tSiqVLl1qrMERFRQH/7E1L8kfwk4ArhBdlZ2fzhS98gW9961uMHDmSq1ev0tDQ\n0KXQ5NixY1m5ciUAu3fv5tZbb3VYiFIMPDKGK4QHLL1O+78nTpzIz372MxYuXEhGRgYLFy7kk08+\ncbofuZ58cJEerhBuqK+v7/Ib9CkmbT366KM8+uijLu3PNlg7I0F54JCAKwYdVxMffCErK4uIiIge\nt+lpaksRXCS1Vwgh3CCpvUK4oS+pvfaqqqp44IEHKCsr60uTxCAhX5oJ0UeujMO6or6+vseCm8eP\nH/fK4wj/kYArhIuampp44IEHmDJlCunp6bz11luAXoft61//OnfccQf33nsvra2tgHspvtBzmq+k\n+A4MEnCFcNHOnTtJTEzk6NGjlJaWcu+99wJw+vRpnnnmGY4fP05MTAzvvPOOpPgKh2QMVwgXpaWl\n8e1vf5tVq1axaNEiZs2aRW1tLWPGjCEtLQ2AO++8kwsXLnD16lWOHDniVoovOL4ETFJ8Bw4JuEK4\n6Pbbb6e4uJgdO3bwwx/+kM9+9rN89atf7XJZV2hoKC0tLQA8/vjjPab4gmtpvpLiO3BIwBXCRWaz\nmZtuuoklS5YQExPDH//4R7761a867JXOnz+/1xRfkDTfwUYCrhAuKisr4zvf+Q4hISGEh4dTUFAA\nOL5KwTbFt7Ozk/DwcNasWdNjMJVrywc+CbhCuGjhwoUsXLiw2/LS0lLr35beKriX4gu9X14mATn4\nScAVg04gpfba6i3NV1J8g5+k9gohhBsktVcIw/nzz/W6zejRvW8jRH+QxAfhtsLCQvbv3+/vZgCB\n1RYheiMBV3DlyhWv7EeGeYTomQwpDHL19fVs2bKF3NxcwsLCnG63e/duSktLiYqKIjo6mlGjRlFX\nV8fGjRtJSkrCbDazZMkSysvLKSkpAWDKlCnMmDEDgLq6OjZt2kRCQgJms5mbb76ZnJwcGhsbHS43\nmUyUlpZy6NAhbty4QWJiIosWLUIp5bAtQgQD6eEOYpWVlZw8eZJp06ZRVFREVVWVw+3MZjPl5eU8\n9dRTLFmyhMuXL1vX1dbWkpWVRV5eHk1NTRw7dozly5fz5JNPUlxc3KW8TE1NDdOnT+cb3/gG4eHh\nHD582OnympoaTpw4QW5uLitWrEApRVlZWY9tESLQSQ93EBszZgzh4eGUlJQwc+ZM2traHG5XVVXF\nhAkTMJlMmEwmxo0bZ10XGxtLYmIiANXV1dbtQL/4v7q6mltuuQWAmJgYkpKSAEhPT+fQoUNMnDjR\n4fLQ0FDMZjPr168H9Bm5hg4dSlNTk9O2CBHoJOAOUO3t7ZSXl3dZpmka4eHhTJo0CfjndZ2WC+49\nuc6zp2GI3ji70N+yPCMjg+zs7C7rDhw44PHjCeFvMqQwQIWFhZGRkdHlJzMz0xps3ZGamkplZSUd\nHR20trZy6tQph9ulpKRYt2tra6OysrJLKuu1a9e4dOkSoKfJJicnO10+evRoKioquH79OgDNzc1c\nu3bN5bYIEYikhyt6lZCQwOTJkykoKCAqKso6hOBou8zMTOswwNSpU63DCQAjRozg8OHDbNu2jZtv\nvpmsrCwaGxsdLjeZTMybN4+NGzeiaRqhoaHcf//9JCYmutQWIQKRZJoJSkpKyMzM7NfHqKur4/XX\nXycvL8+l5UIEqr5kmsmQgiA0NNTfTRBiUJAerhBCuEHmUhCD2urVq1m1ahUgcymIwCZDCsIrvJEe\nLJ9axEAnPVzRZ66mBx87doz9+/ejlCI+Pp65c+d2Sw3WNK3Ll2j79u2jvb2dOXPmOE31FSJYSMAV\nfVJZWUlDQ4M1PXjUqFGkpqZ22+7TTz9lz549PPnkk0RGRtLS0kJLSwu1tbXk5ORYL++qq6tz+Di2\nqb4hISG8//77lJWVkZ6e3q/HJ4Q3ScAVfeJqevD58+eZNGkSkZGRAAwZMoSWlpYuqcE9OXfunMNU\nXyGCiQRc4ZQv0oPthyBCQkK6jOV2dHRY/3aU6itEMJEvzYRT3kwPHj16NOXl5TQ3NwNYf9sbOnQo\n169fp7m5mY6ODk6fPm29v6NUXyGCifRwhU+MHDmSe+65hz/96U+EhIRwyy23MGfOnG7bhYSEMGfO\nHNavX090dDQjRoyw3t9Rqm9MTIyvD0UIj0nig/AKX6QHu0KuwxX9TRIfhN8FSnqwBFMRyKSHK4QQ\nbpAerhBucGXYAfreW25paaGsrIysrCyg55nRNmzYQG5ubp8eTwQ+uUpBiH7S0tJCUVGRS9v6K9iW\nlZX55XEHK+nhCuEiS+XhlJQULl68SHR0NIsXL8ZkMrF//35rteKpU6dy1113sWvXLq5evcq6desY\nM2YMWVlZdHZ2sn379m73t0zA09NjAHz00UeUlZV1qVg8c+ZMj4+ps7PTK+dGuEZ6uEK4oba2lunT\np/P0008zZMgQKioqMJvNXaoVHzlyhE8++YTs7Gzi4uJYsWIFCxYscHp/Vx4D4PLly1RWVpKXl8dX\nvvIVqVgchCTgCuGGuLg44uPjAb2kUF1dXZdqxeHh4dZqxa7e39VtqqurGT9+PKGhoYSHh0vF4iAk\nQwpCuMH28jellPUjuatXzTi7v7vbOONKOrbwH+nhCtFHKSkpnDx5slu14oiICKeT+Xj6OKdOnbI+\njiXt2Zar6dgtLS1dbnuzncI56eEK0UcJCQlkZGQ4rFacnJxMfn4+Y8eOtV4e5qlRo0Yxfvx4CgoK\nGDp0KPHx8URERHi0r+rqahoaGlBKcfjwYRISEkhKSupT+0TvJPFBDDq+ug63P7S1tREeHk57ezt/\n+tOfePDBB7uUondHfX09b7zxBk888USPE8eLrvqS+CABV4gg8u677/Lpp5/S0dFBZmYmd999d5/2\nV1NTY50gSLhGAq4QQviIpPYKYZDZwkQgk6sUhE8VFhayf/9+fzdDCL+QgCv6xBvl0YUYLGRIQXjM\n1fLou3fvprS0tEv+P+Bw/oF9+/ZhMpmYPn06O3fu5MqVKyxdupTz589z9OhRHn744V7nGxAiUEkP\nV3iksrKSkydPWsujV1VVOdzObDZTXl7OU089xZIlS6z5/87mH0hJSbGmxZrNZtra2ujs7KS6urpL\n+XVX5iQQItBIwBUescx+FRoaysyZM0lISHC4XVVVlXWegYiICGv+v7P5B0aNGsXly5dpbW3FZDKR\nlJTE5cuXuwVcV+YkECLQyGcw0Y0vyqNb9mkvJCSE2NhYSkpKSE5OJj4+nvPnz1NbW9vletG+zDcg\nhL9ID1d0483y6KmpqVRWVtLR0UFrayunTp0CnM8/YFm3f/9+UlNTSUlJ4ciRI0570EIEE+nhG4Ab\nJgAAFqxJREFUin6VkJDA5MmTKSgoICoqisTEROtyZ/MPpKamsnfvXpKSkggLC8NkMlmDsRDBTDLN\nRJ8ESnl0IXylL5lmMqQg+iRQyqMLEQykhysGlOee8842QjgjPVwRMJ5//nl+/etf97jNsGHDrH/P\nmjWrx22rqqpIS0tzuO7atWvk5+e730gh/EQC7iBXX19Pe3u70/Wvvfaa1x/TcikZwN69e93a3tbV\nq1dZu3at19olRH+TgDvIHT16FLPZ7HR9R0dHr/t44YUXGD9+PLNnz+bkyZPW5Zs2beKuu+5i6tSp\n5OXlWa+7tR0Osu3t/vSnP2XChAnMnj2bJUuWWHvKHR0dfP3rX+eOO+7g3nvvpbW1FYBVq1Zx7tw5\npk6dyve+9z33DlwIP5CAK/qkuLiYN998k9LSUt5//30OHz4M6Km/mzdvZt++fRQXFxMSEsKmTZu6\n3d/Sez18+DBbtmyhrKyMHTt2UFRUZN3m9OnTPPPMMxw/fpyYmBjeeecdAP7zP/+T2267jeLiYn7+\n85/74GiF6Bu5Dle4XHHWkT179pCTk0NERAQRERF8/vOfB2DXrl0cOXKErKwsNE2jpaWlx1Iw+/bt\n46GHHiIsLIywsDAefPBB67oxY8ZYx3HvvPNOLly44HF7hfAnCbiD0JkzZ9i2bRsAZ8+eZfjw4cTG\nxqKUYunSpV4rufL444/zwgsvdFvubEzWGdtCiaGhod0qzgoRLGRIYRAaO3YsK1euZOXKlSxevJjl\ny5ezcuVKnn32WWuwvXbtWpf7NDY2OtzX7Nmz2bp1K62trTQ0NLB9+3YA5s+fz9tvv82nn34K6F9w\nXbx4Eejao7b8fffdd7N9+3ZaW1tpbGzkvffe67aNvWHDhtHQ0ODJKRDCL6SHO8g5C2Z79uzh8uXL\nhISEsHbtWqZOncqMGTO6bTdlyhQee+wx0tPTiY+PZ/r06QBMnDiRn/3sZyxcuJDOzk7Cw8NZs2YN\nycnJXXq4lr+nTZvG5z//eTIyMoiPjyc9PZ2YmJgu29i76aabuPvuu0lPT+e+++6TcVwR8CTxYZDb\nvXs3t956q8O5Cj7++GMeeughdu/ezWc+85l+b8v169eJioqiubmZ2bNns379ekkbFgFHqvYKjzU3\nNxMREUFIiOPRpZMnTzJ+/HiftOUrX/kK5eXltLa28vjjj/Pd737XJ48rhDsk4AohhI9ImXQh3OCL\n+RZmzZplzaKTL/eEhVylIEQ/sE1Z7u0yuN7Sq48fP+61dgn/koArhAuampp44IEHmDJlCunp6bz1\n1lsA5OTkkJWVRVpaGi+99JJ1e9uU5d70lF798ccfs3TpUpqamvp2ACIgyJCCEC7YuXMniYmJ1uuD\nLUMEL7/8MrGxsbS0tJCVlcUjjzxCXFyc28kdjmzduhWz2UxeXh4FBQVkZWVxzz339Hm/wn+khyuE\nC9LS0vjggw9YtWoVe/futfZgX3zxRTIzM5kxYwaXLl3i9OnTgPvp0o62X7BgAXl5eYSHh/Pss88y\nZcqUvh+I8Cvp4Qrhgttvv53i4mJ27NjBj370I+bPn8/s2bP58MMPOXjwIBEREcybN8/ltGNX0quj\noqKAf44BDx06tH8OTviMBFwhXGA2m7nppptYsmQJMTEx/PGPfyQzM5O4uDgiIiKorKzkwIEDDu/r\nqPdqSa+GnpNPxMAiAVcIF5SVlfGd73yHkJAQwsPDKSgoYPLkyeTn5zN58mTGjx/PzJkzrds7Sl92\nRq4pHzwk4ArhgoULF7Jw4cJuy3fs2OFw+/r6eod/OyIBefCQgCsGnUArIpmVldVlCkp74eHhPmyN\n6E+S2iuEEG4IqtRepVQhcBfQDijgkqZpE4112cDvgWTgIPCEpmnVvm6j8I66ujpef/118vLyXNq+\npaWFsrIysrKy3HqcwsJCIiIirGOo588/1+P2o0f3vF6I/uKP63A14GlN06I1TRtmE2yHA+8APwBu\nAo4Am/3QPuGiK1eueHV/LS0tXWqZCTHQ+GsM11F3/GHguKZp7wIopZ4DapRS4zRNO+XLxone1dfX\ns2XLFnJzcwkLC3O6XWdnJ9u3b+fixYtER0ezePFiTCYTmzdvpr6+no6ODmtl3127dnH16lXWrVvH\nmDFjWLBgAaWlpRw6dIgbN26QmJjIokWLUEqxe/duSktLiYqKIjo6mlGjRvnw6IXwjL8yzVYrpa4o\npfYopeYYyyYDxywbaJrWBJwxlosAUllZycmTJ5k2bRpFRUVUVVU53ba2tpbp06fz9NNPM2TIECoq\nKgB46KGHWL58OcuXL+fgwYM0NzeTnZ1NXFwcK1asYMGCBdTU1HDixAlyc3NZsWIFSinKysowm82U\nl5fz1FNPsWTJEi5fvuyrQxeiT/zRw/0uUA60AV8G/qKUygSGAvafUesB12cBET4xZswYwsPDKSkp\nYebMmbS1tTndNi4ujvj4eAASEhKoq6sD4MCBA1RWVgJ6b7m2ttaaWWVx7tw5zGYz69evB6Cjo4Oh\nQ4fS1NTEhAkTMJlMmEwmxo0b1x+HKYTX+Tzgapp22Obmn5VSi4FFQCMQbbd5DOBwItHnbK7tmTt3\nLnPnzvVqOwer9vZ2ysvLuyzTNI3w8HAmTZoE/PMyJcv1oz1dthQaGmr9WylFZ2cnFy5c4Pz583zt\na1/DZDLxyiuv0NHR4fD+GRkZZGdnd1nmLKNLiP5QWFhIYWGhV/YVSNfhngCWWW4opaKA24zl3TwX\naBdTDhBhYWFkZGT062O0trYSGRmJyWSipqaGS5cuAXo5dNve8ujRo9m8eTMzZsyw1jpra2sjNTWV\nbdu2MWvWLG7cuMGpU6eYNm1av7ZZDF72Hbrnn3/e4335NOAqpWLQLwn7COgAFgP3AM8A14BfKKVy\ngB3Aj4ES+cJs4Bk7dixFRUWsXbuW4cOHk5SUBEBkZCTJycnk5+czduxYFixYwLx589i4cSOaphEa\nGsr9999PYmIikydPpqCggKioKBITE/18REK4xqeJD0qpEejBdDxwA6gEfqhp2ofG+vnAGiAF/Trc\nxx1dhyuJD4GhpKQkIKvqynW4oj9JEUnhF2VlZaSlpfm7GUL4lARcIYTwkb4EXKn4IIQQPiIB1w+8\ndYmJv0j7/UvaH7wk4PpBsL/gpP3+Je0PXhJwhRDCRyTgCiGEjwTtVQr+boMQYvAaVJeFCSFEMJIh\nBSGE8BEJuEII4SMScIUQwkck4AohhI8EVcBVSr2qlDIrpeqUUpVKqSft1mcrpSqUUo1KqV1KqRR/\ntdWeUipcKfWSUuqCUuqaUqpYKXWv3TYB234ApdQ3lFKHlVItSqkNDtYHevvjlFJbjPadV0p92d9t\n6klP5zvQzzX0/poPkmNwGnM8ar+maUHzA0wChhh/jwPMwBTj9nCgDr0YZTjwC2C/v9ts0/bPAP8O\nJBu3F6GXEEoJhvYbbfwC8Hn0KTQ32K0Lhva/bvxEAncb7Z3o73a5e76D4Vwb7XT6mg+iY3AYczxt\nv98PqA8nYjxwGfiicXs5sNfuyW4Cxvm7rT0cwzEgJ9jaD/zUQcAN6PYb7WkFbrNZ9grwH/5um7vn\nO9DPdS/HcgzICcZjsI05nrY/qIYUAJRSa5RS14EK9IPfYawKqqq/Sql49HfM48aioGq/A4He/nFA\nu6ZpZ22WHSNw2ueOQD/XDhmv+dvRy2YFzTE4iTketT/oAq6mad9Ar/A7C3gXvdeCseya3eYBWfVX\nKWUCNgIva5p22lgcNO13ItDbPxS9PbYCqX3uCPRz3Y3Na/5Pml42K2iOwUHMacPD9gdMwFVK/V0p\n1amUuuHgZ7fttppuH5AM5BmL3ar6622utl/ppW43or9RPGOzi6Bofw/82n4XBHr73BFUx+LkNR9U\nx+Ag5njU/oAJuJqmzdM0LUTTtFAHP7Od3M2EXtkX9I8p1gJbvVX99TY32v9HYATwsKZpN2yWB0v7\nnfFr+11wCjAppW6zWZZB4LTPHYF+ru05es0H2zFYmIAx6EOB7rff3wPRbgxYjwQeA6LQ3yg+h/5u\nsshYPwK4ij4gH4H+reE+f7fb7hgKgH3AZxysC4b2hwJDgP8A/my0MzSI2v8asAn9C45ZRnsD+SoF\nh+c7GM61zTE4fM0HwzH0FHM8bb/fD8qNgx8BFAK16JdjHANy7baZjz6wfR34EOOSq0D4Qb8UphP9\nm8wG46ce+HIwtN9o34+NY7hh8/PvQdT+OGAL+sfBC8Bj/m6Tp+c70M+10cYeX/OBfgy9xRxP2i+z\nhQkhhI8EzBiuEEIMdBJwhRDCRyTgCiGEj0jAFUIIH5GAK4QQPiIBVwghfEQCrhBC+IgEXAGAUup3\nSqm/+7sdgc6Yb+Jhf7fDW5RSLyul/uLONkq3TilVY8y14Urqt+3+YpVSnyilRveh3eFKqSql1FRP\n9+EPEnB7YbzYLJO6tCmlziql/q9S6jP+bls/kCwYQw+B6BZgu6/b44gPg/+/Av9ic/t+YBl6iust\nwH5j8qPfuri/HwDva5p2HqyVOLYrpRqUUkeUUhm2GyulfqWU+pntMk3T2tDTaX/h2SH5hwRc13yA\n/sIajf5ieZo+PNFKqTAvtcvdx1VKKXnO+0DTtCuaprX7ux2+pGlag6ZptlNb3g6YNU07qGnap+6c\nD6VUJPAk8JLN4h+gz1cwBfjIdp1S6k7gXuAnDnb3GjBLKTXR5YPxN3/nKwf6D/Ay8Be7ZeuASza3\nJwHvoeeJ/y/6CyHebh/bge8CF4FPnDzWZeBLNrf3GvsMMW7fhp6bPsq4HYtetaAWPV/9A2CSzf2X\noeev3weUoc/jOQn9jfaXxv3+AfwXsBb4sIfzEIL+j3DOeKxTwHds1i9An34vzu5+/wGU2NzOBarQ\n5zN4F32qu043n5MkYKvR9utAud15GwW8YRxfrfHcjLVZ/2PjfDyGPml0PfocCzfZrLfMYWD5PdtY\n14k+6xVAqnH7MfSc+yagGEhDn4j6/xnHuQdItTuGB4EioBk4C/wMCLNZfx49EBWgz7t6Efi23XpL\n+zqBcz2crxXASeOxPgX+2+Y19TLwF/Re7CXjfG3AKCtj/z9g/G17bs45WHYDJ/MKoFdLqLFb9j7w\ndePvCUCD8XcocMRy7p3sbxfwE3/HCVd/pLfjmVb0GYJQSiWgvyuXAtOAbPR3621295mD/o/4OWMb\nRz4C5hr7jTT212L8xlh3RtO0y8btV4As9H/eLPR/+J1KqQibfQ4Bfgh8HT3YVgHfRu9lLAdmor+w\nv9LLMYeg/0N+Ef2f4vvAKqXUE8b6Xej/zI/a3e/LwKvGMc0E1gO/Q5/a7n3gOdwfyshHr0s2xzim\nb6FPLmI5b39HD8T3ADPQ38j+ppQaYrOPW4EvAQ+hv1lMAV4w1v0SeBP4GxAPJKDPeOXMc8Bq45jq\n0Oum/RZYhf68DDFuY7Txc+jzw/4WmIj+JvSIzeNbfAv9dTUF+DnwC6XUXca6LEChP4+3GLe7MXqI\nv0d/ExmHPuHKTrvNZqO/QWQb5yQH+KaTY/1X9N7mJfRzk2Us248eeC3n66KT+89CD6K2jgHzlVKh\n6L3ZUmP5SqBY07Se5mM+hP46CA7+jviB/oNdDxeYDtQArxm3fwJ8YHefOPR3+mk2+/hfwNTLY60A\nKoy/s9Hn1twAfM9Y9irwB+Pv243HuNvm/tHo//C5xu1l6L2NTLvH+Rj4N5vbCr0H5LSH66S9q4G/\n2tz+FfCRze1ZQDuQYNx+Ddhht491wA03H/cY8CMn63KBk3bLQo3nzFL/7sfob05Dbbb5PnDK2fNu\ns9xRD/drNusXGcseslm2DKi3uf0R8AO7/T6E0bMzbp8HNtltcwr4vqO29HCuctCnEYzq4fVdBfpE\nVsayP9g9r/b/Ayux61Gjv8n91oXnbgt6pRPbZdHo02ZeMPYzHn3O2bPoUyT+Dv2TyHZsPjka930G\nqHLn9ePPH+nhuuY+Y0C/Gf1j4t/R39UBpgJzjPUNSqkGoBq912Y72fVxTdM6enmcQmCcUftprvE4\nhcbfoL+TFxp/T0APpgcsd9b0cbYy9F6fRQc2tZeUUtHoPRDb+2nAwV7ahlLqKaWX7b5iHOf/QZ+C\nz2IjcLdSKtm4vQQ9AJtt2nzIbre9Pq4DvwF+pJTap5T6qd031VOBMXbPRx368Ivt81GlaVqjze3L\nwM0etAX0c27xv+jP/XG7ZVE2Pew7gR/YtfE1INJ47i1K6cqTNn6AHlAvKKU2KqWWKqWG2m1TbrwG\n+vI4ropE/9RmpWlavaZpX9E07VZNnwj/JPqnmFXoPe6J6K+dY+jB11azsc+gIAHXNR8B6egfyYZo\nmvaopmk1xroQ9DHCdPQKApaf243lFtd7exDjhfYJ+se+uegB9yP0IDYBfeyy0IX22v7ztNr9M3lE\nKfUY+ljvBmAh+jGuRS8RbWn/UfSe8hKjhtWjGMMJ3qRp2gb0IYEN6Od5n1Lq343VIcBRuj8f49B7\n0xb2X/RoeP7/YLsvrYdlITa/n7drX5rRxk+92UbjTWUq+nNRBfwbUKmUusWbj+OGGvRPgE4ppZYC\nbZqmvYn+v/CO0Vl5DZhnt/lNdD1nAc3k7wYEiSbNuITFgWL0F3O11rVkjqd2o38svRMo1DTtH0qp\nf6B/4WY7fluB/k8xE/3LNUvvNQ29pIlDmqbVK6XM6GObhTarpqP3bJy5GzigaVq+ZYFSaqyD7Tai\njwefQK+s8I7Nukq6jzXehQeM8/AS8JJS6rv8c2yxGFgM/EPr+s26u9rQhyL6QzEwQdO0c33cTzsu\ntFHTtE7057pQKfUccAV4gK5XCvSVq+frKPoQi0NKqZHob0azjEUhgOWqHkvFC1t3oJ/PoCA93L5b\ng1487k2l1HSl1Gil1GeNC8OjPNhfIfrHqDOapv3DZtm/YBMgNU07g/7t8jql1CylVBp6sLuG/qVN\nT34DfFcp9YhSapxS6kX0YYaenAKmKqXuVUqNVUr9CP3LFnub0L+A+Smw3e5j+2+BhUqpbxv7eBL4\ngu2dlVJZSqkKpdQ0nFBKvaiU+pxxrjPRv2ix1JLahP4RfptSarZS6lbj9y9V13pmvbkA3GGcn+HG\nFzquUr2s/wn6p4DnlVKTlVLjjefi5248hqWN2UqpeKVUrMOGKLVIKfWvSqlMpVQK+pvhUPQrO7zp\nAjBdKZVqnC9n5+B/gIlKKWe93F8Dv9I07WPj9l5gmfEJ71vGbVv3oF91ERQk4PaRMT55N/p46n+j\nj939Dn2cqrWHuzpTiP4u/vdelgE8jj4mug19TDYCuFfTtN4e91foX4SsN+5nqarak3Xo39xvMh4z\nBf3b/C40TatG/6dIx244QdO0A+hXRjyDPh73EPr1zLZjep9B/2jdU2JJCHrwPoH+D/wJ+rlA07Rm\n9DeCc0Z7K4xjjUX/8shV6437FqH3CO+2HIbddo6Ga3ocwtE07a/on2Lmoo9hHwS+h/6R3539rkT/\niF2N815eHfqb2gfox/Ms8KSmV6D1pl+i93LL0c9XsqONNE07jv76WWy/Tim1ELhd07Tf2yxeg/7J\n6AAwln9+d2K56iWarp+iApqU2BF+pZT6L2C+pmkZvW4sBgTjsrgX0a8Z9zgAKaXeBI5omubuJwO/\nkR6u8CljOCFDKXWbUuop9EvhnI45i4FH07T/Qe+5Jnm6D6VUOPqnpBe91S5fkB6u8Cml1Bvol7fF\noF9rWqBpmv2lPkIMSBJwhRDCR2RIQQghfEQCrhBC+IgEXCGE8BEJuEII4SMScIUQwkf+P9lTErM2\nHLE2AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b52c110>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_wordshift(wordShift, params={}):\n",
    "    \n",
    "    topN = params.get('topN', 50)\n",
    "    shifts = list()\n",
    "    sortedWords = list()\n",
    "    for w in sorted(wordShift['word_shift'].items(), \n",
    "                    key=lambda x: abs(x[1]['shift']), \n",
    "                    reverse=True):\n",
    "        sortedWords.append(w[0])\n",
    "        shifts.append(wordShift['word_shift'][w[0]]['shift'])\n",
    "    dVals = shifts[:topN]\n",
    "    sortedWords = sortedWords[:topN]\n",
    "    \n",
    "    fig = plt.figure(figsize=params.get('figsize',(4,8)))\n",
    "    #axSummary = fig.add_axes(params.get('rect_summary',[0.,0.9,1.,0.1]))\n",
    "    #axSummary.set_xticks([])\n",
    "    #axSummary.set_yticks([])\n",
    "    #axSummary.set_xlim([-100,100])\n",
    "    \n",
    "    #axShifts = fig.add_axes(params.get('rect_shifts',[0.,0.,1.,0.9]))\n",
    "    axShifts = fig.add_axes(params.get('rect_shifts',[0.,0.,1.,1.]))\n",
    "    barlist = axShifts.barh(range(1,topN+1), dVals,\n",
    "                  align='center', linewidth=0, alpha=0.5, color='r')\n",
    "    axShifts.set_ylim(axShifts.get_ylim()[::-1])\n",
    "    for i,w in enumerate(sortedWords):\n",
    "        case = wordShift['word_shift'][w]['case'].replace('u','\\\\uparrow').replace('d','\\\\downarrow')\n",
    "        shift = wordShift['word_shift'][w]['shift']\n",
    "        \n",
    "        barlist[i].set_color('b' if case[0] == '+' else 'y')\n",
    "        if shift > 0:\n",
    "            axShifts.text(shift, i+1, r' {} ${}$'.format(w, case), \n",
    "                          color='0.0' if '+' in case else '0.5',\n",
    "                          va='center', ha='left')\n",
    "        else:\n",
    "            axShifts.text(shift, i+1, r'${}$ {} '.format(case, w), \n",
    "                          color='0.0' if '+' in case else '0.5',\n",
    "                          va='center', ha='right')\n",
    "    \n",
    "    #axShifts.set_xticks([])\n",
    "    #axShifts.set_yticks([])\n",
    "    #axShifts.set_xlim([-100,100])\n",
    "    xmax = 1.5 * max([abs(x) for x in axShifts.get_xlim()])\n",
    "    axShifts.set_xlabel('Per word avg. sentiment shift (%)', fontsize=14)\n",
    "    axShifts.set_ylabel('Word rank', fontsize=14)\n",
    "    axShifts.set_xlim([-xmax,xmax])\n",
    "    axShifts.set_ylim([topN+1, 0])\n",
    "    axShifts.tick_params(axis='both', which='major', labelsize=12)\n",
    "    \n",
    "    if params.get('rect_cumulative',True) <> False:\n",
    "        axCumulative = fig.add_axes(params.get('rect_cumulative',[.1, 0.1, .17, .2]))\n",
    "        csum = np.cumsum(shifts)\n",
    "        axCumulative.plot(csum, range(len(csum)), c='k')\n",
    "        axCumulative.axhline(topN)\n",
    "        axCumulative.set_yscale('log')\n",
    "        if sum(shifts) > 0:\n",
    "            axCumulative.set_xticks([0,100])\n",
    "            axCumulative.set_xticklabels([0,100])\n",
    "            axCumulative.set_xlim((0,sum(shifts)))\n",
    "        else:\n",
    "            axCumulative.set_xticks([-100,0])\n",
    "            axCumulative.set_xticklabels([-100,0])\n",
    "            axCumulative.set_xlim((sum(shifts),0))\n",
    "        axCumulative.invert_yaxis()\n",
    "        axCumulative.set_xlabel(r'$\\sum_{i=1}^r \\delta h_{avg,i}$',fontsize=12)\n",
    "        axCumulative.tick_params(axis='both', which='major', labelsize=8)\n",
    "        \n",
    "    #if params.get('rect_size', True) <> False:\n",
    "    #    axSize = fig.add_axes(params.get('rect_size',[.7, .05, .25, .125]))\n",
    "    #    axSize.set_xticks([])\n",
    "    #    axSize.set_yticks([])\n",
    "        \n",
    "    if 'title' in params:\n",
    "        axShifts.set_title(params['title'], fontsize=14)\n",
    "        \n",
    "    if 'savefig' in params:\n",
    "        plt.savefig(params['savefig'], bbox_inches='tight', dpi=300)\n",
    "        \n",
    "    \n",
    "#wShift = WordShifts()\n",
    "#wsData = wShift.computeWordShifts(sampleText[-int(len(sampleText)*0.1):],\n",
    "#                                  sampleText[:int(len(sampleText)*0.1)])\n",
    "    \n",
    "plot_wordshift(wsData['happiness'], params={'topN':50, 'rect_cumulative':False})"
   ]
  },
  {
   "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": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}