feature_selection_for_UAR.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "import re\n",
    "from scipy.stats.stats import pearsonr\n",
    "import time\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dataset_path = r'C:\\Users\\ycg3\\Desktop\\feature_selection.arff'\n",
    "dataset = pd.read_csv(dataset_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Read in the features and labels\n",
    "X = dataset.iloc[:, :-1].values  # All the columns but the last one are features\n",
    "Y = dataset.iloc[:, -1].values  # The last column is the label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Split the dataset into train and test dataset\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "            min_samples_leaf=1, min_samples_split=2,\n",
       "            min_weight_fraction_leaf=0.0, n_estimators=30, n_jobs=4,\n",
       "            oob_score=False, random_state=None, verbose=0,\n",
       "            warm_start=False)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train the classifier\n",
    "rfc = RandomForestClassifier(n_estimators=30, n_jobs=4)#, random_state=10)  # Use 4 CPU cores in parallel\n",
    "rfc.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature ranking:\n",
      "1. feature 1 (0.056470)\n",
      "2. feature 77 (0.053959)\n",
      "3. feature 12 (0.037280)\n",
      "4. feature 2 (0.036107)\n",
      "5. feature 23 (0.036080)\n",
      "6. feature 24 (0.033120)\n",
      "7. feature 3 (0.032326)\n",
      "8. feature 8 (0.030829)\n",
      "9. feature 6 (0.029574)\n",
      "10. feature 7 (0.026693)\n",
      "11. feature 5 (0.026344)\n",
      "12. feature 78 (0.024645)\n",
      "13. feature 125 (0.022969)\n",
      "14. feature 86 (0.022478)\n",
      "15. feature 73 (0.021651)\n",
      "16. feature 20 (0.021420)\n",
      "17. feature 119 (0.017861)\n",
      "18. feature 21 (0.017082)\n",
      "19. feature 11 (0.017078)\n",
      "20. feature 31 (0.016996)\n",
      "21. feature 117 (0.016855)\n",
      "22. feature 122 (0.016235)\n",
      "23. feature 81 (0.015339)\n",
      "24. feature 74 (0.015172)\n",
      "25. feature 19 (0.014765)\n",
      "26. feature 126 (0.014542)\n",
      "27. feature 57 (0.014009)\n",
      "28. feature 118 (0.013515)\n",
      "29. feature 14 (0.013028)\n",
      "30. feature 123 (0.013021)\n",
      "31. feature 121 (0.012201)\n",
      "32. feature 10 (0.011113)\n",
      "33. feature 85 (0.010850)\n",
      "34. feature 9 (0.010810)\n",
      "35. feature 103 (0.010293)\n",
      "36. feature 56 (0.010258)\n",
      "37. feature 30 (0.009881)\n",
      "38. feature 120 (0.009812)\n",
      "39. feature 4 (0.009737)\n",
      "40. feature 76 (0.008068)\n",
      "41. feature 59 (0.008029)\n",
      "42. feature 22 (0.008004)\n",
      "43. feature 124 (0.007998)\n",
      "44. feature 27 (0.007883)\n",
      "45. feature 99 (0.007728)\n",
      "46. feature 55 (0.007243)\n",
      "47. feature 64 (0.007058)\n",
      "48. feature 75 (0.006922)\n",
      "49. feature 13 (0.006312)\n",
      "50. feature 92 (0.006008)\n",
      "51. feature 63 (0.005478)\n",
      "52. feature 101 (0.004723)\n",
      "53. feature 91 (0.004371)\n",
      "54. feature 98 (0.004245)\n",
      "55. feature 104 (0.004228)\n",
      "56. feature 29 (0.004179)\n",
      "57. feature 97 (0.003874)\n",
      "58. feature 95 (0.003550)\n",
      "59. feature 32 (0.003487)\n",
      "60. feature 80 (0.003268)\n",
      "61. feature 84 (0.003203)\n",
      "62. feature 96 (0.003080)\n",
      "63. feature 28 (0.002964)\n",
      "64. feature 26 (0.002964)\n",
      "65. feature 102 (0.002803)\n",
      "66. feature 17 (0.002320)\n",
      "67. feature 15 (0.002010)\n",
      "68. feature 34 (0.001734)\n",
      "69. feature 83 (0.001668)\n",
      "70. feature 100 (0.001549)\n",
      "71. feature 82 (0.001450)\n",
      "72. feature 25 (0.001422)\n",
      "73. feature 94 (0.001415)\n",
      "74. feature 67 (0.001242)\n",
      "75. feature 65 (0.001174)\n",
      "76. feature 105 (0.001050)\n",
      "77. feature 66 (0.001040)\n",
      "78. feature 16 (0.001026)\n",
      "79. feature 58 (0.000925)\n",
      "80. feature 68 (0.000902)\n",
      "81. feature 79 (0.000816)\n",
      "82. feature 61 (0.000792)\n",
      "83. feature 87 (0.000780)\n",
      "84. feature 60 (0.000777)\n",
      "85. feature 93 (0.000771)\n",
      "86. feature 62 (0.000702)\n",
      "87. feature 113 (0.000683)\n",
      "88. feature 107 (0.000673)\n",
      "89. feature 112 (0.000531)\n",
      "90. feature 69 (0.000519)\n",
      "91. feature 106 (0.000495)\n",
      "92. feature 33 (0.000473)\n",
      "93. feature 18 (0.000469)\n",
      "94. feature 116 (0.000462)\n",
      "95. feature 110 (0.000455)\n",
      "96. feature 108 (0.000445)\n",
      "97. feature 48 (0.000444)\n",
      "98. feature 41 (0.000442)\n",
      "99. feature 114 (0.000438)\n",
      "100. feature 90 (0.000420)\n",
      "101. feature 71 (0.000408)\n",
      "102. feature 42 (0.000402)\n",
      "103. feature 0 (0.000397)\n",
      "104. feature 70 (0.000389)\n",
      "105. feature 111 (0.000344)\n",
      "106. feature 54 (0.000337)\n",
      "107. feature 72 (0.000334)\n",
      "108. feature 50 (0.000333)\n",
      "109. feature 37 (0.000325)\n",
      "110. feature 36 (0.000318)\n",
      "111. feature 40 (0.000313)\n",
      "112. feature 38 (0.000308)\n",
      "113. feature 47 (0.000303)\n",
      "114. feature 115 (0.000301)\n",
      "115. feature 88 (0.000296)\n",
      "116. feature 45 (0.000261)\n",
      "117. feature 44 (0.000258)\n",
      "118. feature 43 (0.000246)\n",
      "119. feature 109 (0.000223)\n",
      "120. feature 46 (0.000220)\n",
      "121. feature 89 (0.000215)\n",
      "122. feature 39 (0.000203)\n",
      "123. feature 35 (0.000193)\n",
      "124. feature 49 (0.000189)\n",
      "125. feature 51 (0.000164)\n",
      "126. feature 52 (0.000135)\n",
      "127. feature 53 (0.000000)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhoAAAFyCAYAAACz9nOMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3XmcHVWd9/HPLyF0CEviTJwkIIKgRHDQIXGBGUUYhASR\nMObBpSGKgCK7tPAAY0S2UUQHIjoiqMM2QCsij2yasCgGBgLaHRahQwRCwpKEBMgiSS/p/j1/nFO5\n1ZV7b/ft7uq+3f19v1731V11T506Vbdu1e+eOueUuTsiIiIieRgx0AUQERGRoUuBhoiIiORGgYaI\niIjkRoGGiIiI5EaBhoiIiORGgYaIiIjkRoGGiIiI5EaBhoiIiORGgYaIiIjkRoGGiPSame1iZh1m\n9sWBLouIVBcFGiIVMrNj4kW12Os7Oa73UDM7P6/8+8Cgfp6BmZ1kZscMdDlEhpqtBroAIoOUA+cB\nL2bm/yXHdX4SOBm4MMd19Ii7LzWzbYC2gS5LL5wMrAKuH+iCiAwlCjREem6uuzf24/osl0zNxrj7\nht7m4+6tfVGe/mZm27j7xoEuh8hQpVsnIjkys1lm9mcz22Bmr5tZvZm9I5Pmo2Z2i5ktNbNmM1tm\nZpeb2ehUmmsJv7hJ3aZpj9MHxOn9M/lu0W7CzK4zs/VmtpuZ/dbM1gE3pt7/iJnNNbM1ZvaWmT1g\nZv/cje0st66dzeyu+P/LZpZsx95mdr+Z/c3MXjSz2kyeyS2qj5nZ1Wa22szWmtn1ZjauSBlONrO/\nxH34ipn9l5mNzaR5wMyeNLMpZjbfzN4CvmNmS4D3Acm+7DCz38dl3mZm/xmXWx/L8Fsze38m74/H\n5T5jZrPN7CUz22hm95nZ7kXK+5GYzxtxHzxhZqdn0kw2s1vjsbPRzP5kZodn0mxlZueb2eKYZrWZ\nPWhmB3X1uYn0B9VoiPTcWDP7+/QMd389+d/MZgMXAb8Afga8HTgd+KOZ7ePu62LSzwDbAFcCrwMf\nBk4DdgI+F9NcBewIfAI4ms61G07320c44Xs/D3gQOBPYEMv7r8BvgT8DFwAdwLHA783so+7+526u\nI72uEcDvgD8C/zeW/UfxAv9tQpDza+BE4Hoze9jdl2by+S/gTeB8YDIh4HoncGCSwMwuAL4F3EPY\nj0m6D5rZv7h7e6pM4+N2/gK4AVgJ/CGuZz3wH4T9uzIusxswA/gVsASYAHwVeMDM9nL3FZnyngu0\nA98HxgLnxO3cL1Xeg4E7gVeBHwArgD2Bw4AfxjTvAx4CXgYuAd4CPgv8xsxmuvvtMbsL4zp/CvwJ\n2AH4IDAFuB+RgebueumlVwUv4BjCRTj7ak+leSehvcI5mWX3AlqBc1Pzaoqs4xxgE/CO1LwfpdeR\nmv9xwoVt/8z8XWK5vpiad21M+x9F8nkWuDszrwZ4nnCbqNw+Kbeus1PzxhIumJuAI1Pz94jLf6vI\nfn4UGJmaf1bM91NxejzQDPw2U6aTY7pjUvP+EOd9ucg2PAX8vsj8UUXmvRPYCMzOfA4dhHY66fKe\nFte5V5weAbwQ9+v2ZfbpfcBCYKvM/IeARanphcAdA/290EuvUi/dOhHpGQdOItQwJK+DU+//H8Kv\n4l+Z2d8nL+A14K+kfo27e0vyv5mNiekeIVyQ9smp/FelJ8zsn4D3APWZ8m5P+FW8f5E8uuu/k3/c\nfS0hoHnL3W9NzV8MrCHUHmT91As1EgA/IVy4PxmnDwZGEWoG0n5GqKE4LDO/Bbiuu4V3980NXM1s\nhJn9HaEW6FlCrUHWNZnyPkg4FpJt2wfYFfiBu68vtk4zexvhGPkVseYs9ZncA7zHzCbF5GuA95nZ\nu7u7TSL9SbdORHruT166Mei7CYHCc0Xec0KtBgBmtjNwMXA48LZMurH0vU3u/nJm3nvi3xtKLNNh\nZmNjoFCJZk/dTorWEm4HZK2l8/ZD2Aed9qG7v2VmywkXawi1CwCLM+nazOwFQm1L2ivuvql7xQcz\nM+AMQmD5LmBkqmyriyzyUmb6zfg32bbd47JPl1ntuwnBycWEWzlZDvwDsJxwy+g3wGIz+wswF/gf\nd3+qTP4i/UaBhkg+RhCq0afHv1l/g/ALmVBFPo5wH/5Zwq2FnQjdLLtT61iqfcbIEvNbisxL1nMm\n8ESJ5f7WjbJktVc4P5eeNRmV9jBJ2tr8HPgm8AbhM72C4p9PX2xbku9/EtrTFPMcgLs/GBubHgEc\nAhwP1JnZV939mgrWKZILBRoi+XiecGF50d2L1Wok9ibUJnzB3W9KZprZJ4qkLRVQvBnXle2JsWu3\nSxvKC7De3X9fwXJ5M8L++ePmGWbbApOAu+OspPHoZFLjmpjZKEINxL3dXFep/ft/CG03TuhUsNDz\nZVU3805Ljo1/BErt6xfi37bufB7uvoYQmF5vZmMIt2suABRoyIBTGw2RfNxG+NVbdCTPeJ8fCr9+\ns9/FM9jywvdWXHaHzPylMZ9sO4qTi+RRSgPhAnhWvJBnyzu+m/nk4QQzS/8oOplQW/PbOH0foeHt\n6ZnlvkzogXFXN9fzFlsGaxD2bafaCDP7DKHWqScaCb1Xzsh2v024+yrgAeCrZjYx+37680gdS8my\nGwi1HTU9LJ9In1KNhkjPlK0Gd/cXzOybhDEa3kW4h76e0CDw34CrgcuBRYQL/GUWxtdYR/gFXeyC\n1xDX+yMzm0fogfJLd19nZr8CTg/NCXge+BShO223uLub2ZcJF++nLYzb8QrhYnogof3EEd3Nr49t\nDdxvZrcA7yW0lXjQ3e8CcPfVZnYJ8C0zmwvckUr3GHBT8Wy30ACcGLslPwe85u5/IAQq55nZNcDD\nhFqooynUAlUk7uuTYjkfj/t6eSzzXu5+aEx6CqFm4ikz+xmhlmMCoZvsThQaCj9jZg/E8r8BfAg4\nkthNVmSgKdAQ6Zkuawrc/VIzexaoIzTYg9BQMLkY4u6bzOxThIvCuYRumrcBP2bLthK3xXSfpzCW\nxi/je6cRvs9fJbTB+CWhG2ixIdGLlt3d/2hm+xGGVj8F2I4wvsOjhMCoy03u7rrKpM3Od+BUwvZe\nSOhdchPwtUzZLzSz12LaywkX3KsI3U+zbSZKlekiQsPS/0vobfNHQnfY7wBjgKMI41g0EHq8fLdE\neYvpNN/d7zGzAwk1Xl8n1Gg9TxgLI0nTZGYfjGmOAZJeSwtjWRNXEMb5OJhQi7EU+AahfYfIgDP3\nQf0cJBEZoiw84Owa4ENleveISJXrlzYaZnaKmS2Jw+MuMLMPlUk70cxuMrNnzazdzC7vIu/Px2F/\nb+v7kouIiEhv5B5omNnngMsI1X/7EKqD55VpXFZDqB68GHi8i7x3JQzzO7+Piisi1aU/uruKSI76\no0ajDrja3W9w90WEZxpsAI4rltjdl7p7nbvfSGgYV1Qcf+BGwr3vJX1fbBGpArq3KzLI5RpoxH7s\nU0k92MdDo5D7SD1gqIfOB1a6+7W9zEdEqpC7X+/uI9U+Q2Rwy7vXyXhCf/eVmfkrCYPr9IiZfZTw\nVMkPdDP93wPTCIP5NPd0vSIiIsPQaMIAgPOKPFKgS4Oue6uZbUd4HsNX3P3NrtJH0+h+X3oRERHZ\n0tHAzZUulHcbjdWEUfUmZOZPIPTP74ndCQ9JutPM2sysDfgicISZtcbBkbJeBLjxxhtpaGigoaGB\n/fffv+j/XU0PRNrhss5qL5/2yeArn/bJ4Cuf9kl1rDM9feONN3a6llYq1xqN+PTEBuAg4gBF8UmI\nB9HzUeuaCCPzpX2bMLjQ6Wz55ESIt0v23HNPpkwJT3UeO3Zs0f+7mh6ItMNlndVePu2TwVc+7ZPB\nVz7tk+pYZ7Fpetj0oD9unVwOXBcDjscIvVDGANcBxKGDd3T3Y5IFzOwDhG5t2wFvj9Ot7t7k7q3A\nM+kVmNkaQjvTpn7YHhEREemm3AMNd78ljplxEeGWyePAtPjQIICJwM6ZxRZS6NY2hTD071LCcyJE\nRERkkOiXxqDufiVwZYn3ji0yr6K2I8XyEBERkYE38oILLhjoMuTuwgsvnAR89atf/SqTJk3aPH/v\nvfcu+n9X0wORdriss9rLp30y+MqnfTL4yqd9Uh3rTKaXL1/OT3/6U4CfXnDBBcup0LB4qJqZTQEa\nGhoasg1bREREpIzGxkamTp0KMLUnA+gNunE0+kJ9fXgBNDfD0qWwyy4wenSYV1sbXiIiItI7wzLQ\nSAcSjY0wdWoIPFTZISIi0rf65THxIiIiMjwp0BAREZHcKNAQERGR3CjQEBERkdwo0BAREZHcKNAQ\nERGR3Ayr7q1nnAHjxnUeO6OlJbw3d666t4qIiPS1YRVo/OAHIZhIj50B4f/p0we2bCIiIkORbp2I\niIhIbhRoiIiISG4UaIiIiEhuFGiIiIhIbhRoiIiISG4UaIiIiEhuFGiIiIhIbhRoiIiISG4UaIiI\niEhuFGiIiIhIbhRoiIiISG76JdAws1PMbImZbTSzBWb2oTJpJ5rZTWb2rJm1m9nlRdJ82czmm9kb\n8XVvuTxFRERkYOQeaJjZ54DLgPOBfYAngHlmNr7EIjXAa8DFwOMl0nwcuBk4ANgXeAm4x8wm9V3J\nRUREpLf6o0ajDrja3W9w90XAicAG4Lhiid19qbvXufuNwLoSab7g7le5+5Puvhj4MmFbDspnE0RE\nRKQncg00zGwUMBW4P5nn7g7cB+zXh6vaFhgFvNGHeYqIiEgv5V2jMR4YCazMzF8JTOzD9VwKvEII\nYERERKRKbDXQBegtMzsX+CzwcXdvHejyiIiISEHegcZqoB2YkJk/AVjR28zN7CzgbOAgd3+6q/R1\ndXWMHTuWtWuTaZg2rRao7W1RREREBr36+nrq6+s7zVubXDR7yEKTifyY2QLgUXf/Wpw2YBnwQ3f/\nfhfL/gFY6O5fL/Le2cC/A4e4+5+6yGcK0NDQ0MCUKVNobISpU6GhIbyf/D9lSg82UEREZAhrbGxk\n6tSpAFPdvbHS5fvj1snlwHVm1gA8RuiFMga4DsDMLgF2dPdjkgXM7AOAAdsBb4/Tre7eFN8/B7iQ\nUBWxzMySGpO/uftb/bBNIiIi0g25BxrufkscM+Miwi2Tx4Fp7r4qJpkI7JxZbCGQVLVMAY4ClgK7\nxXknEnqZ3JpZ7sK4HhEREakC/dIY1N2vBK4s8d6xReaV7Q3j7u/qo6KJiIhIjvSsExEREcmNAg0R\nERHJjQINERERyY0CDREREcmNAg0RERHJjQINERERyY0CDREREcmNAg0RERHJjQINERERyc2gf0x8\nHurrwwuguRmWLoVddoHRo8O82trwEhERkfIUaBSRDiSSJ73W1+vpriIiIpXSrRMRERHJjQINERER\nyY0CDREREcmNAg0RERHJjQINERERyY0CDREREcmNAg0RERHJjQINERERyY0CDREREcmNAg0RERHJ\njQINERERyY0CDREREcmNAg0RERHJTb8EGmZ2ipktMbONZrbAzD5UJu1EM7vJzJ41s3Yzu7xEus+Y\nWVPM8wkzOzS/LRAREZGeyD3QMLPPAZcB5wP7AE8A88xsfIlFaoDXgIuBx0vk+c/AzcDPgH8Cbgd+\nY2Z79W3pRUREpDf6o0ajDrja3W9w90XAicAG4Lhiid19qbvXufuNwLoSeZ4O/M7dL3f3Z939W0Aj\ncGoO5RcREZEe2irPzM1sFDAV+E4yz93dzO4D9utF1vsRaknS5gFH9DTDujoYOxaam2HpUthlFxg9\nGtau7UUpRUREhrlcAw1gPDASWJmZvxKY3It8J5bIc2JPM5wzB6ZMgcZGmDoV6us7T4uIiEjl8g40\nqkpdXR1jx47dXEtRVwfTptUCtQNaLhERkWpQX19PfX19p3lre1m1n3egsRpoByZk5k8AVvQi3xU9\nyXPOnDlMmTJlcy3FnDlh/uzZvSiJiIjIEFFbW0ttbecf342NjUztRdV+ro1B3b0NaAAOSuaZmcXp\nh3uR9SPpPKOD43wRERGpEv1x6+Ry4DozawAeI/RCGQNcB2BmlwA7uvsxyQJm9gHAgO2At8fpVndv\nikmuAB4ws68DdxPufUwFvtIP2yMiIiLdlHug4e63xDEzLiLc3ngcmObuq2KSicDOmcUWAh7/nwIc\nBSwFdot5PmJmRwHfjq+/Ake4+zN5bouIiIhUpl8ag7r7lcCVJd47tsi8Lm/puPuvgV/3vnQiIiKS\nFz3rRERERHIzrLq39sTcueFvXR3U1HQezAugtja8REREZEsKNLowfXro/pp0hU0P5iUiIiLl6daJ\niIiI5EaBhoiIiORGt056qb4+vGDLB7KB2nCIiMjwpkCjl9KBRPaBbCIiIsOdbp2IiIhIbhRoiIiI\nSG4UaIiIiEhuFGiIiIhIbhRoiIiISG4UaIiIiEhuFGiIiIhIbhRoiIiISG40YFcf0pNeRUREOlOg\n0Yf0pFcREZHOdOtEREREcqMajR6oq4OxY7d8iNratQNdMhERkeqiQKMH5swJt0OyD1FLpkVERCTQ\nrRMRERHJjWo0clbqNguoF4qIiAx9CjRyVuo2i4iIyHDQL7dOzOwUM1tiZhvNbIGZfaiL9AeYWYOZ\nNZvZYjM7pkiaM8xskZltMLNlZna5mdXktxUiIiJSqdwDDTP7HHAZcD6wD/AEMM/MxpdIvytwF3A/\n8AHgCuDnZnZwKs1RwCUxz/cCxwGfBb6d13aIiIhI5fqjRqMOuNrdb3D3RcCJwAZCcFDMScAL7n62\nuz/r7j8Gbo35JPYDHnL3X7r7Mne/D/gF8OH8NkNEREQqlWugYWajgKmE2gkA3N2B+wjBQjH7xvfT\n5mXSPwxMTW7BmNluwCeBu/um5CIiItIX8m4MOh4YCazMzF8JTC6xzMQS6Xcwsxp3b3H3+njr5SEz\ns7iOq9z90j4su4iIiPTSoBxHw8wOAL5BuA2zDzAT+JSZfXMgyyUiIiKd5V2jsRpoByZk5k8AVpRY\nZkWJ9OvcvSVOXwT8j7tfG6efNrPtgKuB/yhVmLq6OsaOHbt5qPC6Opg2rRYYXINZ1NeHF2h8DhER\n6Tv19fXUJxeYaG0vn6+Ra6Dh7m1m1gAcBNwBEG91HAT8sMRijwCHZuYdEucnxgCbMmk6kvxjO5At\nzJkzhylTpmwe0yJ5yurs2d3epKqQDiQ0PoeIiPSV2tpaajO/VBsbG5nai+dr9MeAXZcD18WA4zFC\n75ExwHUAZnYJsKO7J2NlXAWcYmaXAtcQgpIjCY09E3cCdWb2BPAo8B5CLccdpYKMaqGRQkVEZDjJ\nPdBw91tiw82LCLdAHgemufuqmGQisHMq/YtmdhgwBzgdeBk4PnZhTVxMqMG4GNgJWEWoMan6Nhoa\nKVRERIaTfhmC3N2vBK4s8d6xRebNJ3SLLZVfEmRc3FdlFBERkb43KHudiIiIyOCgQENERERyo0BD\nREREcqPHxFcJjY0hIiJDkQKNKlFubIwkCKmv3zII6eU4KiIiIrlSoDEIlAtCkmkREZFqpDYaIiIi\nkhsFGiIiIpIbBRoiIiKSG7XRGCBz54a/dXVQU6MGniIiMjQp0Bgg06eHp8YmT5BVA08RERmKdOtE\nREREcqNAQ0RERHKjQENERERyo0BDREREcqNAQ0RERHKjQENERERyo0BDREREcqNAQ0RERHKjQENE\nRERyo0BDREREcqNAQ0RERHKjQENERERyo0BDREREctMvgYaZnWJmS8xso5ktMLMPdZH+ADNrMLNm\nM1tsZscUSTPWzH5sZq/GdIvMbHp+WyGJ+nqYMSO8DjkEJk8Of5N59fUDXUIREakWuT8m3sw+B1wG\nnAA8BtQB88xsD3dfXST9rsBdwJXAUcAngJ+b2avufm9MMwq4D1gBzAReBXYB1uS9PQK1teEFhUfa\nJ4+4FxERScs90CAEFle7+w0AZnYicBhwHPC9IulPAl5w97Pj9LNm9tGYz71x3vHAOGBfd2+P85bl\nVH4RERHpoVwDjVjzMBX4TjLP3d3M7gP2K7HYvoTairR5wJzU9OHAI8CVZnYEsAq4GbjU3Tv6qPhD\nTn194bZGczMsXQq77AKjR4d56ZoKERGRvpB3jcZ4YCSwMjN/JTC5xDITS6Tfwcxq3L0F2A34V+BG\n4FDg3cBPCNtzcd8UfeDMnRv+1tVBTU3ngGDt2p7n25tbHqWClJaWQpl160RERLL649ZJHkYQgo8T\n3N2BhWb2DuAsygQadXV1jB07dvPFuq4Opk2rBarrZ/z06TB7NsyJdTjpgCAJEPpbqSAlKd/0Es1w\nVYsiIjJ41NfXU59p0b+2N79wyT/QWA20AxMy8ycQGnIWs6JE+nWxNgNgOdAag4xEEzDRzLZy903F\nMp4zZw5TpkzZfKFMLuSzZ3d3c6RSajgqIjJ41NbWUpv59dfY2MjUXvzCzTXQcPc2M2sADgLuADAz\ni9M/LLHYI4TbIWmHxPmJ/2XLaojJwPJSQcZQVe42C6jGQEREBlZ/3Dq5HLguBhxJ99YxwHUAZnYJ\nsKO7J2NlXAWcYmaXAtcQgpIjgU+m8vxJTPND4EfAHsC/Az/IfWuqTLnbLCIiIgMt90DD3W8xs/HA\nRYRbII8D09x9VUwyEdg5lf5FMzuM0MvkdOBl4Hh3vy+V5mUzmxbTPAG8Ev8v1l1WqpjacIiIDG39\n0hjU3a8kDMBV7L1ji8ybT+gWWy7PR4F/rrQsy5Yto6npLWBPli9fzqRJkyrNQvqQ2nCIiAxtw+pZ\nJ8uXL2fPyZOZNetoAI6cOZPly5cPcKlERESGrmEVaKxZs4YNzc2b+782t7awZo1GLRcREcnLsAo0\nEu8a6AKIiIgME8My0BAREZH+MVhHBu0zq1eHB8g2NTUxfvy2wDsHtkAiIiJDyLCv0TjnrLMAmDXr\naPacPFmNQ0VERPrQsA80Wja1AeEBKRuam9U4VEREpA8N+0AjoQaiIiIifU+BhoiIiORm2DcGHe7q\n6mDsWA3/LSIi+VCgMcSVepZIS0uYN20afOMbGv5bRETyoUBjiCv1LBEI/0+fPnBlExGRoU9tNERE\nRCQ3CjREREQkNwo0REREJDcKNERERCQ3CjREREQkNwo0REREJDcKNERERCQ3CjREREQkNxqwSwal\nUiOeavh0EZHqokBDBoVygUVLCyxerOHTRUSqkQKNIajUg9LWri29zNy5hWVraqqvhqDUUOpTphSm\nRUSk+vRLoGFmpwBnAROBJ4DT3P1PZdIfAFwGvA9YBnzb3a8vkfbzwM3Ab9x9Zh8XfVCaM6fzBbg7\nF+Tp02H27LAsDM4HrJUKsF55Jby/0066zSIi0t9yDzTM7HOEoOEE4DGgDphnZnu4++oi6XcF7gKu\nBI4CPgH83Mxedfd7i6T9PjA/vy2QwaJUgJWmp9SKiPSv/qjRqAOudvcbAMzsROAw4Djge0XSnwS8\n4O5nx+lnzeyjMZ/NgYaZjQBuBL4F7A+MzW0LpEulahOg/2sMyt0GKnf7KCvbLuSpp2DTJhg5Etrb\nYautYO+9VTMiIlJOroGGmY0CpgLfSea5u5vZfcB+JRbbF7gvM28eMCcz73xgpbtfa2b791GRpYe6\nU5vQX8rdBqqkPUexdiENDZ3zuftu1YyIiJSTd43GeGAksDIzfyUwucQyE0uk38HMaty9JdZwHAt8\noC8LK531pFGpdE1dc0VkOBl0vU7MbDvgBuAr7v7mQJdnKOtJo9JEsSClpSW8d/TR8J73DN8AplwP\nGgj/z5gR/lcgIiKDXd6BxmqgHZiQmT8BWFFimRUl0q+LtRnvBXYB7jQzi++PADCzVmCyuy8plvFl\nl10W/vI8MAN4HnhbJdsj3VQsSIHw/0039TyAyVup2oYkSJo7N59bJaVqjzRGiIj0p/r6euqTk2C0\ntpe/AnMNNNy9zcwagIOAOwBicHAQ8MMSiz0CHJqZd0icD7AI2Dvz/reB7YDTgZdKlefMM89k1qxZ\nnMnuzOIOQGdu6axUbQOE/6dPz2e9vak9EhHpK7W1tdRmqkwbGxuZ2osTUX/cOrkcuC4GHEn31jHA\ndQBmdgmwo7sfE9NfBZxiZpcC1xCCkiOBTwK4ewvwTHoFZrYmvOVNuW+NiIiIdFvugYa732Jm44GL\nCLdAHgemufuqmGQisHMq/Ytmdhihl8npwMvA8e6e7YmSi9Wrw9AeTU1NjB+/LfDO/litiIjIkNQv\njUHd/UrCAFzF3ju2yLz5hG6x3c1/izx66pyzzgIeZdasoxkzuolbbn0BmNRX2csQU03jh4iIVKNB\n1+skby2b2gC4GDivuZmFCxcCk1i+fHlMMUm1HbJZNY0fIiJSjRRolLA9oSvLeed9E/gkMz/9aUIX\nlwWq7RAREemmEQNdgGq1Hugg1GwAtLa10tLWCnHehuZm1qxZM0ClExERGRwUaHThXd2cJyIiIlvS\nrRMZltSIU0SkfyjQkKqV5yidasQpItI/FGj0QjLmRuiRokahfW2gRukUEZG+ozYavRDG3IAjZ87k\nkUceoakpDExa6AorIiIyvCnQ6IVkzI3m1hYOOuAAZs06GgiBh4INERERBRp9ZmNr6+ausM2tLTz3\n3HNAGMp82bJlA1cwERGRAaRAow+lu70mt1VmzTqaPSdPVg2HiIgMSwo0cpIeylyDe4mIyHClXic5\nS2o5lixZAuypHiplzJ0b/tbVQU1N5/Et1q4d2LKJiEjPKNDI2Wo6PzPlyJkzefavv0QPZNvS9Okw\ne3YY4wI6j2+RdG+VglLjjGjgMRGpJgo0cpZ+Zsp5hIaiYfwNBRrSO6XGGdHAYyJSTdRGo5/o+Sgi\nIjIcqUZDhjW1CxERyZcCjQGwfPly1qxpotA4FGASTU1NjB+/Lbqt0n/ULkREJF8KNAbAZ2bOZGPr\n+4BGZn760xgAC5g162jGjG7illtfQD1TRERkKFAbjQGQHkW0ta2VlrZWQGNuiIjI0KMajQFSrHFo\nsTE3Jk1SzcZgpK6nIiKBAo0qUmzMjVtvuw3dRhl81PVURCTQrZMqkh5zA8KYG7qNIiIig5lqNKpQ\n+rZKGNwL9UgZpsp1vwXdghGR6tcvNRpmdoqZLTGzjWa2wMw+1EX6A8yswcyazWyxmR2Tef/LZjbf\nzN6Ir3vecIhpAAAgAElEQVS7ynOwyj4F9sknnwTQ02CHienTw985c+C734XFi8PfO+4ILwUZIlLt\ncg80zOxzwGXA+cA+wBPAPDMbXyL9rsBdwP3AB4ArgJ+b2cGpZB8HbgYOAPYFXgLuMbMh15gh+xTY\nE7/yFQCOnDmTRx55hKamJkCBh4iIVKf+qNGoA6529xvcfRFwIrABOK5E+pOAF9z9bHd/1t1/DNwa\n8wHA3b/g7le5+5Puvhj4MmFbDsp1SwZQcjslCTyaW1s46IADmDXraCAEHkltR1NTE8uWLRuIYoqI\niHSSa6BhZqOAqYTaCQDc3YH7gP1KLLZvfD9tXpn0ANsCo4A3elzYQSg9Hkdza8vm2o7kNotqOURE\nZKDlXaMxHhgJrMzMXwlMLLHMxBLpdzCzmhLLXAq8wpYBypCXbjiavc2iHisiIjLQBn2vEzM7F/gs\n8HF3by2X9rLLLgt/eR6YATwPvC3vIvY7PSlWKqUBxkQEoL6+nvrkZBCt7eUTJvMONFYD7cCEzPwJ\nwIoSy6wokX6du7ekZ5rZWcDZwEHu/nRXhTnzzDOZNWsWZ7I7s7gD0OhJIqABxkQkqK2tpTbzq6Kx\nsZGpvXjCZK63Tty9DWgg1UjTzCxOP1xisUfYslHnIXH+ZmZ2NjAbmObuC/uqzCIiItJ3+uPWyeXA\ndWbWADxG6D0yBrgOwMwuAXZ092SsjKuAU8zsUuAaQtBxJPDJJEMzOwe4EKgFlplZUgPyN3d/K/ct\nGiTSz0zRMOb5KjewVi9rHUVEBrXcAw13vyWOmXER4RbI44RaiFUxyURg51T6F83sMGAOcDrwMnC8\nu6cbep5I6GVya2Z1F8b1DGvFnpny+wcu54UXxlEIPAAm0dTUREvLmk7v6UFulZs+HWbPDgNrQedb\nD8mtiLo6GDu2cxuIlngzcO5c3aYQkaGpXxqDuvuVwJUl3ju2yLz5hG6xpfJTe8cy0s9MOY/CmBsb\nW98HNDLz05/GAFjArFlHM5KFtLMP0MiRM2fyk6uvBt6vYc/72Jw5nQOPpL3V1KmFEUBFRIYaPVRt\nCEtHY+kxN1rbWmlpCx10TiK01i03HocGAqsudXUwYwYccghMnhz+1sXh7JJbOCIi1WLQd2+V7itW\nDbRjkffS43Gct3nY80eZNetoRtc8w3cvvRf42Ba3YFT70T9UMyIig4kCDSkpO+z5mcCclhbOOONr\nFLsFM2Z0Ez/+yaPA+9UAdYAUaweisTBEZCAp0JBuG0fnth+tbYXx0bK1H101QFXtRz6K1XaokamI\nDCQFGlKxYrdgSj30rVQD1C1vwaj2Q0RkKFJjUMlNqQaoZwKtm2/BFH/kfVIDogaovZM0HN17b9hu\nO3j728Nru+3g/e8P782YUWjnISLS11SjIbkqVvuRvQXTVe3HmNFN3HLrC6jWo3LZWynz54f5U6fC\nddfptoqI5E81GjJgutP9NnkS7cKFYZT5QlsPEREZDFSjIVWjWO3H9mw5yqkGFOsbpXqoaMh0EelL\nCjSkqhUb5TQ9rke6S62GU69MqR4qybSISF9QoCGDQncHFOtqOPXly0cBk9TTpYxyD4iD8uNx1NcX\nGpZ2NZZHOu2SJeG1zTZheuNG2G032HXXrtcpItVNgYYMWtkutScBP6F07cfommfwjn8CFuhBc2UU\ne0DcxIkhWGhuhgsugGuvLR48pP/vaiyPYmnVWFVk6FGgIUNGl8Opt7QAoZFppQ+au/W22xjONSAa\nCExEekq9TmRYKNbQtJIHza1Zsyb/Qg4x6Vsw6QfAaewOkeFFNRoyrHX3QXOrV68G2KKtR7HbLMGk\nou1Cli1ro6npLYbDLZlit2BUEyIy/CjQEOmGc846i2JtPbK3WYoNtZ6knfnpTzNyxONsbNmLYo1V\ntwxYqjcISbrGlmvEWaybrLrUigw/CjREuqFUW49sA9QtHjSXShveaynZWDXbLiTbWLWaaj+6M+Jo\nsW6y6lIrMvwo0BCpQPZWS7HbLKXSFptfrsdMurFqV7Uf1RSE9JV07cdTT8GmTTByJGzYAK2tsP32\nIZ26wopUNwUaIlWgWMCSNFbtTu3HUBwtNVv70dDQefrhh0M6dYUVqW7qdSJSxUrVfmR7xYQgBGbN\nOpo9J0/mySefBPT0WxEZeKrREBlkuhwvJDNa6uiaZ/jupfcCHxuyvWJKNTKF7t9KqWRUUxHpPgUa\nIkNIdrTUM4E5LS2cccbXGMq9Yko1Mk2Ch/r6LYOHV14Jy+6005bvtbTA4sWFEVGXLIGvfAVOPz0s\no3YhIt2nQENkCBtH54fS5dkrJqhsbJG8G7Z2d0j0Ur1gutO7RkTK65dAw8xOAc4CJgJPAKe5+5/K\npD8AuAx4H7AM+La7X59J8xngImBXYDFwrrv/Lo/yiwx2efeKKTeEe7lalL5u2FrugXDJWB3FbrO0\ntBSWV/Ag0rdyDzTM7HOEoOEE4DGgDphnZnu4++oi6XcF7gKuBI4CPgH83Mxedfd7Y5p/Bm4GzgHu\nBo4GfmNm+7j7M3lvk8hw01WvmHRNSSVji3TnQXjp9iVdtSEpNxppqVqKpF3G1Klh+d7qTVsPtROR\noag/ajTqgKvd/QYAMzsROAw4DvhekfQnAS+4+9lx+lkz+2jM594473Tgd+5+eZz+lpkdDJwKnJzP\nZohIVneHcC+VvlzD1mLtSyprQwLdvV1TKm25tiilxvlob4ettoK99y609TjmGPjGN0rsxJRKnn5b\njgIWqSa5BhpmNgqYCnwnmefubmb3AfuVWGxf4L7MvHnAnNT0foRakmyaI3pVYBGpGsXbl3SvDUkl\nt2vKpS3XFuWEE5rYbbcQsMyatSd33bWcvfdu48EH32LWrD35+c9D2vnzJ3Hrrev53e/aWL58NK++\nWrN5yPb1641RozoYMwY6OowRI9rZffdmRo3axEsvbQNsw0knNVNT4zzzzEja242ttjI6Ooy2NmfU\nKGPECGfDBti0aQTbbedAaKy6004tTJzYzFtvjWLx4u0ZN66Z0aNH09wMF1wA1167ZeDRXwGKAqHh\nJe8ajfHASGBlZv5KYHKJZSaWSL+DmdW4e0uZNBN7V1wRqTbZWpDutCGp5HZNubSVtkVJ17ik0y5c\n+PFOads37YsBbW0LaGv7IB3tIbhpaVvA6tX7dkr7+MIDMKClbQEwZXMg1NK2gK1H7dtpnX9bv+/m\ntM8991GWPFfI58nHD+T3D1zOjTe+k3vv3Ylx45ppbg4BzIMPGqee6nR0GFtt5ey5ZzstLcbixaNp\nbd3Iu9/ttLQYp502klNPHVk0uEmCptGjO+joMNrbrVuB0EsvbcPSpdtsUZ7TTy8EVCNHOiNGOM3N\nI4quMwnUygVflZQvnbY76+xu+UaMaGfPPdupqXFefjmkecc7jJYWY+lSY9KkVkaN2sSIESOYNu1N\nZs0aCRQe6jh+/PhO0y0tLdTU1BR9Lz1d7r1i+bzznX074N+w7HWypBvzXq0gbbn5pfKpJG21rHO4\nl28g1jmcy1fJOvO0sbWV44H/pnNQcgRwO2TeaymadiJzWMFa3kUNa3mWN9ouBMLP9wOYxgMtC0mC\nm+L5BscD/12kp1B3ypcNmrIBTDq4eXTBvpuDpBdf/BdWLO86uEmCppYNhXzMHsRsLR0dOwO788Yb\nG2PptuG5517guedeBMYCHy8bUGXLl11nOlCrpHyVpC23zkrK99BDyT4ZBezK4meXAaPp8I+xbNkL\nwHPALsyfvzuzZyf7a6/4mW8g/HZ/F9BGqPPbBqgB1pdJu4kwzN/uZdKOADYxYsQz/Mu/bI+ZMW3a\nmxx77Gh6zd1zewGj4t6YkZl/HfD/SizzR+DyzLwvAW+mppcCp2fSXAAsLJHnFMA//OEP+8gRIxx2\ncDjczcb5ViN3d3CHfXx0TY3XjPrI5umR4LCPg/vWoz7S6b102q1HfcS3qakpmTabTyVpq2mdw7l8\n2icDU77K13mug7vZfB9h9zgscpjrI0bc6SNsfkz7gMPtDnMdFvkIuyf13rkV7JN0Pq84vOaw3GGl\nwytb5BvK93mH22N57onLrYzLvRan58Z8n85sSzbtG6l1rnHYGKdfc1jv8ETM54FCPiPudHg8vp+k\n3eiwtkQ+6fe6Wmc6bWofjLgzluPzPrqmJu7Pz8d5t3dRnmSdK7uxzkrL19203VlnX5dvINa5Mn4O\n1zscFl/TfeTIQ33UqP0dcGBKT2IBixfi3JjZAuBRd/9anDZCl9Ufuvv3i6T/LnCou38gNe9mYJy7\nfzJO/wLYxt2PSKX5X+AJd9+iMaiZTQEaGhoaGD9+/OZ7qHfdFe6hfupTk7jxxiY+9rFteeqpUZun\ns/dfS6XN3pvNpi13H7ertNW0zuFcPu2Tzvl8+MPN7LBDB+vXb2L58q3ZZZdQtf7YY6M5+eRXmDVr\nWZ+Urzf7ZO+92zpVFWfXma4qTr9X6T7JVl3feGM78+a9ja233rriKvFk2Y6ODtrattq8b2tqnNbW\n1i2WLVUlXmk+kF81fH+nVfl6v85Ro0Ztbo80aVJoFD01PF55qrs3Uqk8azRiEPNZQr3MF4H3AlcD\nrwNvj+9fAlyfSr8roV7nUkI7jpMJdTufSKXZD2gBvh7TXAA0A3uVq9FoaGhwd/eGBncIf9P/Z9+r\nhrTDZZ3VXj7tk+Jpb77Z/fDDw+vgg9332CP8Pfxw9/3377/y7b9/8TJMmRJelb6XLntX5RMZ6hoa\nGnpVo5F7Gw13v8XMxhMG15oAPA5Mc/dVMclEYOdU+hfN7DBCL5PTgZeB4939vlSaR8zsKODb8fVX\n4AjXGBoyQLIDRe2xB5x7bueBooaicr0Dku6ZaekuoV3to0rSJmNj9JWk7MXKoMG9RCrTL41B3f1K\nwgBcxd47tsi8+YRuseXy/DXw6z4poEgvpQeKKjW89XC7aBULvmLtLKNHhy6W2TEj0tL7Mt0dMh3I\n9eX+y3a5TJf39ddhu+0K3S/32AMeeAAWLAjT6o4pUtqw7HUiMhDKBSF9MSJlpUrVGPz1r4X3a2rC\ng8WmTSsMRjVxYiEtlL7Ilgu+KpVeRzoggL676CtYEMmHAg2RYapUAJAdqrvaKCAQGVwUaEjVGK7t\nHCpVzbdgit1+6E7th4gMXQo0pGp0p52D9OwWTLkgDvouAKgkn3RQsmQJbLttuEUD4f8vfQl23bVv\nyyci/U+BhsgQkq7tKNa2oqYmBBeTJw/8xXug1y8i/UOBhsgQUq3tKkRk+Box0AUQERGRoUuBhoiI\niORGt05kWFNPFxGRfCnQGIIqGbq5P/KpZurpIiKSLwUaQ1BXAzH1dz4iIjJ8KdCQqlVq8KdqGZyq\nHA1cJSISKNAQoP8GdKpEqXUO5PNBukuBhIhIoEBDgL59AJb0nWoMAEVEKqFAQ3JRzc/jGEwUAIrI\nYKdAQ3JRbY9El/LUpkRE8qJAQ0QUSIhIbhRoSJ8oNeYGbHkRGw7jc4iISKBAQ/pEJW0IND6HiMjw\noWediIiISG5Uo1GF9PyN/FVyq0dERHpOgUYV0vM38qfuoiIi/UOBxiCnAZ1ERKSaKdAY5Loa0Km+\nHmbMCP9r8CwREelvuQUaZvY24L+ATwEdwK+Br7n7W10sdxHwZWAc8L/ASe7+XCrPC4FDgHcCq4Df\nAOe5+7qcNmVQ6+p5IfPmwYIFaqsgIiL5yLNG42ZgAnAQsDVwHXA1MKvUAmZ2DnAq8EXgReA/gHlm\ntqe7twI7ApOArwNNwC4xz0nAZ7sq0BlnwLhx+mWfprYKIiKSp1wCDTN7LzANmOruC+O804C7zews\nd19RYtGvARe7+11xmS8CK4F/A25x96eBz6TSLzGz2cD/mNkId+8oV64f/EDDYsvQoWHDRWQwyKtG\nYz/gzSTIiO4DHPgIcHt2ATN7FzARuD+Z5+7rzOzRmN8tJdY1DljXVZAhMtQokBCRwSCvQGMi8Fp6\nhru3m9kb8b1SyzihBiNtZallzGw88E3C7ZNBRWNliIjIcFBRoGFmlwDnlEniwJ69KlH3y7I9cDfw\nF0ID0S7V1dUxduzYTvP23bcWqOxnYV88q0NjZVSnUrcjhntbHhEZHurr66lPToLR2l7++q20RuM/\ngWu7SPMCsAL4h/RMMxsJ/F18r5gVgBEakKZrNSYA6VswmNl2wDxgDTDT3du7U/g5c+YwJXOVaGwM\nF/xK6FkdQ1dXvXTUlkdEhrLa2lpqMyfBxsZGpvbi4lZRoOHurwOvd5XOzB4BxpnZPql2GgcRAolH\nS+S9xMxWxHRPxnx2ILTp+HEq7+0JQcZGYEbsjSIiIiJVKJc2Gu6+yMzmAT8zs5MI3Vt/BNSne5yY\n2SLgHHdPGof+APimmT1H6N56MfAysfFoDDLuBUYDRxOCmSS7VWoQOrwlt7SWLIFtt4Vp08L8bbeF\nL30Jdt2179u/6JkpIiLl5TmOxlGEAbvuIwzYdSuh+2rae4DNjSbc/XtmNobQuHMc8CBwaKrWYgrw\nofj/c/GvEdqGvAtY1vebIdWgXFfOJHjozpggfX17S+OQiIiUl1ug4e5rKDM4V0wzssi8C4ALSqT/\nI7DFMjL0lasd6Cp40HgTIiIDR886kUGpkuBBgYSIyMBRoNEFPR21Omm/i4gMDgo0utDV01FFRESk\nNAUaOUv3Spg4MfSEGDkS2tvDtGpHRERkKFOgkTPVhAxe3enp0lf6YrRZEZFqpECjD/WmPYd6RlSf\n7D5Pf0a9aa9T7LOuqQnTo0fDIYfA0qW9X4+ISDVQoBHl/fySrujiUf366jPSZy0iw8mwDDS6+kV5\nwQWFC8FQen6Jak1ERKS/DctAY7heUIfrdouIyMAZMdAFEBERkaFrWNZo9CXdjhARESlNgUYv9Ucg\n0Z/dLEVERPqSAo1BoDcPFOsrpYKdlpYwb+5cjRciIiJbUqAh3VIq2EkCnenT+79MIiJS/RRoSK66\nasOyyy6FwanUxkVEZOhRoCG5UqAgIjK8KdCQiqmnjYiIdJcCDamYAgkREekuDdglIiIiuVGgISIi\nIrlRoCEiIiK5UaAhIiIiuRm2gUZ90m0i83+xaeh+2kry7W7a7HunnVbPjBkwY0bo7TFxYj3nnhum\np06tJ528r9ZZbWmHyzpVvsG3TpVv8K2z2stXDfukV9w9lxfwNuAmYC3wJvBzYNtuLHcR8CqwAbgX\neHeZtL8DOoAZXeQ5BfCGhgZPHH744UX/z043NLjD4Z4sWi5tJflWkna4rLPay6d9MvjKp30y+Mqn\nfVId60xPNzQ0OODAFO9BPJBnjcbNwJ7AQcBhwP7A1eUWMLNzgFOBE4APA28B88xs6yJp64B2wsaL\niIhIFcol0DCz9wLTgOPd/c/u/jBwGvB5M5tYZtGvARe7+13u/hfgi8COwL9l8v8noA44DrA8tkFE\nRER6L68ajf2AN919YWrefYTah48UW8DM3gVMBO5P5rn7OuDRmF+SbhvCLZmT3f21vi+6iIiI9JW8\nRgadCHQKAty93czeiO+VWsaBlZn5KzPLzAEecve7KijPaICmpqbNM9auXUtjY+MW/2enwyJraWrq\nOm0l+VaSdriss9rLp30y+MqnfTL4yqd9Uh3rTE+nrp2j6YlKGnQAlxAaX5Z6tQN7AP8ONBVZfiXw\n1RJ57xeXn5CZ/0ugPv4/A1gMjEm9353GoEcRghi99NJLL7300qtnr6N60hi00hqN/wSu7SLNC8AK\n4B/SM81sJPB38b1iVhDaW0ygc63GBCC5BXMgsBuw1qxT04zbzGy+u/9ribznAUcDLwLNXZRfRERE\nCkYDuxKupRWz+Iu/T8XGoE8DH0zaaZjZIcBvgXe4e9Fgw8xeBb7v7nPi9A6EoOOL7v4rM/sHYHxm\nsb8QGpre5e5L+3xjREREpMdyaaPh7ovMbB7wMzM7Cdga+BHhFsjmIMPMFgHnuPvtcdYPgG+a2XOE\n2oeLgZeB22O+r5Fp+xFrNl5SkCEiIlJ98nxM/FHAfxF6m3QAtxK6r6a9BxibTLj798xsDGG8jXHA\ng8Ch7t5aZj19XyUjIiIifSKXWyciIiIiMIyfdSIiIiL5U6AhIiIi+elJn9jB+AKWENqKZPsF/43O\n44AUS9MBtAGrgP8H/BBoisu+BayP/68ldNN1YDXhwXBLgMfjdEcqv1L9lN8ARgELKDzLJV2Ocst6\nXKYtM/1GLFdHLG+x5dbF999MrWMNncdJ6aqP9Y3A/5R4737g810svwp4KpZxQ5HtXxL3dVKeTUXS\ndPV6C3gY+Aqlj4nu7utNcV+3xldzZpkNhHFf1lZYxuyx+UoX6VqAZTH9uhL7pBV4CPgV8HqJbVtH\n4bh9GLgtVYbulnkTocH22tRym2K+S2I5krRt8fNYQXiQYnPclvT3JPl/WRfr7SD0arsqlXe2XK8D\ndxKOsdYu8kuWSZej3H7o7nckm3/2nNAR93+2/F29WjPrb4mf54a4X9dl3k+2rdQ5L71dL5cpTwvh\nGEmOk6viZ7kpk+6t1Pqy72X3YfJ6rERexV4bCeeeUufvrj6fNsK5pSWTLrtcUv52Ct+zZN/8Le7v\n7Pc9u10/pPx3ekm8Zl2VWmf6c2smfMeWx/2a3a7kXPRy3J5KjiOPy/4tlvFx4K9x+g3Cg04/XOn1\ndzjVaHwQqAUuA46h8OF8O/49BpgE/J4QRCQX5vUUDvSVhCCglvCslX8Evk44gXcQnjy7bVzfvYQH\nw40B9gL+G9gH+Eyc/xVCF2Bi3t+NecwCvgl8iHChP43QG8dj2iXA/8a0UDiwTwfOiO+vo3DQPA9s\nn1pPQ3wfOvfg2RD/LiIM8U7c1i8DR8R8OoD/AL4a80nrAD4KTKfw5YDQsPcvwN7AFXG5ptRyyyl8\nAdcTehotJFyYknyfiv+/k/DFOpLQiykt6bnkQD3hopJcuB4kfFkA/i/QCPyU8EV6mtA3fGMsVxth\nfy4jPBhwU2r//BX4c2rbkpP7N2KeG2L678ftrAF2j2V4ltA4ej3hIYNrU3kviuW4K5bpu3Q+Nv8h\n/v0tcEPMD8IouYfHvMYTxrgZHcv0eixPcpxsIhynRwBPEk4cEE4es2Oe2xMaYZ8G7AB8mnA83UPh\nWNkYX8nxdU+c/3vgk3E73g5sR/jck0cPtMTt2CpVpmcpHCsnED7DERSeX7SKQoBeTwhU7yd8NsTt\nJpbxrVjGGamyfQ94jkJQOorw3Xsf4QS6ksJx357K59G4/CtxP15GGIgwfQH5Uep9j2X6COH4eTqW\nuxWYTzj+XiNcYA6n8P1bH7erPZZlE+FzGR3//0bM+yXgW4TPPmkY30H40fNqav83UPhcF6TKtZJw\nHkoHM98B/hi34RdxXmssf7JPIHwWy4Br4vSrsdzEvK4gnOuSESNrgVsIn+MmwrFCfD8Jdlrj/GYK\n57UkGHk/cHCc917Cdym5+CfbXR/3CYTvbkfM76hYto2E73dH3A/LgJ3j/oNwzK6hcAGfAfyBcMyO\nJJzfkqBldXy9Hpetj8sbhQAjHTBeSehl2UE4fyRlmxuXeQY4APj7uF0tqf25jvCdudvMPk04Vpxw\nvl9P4bNrieUbB2xD+HwfTO3DtbEMYyh835LA4aq43x8FPhG3eyPhPNAB/Clu/z6E7/+2cfv+EfgX\nQm/Qe8zs76nEQNc0DGANhwOvxP+3GF0U+A3hCzSezjUee8f/P5pKu32ctzJ+cB2Ei/V3CV/mW4Eb\nipThEQq/gu8AFsf5d8YPfnYq7f2xzHMIgUcSTPy2SHnSF+/9KRycHcB5hC9IB3A8nSP1DsIX/Z/j\nvHtTeX4wzrs1Tv+A8Gs9HbVfEP8+l5o/Iy7bTrhwHJfa/0646N2b/gwIX4rVhIDCY5mToKmdMBYL\nhC/IX+J7l6W24xbCyayOEBhcRLhwOnAp4WTRDBwb83k1pt2bwi+VjXEfr4nL3RHT1hNO8OlfKh+N\n71k8Bo4Ddsrs16dimj8TAqn0L437CLUHyfL3xH27U2o9zYRgMjmmHPhF5vhbHdd3GoWarfR6Dkzt\njzfivKsIF4HkpPkKIcBJpn9JOFEnwU57XH874YR2J2GQvhtS+SS1Bcnn1BjnJxflJAA6Ok7/NPU5\nNMbtcOBz8XPo9P2M69xEqH1Lgpi7CcdN8v27LfNdTi4Gyd9W4NdxW5LP2OP6k/1+MiH4SPZ7S9xv\nK2O+O6TK9zzh2FhO+CX+GnB9TPcU8bsc83k5LvP7+HLCRXU54UR+a8z33+N7J6WWTT7PFakyOHBp\nnN6Lwq/rOsJ3L6kdSMqafM+T4z0J5p4mXKCTPJPXEYRzVfJ9fprCcfn+uN4nKZyfNgLXE4Ks5Di6\nKZZpE3BunHd6arlk/RcRzi2bgHPi/liZWr/HsieB1B2E70pyfmqP271LzHO/WJ7PxjyS2r89KZwn\nd0hdBzooHLfJd+RC4q/81DHVQagddMKP0hfjtv4Z+Fn8e0Xqc07e+wPhPNgal19O4Xz5+/j+T2MZ\nXyN8Z1YSahiXUzh3Hxjfd0Iw8J74/wUUAqhVFH78/DXmc298771xO84nHPOr0p9n5ryfPucm55oD\nK7reDvQFf4CCjFFxZ19XLNAgnFRbCSe6d1O4+GyKB3k7sFcqr7PiAXAO4WTjwI8JX5wb4oG+Jn6g\nX04t9xaFE0d7fH9nwglmHeFX147ABygEFtPiQZH8Gki+yBvjgfuPcXuSXw+Hxbwfi+kuovBL/GoK\nX/LkBJwONFYC74nlPYFCzc+oWIYfppbvIFxwmul8e+YBwq+cTYRfIdcQfvGmA5TkRH8O4WL7OuEE\n9G9x/o3x75qYz3bAoXHefAonA4/vJxe6pNp4RdyvTvjVnFRNz4/b4fEznEO4CGyK2/5qqpxrYznW\nE2qc0ifiU1PHTvLr7/2pz7U9LvcKhdqAR1PLr4v7bGUs1ybCCeN+Cr+U2mO6FsJxkXz2jYTAKn2b\n68zU55mk3UQYmfdBwkkryfdpwjGeTLcSfukk0xuAT6WCw6Tq1mOZk8/6obiP0+XdLi73YCxrcrFI\nPpkAmFQAAA+JSURBVO/kmF5CqA3oSH1mSbokTRKEjqDwHXM67+ObCLUYDlwd03+AcIJeSSGoT9I3\nEQLUdDDWEsvghOBhFeGCmATVyT5ui/kmVeAbUvOz5UoCxS/F/Z2UYxOF468upv0dhYAgCaSOj9uy\nOJN3I6HGMTmuRxFqHDcHAfGzdMJ39bXU8t8mHO9LUvukA9iXUHOUrq6fldruJgrfr2QdB1L4xf1Q\n/HsqnW+fnRf3VTOFWq7bKHwfOgg1Mn+gcGvvLkIQ9Kk4PynPFAoX3LWE80b6mF0JLI3/H0w4B11B\n4TbDmvjZJvt5LCE4T47v7eM+2BQ/4+QWxiJCzURyTD4R/ybzXo3bn9T6zKTw2XcQrid/IAQwLfGz\nT+/nVXEfvUQIADpime8nXBOSHz/NwMcoHAfzUmW6kEIN/C/pHER76r3X4j54ghAAJuW4C9g5dT77\nBIVzbnKtewP4OwUaXQcan4079YslAo2z4wG+ddzxr8cD54Y4/UfCBXw9hQtA8gvyDba8r30/4YT3\nFcIJ6QuxDG2EWoXkpPMY4Yu/LeGXdzqPDsIJ+7N0vh+7kvCr/oHMAXV0TPPH+PolhYt5kiZdzuRA\nSwcat8X5yYmlIbX/WglV9emTbzPhYvwZCifcxYQv1cKY5neEi0pShmZgKoWL3PcJUfe8VJrkJJ5U\n76Xv2T4b5yW//tbEee0UTqwbKFwQ0hfQ01Pb4IRfJU8SApA1hItCcrG4lUKwlpxU2yl8SQ+M+zbZ\nxt+m9t18Qu3MuUU+UyeceFvicsm8l1PTyb3WK+h8Dzm5OHQQfgW1E6o+1xGOlfRJbCXhgrEpbmeS\nx92ZMv2JcDFPLmqtwNsoVOOny50ERU9m5jvhl+aIuM7ku7E8k+4vFC506WWTC8BpFE7mc+J+T75j\nL6bSJwHCAsIv2GQdban/NxAuCG8QTt7JLaCW1DqTvJLl/isu+z+pdC2EW5Tn0zkoSX69ptt+JPv/\nxcxnkVycbs9sd/L6a5y/iELAM4nOAdEKCueT5ALSTiE4bicEfn+O722gUBORvF4hXAzbKQSMxfbv\nWalydlBoh5ak3xg/n+S47SBcNJOgPTmOHiUEKukyJN/t5HudtAdJvu8/i+ecdKDxr6ll2gi3VP+c\n2vaPU6gte4Xw3X0obueThB9jB6c+5zYKF+q2+Pob4YfTLRRqTdPf+fT+mEnh2E6O9fSxkNTQbh0/\ng1UUgv305/8MhfYV+6Xy+CmFc06yzmz7nvT7ybzH49+NFI6p5Dh/k/BD7hkK7d6ccJ5OrkE1cb/e\nT+Fa9xIwteJr7kBf9Aco0Jgbd2ryKykbaDQRqu9+QqFB1eOEKrEXCF/6bQjPXfkihQvZgRSq1X8d\nD5JF8SD7Qsz7CsI9t7nA7XGeE76EYwkXuJ/EA+FlQvXtMxSq8x+LyyYHxopUef6UOnhfpvDFn0Qh\n0Pg3wsk/e2F4kS0DjVUUovDnk+2I67+TQsPRJJ+HgZFxm7KNnZ4iBGl3U7gYOPBA6jN4JR7Q1xBO\nMovivN9SOJl+gdB+5Yo4L7kQfZPOwUWyzmTbksg+uSg2E6qrD4rLLYj7dwPhovw7wokraZNxOeE+\nfbKdyUkyya89lvP/t3dusVZeRRz/nUJDoPTBppE+GMCnVqOtxYQmGBNN4y1pq9Gk1qiJvcTQxgAJ\nqBGEVDE2qbWiD30oWEmlbcTaNlDQmlaghhah2Fparj3AOeAJlHI/wLnt/fnwn3HW3mefwyUc+tD5\nJ1/25VvfWrNmzZo1M2vW3s8jb+oUkh2fxH3IACqT8tzDd4/yiNHvnqx7RPcQi68rrsrGYypShL1W\nx8s07oWXHvZryFjeXdS118akq6i338qUBl0Z8fAF9xaUq1BDoXIfQy/fZ2Nw2t67R+ntHEAe2Enk\nzdaK5+ooTO9yuw7NtxuJOeY5E91EROcx++4p5PnOsnseLdpP5IZ4X99Dc95l2WVqnrWxu+j3o4We\n+C0RXvc/nPxjUfYVIn+rjM50I/nrQNshzcmffUjh/8M+lwuBX50FDa4/phLRHvd0NxF5Ss7LOtqe\nO2z99ohbFyGLZVs/LNqfTWxbVGgRv48weBZa2UXWrs/HPcYrz4XqYHDS5Aqjw+vfh/2BJq0NDe/3\nCuun930CsXVywsp4FPiaYvzcYHqASDb+Dooadxt/VhfjtgvJ33L7vNleH0Oyvd/GzfOjPLpUQzrz\nI8aTTcaD79pYbyjKuezMs9cu4/kMe9YT5X9k93cY7aXx73wqjY89SDe9QMy/Ukdste+mIVm9u+Dr\n1WjuTSXWwKvT0BjeyJhIWIODDA0UkqqhRB5PLHrLGNwBTGyqbyaNGdylYu5H1ug8YKuVn05kUt9i\n9FTAz+3+RjQB+1HU5FkT3oWEMv1q0dZdBS0PFW27kH3R7q1lsHHl4cKqqLs0NBYX7X/I+rHTaP81\ng09Z9BdXKfj/QaHaDWhhmlw880BBT3uL+kpDZlvR14/a97uJhaWGFOrSoi891r5HI1agybULGXRe\nz140Ac80tVkrXn0BdCPuSaPlYeDVQnl5QumbxrexSPn9pql/vrD6Hqp7ZO5BDRSvXm4AyVRFZKdP\nJxTMtqY21xKG1Y9R4p//Wm9l/OpAcriWSOB0Wl2Om08dvGL860Se+VakCP9bjMsEND/K/pZ1HLJn\nltv4eFub7P5tBf1vF2Pvi5fvsdda1D0A7Lby86zeGlpUPk8Y7u8hp2Blizp8fErav17QMZ3Y5nkE\nycdcwmhfQ3j6zxFz8u9E9MmTjvch+dtv9J4m5PiAjafnrlTAxoKGinBaJhVlXBb2IplxXtaRPtlH\nRDJ6UYLiNTYW5XjfQ8jCN2iMrKws9FYps828LOW+z2iaSSzazq860hHVMFdJ33ZCPz1t311PGBpb\niCiLRxuadVQfEe35E6Ffm407X5i/z2CayrL3It35ZlOZ5tMz53KaZqg25trz7xr/r0KGaXtRpll+\nnbcbbdx/YGO+kYjIXG+fdzhfW6yhO9Ffh5zzuvtBOnXiuIvGf4dtxt1oQt+KMnb3Is/zyygBprOp\n/ONocDqRsvE8gi3IK74WDbDz+loiB2C10QOwua2tbTzaLx9l11xkSd6MFM8YJOhfIE4s+OkMiISo\nHpRxXAEH7M/pfLuiRJldfmWL+zd7+1VVefTiKqN9itHvJwTqaMvkBrvnJyPmoxDugNH3XFVVe4s2\nuor3lxWfby/qOWTf1bxgVVV7kGIeb1+NKniwCinnbyOerUd8BSnfTfbcGKRse9Gi6JGin9nrcSL6\n8nLx3v+Z2E/eTAba29rallk9fVbv56qqOlpV1Zmqqg4i+TiJIhagRQ+kcNcVfPBF3g3BmcaXBxHP\nb7RyfhrnOuNdHVjmbaKF9CYr04MiZl9CRpmf2PgEWniPWVk3gp3WBWhMP1vQd8jqHkOcZpiIQsOz\n0bhcYX2eQijVOUROAsjjuxzlIe1EY7YfJSOCFlen/xBAW1vblUSCtv+T5AziVIQvnGusr6CFYzTy\nNM+g6CRIntqRh/yq8fAIMrZAi7N7qh6pmVTw4ZNGP2gebkLz2/NvxiJ5BIWiR1mb4wh9MA7JUwdS\n9uMIQ9H712Xj6REAkNHuNIAWXIh58H9ZQPL/cRq3LCcTp3cGiLD9M0gvOH0VMuy60FwfXfQZoxnE\nR1Bk6yCwhNh6hDi1UaEIyETE877i+2nWx58gfi5BxuZtRp+Xm0/o8DFIT7n8nrDPXnYSiugeR3r6\nhuJymo4i+e4zvryI+Pqa1eFHx3fbc08QzoS38w6xBeWGr0dR/Wj3TcRJw/n2eQtxYg4rVyHdcATN\nwcXWr04ieXc0kssxAFVVeWRtvL2+QTgUO4htur+hxNGxaG3rRZHaD9v98cCnkLy63m/GZd7uOeP9\njjBc4mjGFUhhLDGmPogWtDoKsU6j0eJ9FxkLJ5FBMAUdV1yEBHMiOra6CimwWwkrdgWKWPh+3yPE\n8atD1vZSQvEuQhPvEOGd9yPFPYfwhjqR0C21z+uQQvKchqOEN1FHyvzfhKJ8kggxlqdDPNfhUeSp\n+vOLkQHxPaPtOJEP0FWUq6MFcjWaPE5DBxGWfQYZSWUy5Ra0F+5W90G0SK9HytRzVyqkAP8K3ImU\n1ZminTX2eszK+DN9VqeHJD0HxI2HfxH7ltuNfy8a7zegUzsHbAzdk3YvYSXyovqIbZABK/81G7v7\nUN7KHLSAvUV4Mn8mQvU91pcOG6/dxImB161ffyDC3xVKOPY8gjqSiZMo2nA/kQlf2Ws7ks9vEh6M\nb1VttzZ8v/0dG4c7kCyXY+1Rn58S20g1JJse6eqxMagR2w9P0LgF9LC1UUde4ENEhKBCStcX1+Xo\nuF257eHK8yUit2ApMXd+QURHPJ9itn3XTWxRdSNDx+eILxDLrH5ffL3eXxqt/jsqlY3D74g8A+eJ\nG/7eZ5etuo2BJ+v6iQj3PDtRnlaFoorfojGS8DSax17vIrt8wdllYzsPea71grfume8syvca3fuJ\n00Fe9g1izvyTxpD8YmRceB9PGm0njK/lsUyfr93WvxU0evX9SEYmoK2Bw0infqYYq7rdc74/jxLU\n/aSOG9G/L+ocsLH5FZrPn0Zz13m3EM1Lp38GsUbUkT5pt3oWWJ/92Tqas340+S9EsqcnsZ+x+iYg\nPdWFdMQswghx3e1yebmNXd3G34/xek7MC0TexOsoStasyw9b3ceQAbTe2h0wGm4nElI3E85Vzfg+\nGcnUV5DDOQXpitPAx85r7X2/F/9LbGjMJhRA8zZHq1BZq8ufc+/IM+/9dMNLxP7eKaS8/Lz+20hB\n1ZDR0Vx3HSnU64ZpfzgazxaKO5f+ne35GgoDX8jzs86jnYtRphyv5msrQ/94mde/Einecu/0Qmmp\n2fgcQcriXhrzWJzOTuPvUFsCw9V/Nhn2BWy4egaMrlUM/SNzHgWqn6W9cxmHyniyj8aEzAu5nkLO\nhEcVh+LR+Y5jO5rTdzD4B5CaedCKJ3Uih+G0jYErfN/SKfnfw9lpuljXRrRg9VyEulptNbXi8XB6\nqg4sMH19P1oQe5vaGO7ZE0PcW4TkY7gfHyvrqSFD/RQypNxI8u3LVmPTT/yAY6v6FqDtjceRkTlU\nX7ZZ/z2CXc5BNzSO0+hUNrd1nDAammlslbA8FA/8sxuizwJTznftzT9VSyQSiUQiMWL4IOZoJBKJ\nRCKRuERIQyORSCQSicSIIQ2NRCKRSCQSI4Y0NBKJRCKRSIwY0tBIJBKJRCIxYkhDI5FIJBKJxIgh\nDY1EIpFIJBIjhjQ0EolEIpFIjBjS0EgkEolEIjFiSEMjkUgkEonEiCENjUQikUgkEiOG/wHHPU15\nCyeeUgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2bdbc50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = rfc.feature_importances_\n",
    "std = np.std([tree.feature_importances_ for tree in rfc.estimators_], axis=0)\n",
    "indices = np.argsort(importances)[::-1]\n",
    "print(\"Feature ranking:\")\n",
    "for f in range(len(importances)):\n",
    "    print(\"%d. feature %d (%f)\" % (f + 1, indices[f], importances[indices[f]]))\n",
    "plt.figure()\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(len(importances)), importances[indices],\n",
    "                    color=\"r\", yerr=std[indices], align=\"center\")\n",
    "plt.xticks(range(len(importances)), indices)\n",
    "plt.xlim([-1, len(importances)])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "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": 1
}