Matrixnet borders.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:8743a5eed8caf0894d9cf4d8e53e56d15bae9396c64f41a86dc0217d8a72a65d"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import sys\n",
      "sys.path.insert(0, '/home/mikari/REP/rep_core')\n",
      "#sys.path.insert(0, '/local/userdata/shelf/notebook/axelr/lhcb_trigger_ml/IPythonWorkflow/')\n",
      "sys.path.insert(0, '/home/mikari/topo_code')\n",
      "import rep\n",
      "import topo_utils\n",
      "reload(topo_utils)\n",
      "import pandas\n",
      "\n",
      "\n",
      "train = pandas.read_csv('/local/userdata/shelf/notebook/mikari/topo/learn_data_updated_data/train', sep='\\t')\n",
      "test = pandas.read_csv('/local/userdata/shelf/notebook/mikari/topo/learn_data_updated_data/test', sep='\\t')\n",
      "\n",
      "train,test = topo_utils.append_labels([train, test])\n",
      "train = train[train['is_passed_l0_goodGenB'] == 1]\n",
      "test = test[test['is_passed_l0_goodGenB']==1]\n",
      "train = topo_utils.add_folding_field(train, 1000)\n",
      "test = topo_utils.add_folding_field(test, 1000)\n",
      "train_subset = pandas.concat([train[train['signal']==1][train['fold'] < 1000], train[train['signal']==0]])\n",
      "# test_subset = cern_utils.data_storage.DataStorage.union([test.select('signal==1').select('fold < 300'), test.select('signal==0')])\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "/mnt/w76/venv/py27/local/lib/python2.7/site-packages/pandas/core/frame.py:1712: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
        "  \"DataFrame index.\", UserWarning)\n"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "empty_events = {\n",
      "    11114001: 164  ,\n",
      "    11124001: 49  ,\n",
      "    12103035: 421  ,\n",
      "    12165106: 273  ,\n",
      "    12265042: 53  ,\n",
      "    11874042: 129  ,\n",
      "    11296013: 23 ,\n",
      "    30000000: 107352  ,\n",
      "}\n",
      "\n",
      "test_empty_events = {}\n",
      "for key, value in empty_events.iteritems():\n",
      "    if (key == 30000000):\n",
      "        test_empty_events[key] = value * 0.5\n",
      "    else:\n",
      "        test_empty_events[key] = value * 0.5\n",
      "    \n",
      "print test_empty_events"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "{30000000: 53676.0, 11124001: 24.5, 12265042: 26.5, 11296013: 11.5, 11114001: 82.0, 12165106: 136.5, 11874042: 64.5, 12103035: 210.5}\n"
       ]
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "first_factors = [\"sumpt\",\n",
      "             \"m\",\n",
      "             \"mcor\",\n",
      "             \"ipchi2\",\n",
      "             \"vchi2\",\n",
      "             \"sumipchi2\",\n",
      "             \"fdr\",\n",
      "             \"nlt16\",\n",
      "             \"minpt\",\n",
      "             \"ptau\",\n",
      "             \"q\",\n",
      "             \"eta\",\n",
      "             \"nhlt1\",\n",
      "             \"n\",\n",
      "             \"fdchi2\"]\n",
      "second_factors = [\"sumpt\", \"m\", \"mcor\", \"ipchi2\", \"vchi2\",\n",
      "                  \"sumipchi2\", \"nlt16\", \"minpt\", \"pt\", \"n\", \"nhlt1\", \"fdchi2\", \"eta\"]\n",
      "\n",
      "third_factors = [\n",
      "             \"m\",\n",
      "             \"mcor\",\n",
      "             \"ipchi2\",\n",
      "             \"vchi2\",\n",
      "             \"sumipchi2\",\n",
      "             \"nlt16\",\n",
      "             \"minpt\",\n",
      "             \"ptau\",\n",
      "             \"q\",\n",
      "             \"n\",\n",
      "             \"fdchi2\"]\n",
      "\n",
      "fourth_factors = [\"m\", \"mcor\", \"ipchi2\", \"vchi2\",\n",
      "                  \"sumipchi2\", \"nlt16\", \"minpt\", \"pt\", \"n\", \"fdchi2\", \"eta\"]\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 62
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "reload(topo_utils)\n",
      "metric = topo_utils.metric_generate(test, 3000.0 / 1e6, test_empty_events, metric_plot=True)\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "219\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import pandas"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "train_sgn = train_subset[train_subset['truematch'] == 1][train_subset[\"presel\"] == 1][train_subset['signal']==1]\n",
      "train_bkg = train_subset[train_subset[\"presel\"] == 1][train_subset['signal']==0]\n",
      "train_labels = numpy.array([1] * len(train_sgn) + [0] * len(train_bkg))\n",
      "train_data =  pandas.concat([train_sgn, train_bkg])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from rep.classifiers import EventFilterClassifier\n",
      "base_name = \"mikari-topo-29_12_2014_4:37\"\n",
      "all_intervals = [4, 6, 8, 16, 32, 64]\n",
      "some_intervals = [4, 6]\n",
      "iterations=5000"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 60
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "mn_first = {}\n",
      "for inter in all_intervals:\n",
      "    mn_first[inter] = EventFilterClassifier(formula_name=base_name + \"-\" + str(inter) + \"-first-features\", \n",
      "                                         dataset_name=base_name + \"-\" + str(inter) + \"-first-features\", url_base='w80.h.cern.yandex.net/')\n",
      "    mn_first[inter].set_params(\n",
      "        intervals=inter,\n",
      "        features=first_factors,\n",
      "        iterations=iterations,\n",
      "        features_sample_rate_per_iteration=0.8,\n",
      "        training_fraction=0.5,\n",
      "        regularization=0.05,\n",
      "        sync=False)\n",
      "    mn_first[inter].fit(train_data, train_labels)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object') ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "mn_second = {}\n",
      "for inter in all_intervals:\n",
      "    mn_second[inter] = EventFilterClassifier(formula_name=base_name + \"-\" + str(inter) + \"-second-features\", \n",
      "                                         dataset_name=base_name + \"-\" + str(inter) + \"-second-features\", url_base='w80.h.cern.yandex.net')\n",
      "    mn_second[inter].set_params(\n",
      "        intervals=inter,\n",
      "        features=second_factors,\n",
      "        iterations=iterations,\n",
      "        features_sample_rate_per_iteration=0.8,\n",
      "        training_fraction=0.5,\n",
      "        regularization=0.05,\n",
      "        sync=False)\n",
      "    mn_second[inter].fit(train_data, train_labels)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object') ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n"
       ]
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "mn_third = {}\n",
      "for inter in some_intervals:\n",
      "    mn_third[inter] = EventFilterClassifier(formula_name=base_name + \"-\" + str(inter) + \"-third-features\", \n",
      "                                         dataset_name=base_name + \"-\" + str(inter) + \"-third-features\", url_base='w80.h.cern.yandex.net')\n",
      "    mn_third[inter].set_params(\n",
      "        intervals=inter,\n",
      "        features=third_factors,\n",
      "        iterations=iterations,\n",
      "        features_sample_rate_per_iteration=0.8,\n",
      "        training_fraction=0.5,\n",
      "        regularization=0.05,\n",
      "        sync=False)\n",
      "    mn_third[inter].fit(train_data, train_labels)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object') ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'ptau', 'q', 'n', 'fdchi2']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'ptau', 'q', 'n', 'fdchi2']\n"
       ]
      }
     ],
     "prompt_number": 64
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "mn_fourth = {}\n",
      "for inter in some_intervals:\n",
      "    mn_fourth[inter] = EventFilterClassifier(formula_name=base_name + \"-\" + str(inter) + \"-fourth-features\", \n",
      "                                         dataset_name=base_name + \"-\" + str(inter) + \"-fourth-features\", url_base='w80.h.cern.yandex.net')\n",
      "    mn_fourth[inter].set_params(\n",
      "        intervals=inter,\n",
      "        features=fourth_factors,\n",
      "        iterations=iterations,\n",
      "        features_sample_rate_per_iteration=0.8,\n",
      "        training_fraction=0.5,\n",
      "        regularization=0.05,\n",
      "        sync=False)\n",
      "    mn_fourth[inter].fit(train_data, train_labels)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object') ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'fdchi2', 'eta']\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'fdchi2', 'eta']\n"
       ]
      }
     ],
     "prompt_number": 65
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "test_subset_labels = test['signal']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "eff_first = dict()\n",
      "\n",
      "for inter, mn in mn_first.iteritems():\n",
      "    preds = mn.predict_proba(test)[:, 1]\n",
      "    eff_first[inter] = metric(test_subset_labels, preds)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold'], dtype='object') ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.989257735988"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.989117760591"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.988631121019"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.988657641979"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.989692218286"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'fdr', 'nlt16', 'minpt', 'ptau', 'q', 'eta', 'nhlt1', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.989257735988"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "eff_second = dict()\n",
      "\n",
      "for inter, mn in mn_second.iteritems():\n",
      "    try:\n",
      "        preds = mn.predict_proba(test)[:, 1]\n",
      "        eff_second[inter] = metric(test_subset_labels, preds)\n",
      "    except:\n",
      "        pass"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object') ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.989127101431"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.989878339321"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.988802627007"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.988899534008"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.990197670288"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['sumpt', 'm', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'nhlt1', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.989127101431"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "eff_third = dict()\n",
      "\n",
      "for inter, mn in mn_third.iteritems():\n",
      "    try:\n",
      "        preds = mn.predict_proba(test)[:, 1]\n",
      "        eff_third[inter] = metric(test_subset_labels, preds)\n",
      "    except:\n",
      "        pass"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object') ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'ptau', 'q', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.986384110174"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'ptau', 'q', 'n', 'fdchi2']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.987623404662"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 66
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "eff_fourth = dict()\n",
      "\n",
      "for inter, mn in mn_fourth.iteritems():\n",
      "    try:\n",
      "        preds = mn.predict_proba(test)[:, 1]\n",
      "        eff_fourth[inter] = metric(test_subset_labels, preds)\n",
      "    except:\n",
      "        pass"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object') ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.988771390732"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Index([u'event_id', u'svr', u'signal', u'presel', u'truematch', u'sumpt', u'm', u'mcor', u'ipchi2', u'vchi2', u'sumipchi2', u'fdr', u'nlt16', u'minpt', u'ptau', u'q', u'eta', u'pt', u'nhlt1', u'nmu', u'n', u'bdt', u'fdchi2', u'NUMBER', u'btau', u'is_passed_l0_goodGenB', u'channel', u'channel_label', u'fold', u'bdt_cut', u'bdt_passed', u'bdt_cut_m', u'bdt_passed_modified', u'modify_prediction'], dtype='object')"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " ['m', 'mcor', 'ipchi2', 'vchi2', 'sumipchi2', 'nlt16', 'minpt', 'pt', 'n', 'fdchi2', 'eta']\n",
        "2"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.988357585007"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 72
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def plot_eff(data_dict, p_label='p_empty', total_label='total_empty', alpha=1, move=3, bar_width=1):\n",
      "    plt.subplots(figsize=(13,7))\n",
      "    color = ['b', 'g', 'r', 'cyan', 'pink', 'orange', 'black', 'yellow', 'brown', 'magenta', 'purple', 'indigo', 'orange', 'violet', 'teal', 'grey']\n",
      "    num = 0\n",
      "    bars = len(data_dict) + 1\n",
      "    for key, value in sorted(data_dict.iteritems()):\n",
      "        move = num\n",
      "        error_config = {'ecolor': '0.3'}\n",
      "        modes = sort(list(set(value.keys())))# - {30000000}))\n",
      "        channel_count = len(modes)\n",
      "        index = np.arange(channel_count)\n",
      "        plt.bar(bars * index + move, [1 - value[mode][p_label] for mode in modes], bar_width,\n",
      "                       alpha=alpha, color=color[num], \n",
      "                       yerr=[sqrt(value[mode][p_label] * (1 - value[mode][p_label]) / value[mode][total_label])\n",
      "                                                   for mode in modes], error_kw=error_config,\n",
      "                       label=key)\n",
      "        \n",
      "        num += 1\n",
      "        plt.xticks(bars * index + move * bar_width, modes)\n",
      "    plt.xlabel('Modes')\n",
      "    plt.ylabel('Inefficiency')\n",
      "    plt.title('Classifier Inefficiency for Matrixnet')\n",
      "    plt.legend(loc='best')\n",
      "    plt.tight_layout()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 68
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "first features"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plot_eff(eff_first)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA58AAAHyCAYAAABs5HrRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x5b125510>"
       ]
      }
     ],
     "prompt_number": 69
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "second features"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plot_eff(eff_second)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA58AAAHyCAYAAABs5HrRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAYIH9CdI2vdQ+weMfAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x46679a10>"
       ]
      }
     ],
     "prompt_number": 70
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "eff_total = dict()\n",
      "eff_total[\"4-15-first-set\"] = eff_first[4]\n",
      "eff_total[\"6-15-first-set\"] = eff_first[6]\n",
      "eff_total[\"4-13-second-set\"] = eff_second[4]\n",
      "eff_total[\"6-13-second-set\"] = eff_second[6]\n",
      "eff_total[\"4-11-third-set\"] = eff_third[4]\n",
      "eff_total[\"6-11-third-set\"] = eff_third[6]\n",
      "eff_total[\"4-11-fourth-set\"] = eff_fourth[4]\n",
      "eff_total[\"6-11-fourth-set\"] = eff_fourth[6]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 75
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plot_eff(eff_total)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA58AAAHyCAYAAABs5HrRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xl4FFW+//F3N2uW6tCJJGQhBJBVIjIGXJhAHPCyKZoR\nXIZF0VwGr+iAo44DRILIIOjoBXF+yCjjgoKKAi4szzUKBgQeZjBxBhBwQoBMwpqEdJRsnf79UUmT\nQCekIQ0kfl4+9VBddeqcU9UlT3/5njoFIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi\nIiIiIiIiIiIiIiIiIiIiIiIiIiJXhBTgHR/W/y9gQOV6M+A9wAFsA34JfO/Dti/GvcBRoBDoDfQC\n9gBFwKPA/wNm1KOetcA4H/XxQlQ/j8mXuS8XwgHEXO5OiIiIiIiIZ78B/o4ZSOUDXwC3VO6biW+D\nz+p+BRwGWvuo/jeB2Q1U1yHg9mqf3wb+3EB1X04NfR4pQAXw2Fnbf1e5fWY969kIPNRgvbo4KVy6\n/ydERASwXu4OiIhIg3gceBGYBgQBVwEvA0Mq91suYV+igCyguAHqauZhm6tyuVgWIBLYXW1b1Fmf\nG6uLOY/arvk+YPxZ2+8H9lL/7+N85ZrXsx4REREREbkMgjCHKQ6to0wKNbM8qzCHmxZhDo29rtq+\nO4H9lftygKcqt4cDGyrbyge2VDsmCxiEmdU6DZRXlpsJJGBmQqvEYA5TLQBygT+c1c+VlX3NBx70\ncC5/40zmMwYz8za+sg+FwLPVylory/4HOAV8ghmYt6o8v4rKP38AUiv7fbqyni6cm2X9DeZwVgdw\ngDPXfCM1M3qPVuvPJqBztX0VwG8xg7Yi4HVq/uPA1Mq6HZjDlX8BPFl5XapbCPwv5/ryrPO4Ggiu\nPP4U5vc+p1qbD2B+ly8Bx4BZHuqsypzvBnpWbrsG2FW5vSrzace8R05U9v8LoEPlvjnV+uWo7H/V\n9fifyuvxQ7VtnYCWwLecGTrcrLKvVUOhU4APgLcqz+0H4KZq/Y7B8702FCgBSiv78q2HcxYRERER\nkbMMxQxi6pJCzeDzN5gBWDPgeWo+k3kS6F+5HgjEVq7/GfhL5TEW4IZqxxzAHG4LZjYsrdq+BM4E\nn80wg7enMQPDKODfmAFvVT9Pcyaoa+XhXDwFn69iZs2uxQwqelXun4YZrLStbPtVYHW1uqqCnCpf\nUTPg/RtngtlbgDzMZ1gBQoGuHo77DWaWsGPl56epGdxUYAb//kB7zICvaujvA5hB6zWVn6Mry7TD\n/I6DKrc3xwwi++DZ2eexEliBORQ6HDNorAroHgDKOBM8e7rmVcHnHzHvF4D5led2dvA5AvNa+1Xu\nW19Hv8C8Hp9g3mstq22r+l6uwbzu3YHpwDecCZxTMO+XQZWf/wT8o3L9fPfaTMzhySIicolo2K2I\nSOMXgvnj3BvvYQZpTuA5zCCqbeW+IszsllG5/s9q28MxM1kuYHstddc1xPeXmEHX85gBRjZm5u/u\namU2cyZgKann+VRl1b4D0jEnDwJIApKB45w519uo+3nU2vo/AXitsn9gBo37PJT7b8zzO1D5eT7m\n9e1SrcwLwE+YQflXmEEzmIHZXMzgEMxnUg8DRzAD+tGV24dWnlNdGbuq8/ADRmIGbsWYGcAXgTHV\nyh4E3qhc93TNq+paBtyHGfzeU/m5unzgc8xrfRqYx5mJqM6uq7p5mPdXqYd9uzC/tzWYw8vHUXP4\nbhpm1rqqf1XX8nz3mqWWvoiIiI8o+BQRafxOYg6rrK+WmMM1D2IORzyM+WM+sHL/3ZjBykHMQCu+\ncvuLmMHQF5jZuekX0NcoIAIzSKla/gi0qVbmyAXUW/2YnziTvWuPmWWsams3ZoATUkddtT2X2A7I\nrEdfooAF1do8Wbm9bbUytfW3rjbeAsZWro/l/JPlVJ1HCGaweKjavsNAWLXPueepq6q+w5hDW+di\nBt7ZZ5UJwhyq/B/Me2sL5rlVD/I8Xd/ztf82ZhZ4LWb2srqj1dZ/wsx4VmU6z3eviYjIJaTgU0Sk\n8dta+Wddz3xWNx5ziGx/zB/iUdTMAm3HHAZ6FfAh5jN1YD4b9zvM4ZDDMGc+/S8v+3oEM2ixV1ts\nwPDK/Q0xkVB1uZhDMqu3548ZHHkrh5pDdOtqc8JZbQZgDhe9mDbWYGb1emEObX23HvWBGfw6OfPs\nJZhBubdBftX98TZmBrL6kNWq7+1JzEmcemPeW/2peW9d6Pf7F+AzzHu8f7XtddV3qe81ERE5DwWf\nIiKN3yngGcwhhbdi/t3eAjNAnOehvD9mMHIKc/jpc9X2tcDMfAZwZjKeisp9Qznz7sWiyjq8/QG/\nqbJ/kzEzsBagG+akOlC/YZDeDJVcgjkkN7zys50zwUd96q8eOL0JTARurvwcRs2htNXbnIY50Q+Y\nGeU7PZTz1MbfMCfFqZrUJwYzUARzGOtHmEOmt3Nu1rG28ziN+UzlbM488/k4sPw8x9fmfcz77EMP\n/ffHfH7UgRnoJZ91bB5nnoWtr3GYz7bej/kPHm9h3p9VbdfmfPfaScxrq6G3IiKXiIJPEZGm4SXM\nWWn/hDnc8RjmrKmfV+6v/nqSNzEzbEcxn6fbSc0gMgkzsPkR84d71bOBPYGvK7fvwHxG8P889MXT\nq1CqPpdjvv5lUGX7BZgZNHsdx56v/rrKz8EcOrwdc+bXnZwZRlzbsWfXXfV5I2bw8yZmcLUVz1nK\nZZgB6LrKNvdSM/j0dG2qtr0FLKo81oE5zLT6EOG3MDOf9Xk/ZfV2JmIGYEeBDMws4ise2q+rrqoy\nxZgz6hZ72Pcy5tDbfMxZlFPPqvsVzCHDp/A8U+/ZfY+urHM85pDa5Zjvsn2pjr7X9177EPN52FOV\ndYqISCO3FPMv/H/WUWYhZ3781DZrn4iIiJiZuh8583yuiIiIVIrHDChrCz7v4syU930wZygUERER\nz17EHJorIiIiHsRQe/D5BmYAWuVfmBNfiIiIyBkBmEN4/4k5FFVERKTRaX6Z24/izIvHwXzGKIrz\nT6IgIiLyc/Ij5gQ+IiIijdblDj7h3Fnmzpn0oHfv3q6MjIxL1B0RERERERGplAFc1xAVXe7ZbrM5\nM4U81JL1zMjIwOVyaWnAZebMmZe9D1r0vf1cFn1vjXPR99Y4F31vjXPR99Y4F31vjXPx9nvDfHdz\ng7jcmc+1mFOur8R875aTC3vxt4iIiIiINDE2WzAORz4AHTt2pF27dueUOXLkCAcOHDA/WHG/nbo+\n5Vtgvpy43vW3aAFlZfUvD7Ro3pyy8vL696kZlDm96FM19SlvtcKsWbMuqP6L5evgczkwELgK89nO\nmZjfMcBrmC/LvgXzVSslwAQf90dERERERBoJM/A0n8o7cMBc6lRhgRRz9UDlf3UpS6n2zF89GrCU\nlYGrsj+Vy/mUWSy4Nu6oR8nKNhL64nrX3anztmIZc2b9wIED5w0aKyrcp1C/+s9+SPIi+Dr4vK8e\nZSb7uA/iQUJCwuXuglwAfW+Nk763xknfW+Ok761x0vfWOOl7E281YBzrUy6X65x5iEREREREpAmz\nWCx4mI+0riPcmc96SfG69uppw3oedDGZz3qUH3P+Mmfz5hQqM58NEjde7mc+L0pwcDD5+fmXuxvS\nCNjtdvLy8i53N0REREREfrYadfCZn5+PMqJSH5aGHKwuIiIiIiJeu9yvWhEREREREZGfAQWfIiIi\nIiIi4nMKPkVERERERMTnFHyKiIiIiIiIzyn4bIJ+//vfExQURERExCVpLyUlhXHjxl2StkRERERE\npHFqcsGnzRaMxWLx2WKzBXvVn/3799O6des6g7Pk5GRiY2Np0aIFs2bNqrHvyJEjjBw5ksjISKxW\nK4cOHaqzvczMTP7yl7+QmZlJTk6OV32tj40bN9K+ffsa2y7VTLIJCQm88cYbl6QtERERERFpWE0u\n+HQ48jFfFeubxay//h555BH69etXZ4DWpUsXXnjhBUaMGHFOOavVyvDhw/noo4/q1V52djZXXXUV\nISEhXvWzPsrLyz1uv1Svu9HrUkREREREGq8mF3xeSVasWIHdbmfQoEF1Bmjjx49n6NChGIZxTrnQ\n0FAmTZpEXFzcedv74osvGDp0KDk5ORiGwYMPPgjA8uXL6dSpE4ZhcNNNN5GRkeE+xmq1kpmZ6f78\nwAMPkJycDJhZzqioKObPn09kZCS/+c1vGD58uLt+m81Gbm4uFouF0tJS7r//foKCgrj66qvZunVr\nrf3csmUL1157LQEBAYSGhjJlyhT3vi+//JI+ffpgs9no3r0769evB2D69OmkpaUxefJkDMPgscce\nO+/1EBERERGRK4eCTx8pLCxk5syZvPzyy5csMzh48GDWrVtHREQEDoeDpUuXkpGRwcSJE3nnnXdw\nOBzcd999jBgxgtLSUo91VA0vrnL06FF++uknDh8+zDvvvFOj/sLCQsLDw3G5XHzyySeMHz+eU6dO\ncffddzN58uRa+/nYY4/xhz/8gR9//JHDhw+7hyT/+9//5q677uLFF1+ksLCQ1157jXvvvZfc3Fzm\nzJlDfHw8r776Kg6Hg4ULFzbsxRMREREREZ9S8OkjycnJJCUlERERcUmHi54d6H7wwQfceeed9O/f\nHzADv+bNm7Nx48Z61dGiRQtmzJiB1WqlVatWtQbS8fHxDBo0CICxY8fy3Xff1Vp/YGAgP/zwAydP\nnqRVq1Zcf/31ACxbtozbb7/dXc/AgQO58cYb+fTTT2s9PxERERERaRwUfPpAeno6qamp7uGk1QOm\na665BsMwMAyDLVu2XFQ7aWlp7rpiY2M9ljl27Ng5EwS1b9+eo0eP1quNkJAQmjdvft5yYWFh7nV/\nf3+cTicVFRW8++677j6OGDECgCVLlrB792569OjBL37xC1avXg2Yz6t++OGH2O1297Jlyxby8vLc\ndeu5TxERERGRxun8UYV4bdOmTWRlZREdHQ1AUVERTqeTPXv2sGvXrjqP9Sa4io+Px+Fw1FkmLCyM\nrKysGtuys7PdwWKLFi346aef3PtOnDhxTrB6vv7V1ecxY8YwZsyYGtu6devG+++/D8CqVau45557\nOHHiBOHh4Tz44IO8+uqr9W5bREREREQaB2U+fWDixIlkZmaSkZFBeno6kyZNYsSIEWzYsMFj+fLy\ncoqLi3E6nZSVlVFcXExFRYV7f3FxMcXFxees18eoUaNYs2YN33zzDS6Xi0WLFlFWVsbAgQMB6N27\nN++++y4VFRV8+eWXfPXVV3XWFxwcTH5+fo2g19uhsO+//z75+easwYZhYLVasVqtjBs3jlWrVvHV\nV1/hcrkoKytjy5Yt7lfGBAcHc+DAAa/aEhERERGRK0OTCz4Nww5YfLaY9dfNz8+P0NBQQkNDCQsL\nIzAwED8/v1pff5KUlIS/vz8rVqxgzpw5+Pv7s2zZMvd+f39/bDYbFouF7t27ExAQUGf71TOE1113\nHa+99hpjx47FZrOxbNkyPvvsM1q1agXAggUL3ENd33zzTe64445a6wKIjY1l5MiRREVFERwc7J7t\n9uxydWUp16xZQ9euXQkICGDy5Mm8/fbbBAQE0KVLF5YvX860adMICgqiXbt2PPfcc+5A/NFHH2XZ\nsmUEBQXVmCFXRERERESufI1lHKPLU3bNYrFoAhqpF90rIiIiIo2PmdDw5jecBVK8KJ7ide3g7W9K\niwXXxh31L57QF9e7XlQ/5vxlzubNKVTmlBokbmxymU8RERERERG58ij4FBEREREREZ9T8CkiIiIi\nIiI+p+BTREREREREfE7Bp4iIiIiIiPicgk8RERERERHxOQWfIiIiIiIi4nMKPkVERERERMTnFHz+\nDDzwwAMkJyfXut8wDLKyshqsPhERERERkbM1ueDT1saGxWLx2WJrY/OqP/v376d169aMGzeu1jLJ\nycnExsbSokULZs2aVWPfkSNHGDlyJJGRkVitVg4dOlRne2+++Sbx8fE1tlX1vTYOh4OYmJjzn0w9\n67sQWVlZWK1WKioqGrReERERERG5MjS/3B1oaI5TDkjxYf0pDq/KP/LII/Tr16/OYK1Lly688MIL\nLF68+JxyVquV4cOHM23aNG6++eYL6vPFqKiowGo9998oXC6XT9rzVb0iIiIiInJ5NbnM55VkxYoV\n2O12Bg0aVGdQNX78eIYOHYphGOeUCw0NZdKkScTFxZ23vT179vDwww+zdetWDMMgODjYvS8vL4/b\nb78dwzC47rrr2Ldvn3uf1WolMzMTMIfUPvzww4wYMQKbzcbGjRvZunUrPXr0ICgoiHvvvZfi4uI6\n+5GSkkJYWBiGYdClSxdSU1MBM5BNTk4mMjKSoKAgRo4cyYkTJwAYMGAAAG3atMEwDLZv337e8xUR\nERERkcZDwaePFBYWMnPmTF5++eVLls3r0aMHixcv5qabbsLhcJCXlweY2cQVK1YwZ84cCgoK6NWr\nF3/84x9rreeDDz5g9uzZFBYWcv3115OYmMjvfvc7Tp06xdixY/n4449rzeT+85//ZOnSpWRkZOBw\nONi0aROdO3cG4Pnnn+fLL78kPT2dvLw82rdvT1JSEgBpaWkAnDp1CofDwQ033NCQl0ZERERERC6z\nJjfs9kqRnJxMUlISERERDf58ZF08BboWi4Vf//rXXHvttQCMGTOGRx991OPxVWV/8YtfALB9+3Za\ntWrFpEmTALjtttvqHP7r5+dHSUkJu3fvJiQkhIiICPe+119/nddff522bdsCMGPGDNq3b09xcbGG\n24qIiIiINHHKfPpAeno6qampTJkyBagZEF5zzTUYhoFhGGzZsuWi2klLS3PXFRsbW2fZsLAw93pV\ngFibdu3audePHTtGZGRkjf1RUVHu9WHDhrn7sHz5cq6++mr+/Oc/k5ycTFhYGKNGjSI7OxuAw4cP\nk5iYiN1ux26307NnT1q2bMnJkye9Om8REREREWl8lPn0gU2bNpGVlUV0dDQARUVFOJ1O9uzZw65d\nu+o81pssaXx8PA5HzQmQGjrLGhYWxn/+858a2w4fPkzHjh0BWLdu3TnHjB07lrFjx1JUVMRvf/tb\nnnzySZYvX054eDgff/yxx+dXDx482KD9FhERERGRK4synz4wceJEMjMzycjIID09nUmTJjFixAg2\nbNjgsXx5eTnFxcU4nU7KysooLi6u8cqR4uJi9yQ/1dc9CQkJITc3l7KyMvc2b4a0nl02Pj6ekpIS\nXnvtNQA+//xztm3bVuvx+/fvJy0tjfLyclq2bEmrVq3cs+VOnDiR6dOnk5ubC0B+fj5r164FzImG\nLBYLBw4cqHdfRURERESk8WhymU8jyPD6dSje1n8+fn5++Pn5uT8HBgbi5+dHSEiIx/JJSUm8/fbb\n7s9z5szhzTffZPz48QD4+/sDZlaze/fuWCwWnE6nx7oGDRpEp06dCAkJoXXr1hw7dszjezmrfz57\nvfrn1q1b89FHH5GUlMRTTz3FsGHDSExMrPXci4uLmTp1Knv37sVisXDzzTfz+uuvAzB9+nSee+45\nbrjhBgoKCggJCeHee+9l+PDhBAUF8fjjjxMXF4fFYmH9+vX069ev1nZERERERKRxuXQz4VwcV20T\n6WiiGqkP3SsiIiIijY+ZFPHmN5wFUrwonuJ17eDtb0qLBdfGHfUvntAX17teVD/Gu+6Ad6dQmZdq\nkLhRw25FRERERETE5xR8ioiIiIiIiM8p+BQRERERERGfU/ApIiIiIiIiPqfgU0RERERERHxOwaeI\niIiIiIj4nIJPERERERER8TkFnyIiIiIiIuJzCj7lkrJarWRmZl7uboiIiIiIyCXW5ILPYJsNi8Xi\nsyXYZvOqP/v376d169aMGzeu1jLJycnExsbSokULZs2aVWPfV199Ra9evQgKCsIwDBISEkhPT7+g\na9NUpaSk1Hl9RURERETk8mtywWe+w4ELfLbkOxxe9eeRRx6hX79+WCyWWst06dKFF154gREjRpxT\n7pprruH//u//OHXqFHl5edxwww08+OCDXvVBRERERETkcmtyweeVZMWKFdjtdgYNGoTL5aq13Pjx\n4xk6dCiGYZxTLjQ0lPDwcAAqKiqwWq106NCh1rr27NnDzTffTGBgICEhIYwePdq9Lz09nfj4eGw2\nGx06dODtt9927/vxxx+ZNGkSoaGh2Gw2br75ZoqLiwFYvnw5nTp1wjAMbrrpJjIyMtzHxcTE8Oc/\n/5k+ffoQEBDAnXfeyenTp937Z82aRXBwMNHR0SxdurTO65Wbm8uQIUMwDAO73U7//v3d1yMrK4vh\nw4fTpk0bwsPDmTdvHgDr169n7ty5vP/++xiGQZ8+fepsQ0RERERELg8Fnz5SWFjIzJkzefnll+sM\nPOvj0KFD2O12AgIC+Pzzz3n99ddrLTtjxgxuv/12ioqKOHr0KE8++SQABQUFDBkyhEmTJlFYWMi6\ndet4/PHH2blzJ2BmaA8ePMiePXsoLCxk4cKFWK1WMjIymDhxIu+88w4Oh4P77ruPESNGUFpaCoDF\nYmHlypV88cUXZGdns2/fPnf/Vq1axZIlS/j73//ODz/8wNdff13neb744ot07tyZgoIC8vLyeOml\nl7BYLDidToYNG8aAAQPIy8tjx44dLFmyhNWrVzN06FCmTZvGvffei8Ph4Ntvv72oay0iIiIiIr6h\n4NNHkpOTSUpKIiIios4ht/URHR1Nfn4++fn59OvXr85ht4GBgRw8eJCcnByaN29Ov379AFizZg3d\nunVjzJgxAPTs2ZO77rqLlStXUlxczPLly/nf//1fQkJCAIiLi6Nly5Z88MEH3HnnnfTv3x+Axx57\njObNm7Nx40Z3m48++ighISHY7XZuv/12d2b0ww8/JCkpiU6dOtGyZctznmf11Pfc3FwOHjyIxWLh\nhhtuAGDz5s389NNPPP3001itVqKiokhKSuKDDz4AwOVyXXSALyIiIiIivqXg0wfS09NJTU1lypQp\nADUCo2uuuQbDMDAMgy1btnhVr2EYzJs3j08//ZTCwkLS0tLcdcXGxgLw/PPPU1paSt++fenRowdL\nliwBIDs7m+3bt2O3293Le++9R35+Pnl5eZSVldGpU6dz2jx27Bjt27evsa19+/YcPXrU/bldu3bu\ndT8/P3dW9Pjx40RFRbn3RUZGutcPHTrk7rutchKnJ554gujoaAYPHkxMTAxz5sxx9z0nJ6dG3+fO\nnUtBQYFX109ERERERC6f5pe7A03Rpk2byMrKIjo6GoCioiKcTid79uxh165ddR57vixpeXm5ez0+\nPh7HWRMghYeHu5+t3Lp1K7fccgsDBw4kPDycwYMH8/nnn59TZ3FxMS1btiQzM5Nu3brV2BcWFkZW\nVlaNbdnZ2YSFhdXZTzCfV83Ozq5xXJXo6Ohz+m4YBgsWLGDBggXs2bOHhIQE4uLiCA8Pp2vXrrVe\nu4vNLIuIiIiIiO8p8+kDEydOJDMzk4yMDNLT05k0aRIjRoxgw4YNHsuXl5dTXFyM0+mkrKyM4uJi\nKioqAPjss884cOAAYD63+cQTT/CrX/3KnS082+rVqzly5AgANpsNq9WK1WolMTGR9PR0Vq5cidPp\npKKigm+//Za9e/fSunVr7rvvPh5//HFOnjyJy+Vix44dlJaWMmrUKNasWcM333yDy+Vi0aJFlJWV\nMXDgwFrPvyrTO2rUKN544w0yMzMpKSnh2WefrfO6bdiwwR3oBgYG0qxZM6xWKwMHDqSiooJFixZR\nWlqKy+Vi79697udVQ0JCOHz4sIbeioiIiIhcwZpc8Gk3DCzgs8VuGOftg5+fH6GhoYSGhhIWFkZg\nYCB+fn7u5ynPlpSUhL+/PytWrGDOnDn4+/uzbNkyADIzM7nlllsIDAykS5cuWCwW3n///Vrb3rx5\ns3vm2eHDhzN//ny6dOmC3W5n/fr1LF68mODgYEJCQpg6dap7RttXX32VqKgounXrRps2bXj88cdx\nuVxcd911vPbaa4wdOxabzcayZcv47LPPaNWqlcf2q96HCpCYmMhDDz3E9ddfT5cuXYiPj68zS7lr\n1y4GDBhAQEAAffv25aGHHuLWW2+lWbNmbNiwgdTUVMLCwmjTpg3jx48nPz8fgNGjR3P69GmCgoKI\ni4s77/cjIiIiIiKXXmMZr+jylNWyWCzKdkm96F4RERERaXzMxIU3v+EskOJF8RSvawdvf1NaLLg2\n7qh/8YS+uN71ovox3nUHvDuFytxRg8SNTS7zKSIiIiIiIlceBZ8iIiIiIiLicwo+RURERERExOcU\nfIqIiIiIiIjPKfgUERERERERn1PwKSIiIiIiIj6n4FNERERERER8TsGniIiIiIiI+JyCzybG6XTy\nm9/8BsMwuPHGG9m8eTPdu3e/3N0SEREREZGfuSYXfNqCg7FYLD5bbMHBXvVn//79tG7dmnHjxtVa\nJjk5mdjYWFq0aMGsWbNq7Nu4cSNWqxXDMNzLO++8U2tdmzZtIi0tjePHj7Nt2zZ++ctf8v3333vV\n5yopKSl19vtiPPDAAyQnJ/ukbhERERERufI0v9wdaGiO/HxwuXxXv8XiVflHHnmEfv36YanjuC5d\nuvDCCy+wePFij+UiIyM5fPhwvdrLzs4mJiaG1q1bn7dseXk5zZs3uVtARERERESuQE0u83klWbFi\nBXa7nUGDBuGqIyAeP348Q4cOxTCMOsudzxtvvMFvf/tbtm7dimEYzJo1i40bN9K+fXt3mZiYGObP\nn0/v3r2x2Ww4nU5SUlIICwvDMAy6dOlCamoq69evZ+7cubz//vsYhkGfPn08trlnzx5uvvlmAgMD\nCQkJYfTo0e596enpxMfHY7PZ6NChA2+//TYAS5Ys4b333mP+/PkYhsEdd9xxwecsIiIiIiKNg9Je\nPlJYWMjMmTP56quvWLJkyUXVdezYMdq1a4fFYuHOO+/kxRdfJCAg4JxyDz30EM2bN+f1118nLS0N\nMIftVmexWPjoo4/44osvaNOmDbt372bp0qVkZGTQrl07cnJyKC0tJSYmhmnTpvHvf//bHTR6MmPG\nDG6//XagHOv0AAAgAElEQVS++eYbysvL2blzJwAFBQUMGTKEl156iTFjxrB7924GDBhAr169mDhx\nIlu3bqV9+/Y8++yzF3VtRERERESkcVDm00eSk5NJSkoiIiKiziG359OzZ0927drFkSNH+Mc//sH+\n/fv5n//5n1rL1ydz+thjj9G2bVtatGiBn58fJSUl7N69m7KyMiIiIoiJiXHXdb76AgMDOXjwIDk5\nOTRv3px+/foBsGbNGrp168aYMWPc53HXXXexcuXKetctIiIiIiJNh4JPH0hPTyc1NZUpU6YANQPC\na665xj1x0JYtW85bV2hoKFdffTUAERERzJ8/n48++giAd999113XiBEj6t2/8PBw9/rVV1/Nn//8\nZ5KTkwkLC2PUqFFkZ2d7PC4tLc3dXmxsLADPP/88paWl9O3blx49erizvNnZ2Wzfvh273e5e3nvv\nPfLz8wEuKiAXEREREZHGR8NufWDTpk1kZWURHR0NQFFREU6nkz179rBr1646jz1fUFY9YzhmzBh3\nZvFijB07lrFjx1JUVMRvf/tbnnzySZYvX35OX+Lj43E4HDW2hYeHs3TpUgC2bt3KLbfcwsCBAwkP\nD2fw4MF8/vnnHttU8CkiIiIi8vOizKcPTJw4kczMTDIyMkhPT2fSpEmMGDGCDRs2eCxfXl5OcXEx\nTqeTsrIyiouLqaioAGDz5s3k5OQAcPToUf74xz826AQ9+/fvJy0tjfLyclq2bEmrVq2wWs3bIiQk\nhMOHD9c5PHb16tUcOXIEAJvNhtVqxWq1kpiYSHp6OitXrsTpdFJRUcG3337L3r17AQgODubgwYMN\ndh4iIiIiInJla3LBp2G3g8Xis8Ww28/bBz8/P0JDQwkNDSUsLIzAwED8/PwICQnxWD4pKQl/f39W\nrFjBnDlz8Pf3Z9myZQBs376duLg4AgMDufbaa4mJieG1116rte2q95Geva02xcXFTJ06FbvdzlVX\nXUVOTg7z5s0DYPTo0Zw+fZqgoCDi4uI8Hr9582b69OlDQEAAw4cPZ/78+XTp0gW73c769etZvHgx\nwcHBhISEMHXqVIqLiwFzcqS///3v2Gw2fv3rX9d+MUVEREREpEloLGMfXZ6ybxaLRZPWSL3oXhER\nERFpfMwkije/4SyQ4kXxFK9rB29/U1osuDbuqH/xhL643vWi+gt4Cs+bU6jMYzVI3NjkMp8iIiIi\nIiJy5VHwKSIiIiIiIj6n4FNERERERER8TsGniIiIiIiI+Jyvg8+hwD+B3cAfPOzvDmwH/lVZpuHe\nISIiIiIiIiJXDF8Gn62A/4cZgF4LjAL6nFVmBrAU6AXcBSzyYX9ERERERETkMvFl8HkDsAv4D1AO\nvA+MOKvMYSCocr0NcNCH/REREREREZHLpLkP647CDC6rZAMJZ5WZC2wFHgUCgEE+7I+IiIiIiIhc\nJr4MPuvz6tKXgNeBl4EbgWXANZ4KpqSkuNcTEhJISEi46A42VS+99BLPPfcc5eXlHDx4ELvd7tP2\n3nzzTd544w3S0tJ82o6IiIiIiPjWxo3m4gu+HHabDbSv9rk9NTOhAL8EPqhc3wa0BkI9VZaSkuJe\n6go8g+12LBaLz5ZgLwK5FStW0LNnTwzDoGPHjrUGZ4sWLSIuLo7WrVszYcKEGvvKysoYNWoUHTt2\nxGq1smnTpjrbPH36NNOmTePrr7+msLCwwQPPrKwsrFYrFRUVDVpvfTzwwAMkJydf8nZFRERERH4u\nEhIgJeXM0pB8mfncgTmRUCRwDLgb+O1ZZf4NDAbeAnpgDr09eTGN5hcU4Nq442KqqJMloW+9yn3y\nySdMnz6d1atXExsby/HjxykvL/dYNjIykuTkZDZs2MDp06fP2T9gwACmTp3K6NGjsVgsdbZ77Ngx\nSktL6dGjR7366Y3q/Xe56pPYFhERERERMfky81kMPAxsADKAj4GdwCzg9soyjwOTMCcm+ghIApw+\n7NMlk5KSwuzZs4mNjQWgbdu2hIeHeyybmJjIHXfcQUhIyDn7WrRowWOPPUb//v1p1qxZnW3u27eP\na64xRy23adOGwYMHA5CamkqvXr0wDIPY2Fi+/PJL9zExMTGkpqbW6Pe4ceOAM1nOpUuX0rFjRwYP\nHszAgQPd9dtsNrZt2+YOiJ988klCQkKIjIxkzZo1tfZzz5493HzzzQQGBhISEsLo0aPd+9LT04mP\nj8dms9GhQwfefvttAJYsWcJ7773H/PnzMQyDO+7QW3lERERERBoTX2Y+AdZVLtXNrLa+F7jJx324\n5IqKisjIyCArK4uuXbtSUFDAyJEjeeWVV/Dz86v1uIvNJnbt2pVdu3bRsWNHTp06hdVq5ciRIyQm\nJrJ06VJGjRrFqlWrSExMZN++fYSFhbmHE1fxlFndvn07e/fuBSA3N7dG/QDff/8927dvJykpiZMn\nT7J48WL++7//u9YAccaMGdx+++188803lJeXs3PnTgAKCgoYMmQIL730EmPGjGH37t0MGDCAXr16\nMXHiRLZu3Ur79u159tlnL+o6iYiIiIjIpefLzOfPVn5+Pi6XizVr1rBt2zb279/Pvn37zvu84vmG\n1NbH2QHsp59+Su/evRk1ahRgZlmvu+46Vq9eXa/jAZ555hlatmxJy5Ytaw2QO3TowP333w/A+PHj\nOXHiBP/5z388lg0MDOTgwYPk5OTQvHlz+vXrB8CaNWvo1q0bY8aMAaBnz57cddddrFy50t03DfcV\nEREREWmcFHz6QGBgIACTJ08mODiYoKAgpkyZwtq1axk+fDiGYWAYBsuXL69xnLeB1aFDh9x12Ww2\nj2WOHj1K+/bta2yLjo7m2LFjHst7CoBrGy5cXbt27dzr/v7+AJSUlJCWlubuY9UQ5Oeff57S0lL6\n9u1Ljx49WLJkCQDZ2dls374du93uXt577z3y8/Nr7ZuIiIiIiDQOvh52+7Nkt9uJioryuG/t2rW1\nHudtcBUdHY3D4aizTLt27Vi/fn2NbYcOHSI+Ph6Ali1b8uOPP7r3nThxos76vO1jfHz8OX0MDw9n\n6dKlAGzdupVbbrmFgQMHEh4ezuDBg/n8888bpG0REREREblyKPPpIxMmTODVV1+loKCAwsJCFi5c\nyG233eaxrNPppLi4mPLycpxOJyUlJTidZ+ZdKikpobi4+Jz1+rjtttv47rvv+PjjjwFYvXo13377\nLSNHjgSgd+/erFixAqfTyXfffcfKlSvrDPLatGmDxWLhwIED9e7D2VavXs2RI0cAsNlsWK1WrFYr\niYmJpKens3LlSpxOJxUVFXz77bfu502Dg4M5ePDgBbcrIiIiIiKXT5PLfNrbtKn361AutP76SE5O\n5sSJE3Tu3BmA0aNHM3v2bI9lZ8+eXWMSnWXLlpGSksIzzzwDQLdu3Th06BAWi4UhQ4a4g7/o6GiP\n9VUPHtu1a8dHH33E1KlTGT9+PB07dmTVqlXuYbJz5sxh9OjRBAUFkZCQwD333ENeXp7HugCCgoJ4\n/PHHiYuLw2KxsG7dunMmLfJ0XHWbN2/m4YcfprCwkKuuuor58+fTpUsXANavX8/UqVN56KGHsFqt\n9O7dmwULFgDw0EMPcdddd2Gz2Rg8eLA7oBYRERERkStfYxnH6PL0PKTFYtEENFIvuldEREREGh8z\noeHNbzgLpHhRPMXr2sHb35QWC66NO+pfPKEvrne9qH6Md90B706hMqfUIHGjht2KiIiIiIiIzyn4\nFBEREREREZ9T8CkiIiIiIiI+p+BTREREREREfE7Bp4iIiIiIiPicgk8RERERERHxOQWfIiIiIiIi\n4nMKPkVERERERMTnFHz+DKSkpDBu3Lha9/fq1Yuvv/66weoTERERERE5W5MLPoPtNiwWi8+WYLut\n3n1ZsWIFPXv2xDAMOnbsSFpamsdyixYtIi4ujtatWzNhwoQa+8rKyhg1ahQdO3bEarWyadOmOtvc\nuHEj7du3r7HNYrHUecy//vUvBgwYUI8zql99F8pqtZKZmemTukVERERE5PJqfrk70NDyCxy43vVd\n/ZYxjnqV++STT5g+fTqrV68mNjaW48ePU15e7rFsZGQkycnJbNiwgdOnT5+zf8CAAUydOpXRo0df\nUODncrm8PqaK0+mkWbNmF3y8ty6mryIiIiIicuVqcpnPK0VKSgqzZ88mNjYWgLZt2xIeHu6xbGJi\nInfccQchISHn7GvRogWPPfYY/fv3P28Q+OOPPzJs2DBycnIwDAObzUZubi4Wi4XS0lLuv/9+goKC\nuPrqq9m6dav7uJiYGL788kt3v0eNGsW4ceOw2+289dZb7N27l759+2Kz2fiv//ovTpw4UWc/Fi9e\nTHR0NIGBgXTo0IFly5a5973yyivExMRgs9kYOHAg//73vwHcmdfevXtjGAYffvhhnW2IiIiIiEjj\nouDTB4qKisjIyCArK4uuXbsSGhpKUlKSx6xmdReb9QsICGD9+vVERETgcDgoLCwkPDwcl8vFJ598\nwvjx4zl16hR33303kydPdh93djb1888/Z8yYMeTn5zNmzBjuuecebr31Vk6dOsWcOXN45513as3A\nFhQU8NRTT5GamkpRURE7d+4kLi4OgPfee49XXnmFr776isLCQoYNG8aoUaMA3M+cfvfddzgcDkaP\nHn1R10JERERERK4sCj59ID8/H5fLxZo1a9i2bRv79+9n3759JCcn13lcQzxLWVsAGx8fz6BBgwAY\nO3Ys3333Xa11/PKXv2To0KEAHDx4kO+//56ZM2disVjo27cviYmJtbbTokULmjVrxq5duzh9+jQh\nISF0794dgL/+9a88/fTTdOzYEYCnnnqKffv2sX///gs+XxERERERaRwUfPpAYGAgAJMnTyY4OJig\noCCmTJnC2rVrGT58OIZhYBgGy5cvr3Gct5nPQ4cOueuy2eqeCCksLMy97u/vj9PppKKiwmPZdu3a\nudePHTtGcHAwrVq1cm+LjIx0r0+aNMndh+eff56AgAB3hjMiIoKhQ4eya9cuALKzs/nd736H3W7H\nbre7hxkfP37cq/MWEREREZHGp8lNOHQlsNvtREVFedy3du3aWo/zNvMZHR2Nw1FzAiRPdVxMRjU0\nNJS8vDyKi4tp3bo1YAaRVRYvXszixYtrHDNs2DCGDRtGaWkp06dPJykpia1btxIeHs7cuXPdQ21F\nREREROTnQ5lPH5kwYQKvvvoqBQUFFBYWsnDhQm677TaPZZ1OJ8XFxZSXl+N0OikpKcHpdLr3l5SU\nUFxcfM66J8HBweTn59cISi/mWdKuXbvSrVs3nnvuOSoqKvj73//OmjVrag1ojx07xrp16ygpKaF5\n8+b4+/tjtZq32cSJE/nTn/7EDz/8AJjPxq5evbpG3w8cOHDBfRURERERkStXk8t82tsY9X4dyoXW\nXx/JycmcOHGCzp07AzB69Ghmz57tsezs2bN59tln3Z+XLVtGSkoKzzzzDADdunXj0KFDWCwWhgwZ\ngsVi4cCBA0RHR59TV2xsLCNHjiQqKsr97GXVO0qrqy149FT2/fffZ9y4cSxcuJAbb7yRcePGUVBQ\n4PF4p9PJnDlzuPfee3E6nVx33XXuzOjYsWMpKipi2LBhHD16FMMwuPXWW7nzzjsBmDFjBvfccw9O\np5O//vWvypCKiIiIiDQhFz/DzaXh8pS9s1gsei+k1IvuFREREZHGx0yKePMbzgIpXhRP8bp28PY3\npcWCa+OO+hdP6IvrXS+qH+Ndd8C7U6jMSzVI3KhhtyIiIiIiIuJzCj5FRERERETE5xR8ioiIiIiI\niM8p+BQRERERERGfU/ApIiIiIiIiPqfgU0RERERERHxOwaeIiIiIiIj4nIJPERERERER8TkFn3JJ\nxcTEkJqaerm7ISIiIiIil1iTCz5tNhsWi8Vni81mq3dfVqxYQc+ePTEMg44dO5KWluax3KJFi4iL\ni6N169ZMmDChxr7du3fzi1/8gjZt2hAYGEjfvn358ssvL+oaXU5V17Ehvfnmm8THxzdonSIiIiIi\n0rCaX+4ONDSHw3FF1P/JJ58wffp0Vq9eTWxsLMePH6e8vNxj2cjISJKTk9mwYQOnT58+Z9+qVavo\n0KEDLpeLRYsWMXr0aE6ePHnR5yIiIiIiInKpNLnM55UiJSWF2bNnExsbC0Dbtm0JDw/3WDYxMZE7\n7riDkJCQc/YFBQXRoUMHAJxOJ1arlejo6Frbzc3NZciQIRiGgd1up3///rhcLgCysrIYPnw4bdq0\nITw8nHnz5rmPKy8vZ/r06URGRmIYBn369CE7OxuA1NRUevXqhWEYxMbG1si8JiQk8MwzzxAfH09g\nYCADBgzg+PHj7v1LliyhXbt2hIWF8ac//anOa1ZUVMQ999xDUFAQQUFBXH/99Rw7dgyAkydPct99\n9xEcHMxVV13F73//eyoqKtizZw8PP/wwW7duxTAMgoOD62xDREREREQuDwWfPlBUVERGRgZZWVl0\n7dqV0NBQkpKSzslqnq0qSPSkTZs2+Pn5MX/+fFauXFlruRdffJHOnTtTUFBAXl4eL730EhaLBafT\nybBhwxgwYAB5eXns2LGDJUuWsHr1agCee+451q1bx7Zt23A4HCxfvhx/f3+OHDlCYmIiKSkpOBwO\nnn32WRITEzl69Ki7zeXLl/Puu+9y4sQJmjVrxvPPPw/Azp07eeKJJ/jss8/Izc3F4XC4A1pP/va3\nv3H69GmOHj3KqVOneOutt/Dz8wPgvvvuIyQkhCNHjrB//362bNnCwoUL6dGjB4sXL+amm27C4XCQ\nl5dX5zUWEREREZHLQ8GnD+Tn5+NyuVizZg3btm1j//797Nu3j+Tk5DqPq+tZyIKCAoqKirj//vsZ\nPXp0rYFqYGAgubm5HDx4EIvFwg033ADA5s2b+emnn3j66aexWq1ERUWRlJTEBx98AMDSpUuZO3cu\n7du3B6B79+4EBwfz6aef0rt3b0aNGgWYWdrrrrvOHbRaLBYmTJhAdHQ0rVu35u677yYjIwOAlStX\nkpiYSFxcHFarlZSUFJo3r32kd2BgICdPnuSHH34AcGdbDx48yNdff82LL75Iy5Ytsdvt/O53v3P3\nva6gXURERERErgwKPn0gMDAQgMmTJxMcHExQUBBTpkxh7dq1DB8+HMMwMAyD5cuX1zjufEFUq1at\nmDVrFpmZmfzzn//k0KFD7rqqJkJ64okniI6OZvDgwcTExDBnzhwAsrOzycnJwW63u5e5c+dSUFAA\nwJEjR+jUqdM5bR49etQdkFaJjo52D4cFaNeunXvdz8+PkpISAI4dO0ZkZGSN/l911VU1rlNV37Oz\nsxk3bhyDBg3i7rvvJjw8nMcff5zS0lKys7MpKysjPDzc3fdJkyZx6tSpOq+XiIiIiIhcOZrchENX\nArvdTlRUlMd9a9eurfW4+swCW1FRQUVFBWAGgWdPgGQYBgsWLGDBggXs2bOHhIQE4uLiCA8Pp2vX\nruzatctjvREREWRmZtKlS5ca29u1a8f69etrbDt06FC9ZpcNCwurMcy2uLiYEydOuD8XFRWdc8yz\nzz7Ls88+y+HDhxk6dCjdunXj1ltvJTAwkLy8PI/XqKFnzxURERERkYanzKePTJgwgVdffZWCggIK\nCwtZuHAht912m8eyTqeT4uJiysvLcTqdlJSU4HQ6Afjqq6/cAeNPP/3EtGnTiIiIcE9kdLYNGzaQ\nlZUFmJnFZs2aYbVaGThwIBUVFSxatIjS0lJcLhd79+5l586d7v5Onz6dw4cPA+YrXvLy8rjtttv4\n7rvv+PjjjwFYvXo13377LSNHjnS3WVvGNjExkVWrVvGPf/wDp9PJ7Nmza53xF+Drr79mz549AAQE\nBNCiRQusViudOnWib9++TJs2jR9//BGAgwcPsmXLFgCCg4PJzc2lrKys1rpFREREROTyanLBp2EY\nV0T9ycnJXH/99XTu3JmOHTvSvXt3Zs+e7bHs7Nmz8ff3Z968eSxbtgw/Pz/3cNnjx4+TmJhIYGAg\nkZGR7N27l3Xr1tWa7du1axcDBgwgICCAvn378tBDD3HrrbfSrFkzNmzYQGpqKmFhYbRp04bx48eT\nn58PwIwZM7j11lvp27cvNpuNcePGUVxcTLt27fjoo4945plnCAwMJDk5mVWrVtUYalu9L9Xf4xkX\nF8cLL7zA8OHDiYiIICAg4JwhvNVlZ2czcuRIAgMD6dKlCzfddBMPPPAAAB9++CE5OTl06NABm83G\n7bffzqFDhwAYPHgwnTp1IiQkhNDQ0Hp9PyIiIiIicmk1lvGKLk/ZNYvFoslmpF50r4iIiIg0PmZS\nw5vfcBZI8aJ4ite1g7e/KS0WXBt31L94Ql9c73pR/RjvugPenUJlXqlB4sYml/kUERERERGRK4+C\nTxEREREREfE5BZ8iIiIiIiLicwo+RURERERExOcUfIqIiIiIiIjPKfgUERERERERn1PwKSIiIiIi\nIj6n4FNERERERER8TsFnE/T73/+eoKAgIiIiOHz4MIZh4PL2ZbgiIiIiIiINqMkFn8HBNiwWi8+W\n4GBbvfuyYsUKevbsiWEYdOzYkbS0NI/lFi1aRFxcHK1bt2bChAk19mVlZWG1WjEMw73MmTOn1jYz\nMzP5y1/+QmZmJjk5ObRv3x6Hw4HFYql3v6u8+eabxMfHe31cfaSkpDBu3Dif1C0iIiIiIlee5pe7\nAw0tP9+BL5N8FoujXuU++eQTpk+fzurVq4mNjeX48eOUl5d7LBsZGUlycjIbNmzg9OnTHssUFhbW\nK4DMzs7mqquuIiQk5Lxlq7KhFxKYioiIiIiIeKPJZT6vFCkpKcyePZvY2FgA2rZtS3h4uMeyiYmJ\n3HHHHXUGjBUVFedt84svvmDo0KHk5ORgGAYPPvigO3NadXxCQgIzZsygf//+GIZBZmYmixcvJjo6\nmsDAQDp06MCyZcv4/vvvmTRpElu3bsUwDIKDgz22mZuby5AhQzAMA7vdTv/+/d1BbVZWFsOHD6dN\nmzaEh4czb948ANavX8/cuXN5//33MQyDPn36nPfcRERERESkcVPw6QNFRUVkZGSQlZVF165dCQ0N\nJSkpqdasZpW6nsvs0KEDbdu25b777uPYsWMeywwePJh169YRERGBw+Fg6dKlHsstX76cd955B4fD\nQVBQEE899RSpqakUFRWxc+dO4uLi6N69O6+99ho33XQTDoeDvLw8j3W9+OKLdO7cmYKCAvLy8njp\npZewWCw4nU6GDRvGgAEDyMvLY8eOHSxZsoTVq1czdOhQpk2bxr333ovD4eDbb7+t87qIiIiIiEjj\np+DTB/Lz83G5XKxZs4Zt27axf/9+9u3bR3Jycp3HeRr+2rZtW9LT08nOzmbv3r1YrVZGjx5dax3n\nm1jIYrHw4IMP0qlTJywWC35+fjRr1oxdu3Zx+vRpQkJC6N69e73qAggMDCQ3N5eDBw9isVi44YYb\nANi8eTM//fQTTz/9NFarlaioKJKSkvjggw/cdWsSJBERERGRnw8Fnz4QGBgIwOTJkwkODiYoKIgp\nU6awdu1ahg8f7p44aPny5TWO8xSMBQQEcO211wIQHBzMK6+8QlpaGidPniQtLc1dV9Xw3vqoPvw3\nICCA9957j1deeYWIiAiGDh3Krl27PB536NAhd3s2mznx0hNPPEF0dDSDBw8mJibGPRlSdnY2OTk5\n2O129zJ37lwKCgrq3U8REREREWk6mtyEQ1cCu91OVFSUx31r166t9bj6TPxTFaBWVFQQHx+Pw1G/\nCZDqMmzYMIYNG0ZpaSnTp08nKSmJrVu3ntOf6Ojoc9ozDIMFCxawYMEC9uzZQ0JCAnFxcYSHh9O1\na9daA1lNciQiIiIi8vOizKePTJgwgVdffZWCggIKCwtZuHAht912m8eyTqeT4uJiysvLcTqdlJSU\n4HQ6AfjHP/5BZmYmAKdOnWLq1KnceOONtG3b9oL7Vj3DeuzYMdatW0dJSQnNmzfH398fq9W8LYKD\ng8nNzaWsrKzWujZs2EBWVhZgZnybNWuG1Wpl4MCBVFRUsGjRIkpLS3G5XOzdu5edO3cCEBISwuHD\nhzX0VkRERETkZ6LJBZ92u4HFgs8Wu92oVz+Sk5O5/vrr6dy5Mx07dqR79+7Mnj3bY9nZs2fj7+/P\nvHnzWLZsGX5+fu7hq3v37uVXv/oVgYGBdOrUiZKSEj7++OM62z47q1jXZ6fTyZw5cwgNDcVms5Ga\nmsrixYsBcwKjTp06ERISQmhoqMe2du3axYABAwgICKBv37489NBD3HrrrTRr1owNGzaQmppKWFgY\nbdq0Yfz48eTn5wMwevRoTp8+TVBQEHFxcXWej4iIiIiINH6NZeyjy1OGzGKxKHMm9aJ7RURERKTx\nMZMm3vyGs0CKF8VTvK4dvP1NabHg2rij/sUT+uJ614vqx3jXHfDuFCrzVg0SNza5zKeIiIiIiIhc\neRR8ioiIiIiIiM8p+BQRERERERGfU/ApIiIiIiIiPqfgU0RERERERHxOwaeIiIiIiIj4XPPL3YGL\nYbfbz3mHpYgndrv9cndBRERERORnrVEHn3l5eZe7CyIiIiIiIlIPGnYrIiIiIiIiPqfgU0RERERE\nRHxOwaeIiIiIiIj4nIJPERERERER8TkFnyLy/9u79zDJzoLO49+a6ZBAptp0nyQDZiaOl23UprgE\nQlQQWlYwGKchctmw2d0EVC4usq7o4mUfmQTXiUYiFx+5PCuKmCZgBAn37CqzAQJrSAI5SYyiy8gk\nwQxV05MuIFx6pveP99T06apTPae66+2urvp+nqeePuet06fe7rduv/O+5z2SJElSdIZPSZIkSVJ0\nhk9JkiRJUnSGT0mSJElSdLHD54VACtwNvKbLNi8EbgfuAOYi10eSJEmStAnGIu77VOAtwFOBB4DP\nADcSgmbL44BXZ9t8HZiMWB9JkiRJ0iaJ2fN5AXAXcB+wCLwHuKhtmxcDf0QIngBHItZHkiRJkrRJ\nYoi6zXsAACAASURBVIbPXcCh3Pq9WVneo4HHA58DbgVmI9ZHkiRJkrRJYg67XSqxzTZgD6GXdDdw\nM/Ap7AGVJEmSpKESM3zeSwiULbtZ2RNKtv4p4BhwkDAx0RTw2fad7du378TyzMwMMzMz/ayrJEmS\nJI28AwfCLYZKnN0CcBpwD/AU4DChV/NlwG25bS4GngNcDpwJfIEwDPerbftaWloq05EqSZIkaVhU\nKhXKDag88Ruwr4fN9/W8d+g1l1QqLB24pfzmM+ezdG0Pu7+0t+pAb39CJSTGvuTGmOd8fhN4BfBx\nQqh8HyF4XgHszbZ5P9AgTEz0KeDX6QyekiRJkqQtLuawW4CPZre817atvzq7SZIkSZKGVMyeT0mS\nJEmSAMOnJEmSJGkDGD4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRxZ7tVpI0ZNI0JU3TE8u1Wg2AWq12\nYlmSJKmd4VOS1JN8yJydnWX//v2bXCNJkrQVOOxWkiRJkhSdPZ+SNGQcFitJkgaR4VOShozDYiVJ\n0iBy2K0k6aQmx8epVCodN6CwfHJ8fJNrLEmSBo3hU5J0UvPNJkvQcaOgbCnbXpIkKc/wKUlDYHx8\nsqeeyfHxyU2usSRJGjVlwudsye0kSZuk2ZynuA+SwvKwvSRJ0sYpM+HQvwPeAFwPvAO4J2qNJEkD\nLU0S0iQBYLpeZ25qCoBao0Gt0djMqkmSpAFWJnxeCnwX8CLgzwiHzf8UeDfgST2SNGIMmZIkaS3K\nXmrlQULP58OBXwYuBn4deDPwh3GqJkkaFl57VJIklQmfzwEuB/4N8OfA+cBhQhBNMXxK0kBJkpQk\nCUGvXp9mamoOgEajRqOxOUHPa49KkqQy4fNnCQHzprbyh4CX971GkqR12cyQKUmS1E2Z8HkF8JXc\n+sOBncBB4H9HqJMkSZIkaciUuYTK9cC3c+uLwF/FqY4kSZIkaRiVCZ/bgWO59e8Ap8SpjiRpKIyN\nUalUCm9AYfn45OQmV1qSJMVUZthtE7gI+HC2/jN4iRVJGhpRZqJdXISlpeL7ZmcL72tmwVSSJA2n\nMuHzF4DrgLdl618lXPNTkjQEnIlWkiRthDLh8++BxwFnZuv1eNWRJEmSJA2jMuHzdOAFwG6WzxFd\nAq6MVSlJkmKKMtRYkiStqkz4/DDwAHArKycekiSplCRNSbKwV5+eZmpuDoBGrUajIOzFDocONZYk\naeOVCZ9nAjOR6yFJ2kjbODHzbN7evXsLy9erW8jsxnAoSdLwKRM+PwU8Brgzcl0kSRvlOLCvoPzW\nLuVFZQPEYbSSJA2+MuHz6cDPAV8CvpWVLQGPjVUpSZJ6YU+pJEmDr0z4fHb0WkiSFNn4+DjNZudl\nqrsNNa5WqywsLGxE1SRJGgnbTr4JB4EfAH4yW/4a4JXAJWlIJM2EqfunmLp/ivqO+onlpJlsaD3G\ntm+nUql03IDC8smJiZ72XxQ8+7m9JElaXZmez/2Ecz4fDfxPYDtwHXBBxHpJkjZIo9qgUW1sdjVY\nPHaMpQO3dJTPvv7KwvJTnnH+hk6aJEmS1qdM+HwuME2YhgLCZVdOjVYjSZJKWDwOS9d2ls9eV1xe\nuTR+nSRJUndlht1+hzAvYstpwMPiVEeSJEmSNIzK9HxeD7wNOAN4CfBi4J0xKyVJUkxJkpAk4ZzW\ner3O1NQUAI1Gg0ajcwiyl3KRJGn9yoTPK4FZQg/oecDvAx+MWSlJ0mhLDx0kPXQQgOld5zJ38wEA\narv3UNu9Z9377xYyu/FSLpIkrV+Z8AlwQ3aTJCm6XkNm+kBCejj0ZE6fVWcuDT2ZtbMb1HZu/mRK\nkiRp9fD5aeAphEurLLXdtwSMx6qUJEm9qO00ZEqSNOhWC59PyX7u2IiKSJIkSZKGV5nZbn8UqObW\ndwA/Eqc6kiQNhrExqFQqHTcoLq9UKkxOOihIkqRuyoTPtxKG3rZ8A3hLnOpIkjQYFhdhaanzBsXl\nS0swP9/c3EpLkjTAyoTPbaw85/M45ScqkiRJkiSpVPi8D/jPwCnAw4BXAvfHrJQkSZIkabiUCZ8v\nBp4FNICvAv8WuCxmpSRJkiRJw6XM8NmvAM+JXRFJkgZVmiakaXYd0ek6c3PZdURrDWo1L/EiSVIZ\nq4XP1wC/B7w5V7YEVLKfr4pYL0mSBoYhU5Kk9VstfN6d/byVlRMOVdrWJUmSJEla1Wrh84XAB4Ez\ngDdsTHUkSZIkScNotQmHngx8N/ASYLLgJkmSJElSKav1fL4V+Bvg+whDb/OWsnJJkiRJkk5qtZ7P\nDwA/BLwD+N62m8FTkiRJklTaauHz+uznozeiIpIkSZKk4bXasNtTgd8CpoBfIcxy27IEXBOxXpIk\nSZKkIbJaz+fzgGPAdqAK7MjdqvGrJkmSJEkaFqv1fN4DXAXcAXxkY6ojSZIkSRpGq/V8tnwB+Avg\nf2XrjwZeGq1GkiRJkqShUyZ8/gXwQWBntv7PwKui1UiSJEmSNHTKhM8EeA/h/E+AxewmSZIkSVIp\nZcLn1wkBtOUJwLfiVEeSJEmSNIxWm3Co5VeAG4HvA24CzgVeELNSkiRJkqThUiZ8fgb4UeCx2fod\nwLej1UiSJEmSNHTKhM8K8GPAbsIw3R/Oyv88VqUkSZIkScOlTPh8L3AO8HmWJx0Cw6ckSZIkqaQy\n4fNxhGt7LkWuiyRJkiRpSJWZ7fY24OzYFZEkSZIkDa8yPZ+PBP4B+DuWL7GyBMzGqpQkSZIkabiU\nCZ/7YldCkiRJkjTcyoTPA7ErIUmSJEkabquFz6/RfZKhJWC8/9WRJEmSJA2j1cLnjg2rhSRJkiRp\nqJWZ7VaSJEmSpHUxfEqSJEmSojN8SpIkSZKiM3xKkiRJkqIzfEqSJEmSojN8SpIkSZKiW+1SK5Ik\nSeqjNE1J0/TEcq1WA6BWq51YlqRhZfiUJEnaIPmQOTs7y/79+ze5RpK0cWKHzwuBq4HtwDuB3+uy\n3fOAvwSeBNwWuU6SJElaI3tvJa1VzPB5KvAW4KnAA8BngBuB29u2qwL/BfhsxLpIkiT1XZkgNn7G\nOM0Hmx2/u3fvXiqVSkd59buqLBxdiFjr9bH3VtJaxQyfFwB3Afdl6+8BLqIzfL4OuAr4NaDzHViS\nJGlAlQlizQebsK/gl2+lsLy5rzOoStIwiBk+dwGHcuv3AjNt25wHnAN8hBA+lyLWR5IkaUsZlCGu\nkxPjzB8t33s7cUaVI/OD23sraXPEDJ8nC5LbgGuAy3Jl9nxKkqSBNj4+SbM531HeLYjlJc2EpJkA\nUN9RZ+r+KQAa1QaNaqNj+40Y4lom4M4fbbJ0befvzl5HYXnlUntvJXWKGT7vBXbn1nezsie0CkwD\nB7L1RwI3AHspmHRo3759J5ZnZmaYmZnpZ10lSZJKCcGz6Bj7bJfy5UDaLWRuptgBd1B6byWVc+BA\nuMUQM3zeAjyGMKz2MPBC4GW5+x8EzsqtfwJ4NV1mu82HT0mSpGE1BoU9qF2HuFarHFnYuCGu6QMJ\n6eHQezt9Vp25NPTe1s5uUNu5Ob23kvpnZibcWq64on/7jhk+vwm8Avg4YYjtuwjB8grgc8AHIz62\nJElSdEmSkiShV69en2Zqag6ARqNGo7G2Xr1FeuxXba4c4hq7p7G2szhkStLJxL7O50ezW95ru2z7\nE5HrIkmS1FfrCZmxlOlpnJyYYP7o0Y7yMuetltVL7221WmUh6711mK40vGKHT0mSJK1RmiSkSTbE\ntV5nbiob4tpoUGusvfdx/uhRlg7c0lE++/orC8srM+ev+bHKaOZ6bx2mKw0vw6ckSdKAKhUyx8a6\n9lb2sydTktbL8ClJkrSVLS7CUpcr3M3OFt+XC6TpoYOkhw4CML3rXOZuPgBAbfcearv3rLt6SZKQ\nZL239Xqdqaz3ttFo0FhH762krcfwKUmSNML6FTK7KRMyx8Z6nOF3osqRIxs3w6+k/jB8SpIkDZEk\nTUmyCXvq09NMzWUz8NZqNAZ0wp5unbfdO26bnYWSBp7hU5IkaYgMcsiUNNoMn5IkSRoYaZqQptkM\nv9N15uayGX5rDWo1zxGVtjLDpyRJkgaGIVMaXts2uwKSJEmSpOFn+JQkSZIkRWf4lCRJkiRFZ/iU\nJEmSJEVn+JQkSZIkRWf4lCRJkiRFZ/iUJEmSJEVn+JQkSZIkRWf4lCRJkiRFZ/iUJEmSJEVn+JQk\nSZIkRWf4lCRJkiRFZ/iUJEmSJEVn+JQkSZIkRWf4lCRJkiRFZ/iUJEmSJEVn+JQkSZIkRWf4lCRJ\nkiRFZ/iUJEmSJEVn+JQkSZIkRWf4lCRJkiRFZ/iUJEmSJEVn+JQkSZIkRWf4lCRJkiRFN7bZFZA0\nutI0JU3TE8u1Wg2AWq12YlmSJEnDwfApqavY4TC/n9nZWfbv37/ufUqSJGkwGT4ldWU4lCRJUr94\nzqckSZIkKTp7PiVtKZ4nKkmStDUZPiVtKQ4FliRJ2pocditpQ02Oj1OpVDpuQGH55Pj4JtdYkiRJ\n/WD4lLTC+PhkT+FwfHyyp/3PN5ssQceNgrKlbHtJkiRtfYZPSSs0m/MUx0AKy8P2kiRJ0uo851ND\nwUloJEmSpMFm+NSGiB0OnYRmiI2NnRj2227v3r2F91UnJlg4ciR2zSRJktQDw6c2hOFwNEQ5yLC4\nCEtLxffNzhbe18wFUnvFJUmSBoPhU1Lf9HqQIU0S0iQBYLpeZ25qKuyn0aDWaGxKnXpluJUkSSrH\n8CmpqyRJSZIQrOr1aaam5gBoNGo0Gn0YLt3HkLlZ7NWXJEkqx/Apqat+hUxJkiTJ8KmBVGYo4+T4\neOE1ILtNQjNRrXJkYaGnx5AkSZLUH4ZPRTM+Pll4DciuM5RWJ1hYCDOUlhnKON9sUjQNzSwUllfa\nguqgDZfcsmF4G4Xt2a2d1ytJU5Ls/1SfnmZqLhsKXKvR6MP/acu2gyRJ0oAzfCqaEDzLx8Nms/9B\nZSsZtDBc2nFgX0H5rV3Ki8p6UCZkjm3f3lMgnjjjDI7MhwMlW7YdJEmSBpzhU4rEHrTNs3jsGEsH\nbukon339lYXllZnze9r/+Pg4zR6GfFerVRZyQ74lSZJGkeFTisQetOFVFDz7ub0kSdIw2rbZFZAk\nSZIkDT97PjU41jFxTZokpEkCwHS9ztzUFNB2Hcmxsa776TpccmKChSNHevkrpFU5HFuSJI0qw6c2\nRJKkJEk2Q2l9mqmpbIbS/HUk1zFxzYqQ2c3iIiwVTYAEzM4W3tfMBdIYoWE95w4OYohJmglJMxwE\nqO+oM3V/OAjQqDZoVE/SPptobANn7HU4tiRJGlWGT22IFSFzi4oRGtZz7uAghphBDJnpoYOkhw4C\nML3rXOZuPgBAbfcearv3ALB4HJau7fzd2euKyyuXxqipJEnScDN8Sn00OTHO/NHyPZmKLx8yY0iS\nhCQb8l2v15nKhnw3Gg0aJ+uNlyRJGiGGT6mP5o827UEbMYZMSZKkcpztVpIkSZIUnT2fGkgbMXFN\nkqYk2YQ99elppuaySZBqNRrOOqo+GhvrfUKjiYkqR44sxK6aJEnShjF8aiBtxMQ1ZULm2PbtGzYL\naplzBw0x8aUPJKSHs8v2nFVnLs0u23N2g9rOtT0nu0203GWSZQAqld4mo5IkSRp0hk9pFYvHjrF0\n4JaO8tnXX1lYXpk5f82PVebcQUNMfLWdaw+ZkiRJ6s7wKUUSowdNkiRJ2qoMn1Ik9qBJkiRJywyf\nkrSB0jQhTbMe8ek6c3NZj3itQa3mwQpJkjS8DJ+StIEMmZIkaVQZPqWS0kMHSQ8dBGB617nM3XwA\ngNruPdR279n4+tiDJkmSpC3E8CmVtFkhsxtDpiRJkraSbZtdAUmSJEnS8DN8SpIkSZKiM3xKkiRJ\nkqIzfEqSJEmSojN8SpIkSZKiM3xKkiRJkqIzfEqSJEmSojN8SpIkSZKiM3xKkiRJkqIzfEqSJEmS\nojN8SpIkSZKiG9vsCkiSpN6laUqapieWa7UaALVa7cSyJEmDZCPC54XA1cB24J3A77Xd/2vA5cAS\ncAS4DPjSBtRLkqSB0WuYzJfPzs6yf//+vu5fkqR+ix0+TwXeAjwVeAD4DHAjcHtum88CbwK+Bbwc\nuAa4OHK9JEnadOPjkzSb8x3le/fu5aqrrur8hW3A8c7ivXv3UqlUOsqr31Vl4egC0HtYBQOrJKm/\nYofPC4C7gPuy9fcAF7EyfH4yt/xp4CWR6yRJ0kAIwXOp4J7Z4vLjFdhXsPmtFJY/tK9ZGEq7hdWJ\napUjCwsn1tcSWCVJ6iZ2+NwFHMqt3wvMrLL9y4APxKyQJEmDKElSkiT0Mtbr00xNzQHQaNRoNDp7\nGZNmQtJMwvY76kzdPxW2rzZoVBsALLIcYdMkIU3C9mm9ziVTYftao0GtEbavPPRQYSiFVXpXJyZY\nOHJkDX+xJGnUxA6fRYdzu7kUOA94etGd+/btO7E8MzPDzMzMeuolSdJA6RYyu26fC5ll5ENmV4uL\nsLT80Z2kKUk27Laepkxdckl47FqNRtYj2swF0tjDdB0GLEnxHTgQbjHEDp/3Artz67tZ2RPa8pPA\nbwFPA75TtKN8+IzJDzZJkoJ8yCwj9iRIDgOWpPhmZsKt5Yor+rfv2OHzFuAxwDnAYeCFhKG1eU8A\n3gr8FFA/2Q5jh0M/2CRJKm9s+/bezis94wyOzIdJlsp85o6Pj9NsNkvvv1qtspCdt+oBZUkaLLHD\n5zeBVwAfJ8zR9y7gNuAKQjD9EPD7wOnA9dnv/Avw3G47jBEOJ8fHme/hg619QgY/3CRJo2rx2DGW\nDtwCQHroIOmhgyeWL/ndqwGo7d5DbfceAE55xvk9hdVe5YOqB5S1WfxuKBXbiOt8fjS75b02t/zM\nDajDquabzV7mGqTSFlT9cJMkaWXI7GbxOCxd21k+e11xeeXS5eUkSUiySZPq9TpT2aRJjUaDRnY+\n69gYPYfbiYkqR44sFN4ngUPEpX7ZiPDZF718kFSrEywsRJx5b2zM2QAlSVqn9IGE9HAIk9Nn1ZlL\nsxl4z25Q29k5OVI+ZHaTnzMpTRPSNJvhN61zySXZ/msNarXl/VQqnaOfpDzDpNQfWyZ89nIdtGZz\n7cN28lPRT9frzBVMRd8+G+DKKs0W3tfsw1AiSZKGSW1nccjs2/7bQqYU03pO43KYrkbFFgqfQa/X\nQetVqanoB5xvYJIkSd2V+a40fsY4zQfLh0nopatk5Wlc9qxqVGy58NmvkNlPK66DNj3N1FwWiLtM\nUe+MvZIkSZunzHel5oNN2Ffwy7dSXF5UtoXZmaEYtlz4HESxr4MmSZKkraPUaVxd5hBZrWd1PXOI\nOGmSBsHQh89BOWqznuuglbGe66BJkiQNu/HxSZrNzu9W/brMT16p07i6zSHSZf4QWN8cIoZJDYKh\nD5+D8kLLXwctb/b1VxaWV2bO72n/RcGz7PaDEtAlSZJiCcGzp7MyTywlzYSkmV3mZ0edqfuzy/xU\nGzSqmzNXSIzvb3ZmKLbhDJ/ber/G10bKX4R7ete5zN18AOh+fbRe31zKXActb1ACuiRJ0iCKHTJ7\nnT8Eyn1/m5wYZ/5ob5Mm9cLODPVqOMPncQb6BPEyF+FesX2P4bDMddC8CLckSRpVsa+e0Kte5w8p\na/5ok6VrO8tnr6OwvHLp2h/LzgyVMZzhM2cQh0mUMRa593YNpxl4EW5JkjQUBvHqCWX0OofIevQ6\nkk4qY+jD56CHzG4WjxcfkYpxpCpNE9I0m5Ftus7cXDYjmxfnliRJGhjrmUMkfSAhPZx93zurzlya\nfd87u0FtZ+f3vRgj6RxFp6EPnzo5Q6YkSdJwq+0sDpnr0etIOkfRyfC5BfR6pEqSJEnDrdcJLKVB\nYPjcAmIcqZIkSdLWNWgh09O4VIbhU5IkSdK6GDJVxrbNroAkSZIkafgZPiVJkiRJ0Rk+JUmSJEnR\nGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZ\nPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+\nJUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4l\nSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJ\nkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmS\nJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIk\nSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ\n0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnR\nGT4lSZIkSdEZPiVJkiRJ0Rk+JUmSJEnRGT4lSZIkSdHFDp8XAilwN/CagvtPBd6TbfNp4Hsi10eS\nJEmStAlihs9TgbcQAuhjgecDT2jb5pXAV4AacDXwpoj1kSRJkiRtkpjh8wLgLuA+YJHQw3lR2zY/\nDbwrW74B+DGgErFOkiRJkqRNEDN87gIO5dbvzcq6bXMcaABnR6yTJEmSJGkTxOxlfBHwNOAV2fol\nwAzw8tw2/wD8OHA4W78HeDrwQNu+Pg88LlZFJUmSJEmFvgA8vh87GuvHTrq4F9idW9/Nyp7Q1jbn\nEsLnNiABvlqwr778sZIkSZKk4XMacBA4BzgFuAU4r22bVwNvyJYvJpz3KUmSJElST54N3Em41Mpv\nZGVXAHuz5VOB9xIutXIzsGeD6ydJkiRJkiRJkiRJiukdhImX0lzZCwiXrznGyiHMk8AngCbw5rb9\n/A/gy9l9RZ5HmGk4v7/fIPRWp8CzcuUXZmV3A6/Jlb8S+KdsP5Or/1lDrx/t9nDg49nv/ANwDZ2T\ng9lu/XWydntirvxZwG3AHdn2P5W77zLCxGl3AVe1PcYLgduz35tr+527stt/ypV3e+3+IuHE/zuA\nz7XVbdT08no7Hbge+Hvgi8C+rPzRhHZp3R4EXtX2OK+m83XypuxxbqPzGtbjhDkNWq/rR3Dy1/Qo\nKWq3awjvUXcDHyLMAQGrf749kdBmdwFvzJX/DuE1cidwE/B9ufuK2u17svXbgX8E/ltu+wOE13Tr\n+XFmD3/nsOlXuz0MeDvhPezvgZ/Nyi8nzPvR+l//XO53PgbMAx9s29f3Ap/J6nQd4TSvlm7vuaOm\nl3Zb7fOtbLu9JPc73T7fWm4oWa9RdBrhlMXW+9IfZuXdnvOnEi5rmQKfJryvtfT6/bCfj6Et5McJ\nH4z5F+UPAlOEN/T8l6pHAE8BXkbnm/yTgUdSHD6rhA/mm3P7eyLhyb6dcL7ulwhPulOz5XMIE1Xd\nwvIH9+MJT8AvYYjpR7s9PCuH8L+/CXhO7n7brf96abfHAmdly9PAvxKCxE7CdY1bH5Z/BvxMtvw4\n4P8SAhAs/78fRTgAsCO7/VO2HwjXSi567e7ILe8F/k+pv3A49dJuPw+8O1t+OOF5nw8lECa++wqd\nk+V9jJWvk+cBf50tP4EwI3veG4FrWX5dn+w1PWqK2u0nWL7821Usf9Fa7fPtDpbfz/6aMHcErHyN\n/BLwzmy5W7udwvIEjKcT5qvYk623P49GWb/a7WqWT8UCmMh+XkY4OFDkGYT30/bw+UHgudnyG4D/\nmi13e88dRb20W7fPN+i93Vb7fIMQXq8lvI5PVq9R9fDs5xjwWcL/p9tzPj+PznOBD2TLa/l+2I/H\neFi3PyrmdT61Pp8kHOXLu4dw9KPdNwhHIL5VcN/fEd48iryO8OLO/95FhKMcxwhfpO8ifAm+IFu+\nD1gkHPm4KPudzwP/supfMzr60W4PZeUA3yG04Xfn7rfd+q+XdruD5Vm57yK8j54G/ADhaHAju+9v\ngNls+cXAHwFfz9aPZD+fCXwU+Fp2+1hWBuGLU9Fr92u55R2EsDSqemm3Q4Qvotuzn99muR1afhL4\nZ1bOzH4NK3vCAH4aeFe2fDvhg7t1HesnEq5XfWNu+5O9pkdNUbt9gtC7DOF/dU623O198lzCa+/2\nbP0vWH5v6/YauYjidvsO4f0Rwpe97wALuX2Mci91Xj/arQL8e+APcmXzufu6/a//lpXtCqH9foTl\nAwr550C399xR1Eu7dft8W0u7rfb5toMQaH6n7Xe71WtUPZT9fBjhs+sw3Z/z+c+lG4AfI7Rfr98P\nV3td9fIYT+72Rxk+h8tSD9ueR3hRf6St/BzCcLGWewkfzuew8gtZq1zrt1q7nUE4mv+xbN12GyzP\nJwzve4gwnHKa0Js8Rjgq2OpBezShp/lzwK0sh9Ju7XYyv0g4inwNK49Eq7uPEwLFVwg9W1cDR9u2\nuYSVw/OeQ2iTO9q220Xn6+ocwmfqHxCODnfT/ppWp5eyfES9pf19sr0N7mPla6c1bP1yYH9Wttr7\n4W5CO3+Z0NuSDyt/RvgydSUG0dWUabedwDcJvaF3Er7APjK37c8S/tc3sHJIX5GzgXpuPf8c6Pae\nq05F7QYrP9/W0m6rfb69jvBe+Y011GuUbCN0FDxACObzdH/O598TjxMOhJ9N798Pz+rjY3T9ozR6\nthG+tP5qrswP1MEzRvgi/EbCEAbbbbD8MKEH+qXZep1wHu37Cb2WX2b5i9d2wjC+CwhD/97K+oaB\n/TGhp/VXCOfz6OT+A6FX61GE81l+NfvZ8jDCMOa/zNYfAfwm8NrcNpUuy631XyQcGLq/4H7ofE2r\n028ReqWv7cN+zgX+lOVhYtDZLq3X6CHCkMPvB36Z8PqCcECiRnjtPomV5yFqWdl220Z43d0IU2bO\nHwAABctJREFUPIbwhbp1vm4ruEwTQsd6ngP9fs8dVt3arf3zrZ/t9njCKQ8foPt3mH69D2x1xwn/\nr13A04CZTa1Nnxg+R1OV8CZxgPAF6EcIbwJPJBytyJ/v1DrK0V6+m5VHTNR/bydMjNI6l8J2Gxy7\nCCHzP7IyRPwVoXf6iYQjxvdk5V8mnENxjNDrdjfhyPx62+c9wPk91340PZXQZscIw8o+zcphQc8m\n9JC0hpx9P+HL6xcIbbwru38nxa+3ewmvyVdm219NmGDjd3Pbtb+mtdJlhOFbl5bYtlsbtJsDfrSH\n3/kKYZjik7L1B7KfXyMMN/P11qmXdjtM6EF7X7Z+PeHLNYRendbw5z8hnLeZ196LepiVE0Dl27Po\nPXeqRP1GSbd2K/p8W0u7FX2+td4nn5Tt+5OEdvnbEvUaZQ8CHyaE9m7P+XsJB9wg5LuE8HnW6/fD\n1V5XvT6GtqA9rDxBvOUTFM9weTmdJ/a3dJvttrW/9olrWufCHCScmHxatnxOtn4LnZMwfInRnpms\nZQ/rb7ffIQSZ1Xo2bbf+2kO5djuDEEguLti29YZdJfyvH5utX0wYutfa5j7C0JbvJgyfrWa3f2bl\nhAzQ+drdk1vemz3OKNtDuXb7NZZ7iU8nfBnN338d4UtPN+0TDr0/Wz6P8HxodxkrX9dlXtOjZA8r\n2+1CwrC9brPJXs7JJxxqzb6Z79H+JZZ7s7u126MIk29AmETlHwm9O9tZbvNTCM+R9pmQR80e1t9u\nHyBMnNK6v3Vu2dm5bfYSZl3Nm2H1CYfeSBgNAt3fc0fVHsq122qfb722W5nPt+8pWa9RlBD+bxBG\n7dzE8qRbRc/5/GRAFxN6pGFt3w/79RjaYt5NGLr1bcLRg5cQngiHCOPv/5VwInfLQcLY62a2zQ9m\n5b+frS9mP3+74LHaZ/P7TcIXsztZOc32s7Oyu1l5ntmrsn1/m/AG//Ye/s5h049220UYanEXxVOX\nt9hu/dNLu/13Qi9I/vIcZ+b2kxLO0Xhh22O8ntCm9xCOKLe8mOWp5fPhp9tr982EL913Ema6/aE1\n/cXDoZd2ewTwXsK5uV8ktGPL6YRh01W6+3+sHLb3RyxfsqNoNtT8DJBlX9OjoqjdvkiYAK31//nj\n3PYHKf58y19qJd+b/D7CF+i7Cb0Fj8rdV9RuzyS8plq/87Ks/HRCb/fnCYH0LSzPijuK1tNuX2a5\n3c4lvHfdSej52pOVX0Voh7uATxFG+rR8ktAj843ssVsT16x2qZVu77mjppd2W+3zbS3t1u3zrWUP\nK8+pX+35NGpqhP/B5wnP4dZ3gNUug/LerPxmVh6o7vX7YT8fQ5IkSZIkSZIkSZIkSZIkSZIkSZIk\nSZIkSZIkSZIkSZIkSZIkSZIkSdJwOg68K7c+BnyVzoven8xBVl4vVJKkkbFtsysgSdIW8HXCxdNP\ny9afCdwLLPW4n163lyRpaBg+JUkq5yPARdnyi4B3A5Vs/Uzg40AK3Aqcl5WfBXwS+Dzw9tz2AL8A\nfAG4C3gHoTd1jNDDmgJ3AK+O86dIkiRJkgZRE6gBfwmcCtwOPJ3lYbdvB34jW34acHe2/LZc+bMI\nw3cngccBHwK2Z/f9MfDzwPmEkNuyo89/hyRJm8aeT0mSykmBPYRezw+33fcUQk8owE2E0Hgm8NRc\n+Y3APKH385nAE4DPEYLsM4DdwBeBHwDeBPw0YbivJElDYWyzKyBJ0hZyA/AHhF7Ps9ruq3RuzlKX\ncoA/AX67oPzxwIWEntDnAy9ZU00lSZIkSVtOM/t5DvDKbHmG5WG3b2V5eO2PE87jbJX/erb8TFYO\nu70HmMjuGwd2Zfe1hto+htDbKkmSJEkaEQsFZU8n9IQCJIQJh+6gc8KhmwgTDr0N+BLLl1q5DLgz\nu+824MmEUHp77taa4EiSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmS\nJElD4P8DGgAglah3ymIAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x46654750>"
       ]
      }
     ],
     "prompt_number": 76
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "eff_final = dict()\n",
      "eff_final[\"default-first-set\"] = eff_first[64]\n",
      "eff_final[\"default-second-set\"] = eff_second[64]\n",
      "eff_final[\"6-first-set\"] = eff_first[6]\n",
      "eff_final[\"6-second-set\"] = eff_second[6]\n",
      "eff_final[\"6-11-first-set-(without bad features)\"] = eff_third[6]\n",
      "eff_final[\"6-11-second-set-(without bad features)\"] = eff_fourth[6]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 85
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plot_eff(eff_final)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA58AAAHyCAYAAABs5HrRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x46653290>"
       ]
      }
     ],
     "prompt_number": 86
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "6 - first_factors"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "7\n",
      "sumpt\n",
      "6948.07519531\n",
      "23135.8515625\n",
      "4703.31494141\n",
      "5573.71972656\n",
      "43223.8984375\n",
      "9492.609375\n",
      "4059.85009766\n",
      "\n",
      "---\n",
      "\n",
      "11\n",
      "m\n",
      "279.179016113\n",
      "3006.35498047\n",
      "2322.66503906\n",
      "5727.72509766\n",
      "4637.91503906\n",
      "3674.4699707\n",
      "2458.47998047\n",
      "1530.11999512\n",
      "4433.58984375\n",
      "3548.20507812\n",
      "1368.93505859\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "mcor\n",
      "3597.75488281\n",
      "4552.12011719\n",
      "2495.47509766\n",
      "8499.10546875\n",
      "3996.34521484\n",
      "6998.0546875\n",
      "5187.08984375\n",
      "5809.40039062\n",
      "7329.87011719\n",
      "5497.25488281\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "ipchi2\n",
      "1671.9699707\n",
      "151.517501831\n",
      "15.8388004303\n",
      "56.1777992249\n",
      "425.950012207\n",
      "13903400.0\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "vchi2\n",
      "1.08367502689\n",
      "33668.8984375\n",
      "64.3731994629\n",
      "5.97227478027\n",
      "100975.5\n",
      "67343.796875\n",
      "5804.09472656\n",
      "689.443969727\n",
      "\n",
      "---\n",
      "\n",
      "5\n",
      "sumipchi2\n",
      "801.427978516\n",
      "1794.45507812\n",
      "15499.9501953\n",
      "313.480499268\n",
      "4260.64501953\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "fdr\n",
      "3.19582509995\n",
      "1.78330504894\n",
      "0.712935507298\n",
      "0.402131974697\n",
      "1.1380699873\n",
      "1920.56005859\n",
      "\n",
      "---\n",
      "\n",
      "1\n",
      "nlt16\n",
      "0.5\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "minpt\n",
      "1162.35998535\n",
      "1793.63000488\n",
      "6365.70996094\n",
      "813.568481445\n",
      "424.366516113\n",
      "3327.08496094\n",
      "9434.37011719\n",
      "586.171508789\n",
      "\n",
      "---\n",
      "\n",
      "5\n",
      "ptau\n",
      "0.197485506535\n",
      "0.0755237489939\n",
      "0.61455899477\n",
      "0.314050495625\n",
      "0.127310007811\n",
      "\n",
      "---\n",
      "\n",
      "7\n",
      "q\n",
      "-1.5\n",
      "-2.5\n",
      "-0.5\n",
      "-3.5\n",
      "2.5\n",
      "1.5\n",
      "0.5\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "eta\n",
      "3.44897508621\n",
      "3.14533996582\n",
      "2.50030517578\n",
      "3.74745512009\n",
      "2.7830350399\n",
      "4.49971485138\n",
      "4.09770488739\n",
      "3.99977493286\n",
      "3.00011491776\n",
      "3.5\n",
      "\n",
      "---\n",
      "\n",
      "4\n",
      "nhlt1\n",
      "2.5\n",
      "3.5\n",
      "0.5\n",
      "1.5\n",
      "\n",
      "---\n",
      "\n",
      "2\n",
      "n\n",
      "2.5\n",
      "3.5\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "fdchi2\n",
      "5736.80957031\n",
      "5613029888.0\n",
      "23147.5\n",
      "905.598999023\n",
      "2233.35498047\n",
      "310.874511719\n",
      "\n",
      "---"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "ERROR: Cell magic `%%` not found.\n"
       ]
      }
     ],
     "prompt_number": 87
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "6 - second_factors"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "7\n",
      "sumpt\n",
      "6948.07519531\n",
      "23135.8515625\n",
      "4703.31494141\n",
      "5573.71972656\n",
      "43223.8984375\n",
      "9492.609375\n",
      "4059.85009766\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "m\n",
      "3006.35498047\n",
      "2322.66503906\n",
      "5727.72509766\n",
      "4637.91503906\n",
      "3674.4699707\n",
      "2458.47998047\n",
      "1530.11999512\n",
      "4433.58984375\n",
      "3548.20507812\n",
      "1368.93505859\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "mcor\n",
      "3597.75488281\n",
      "4552.12011719\n",
      "2495.47509766\n",
      "8499.10546875\n",
      "3996.34521484\n",
      "6998.0546875\n",
      "5187.08984375\n",
      "5809.40039062\n",
      "7329.87011719\n",
      "5497.25488281\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "ipchi2\n",
      "1671.9699707\n",
      "151.517501831\n",
      "15.8388004303\n",
      "56.1777992249\n",
      "425.950012207\n",
      "13903400.0\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "vchi2\n",
      "1.08367502689\n",
      "33668.8984375\n",
      "64.3731994629\n",
      "5.97227478027\n",
      "100975.5\n",
      "67343.796875\n",
      "5804.09472656\n",
      "689.443969727\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "sumipchi2\n",
      "801.427978516\n",
      "1794.45507812\n",
      "15499.9501953\n",
      "313.480499268\n",
      "4260.64501953\n",
      "63790252.0\n",
      "\n",
      "---\n",
      "\n",
      "1\n",
      "nlt16\n",
      "0.5\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "minpt\n",
      "1162.35998535\n",
      "1793.63000488\n",
      "6365.70996094\n",
      "813.568481445\n",
      "424.366516113\n",
      "3327.08496094\n",
      "9434.37011719\n",
      "586.171508789\n",
      "\n",
      "---\n",
      "\n",
      "7\n",
      "pt\n",
      "22296.8007812\n",
      "3890.96484375\n",
      "42685.0\n",
      "3131.125\n",
      "9129.32421875\n",
      "6417.58007812\n",
      "4867.38476562\n",
      "\n",
      "---\n",
      "\n",
      "2\n",
      "n\n",
      "2.5\n",
      "3.5\n",
      "\n",
      "---\n",
      "\n",
      "4\n",
      "nhlt1\n",
      "2.5\n",
      "3.5\n",
      "0.5\n",
      "1.5\n",
      "\n",
      "---\n",
      "\n",
      "7\n",
      "fdchi2\n",
      "5736.80957031\n",
      "5613029888.0\n",
      "23147.5\n",
      "11280799744.0\n",
      "905.598999023\n",
      "2233.35498047\n",
      "310.874511719\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "eta\n",
      "3.44897508621\n",
      "3.14533996582\n",
      "2.50030517578\n",
      "3.74745512009\n",
      "2.7830350399\n",
      "4.49971485138\n",
      "4.09770488739\n",
      "3.99977493286\n",
      "3.00011491776\n",
      "3.5\n",
      "\n",
      "---"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "6 - third_factors"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "11\n",
      "m\n",
      "279.179016113\n",
      "3006.35498047\n",
      "2322.66503906\n",
      "5727.72509766\n",
      "4637.91503906\n",
      "3674.4699707\n",
      "2458.47998047\n",
      "1530.11999512\n",
      "4433.58984375\n",
      "3548.20507812\n",
      "1368.93505859\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "mcor\n",
      "3597.75488281\n",
      "4552.12011719\n",
      "2495.47509766\n",
      "8499.10546875\n",
      "3996.34521484\n",
      "6998.0546875\n",
      "5187.08984375\n",
      "5809.40039062\n",
      "7329.87011719\n",
      "5497.25488281\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "ipchi2\n",
      "1671.9699707\n",
      "151.517501831\n",
      "15.8388004303\n",
      "56.1777992249\n",
      "425.950012207\n",
      "13903400.0\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "vchi2\n",
      "1.08367502689\n",
      "33668.8984375\n",
      "64.3731994629\n",
      "5.97227478027\n",
      "100975.5\n",
      "67343.796875\n",
      "5804.09472656\n",
      "689.443969727\n",
      "\n",
      "---\n",
      "\n",
      "5\n",
      "sumipchi2\n",
      "801.427978516\n",
      "1794.45507812\n",
      "15499.9501953\n",
      "313.480499268\n",
      "4260.64501953\n",
      "\n",
      "---\n",
      "\n",
      "1\n",
      "nlt16\n",
      "0.5\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "minpt\n",
      "1162.35998535\n",
      "1793.63000488\n",
      "6365.70996094\n",
      "813.568481445\n",
      "424.366516113\n",
      "3327.08496094\n",
      "9434.37011719\n",
      "586.171508789\n",
      "\n",
      "---\n",
      "\n",
      "7\n",
      "ptau\n",
      "0.197485506535\n",
      "0.0755237489939\n",
      "296.51550293\n",
      "0.61455899477\n",
      "0.314050495625\n",
      "0.127310007811\n",
      "0.00603420007974\n",
      "\n",
      "---\n",
      "\n",
      "7\n",
      "q\n",
      "-1.5\n",
      "-2.5\n",
      "-0.5\n",
      "-3.5\n",
      "2.5\n",
      "1.5\n",
      "0.5\n",
      "\n",
      "---\n",
      "\n",
      "2\n",
      "n\n",
      "2.5\n",
      "3.5\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "fdchi2\n",
      "1.73073005676\n",
      "5736.80957031\n",
      "5613029888.0\n",
      "23147.5\n",
      "11280799744.0\n",
      "905.598999023\n",
      "2233.35498047\n",
      "310.874511719\n",
      "\n",
      "---"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "ename": "SyntaxError",
       "evalue": "invalid syntax (<ipython-input-88-4547287433aa>, line 14)",
       "output_type": "pyerr",
       "traceback": [
        "\u001b[1;36m  File \u001b[1;32m\"<ipython-input-88-4547287433aa>\"\u001b[1;36m, line \u001b[1;32m14\u001b[0m\n\u001b[1;33m    ---\u001b[0m\n\u001b[1;37m       ^\u001b[0m\n\u001b[1;31mSyntaxError\u001b[0m\u001b[1;31m:\u001b[0m invalid syntax\n"
       ]
      }
     ],
     "prompt_number": 88
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "6 - fourth_factors"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "11\n",
      "m\n",
      "279.179016113\n",
      "3006.35498047\n",
      "2322.66503906\n",
      "5727.72509766\n",
      "4637.91503906\n",
      "3674.4699707\n",
      "2458.47998047\n",
      "1530.11999512\n",
      "4433.58984375\n",
      "3548.20507812\n",
      "1368.93505859\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "mcor\n",
      "3597.75488281\n",
      "4552.12011719\n",
      "2495.47509766\n",
      "8499.10546875\n",
      "3996.34521484\n",
      "6998.0546875\n",
      "5187.08984375\n",
      "5809.40039062\n",
      "7329.87011719\n",
      "5497.25488281\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "ipchi2\n",
      "1671.9699707\n",
      "151.517501831\n",
      "15.8388004303\n",
      "56.1777992249\n",
      "425.950012207\n",
      "13903400.0\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "vchi2\n",
      "1.08367502689\n",
      "33668.8984375\n",
      "64.3731994629\n",
      "5.97227478027\n",
      "100975.5\n",
      "67343.796875\n",
      "5804.09472656\n",
      "689.443969727\n",
      "\n",
      "---\n",
      "\n",
      "5\n",
      "sumipchi2\n",
      "801.427978516\n",
      "1794.45507812\n",
      "15499.9501953\n",
      "313.480499268\n",
      "4260.64501953\n",
      "\n",
      "---\n",
      "\n",
      "1\n",
      "nlt16\n",
      "0.5\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "minpt\n",
      "1162.35998535\n",
      "1793.63000488\n",
      "6365.70996094\n",
      "813.568481445\n",
      "424.366516113\n",
      "3327.08496094\n",
      "9434.37011719\n",
      "586.171508789\n",
      "\n",
      "---\n",
      "\n",
      "8\n",
      "pt\n",
      "22296.8007812\n",
      "3890.96484375\n",
      "42685.0\n",
      "3131.125\n",
      "9129.32421875\n",
      "62497.8007812\n",
      "6417.58007812\n",
      "4867.38476562\n",
      "\n",
      "---\n",
      "\n",
      "2\n",
      "n\n",
      "2.5\n",
      "3.5\n",
      "\n",
      "---\n",
      "\n",
      "6\n",
      "fdchi2\n",
      "5736.80957031\n",
      "5613029888.0\n",
      "23147.5\n",
      "905.598999023\n",
      "2233.35498047\n",
      "310.874511719\n",
      "\n",
      "---\n",
      "\n",
      "10\n",
      "eta\n",
      "3.44897508621\n",
      "3.14533996582\n",
      "2.50030517578\n",
      "3.74745512009\n",
      "2.7830350399\n",
      "4.49971485138\n",
      "4.09770488739\n",
      "3.99977493286\n",
      "3.00011491776\n",
      "3.5\n",
      "\n",
      "---\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}