jtg567/HTTP response throttling v2 (Pref Flip).ipynb Secret

## HTTP response throttling v2 (Pref Flip).ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/mnt/anaconda2/lib/python2.7/site-packages/IPython/core/magics/pylab.py:161: UserWarning:\n",
      "\n",
      "pylab import has clobbered these variables: ['rand', 'cbrt', 'rank', 'cosh', 'hypot', 'array', 'tan', 'size', 'isnan', 'randn', 'log', 'floor', 'sum', 'sqrt', 'degrees', 'split', 'rint', 'log10', 'sin', 'log2', 'cos', 'ceil', 'broadcast', 'sinh', 'repeat', 'trunc', 'expm1', 'tanh', 'radians', 'exp', 'log1p', 'mean']\n",
      "`%matplotlib` prevents importing * from pylab and numpy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import plotly.plotly as py\n",
    "from scipy.stats import chi2_contingency\n",
    "from scipy.stats import ttest_ind\n",
    "from collections import defaultdict as dd\n",
    "\n",
    "import datetime as DT\n",
    "\n",
    "from pyspark.sql import Row\n",
    "from pyspark.sql import SQLContext\n",
    "from pyspark.sql.types import *\n",
    "from pyspark.sql.functions import *\n",
    "import itertools\n",
    "\n",
    "from moztelemetry.dataset import Dataset\n",
    "from moztelemetry import get_pings_properties\n",
    "\n",
    "from scipy.stats import mannwhitneyu\n",
    "from __future__ import division\n",
    "\n",
    "py_max = __builtin__.max\n",
    "py_map = __builtin__.map\n",
    "\n",
    "%pylab inline\n",
    "\n",
    "## these ones pull the data\n",
    "#\n",
    "def recursive_get(d, keys):\n",
    "    if len(keys) == 1:\n",
    "        return d.get(keys[0],{})\n",
    "    return recursive_get(d.get(keys[0],{}), keys[1:])\n",
    "\n",
    "def extract_probes(p):\n",
    "    branch = p.get(\"environment\",{}).get(\"experiments\",{}).get(EXPERIMENT_SLUG, {}).get(\"branch\", \"warning\")\n",
    "\n",
    "    output = []\n",
    "    for path, probe_names in experiment_probes.iteritems():\n",
    "        for probe_name in probe_names:\n",
    "            probe = recursive_get(p, path.split(\"/\")).get(probe_name, {})\n",
    "            #take all entries in histogram right now. this is probably problematic. inspect more\n",
    "            for k,v in probe.get(\"values\",{}).iteritems():\n",
    "                if v <= sys.maxint:\n",
    "                    output.extend([{\"probe\": probe_name, \"branch\": branch, \"val\": float(k)}] * int(v))                \n",
    "    return output\n",
    "\n",
    "def exptPings(slug, samp, start= DT.date.today().strftime(\"%Y%m%d\"), end= DT.date.today().strftime(\"%Y%m%d\")):\n",
    "    # returns a spark df with values from probe dictionary by branch and nothing else, default start/end dates are today\n",
    "    cohorts = Dataset.from_source(\"telemetry-cohorts\")\n",
    "    main_pings = cohorts.where(submissionDate = lambda x: x >= start and x <= end).where(experimentId= slug).where(docType= \"main\").records(sc, sample= samp)\n",
    "    #main_pings.cache() this was in ilana's ognb, mreid advised to remove it for the exception I was hitting\n",
    "    \n",
    "    probe_dicts = main_pings.flatMap(extract_probes)\n",
    "    return sqlContext.createDataFrame(probe_dicts.map(lambda d: Row(**d)))\n",
    "\n",
    "## these ones do the analysis\n",
    "#\n",
    "\n",
    "###### get the bins we use for the histogram for a probe by looking at all branches\n",
    "def get_bins(probe_name, logscale=False):\n",
    "    \n",
    "    all_branches = [r.val for r in df.where(df.probe == probe_name).collect()]\n",
    "\n",
    "    #remove top 0.5%, bottom 0.5% for easy outlier\n",
    "    trim = int(len(all_branches)/200.0)\n",
    "    all_branches_trimmed = sorted(all_branches)\n",
    "    all_branches_trimmed = all_branches_trimmed[trim:-1*trim]\n",
    "    \n",
    "    if logscale:\n",
    "        if all_branches_trimmed[0] < 1:\n",
    "            all_branches_trimmed = py_map(lambda d: d+1, all_branches_trimmed)\n",
    "        return list(np.linspace(np.log10(all_branches_trimmed[1]), np.log10(all_branches_trimmed[-1]), 10))\n",
    "    \n",
    "    n,b = np.histogram(all_branches_trimmed,10)\n",
    "    return b\n",
    "\n",
    "# get values for branch of experiment for pref, and trim off outliers\n",
    "def get_vals(pref_name, branch, samp, seed=None):\n",
    "    x_vals = [r.val for r in df.where(df.probe == pref_name)\\\n",
    "                               .where(df.branch == branch)\\\n",
    "                               .sample(False, samp, seed).collect()] \n",
    "\n",
    "    trim = int(len(x_vals)/200.0)\n",
    "    x_trimmed = sorted(x_vals)[trim:-1*trim]\n",
    "    return x_trimmed\n",
    "\n",
    "def median(lst):\n",
    "    return lst[(len(lst))/2]\n",
    "\n",
    "# return (pval, direction) if significant p value for mannwhitneyu vs control\n",
    "def test_unequal(branch_vals, control_vals, p_threshold=.05):\n",
    "    \n",
    "    try:\n",
    "        r = mannwhitneyu(branch_vals, control_vals)\n",
    "    except:\n",
    "        return None, None\n",
    "    \n",
    "    prefix = \"\"\n",
    "    if r.pvalue < p_threshold:\n",
    "        prefix = \"***\"\n",
    "    \n",
    "    if median(branch_vals) > median(control_vals):\n",
    "        return (prefix + str(r.pvalue), \"> control\")\n",
    "   \n",
    "    return (prefix + str(r.pvalue), \"< control\")\n",
    "\n",
    "# return (proceed bool, reason)\n",
    "def can_chart_pref(pref_name):\n",
    "    n = df.where(df.probe == pref_name).count()\n",
    "    if n==0:\n",
    "        return (False, \"0 entries for pref %s\"%pref_name)\n",
    "    elif n>100000000:\n",
    "        return (False, \"%i values for pref %s\"%(n,pref_name))\n",
    "    return (True, None)   \n",
    "\n",
    "# chart histograms for all branches of a probe, log/std, and calculate if any branches vary from the mean\n",
    "def chart_pref(pref_name, logscale, samp):\n",
    "   \n",
    "    sig_branches = []    \n",
    "    fig, axarr = plt.subplots(n_branches, 1, sharex=True, sharey= 'col')\n",
    "\n",
    "    b = get_bins(pref_name, logscale)\n",
    "    plt.tight_layout()\n",
    "\n",
    "    control_vals = get_vals(pref_name, \"Control\", samp, 666)\n",
    "    \n",
    "    for i in range(n_branches):\n",
    "        \n",
    "        if branches[i] == \"Control\":\n",
    "            x = control_vals\n",
    "            \n",
    "        else:\n",
    "            x = get_vals(pref_name, branches[i], samp, 666) \n",
    "        \n",
    "        if logscale: #always assume 0 as lowest val for now\n",
    "            x_trans = py_map(lambda d: d+1, x)\n",
    "            ap,bp,cp = axarr[i].hist(np.log(x_trans), bins=b)\n",
    "            \n",
    "        else: \n",
    "            axarr[i].hist(x, bins=b)\n",
    "            \n",
    "        axarr[i].set_title(branches[i])\n",
    "        \n",
    "        if branches[i] == \"Control\": continue\n",
    "        \n",
    "        print \"len(branch_vals) = \" + str(len(x)) + \", len(control_vals) = \" + str(len(control_vals))\n",
    "        if len(x) != 0 | len(control_vals) != 0:\n",
    "#            p, direction = test_unequal(x, control_vals)\n",
    "#            if p is not None and p.startswith(\"***\"):\n",
    "#                print branches[i], p #, direction\n",
    "            print pref_name + \" (logscale=\" + str(logscale) + \")\" + \":\"\n",
    "        else:\n",
    "            print pref_name + \" branch with no values\"\n",
    "            continue\n",
    "            \n",
    "    plt.show()\n",
    "    \n",
    "from statsmodels.distributions.empirical_distribution import ECDF\n",
    "\n",
    "def chart_ecdfs(pref_name, samp, *percentiles):\n",
    "    \n",
    "    legend = []\n",
    "    \n",
    "    for i in range(n_branches):\n",
    "        legend_entry = \"\"\n",
    "        v = get_vals(pref_name, branches[i], samp, 666)\n",
    "        if len(v) == 0: continue\n",
    "        cdf = ECDF(v)\n",
    "        curr_plot = plt.plot(cdf.x, cdf.y)\n",
    "        curr_color = curr_plot[0].get_color()\n",
    "        legend_entry += branches[i]\n",
    "        for pct in percentiles:\n",
    "            p = np.percentile(np.array(v), pct)\n",
    "            plt.scatter(p, pct/100.0, facecolors = \"none\", \\\n",
    "                     edgecolors = curr_color, label=\"_nolegend_\")\n",
    "            legend_entry += \", \" + str(pct) + \"th percentile=\" + str(p)\n",
    "            \n",
    "        legend.append(legend_entry)\n",
    "        \n",
    "    plt.legend(legend, bbox_to_anchor=[1, .5], loc='center left')\n",
    "    plt.show()\n",
    "    \n",
    "EXPERIMENT_SLUG = \"pref-flip-http-response-throttling-algo-v2-beta-1434388\"\n",
    "branches = ['Control', 'Variant']\n",
    "n_branches = len(branches)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fetching 25490.23603MB in 258 files...\n",
      "len(branch_vals) = 10226848, len(control_vals) = 11746096\n",
      "TIME_TO_DOM_INTERACTIVE_MS (logscale=True):\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmYAAAGUCAYAAACfnu4FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa463799e50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABCYAAAFkCAYAAAD8Azr/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nACBqSEwAAAAAAICoITEBAAAAAACi5v8D/ZyhJxNmSUEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa44692a350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "experiment_probes = {}\n",
    "experiment_probes[\"payload/processes/content/histograms\"] =  [\"TIME_TO_DOM_INTERACTIVE_MS\"]\n",
    "probe_names = list(itertools.chain(*experiment_probes.values()))\n",
    "\n",
    "df = exptPings(EXPERIMENT_SLUG, 0.05, \"20180220\", \"20180305\")\n",
    "\n",
    "for p in probe_names:\n",
    "    chart_pref(p, True, 1.0)\n",
    "    chart_ecdfs(p, 1.0, 50, 95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fetching 50628.56669MB in 517 files...\n",
      "len(branch_vals) = 2052169, len(control_vals) = 2175971\n",
      "TIME_TO_NON_BLANK_PAINT_NETOPT_MS (logscale=True):\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlsAAAGUCAYAAAACk+eNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njE4HNgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa44692fb90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDsAAAFkCAYAAAAuUzCnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3XlcFOUfB/DPMwtygyB4I4gnpmWgWR6IVx6ZZuIZeWbe\nGpX+tLQ8MrW8NS2PPDKtvMqry9IsNUvU8r5vzVuUS2D3+/tj2ZWF5VJwdfm8fe0L5plnZr47zAjz\n3edQIgIiIiIiIiIiInuh2ToAIiIiIiIiIqK8xGQHEREREREREdkVJjuIiIiIiIiIyK4w2UFERERE\nREREdoXJDiIiIiIiIiKyK0x2EBEREREREZFdYbKDiIiIiIiIiOwKkx1EREREREREZFeY7CAiIiIi\nIiIiu8JkBxERERERERHZlXxNdiil6iml1iqlLiilDEqpVtnUb6OU+kkpdUUpFaOU2q6Uej4/YyQi\nIiIiIiIi+5LfLTvcAOwF0A+A5KB+GICfADQHEAJgM4B1Sqmn8i1CIiIiIiIiIrIrSiQnOYg8OJBS\nBgAvicjaXG63H8BXIvJB/kRGRERERERERPbkkR6zQymlAHgAuGHrWIiIiIiIiIjo8eBg6wCyMQTG\nrjDfZFZBKVUEQFMApwEkPpywiIiI7IIzgEAAP4rIdRvHQkRERJRnHtlkh1KqM4CRAFqJyLUsqjYF\n8OXDiYqIiMguvQJgma2DICIiIsorj2SyQynVEcBcABEisjmb6qcBYOnSpQgODs7v0B55UVFRmDp1\nqq3DsDmeh3t4Lox4Hox4Hu7huQAOHTqEyMhIIPV3KREREZG9eOSSHUqpTgDmA+ggIj/kYJNEAAgO\nDkZISEi+xvY48PLy4nkAz0NaPBdGPA9GPA/38FxYYDdQIiIisiv5muxQSrkBKA9ApRYFpU4je0NE\nzimlxgMoKSJdU+t3BrAIwCAAfyuliqVulyAit/MzViIiIiIiIiKyD/k9G0sNAHsARAMQAJMB7AYw\nOnV9cQD+aer3AqAD8AmAi2le0/I5TiIiIiIiIiKyE/naskNEfkMWCRUR6Z5uuUF+xkNERERERERE\n9i+/W3bQQ9apUydbh/BI4Hm4h+fCiOfBiOfhHp4LIiIiIvulRMTWMTwQpVQIgOjo6GgONEdERJQL\nu3fvRmhoKACEishuW8dDRERElFfYsoOIiIiIiIiI7AqTHURERERERERkV5jsICIiIiIiIiK7wmQH\nEREREREREdkVJjuIiIiIiIiIyK4w2UFEREREREREdoXJDiIiIiIiIiKyK0x2EBEREREREZFdYbKD\niIiIiIiIiOwKkx1EREREREREZFeY7CAiIiIiIiIiu8JkBxERERERERHZFSY7iIiIiIiIiMiuMNlB\nRERERERERHaFyQ4iIiIiIiIisitMdhARERERERGRXWGyg4iIiIiIiIjsCpMdRERERERERGRXmOwg\nIiIiIiIiIrvCZAcRERERERER2RUmO4iIiIiIiIjIrjDZQURERERERER2hckOIiIiIiIiIrIrTHYQ\nERERERERkV1hsoOIiIiIiIiI7AqTHURERERERERkV5jsICIiIiIiIiK7wmQHEREREREREdkVJjuI\niIiIiIiIyK4w2UFEREREREREdoXJDiIiIiIiIiKyK/ma7FBK1VNKrVVKXVBKGZRSrXKwTbhSKlop\nlaiUOqqU6pqfMRIRERERERGRfcnvlh1uAPYC6AdAsquslAoEsB7ALwCeAjAdwHylVJP8C5GIiIiI\niIiI7IlDfu5cRH4A8AMAKKVUDjbpC+CkiAxNXT6ilKoLIArAz/kTJRERERERERHZk0dtzI5nAWxK\nV/YjgOdsEAsRERERERERPYYetWRHcQCX05VdBuCplHKyQTxERERERERE9JjJ124sRET2SMTyZTCY\nvgoMIkjRG8zfpy1L0Ruw/q/9SEi6C4MIRAQGMRjrmZfTfzXcW0ZqmSHNehgs6osIjl09i7ikOCil\njGWp21l8hTH49GWm72GlHMikHgTnDDvhrPcDlHF4prTr0n41f6/SliMHddLVT7MuRXcL8PgPTjpn\nODkUSvNzyjhU1L3jIdf1clN33ovzEPlkpNV9EBEREVH+e9SSHf8BKJaurBiA2yJyN6sNo6Ki4OXl\nZVHWqVMndOrUKW8jJKIHtu/ENZy/eht6g/FBX28wJgfSLpu/1xsf5tOuMy2fv34D6w7/gEI6J4gY\n7iUDYDA/yGf2vaQmCczfpz5M/6fbBb3HSSDJDVAG80P1va9WyjSDTc9nBgqAE+AQXwoQlVqgoKAy\nLpu+F82izLxOrJRBwTgM070yR50DRHcHfrryqSHcq2f6h7Tfm7ZOsx+Vfr21dVbqKCikOFxH27Bg\nFNJZ/lqzNlyUadv7qZfTuqElQq1ua0vLly/H8uXLLcpiYmJsFA0RERFR/lLWPqXKlwMpZQDwkois\nzaLOBADNReSpNGXLABQWkRaZbBMCIDo6OhohISF5HTaRXdpz9CpuxyUixZRUMNxLMKToDfe+Nxig\nT7OcrNdj9KYJUEoDUhMExkRBauJApfkeBogyJRcMqesESdpNJBc+mKfvp1BsEHTiDOMjZ5qHdtHu\nPWSbvk+tk+F7paClbnsXt9GoWEcolbpWGV+a0qyWaWkfxJWCLrWecX2aOkpBKaB1zZrmZU0pOOju\nrdc00zbG71Wa701105el30ZTCu7OznDQ6fL0PJP92b17N0JDQwEgVER22zoeIiIiorySry07lFJu\nAMoD5o+8gpRSTwG4ISLnlFLjAZQUka6p6z8F0F8pNRHA5wAaAYgAYDXRQVTQ9Z61DOdv/Qe9QQ+9\nGFK7RFi+JPWrXvQwiAGHEn5Dgs9f93/QIoCW4Au/5GdS0wQaFBxSkwCacVlpqekGzZgggEotMy4n\nxT2FyJB2KO7lDZ1mfNDXaRqUpuCgGR/m05abHuDNZbp760oX8UbFUukbhBERERERUUGW391YagDY\nDFNna2ByavliAD1gHJDU31RZRE4rpV4AMBXAIADnAfQUkfQztBA9lkzdJlL0eiTr9UjRG5CUrEeK\nwYC5P/yOKTsnWraWgAGi9BAYgHTLKY43Ia5XAADqbmFjKwbRAKR+FZ2xlYNoll8dBY6xgRj77GwU\ncnCATtPuvXSaRZLB9HJIU+7l5oKwJypYbcpPRERERET0KMjXZIeI/IYsZnwRke5WyrYCePQ6OxOl\nYTAIbsbfQVKKHknJeiSnGBMWySl6JKUYl5NT9Hh96Yc47LwI0DsYx3rIbmwHb6BcbBdo0KV2mdCg\nQYNO6Yzfpy5rmg4aNLjABUv6vokyRb2y3i8REREREVEB8qgNUEpkcyKCFEMK9KJP7R6iR4remMQw\nJTIaTemPMy7fZr8zZ+OXdh4zoCkNDpru3lfN+FWn3Vt+tkIFdAh/Mn/fIBERERERkZ1jsoMonWrv\n9sIBpwVZV3IBkOiJAWUWwEGng4OmM37V6eCY5nudpqFpaDCqlvV9KLETERERERERkx1kR1L0BrSf\nPBPrLs9EivtpON2pjMigtzG3X1do2r3xJWq9/wYOx22HASkQ6GFQxq+S+jXJ6SLcbjyHNv59jUkM\nTQcHncO9hEbq15drV0fNYA6MSURERERE9KhhsoPsRq2Rb2N3oWkIlFfgdrMjTt06gQW35+DHPnvQ\nsFoVpOhTkGxIwV/Jn8M5uSJKa7WgUzrolAM0pYMDHKBpOuiSdejbvA26N+NUxkRERERERI8jJjvI\nLuw6egG7HWegqcOHaPt8fby+szZQxLjuPP7Ckms6wOAAGHSAOKJP1WGY2ivCtkETERERERFRvmCy\ngx5bz73/NvbF/gqDSkaS/i6AkvjF4TP8vGUK4AKsbHAQN2Pj0WvHc5j9zBb0aVkbnC2ViIiIiIjI\n/jHZQY88vUGPWwm3EX83GXeTUxB/NxkJScn4O/5rOOr9EOTQENfjEnA58RQCPKuhqJcn/Bx80bpO\nJYz7+gfAIRnFfTyZ6CAiIiIiIiogmOygR16pIS1x2fOHjCvcgTD0xq/vj0B8YjI83gvA9SQ9fuy/\nEuVK+mD3sYuY8Pc7cMZTaP3cEw8/cCIiIiIiIrIJJjvokXLl9i1cvX0bcYlJSEwytuC4oY6jyI0W\neDW4Nxx1DnBydISTgyNcChVClybGQURdnR0xufYyRP39IsrPLg3nuIpI9DgIVagwvmm5yWI2FiIi\nIiIiIrJvTHbQI+PPQ2fw3FflAE1vucIDqJTSAVP7tMpy+zdeCkfjp45j6LLFOJt8GpU9emJy/1cR\nUKxwPkZNREREREREjxomO+iRcebKTUDTo5XDLISWqQInR0e4OBaCcyFHRITlrBtK1bLFsPHdofkc\nKRERERERET3KmOwgm9AbDCg2LAwxuuMwqCSIlgTR7gI6oHOduugQ/pStQyQiIiIiIqLHFJMdZBMJ\nd1Nw3W0bil5ri4puz6CQKoRCyhG+Lj54qXY1W4dHREREREREjzEmO+ihuBUfi0kb1uB2QhwSk5Nw\nJzEeANAmuDU+7feqjaMjIiIiIiIie8JkBz0U736xBrP/6wIYNEDvBOgLAUmlUPWpirYOjYiIiIiI\niOwMkx30UMTfTQYAnOiWDDdXDYUKAS4ugLOzjQMjIiIiIiIiu8NkB+WLKzG30W/xDMQk3kZiSiIO\n3/wXcAfKlAEcdLaOjoiIiIiIiOwZkx2UL2at34JVN0cCN4OgpbhCEyf43XkFmlK2Do2IiIiIiIjs\nHJMdlC9S9AIAOPjmnwgu42fjaIiIiIiIiKggYbKD8syhi2fx++GDiE1MxLZz22wdDhERERERERVQ\nTHZQnqkxuTXiPffeK4gvAl8Pd9sFRERERERERAUSkx2UZ1KQgPK3euPT9h/A09UZ/sVd4OfN0UiJ\niIiIiIjo4WKyg/KUl6s7Gj3na+swiIiIiIiIqABjsoPum4ig2ZQhOBlzFHcNCUhyOQuIraMiIiIi\nIiKigo7JDrpvt+Pv4qfYyXC48TRcE8ujiGqHyCYdbR0WERERERERFXBMdtB9k9RWHK898Sbm9I20\nbTBEREREREREqTRbB0BERERERERElJfYsoNybcCSmfjj7HbEJccyXUZERERERESPHCY7KNfm7B8P\nSXKB0+0q8ERnvNiknq1DIiIiIiIiIjJjsoPuSwOfbvhl2khbh0FERERERESUATshEBEREREREZFd\nYcsOyrGYxBjEJsUCWoqtQyEiIiIiIiLKFJMdlCNvLFqI6Wd6GBdcAGfNzbYBEREREREREWWCyQ7K\nkf1nLwCJXujptRxeru6Iav+MrUMiIiIiIiIisuqhjNmhlOqvlDqllEpQSv2plKqZTf1XlFJ7lVJx\nSqmLSqkFSimfhxErZU4zOGP+sOaYPKgeShd3snU4RERERERERFble7JDKdUBwGQA7wN4GsA/AH5U\nSvlmUr8OgMUA5gGoAiACwDMA5uZ3rERERERERET0+HsYLTuiAHwmIktE5DCAPgDiAfTIpP6zAE6J\nyCcickZEtgP4DMaEBz1kIoLElEQkI8HWoRARERERERHlSL4mO5RSjgBCAfxiKhMRAbAJwHOZbLYD\ngL9SqnnqPooBaAdgQ37GStZVGd0WLuNcsFU+hEpxsXU4RERERERERNnK7wFKfQHoAFxOV34ZQCVr\nG4jIdqVUJICvlVLOMMa4FsCA/AyUrDsdcwLud59Hk8J9ER4abOtwiIiIiIiIiLL1yM3GopSqAmA6\ngFEAfgJQAsAkGLuyvJbZdlFRUfDy8rIo69SpEzp16pRvsRYUZT0rYvWHL9k6DCIiegDLly/H8uXL\nLcpiYmJsFA0RERFR/srvZMc1AHoAxdKVFwPwXybbDAOwTUSmpC7vV0r1A/C7UupdEUnfSgQAMHXq\nVISEhORFzERERHbH2gcAu3fvRmhoqI0iIiIiIso/+Tpmh4gkA4gG0MhUppRSqcvbM9nMFUBKujID\nAAGg8iFMIiIiIiIiIrIjD6MbyxQAi5RS0QD+gnF2FlcAiwBAKTUeQEkR6Zpafx2AuUqpPgB+BFAS\nwFQAO0Uks9YglMc2Hd6BrUf/QXKhq7YOhYiIiIiIiChX8j3ZISLfKKV8AYyBsfvKXgBNRcT0FF0c\ngH+a+ouVUu4A+sM4VsctGGdzGZbfsdI9zRdGIMX5P0B5ooxjdVuHQ0RERERERJRjD2WAUhGZDWB2\nJuu6Wyn7BMAn+R0XZU6PZNSK+wDzug5HMCdhISIiIiIiosfIIzcbCz06vAoD1arZOgoiIiIiIiKi\n3MnXAUqJiIiIiIiIiB42JjuIiIiIiIiIyK6wGwtZOB9zAX+d/QdwSLR1KERERERERET3hckOslB9\nXAdcd9sGFAL8CvlnvwERERERERHRI4bJDrIQn5yAkle6YVbzWWjWyM3W4RARERERERHlGpMdlIG3\nuwvatGSig4iIiIiIiB5PHKCUiIiIiIiIiOwKkx1EREREREREZFeY7CAiIiIiIiIiu8JkBwEAfjsW\njWenRiDR9ZitQyEiIiIiIiJ6IEx2EABg9FdrsfP6BsiRF9CoaGdbh0NERERERER03zgbCwEA9HpA\nd9cXSd8sh8YUGBERERERET3G+FhLFpjoICIiIiIioscdH22JiIiIiIiIyK4w2UFEREREREREdoXJ\nDiIiIiIiIiKyK0x2EObt+BrHk7baOgwiIiIiIiKiPMHZWAq4OwmJeP2njoC+CApfbW3rcIiIiIiI\niIgeGFt2FHDJyQIAaO8xA6emLbBxNEREREREREQPjskOAgD4+gKFC9s6CiIiIiIiIqIHx2QHERER\nEREREdkVJjuIiIiIiIiIyK4w2UFEREREREREdoXJDiIiIiIiIiKyK0x2FGCbDu7Ck9PCAABKKRtH\nQ0RERERERJQ3mOwowOb/sBMX9Lvhe+A9dAhtautwiIiIiIiIiPKEg60DINsxGAAYHHH1m9G2DoWI\niIiIiIgoz7BlBxERERERERHZFSY7iIiIiIiIiMiuMNlBRERERERERHaFY3YQERFRlpRSZQD42joO\nIiIiIgDXRORsdpWY7CAiIqJMKaXKaJp2xGAwONs6FiIiIiJN0xKVUpWyS3g8lGSHUqo/gLcBFAfw\nD4CBIvJ3FvULAXgfwCup21wEMEZEFuV/tAXD6t2/YeuN5YDO1pEQEdEjztdgMDgvXboUwcHBto6F\niIiICrBDhw4hMjLSGcYWp7ZNdiilOgCYDOB1AH8BiALwo1Kqoohcy2SzFQD8AHQHcAJACXB8kTz1\n5vLZuKw7gCJn+9g6FCIiegwEBwcjJCTE1mEQERER5cjDaNkRBeAzEVkCAEqpPgBeANADwEfpKyul\nmgGoByBIRG6lFmfbH4dyx6AHfJJr4uoX02wdChEREREREVGeytfWEkopRwChAH4xlYmIANgE4LlM\nNnsRwC4A/1NKnVdKHVFKfayUYl/hfKCUrSMgIiIiIiIiylv53bLDF8ZRIS6nK78MoFIm2wTB2LIj\nEcBLqfuYA8AHQM/8CZOIiIiIiIiI7MWjOBuLBsAAoLOIxAKAUupNACuUUv1E5K61jaKiouDl5WVR\n1qlTJ3Tq1Cm/4yUiInrkLV++HMuXL7coi4mJsVE0RERERPkrvwf9vAZAD6BYuvJiAP7LZJtLAC6Y\nEh2pDgFQAEpndqCpU6di7dq1Fi8mOoiIiIw6deqU4ffk1KlTbR0W5aPw8HA0bNjQ1mHkSGBgIFq1\namXrMB5bmqZhzJgx5uVFixZB0zScPcth7woSTdMwaNAgW4fxWDpz5gw0TcOSJUvMZaNGjYKmcY6M\nx1m+/vREJBlANIBGpjKllEpd3p7JZtsAlFRKuaYpqwRja4/z+RQqERERFVAnT55E7969Ua5cObi4\nuMDLywt169bFjBkzkJiYmG/HPXToEEaPHp1vD6TqAQbmWrx4MTRNy/DS6XS4cuVKhvrbt29H3bp1\n4ebmhhIlSmDw4MGIi4uzqJPV+32QWAuK77//HqNHj7a6TillcQ7TLz8MR48eRVRUFOrUqQMXF5cs\nky1RUVEIDQ1FkSJF4ObmhipVqmD06NEZrhkAOH78ODp27Ah/f3+4ubkhODgYY8eORUJCQoa6hw8f\nRrNmzeDh4YEiRYqgS5cuuHYts8kfM8rJdZxbX331FUJDQ+Hi4oKiRYvitddew/Xr1zPUCwwMtHrP\n9evXz6Lejh07MHr0aNy+ffuB4iqoli9fjunTp1tdl/6escV99Pfff6Nfv36oUaMGChUqBJ1Ol2X9\nBQsWoEqVKnBxcUHFihUxa9Ysq/ViYmLw+uuvo2jRonB3d0fDhg2xZ8+eLPcdExODokWLQtM0rF69\n2mLdpUuXEBkZicqVK8PT0xPe3t6oVauWRbIoO0lJSfjf//6HUqVKwdXVFc8++yw2bdqU4+1z4mF0\nY5kCYJFSKhr3pp51BbAIAJRS4wGUFJGuqfWXARgBYKFSahSMU9B+BGBBZl1YKHdEBKIMtg6DiIjI\n5jZs2ID27dvD2dkZXbp0QdWqVZGUlIQ//vgDQ4cOxcGDB/Hpp5/my7EPHjyI0aNHo0GDBihTpky+\nHONBKKUwduxYBAYGWpQXLlzYYnnv3r1o3LgxqlSpgqlTp+L8+fP4+OOPcfz4cWzYsMFc71F/v4+6\njRs3Yvbs2Xj//fczrEtISICDg217p+/YsQOzZs1ClSpVUKVKFezduzfTutHR0QgLC0OPHj3g7OyM\nPXv2YMKECfjll1+wdetWc73z58+jZs2a8Pb2xsCBA+Hj44MdO3bg/fffx+7du7FmzRpz3QsXLqBe\nvXrw9vbGhAkTcOfOHXz88cfYv38//vrrr2zPT06v49yYM2cO+vfvjyZNmpj3OW3aNERHR2Pnzp0o\nVKiQua5SCk8//TTeeusti31UrFjRYnn79u0YM2YMunfvDk9Pz/uKqyBbtmwZDhw4gMGDB1uUBwQE\nICEhAY6OjjaKzGjjxo34/PPP8eSTT6JcuXI4evRopnU/++wz9O3bF+3atcNbb72F33//HYMGDUJC\nQgKGDBliriciaNGiBfbt24ehQ4eiSJEimD17NsLDw7F7926UK1fO6v5HjhyJxMREqwmfa9eu4eLF\ni2jXrh3KlCmD5ORk/Pzzz+jWrRuOHj2KDz74INv32rVrV6xevRpRUVEoX748Fi1ahBYtWmDLli2o\nXbt2Ds5WDohIvr8A9ANwGkACgB0AaqRZtxDAr+nqVwTwI4BYAGdgTHY4ZbLvEAASHR0tlDMhH3QW\njIL4DXrR1qEQEZENRUdHCwABECKZ/w6329+zp06dEg8PD3niiSfk8uXLGdafOHFCZsyYkW/HX7Fi\nhWiaJr/99luO6ickJORq/+Hh4dKgQYP7CU0WLVokmqbl6OfevHlzKVWqlMTGxprL5s+fL5qmyc8/\n/2wuy+r9BgYGyosvPtp/l8TFxdn0+P379xdN03JU1/TzO3PmTD5Hdc/NmzfN18CkSZNyffzJkyeL\npmmyc+dOc9m4ceNE0zQ5dOiQRd2uXbuKpmly69Ytc1nfvn3Fzc1Nzp8/by7btGmTKKVk3rx52R4/\np9dxTiUlJYm3t3eGe3D9+vWilJJZs2ZZlOf0Hvj4448zPbdKKRk4cGCuY32YbH0ftWzZUsqWLZuj\nuqNGjcrxPZdXrly5IomJiSIiMmDAgEyPn5CQIL6+vtKqVSuL8sjISPHw8LC4N77++mtRSsnq1avN\nZVevXhVvb2955ZVXrO5/37594ujoKB988IFomiarVq3KUfwvvviieHh4iMFgyLLezp07RSklU6ZM\nMZclJiZK+fLlpU6dOllum5O/XUyvh9IJSURmi0igiLiIyHMisivNuu4i0jBd/aMi0lRE3EUkQESG\nClt15Jkj147D7b/GmNZktq1DISIispmJEyciLi4OCxYsQNGiRTOsDwoKwsCBA83Ler0eY8eORfny\n5eHs7IyyZcvi3XffRVJSksV2pvEntm3bhlq1asHFxQXlypXDF198Ya6zePFitG/fHoBxbA1TFxHT\np9qmffz000+oWbMmXFxcMHfu3FzFkVdiY2NhMFhvEXrnzh1s2rQJr776Ktzc3MzlXbp0gZubG775\n5pscvV+TrM5ZZkx97adMmYJp06YhMDAQrq6uCA8Px4EDBzLUP3LkCCIiIlCkSBG4uLigZs2aWLdu\nnUUdUzeerVu3ol+/fihWrBj8/f3N6y9evIiePXuiVKlScHZ2RlBQEPr164eUlBRznZiYGLzxxhso\nU6YMnJ2dUaFCBXz00UemJGKG2OfNm2f+mT7zzDPYtcv85zK6d++O2bONf7el7VJkkn7Mjsx8//33\nCAsLg7u7Ozw9PdGyZUscPHgw2+1yonDhwhbXQG4FBARARHDr1i1z2Z07dwAgw/1ZvHhxaJpm0TJi\n9erVaNmyJUqVKmUua9SoESpWrGi+DjOT0+s4N/bv349bt26Zr3uTF154Ae7u7vjqq6+sbpecnIz4\n+Hir60aPHo2hQ4cCuNftRafTZegu9N1336FatWpwdnZG1apV8eOPP2Yb72+//QZN0/DNN9/gnXfe\nQYkSJeDu7o7WrVvj/PmMIwns3LkTzZo1M//cw8PDsX275SgFpjEvDh06hM6dO8PHxwf16tUzrz9y\n5Ajat2+PokWLwtXVFZUrV8aIESMs9nHx4kX06NEDxYsXN7+fhQsXWo19xYoVGDduHPz9/eHi4oLG\njRvjxIkT5noNGjTAhg0bzPedpmkICgoCYH3MjswsXboUNWrUgKurK4oUKYJOnTpZPUf3w8/PD05O\nTtnW27x5M27cuJGhm1P//v0RGxtr0Rpp1apVKF68ONq0aWMu8/X1Rfv27fHdd98hOTk5w/4HDx6M\ntm3bom7duhb/Z2UnICAA8fHx2f4+WrlyJRwcHNCrVy9zmZOTE3r27IkdO3bgwoULOT5mVh7F2Vjo\nISjtHojOLTMd75WIiMjurV+/HkFBQahVq1aO6vfs2RNLlixB+/bt8fbbb2Pnzp0YP348Dh8+jFWr\nVpnrKaVw7NgxtGvXDj179kS3bt3w+eefo3v37qhRowaCg4MRFhaGQYMGYebMmRgxYgQqV64MAAgO\nDjbv4/AeGpAlAAAgAElEQVThw+jcuTN69+6N119/HZUqVcpVHA9KRBAeHo7Y2FgUKlQITZs2xeTJ\nk1G+fHlznX379iElJQWhoaEW2zo6OqJ69ermPuHZvV8A2Z6z7CxevBixsbEYMGAAEhMTMX36dDRq\n1Aj79u2Dn58fAODAgQOoW7cuSpcujeHDh5sfZF966SWsXr0arVu3tthnv379ULRoUbz//vvmsRsu\nXbqEmjVr4vbt2+jduzcqVaqECxcuYOXKlYiPj4enpycSEhIQFhaGS5cuoU+fPvD398f27dsxfPhw\n/Pfff5gyZYrFcb788kvExsaiT58+UEph4sSJaNu2LU6ePAmdToc+ffrg4sWL2LRpE7788stcPXyY\nfPHFF+jWrRuaNWuGjz76CPHx8ZgzZw7q1auHPXv2mLsWJSUlmZMM2SlSpEiu4zDR6/W4desWkpKS\nsG/fPowcORJeXl545plnzHXCw8MxceJE9OjRA6NHj0aRIkWwbds2fPrppxg8eDBcXFwAGB+Ir1y5\ngho1amQ4zjPPPIPvv/8+y1hyeh3nxt27xs9pTTGm5eLiYnWfv/76K1xdXaHX6xEQEICoqCiLAUfb\ntm2Lo0eP4quvvsL06dPN5990fQPA77//jtWrV6Nfv37w8PDAjBkzEBERgbNnz8Lb2zvbuMeNGwdN\n0zBs2DBcuXIFU6dORZMmTbB3717zQ/ivv/6KFi1aoEaNGuaExsKFC9GwYUP88ccf5p+DqftDu3bt\nULFiRYwfP9587f7777+oV68enJyc0Lt3bwQEBODEiRNYv369uQvElStXUKtWLeh0OgwaNAi+vr74\n/vvv0bNnT9y5cyfDYKwTJkyATqfDkCFDEBMTg4kTJyIyMhI7duwAAIwYMQIxMTG4cOECpk2bBhGB\nu7t7tuck/fl577330LFjR/Tq1QtXr17FjBkzUL9+fezZs8fctSghISHTpFVaOp0uQ9fAnDBdP+mv\n2dDQUGiahj179qBz587muiEhIRn28cwzz2DevHk4evQonnjiCXP5ihUr8Oeff+Lw4cM4efJklnEk\nJiYiLi4OsbGx2LJlCxYtWoTatWtnm7DZu3cvKlasmOH8m+7/vXv3WiQu71t2TT8e9RfsuHltfnF7\n4xmp9PZrtg6DiIhsLK+7scTFiURH5/8rL1pB3759W5RS0qZNmxzV/+eff0QpJb1797YoHzJkiGia\nJlu2bDGXBQYGiqZpsm3bNnPZ1atXxdnZWYYMGWIuW7lyZZbdOqw1n89NHA/SjeWbb76RHj16yBdf\nfCHfffedvPfee+Lm5iZFixa16CZgeg9//PFHhn20b99eSpYsmav3m905s+b06dOilBI3Nze5dOmS\nufyvv/4SpZS89dZb5rJGjRpJ9erVJTk52WIfderUkUqVKpmXFy1aJEopqV+/fobm2F26dBEHBwfZ\nvXt3pjGNHTtWPDw85MSJExblw4cPF0dHR/M5NMXu5+cnMTEx5npr164VTdNkw4YN5rKsmrQrpWT0\n6NEW8aft6hAbGyve3t7Sp08fi+2uXLkihQsXtrieTO89u1dWzftz0o3lzz//tNhfcHCwbN26NUO9\nDz74QFxdXS2OO3LkSIs6u3btEqWULF26NMP2Q4cOFU3TJCkpKdNYcnMd59S1a9dE0zTp1auXRfnh\nw4fN7+PGjRvm8tatW8vHH38sa9eulYULF0r9+vVFKSXDhg2z2D6rc6uUEmdnZzl16pS57N9//xWl\nlHzyySdZxrtlyxZRSom/v79FV5MVK1aIUkpmzpxpLqtYsaK0aNHCYvvExEQJCgqSpk2bmstGjRol\nSimJjIzMcLywsDDx8vKy+P8kvZ49e0qpUqXk5s2bFuWdOnUSb29vc3cPU+xPPPGEpKSkmOvNmDFD\nNE2TAwcOmMsy68ZiuhcXL15sEX/a6/zMmTPi4OAgEyZMsNj2wIED4ujoKOPHj8/w3rN7ZdWlJqt7\nfsCAAeLo6Gh1XdGiRaVz587mZXd3d3nttYzPfhs3bhRN0+Snn34ylyUkJEhAQICMGDFCRO6d28y6\nsUyYMMHi/TRp0iTLn6lJ1apVpXHjxhnKDx48KEopmTt3bqbb5qYbC1t2EBERUZ44fBhI9yFTvoiO\nBqx8SJUrppkMPDw8clR/48aNUEohKirKovytt97CpEmTsGHDBtSvX99cXqVKFYsB1nx9fVGpUqVs\nPyVLq2zZsmjcuPEDxXG/2rVrh3bt2pmXW7Vqheeffx5hYWEYN26cuUuFaUYMa5/iOTs7W50xIzMP\nes7atGmD4sWLm5dr1qyJWrVqYePGjZg0aRJu3ryJzZs3Y+zYsYiJibHY9vnnn8fo0aNx6dIllChR\nAoDxU+levXpZDM4nIvjuu+/QqlUrPP3005nGsnLlStSrVw9eXl4WM280atQIEyZMwNatW9GpUydz\neceOHS0Gm6xXrx5EJFfXS1Z++uknxMTEoGPHjhbxKKVQq1YtbN682VzWrFmzPJ8RwZoqVapg06ZN\niIuLw/bt27Fp0yarM4wEBgaifv36iIiIgI+PDzZs2IBx48ahWLFi6N+/P4Dsr0NTncwGn8zL69ik\nSJEiaN++PRYvXozKlSujTZs2OH/+PAYNGoRChQohOTkZCQkJ5tYW3377rcX23bp1Q/PmzTFlyhQM\nHDgQJUuWzNFxmzRpYjGocLVq1eDp6Znja6lr165wdb03KWZERARKlCiBjRs3YsCAAdizZw+OHTuG\nkSNHWlxLIoJGjRph6dKlFvtTSqF3794WZdeuXcPvv/+OqKioLD+9X716NTp06AC9Xm9xrOeffx5f\nf/01du/ejeeee85c3qNHD4vuXWnvoypVquTo/Wdl1apVEBG0a9fOIp6iRYuiQoUK2Lx5M4YNGwbA\neB7TdtnJjLWWPzmRkJBg0Y0rrfTXbEJCQqbXtohY1B0/fjxSUlIwfPjwHMXRuXNn1KxZE1evXsX6\n9etx+fLlHLVoySom0/q8wGQHERER5YnKlY2JiIdxnAdlerDMaXN9U3/utF04AKBYsWIoXLgwzpw5\nY1FubbYRb29v3Lx5M8cxli1b9oHjyEt16tRBrVq1LB6ETX+om5rsp5WYmJirP+Qf9JylPyeAcSaL\nFStWADBOYSoiGDlyZIZxAQDjQ9mVK1fMyQ4AGWaiuXr1Km7fvm3R5NuaY8eOWXSfsXactNKOBwLc\nm/EmN9dLVkzvvUGDBlbj8fLyMi8XK1YMxYoVy5PjZsXDwwMNGxqH7XvxxRfx5JNPonXr1tizZw+q\nVasGwDht6+uvv47jx4+bfy4vvfQS9Ho9hg0bhs6dO8Pb2zvb6xDI+qEyL6/jtD777DMkJiZiyJAh\nePvtt6GUQmRkJMqVK4c1a9Zk24UiKioKP/74I7Zs2WLukpCd9NcS8OD3Ufny5XH69GkAxmsJMI5n\nYo2maYiJibG4ptL/X2ZKvGR1H129ehW3bt3C3Llz8dlnn2VYn5P7yJRIysv7yGAwWD1HSimL5ENg\nYGCG/z/ykouLS6bjYqS/Zl1cXDK9tpVS5rqnT5/GpEmTMGfOHIuEV1b8/f3N571Dhw7o3bs3Gjdu\njKNHj2bZlSWrmEzr8wKTHURERJQnXF0fvMXFw+Lh4YGSJUti//79udrO2hR81qT9dDEtycVYC1n9\nsZfTOPKav7+/xVSIJUqUgIjg0qVLGepeunQpx59GA3lzzrJiGmT17bffRtOmTa3WSf8Qc79/cBsM\nBjRp0gT/+9//rMaffjrRh/HelVJYunSp1URG2mlZExMTM7R8yUxeJkVefvllvPrqq/jqq6/MyY45\nc+YgJCTEIgEFGFsaLV68GHv27EHDhg3N6zO7Dn18fLKcUjQvr+O0PD09sWbNGpw/fx6nT59GQEAA\n/P39UadOHfj5+WU7dazpIfLGjRs5PubDuo8mT56Mp556ymqd9Emc+7mPTMeJjIxE165drdZ58skn\nLZYfxnvXNA0//PADNC3jPB9p37dpHIvs6HQ6+Pr65jqWEiVKQK/X49q1axbbJycn4/r16xbXbIkS\nJTK9tgGY67733nsoXbo0wsLCzIlzU52rV6/izJkzKFOmTJa/fyIiIjB//nxs3boVTZo0yTL+ixcv\nZhvTg2Kyg4iIiAqkli1bYt68edi5c2e2g5QGBATAYDDg2LFj5oFCAeMAerdu3UJAQECuj38/CYv8\niCM3Tp48adFaoWrVqnBwcMCuXbsQERFhLk9OTsbevXvRoUMHc1l+J2iOHTuWoezo0aPmT1dNsy44\nOjqaWxTklukBNbskWbly5RAbG2u1JcX9epDzV65cOYgI/Pz8sn3vX3/9Nbp3756jePR6/X3HlN7d\nu3dhMBgsEi2XL1+Gj49Phrqm2SNMs9+ULFkSfn5+FjPYmPz111+oXr16lsfOzXV8P0qXLo3SpY0T\nA9y6dQvR0dEW3cQyY5pJJO09Z4v76Pjx4+bERrly5QBYtszJLdO9mNV95OfnBw8PD+j1+vs+jjV5\ncR8FBgZabd2R1qRJkzB69Ohs9xkYGHhf3dWqV68OEcGuXbvQrFkzc/nff/8Ng8Fgcc1Xr14df/zx\nR4Z9/Pnnn3B1dTUnX8+dO4fjx4+bfz4mSin07dsXSincvHkzyyRdQkICRCTbhGn16tWxZcsWxMbG\nWiSJ/vzzTyilsr1nc+qhTD1Lj47ElEQYtERbh0FERGRzQ4cOhaurK1577bUMzaEB44PGjBkzAAAt\nWrSAiGDatGkWdSZPngylFF544YVcH9/NzS3DVJvZyY84rLl27VqGso0bNyI6OhrNmzc3l3l6eqJx\n48ZYunSpebYSAFiyZAni4uIspt28n/ebG99++63FJ4V//fUXdu7ciRYtWgAwPjyFh4fjs88+w3//\n/Zdhe2vvOT2lFF566SWsW7cOu3fvzrRe+/btsWPHDvz0008Z1sXExNxXksA0Jaq1cS2y07RpU3h6\neuLDDz+0mB7XJO17N43Zkd3r559/znUcgPH9W4th3rx5UEqhZs2a5rKKFStiz5495q4TJsuWLYOm\naRaf7Ldt2xbr16+3mLLyl19+wdGjRzNM/3rkyBGcO3fOvJyb6/hBDR8+HHq93mLcnZs3b2aY3jkl\nJQUTJkyAk5OTRdLMdB3k1320ZMkSixYJK1aswKVLl8z3UWhoKMqVK4dJkyZZnCuTnNxHvr6+CAsL\nw+eff27xc0hL0zS0bdsWq1atsjqFdE6OY42bm1uOWy6l9/LLL0PTtEyTGGlb4HTt2jVH99GXX355\nX7E0bNgQPj4+mDNnjkX5nDlz4ObmZvG7ICIiApcvX8bq1avNZdeuXcPKlSvRqlUrc6uncePGYc2a\nNfj222/NL9PMOP/73/+wZs0a8/WX2fmfP38+NE2zmP3l+vXrOHLkiMU4HBEREUhJSTFPqQ4YZ4Ja\ntGgRnn322byZiQVs2VHgBI9ugwTPf1EotpGtQyEiIrKpoKAgLFu2DB07dkRwcDC6dOmCqlWrIikp\nCdu2bcPKlSvNn3A/+eST6Nq1K+bOnYubN2+ifv362LlzJ5YsWYKXX375vgYFrV69OnQ6HSZOnIhb\nt27ByckJjRo1yrJJ84PGER4ejq1bt2Z4sEqvdu3aePrpp1GjRg14eXkhOjoaCxcuREBAQIaB68aN\nG4c6deogLCwMr7/+Os6dO4cpU6agadOmFs2Y7+f95kb58uVRt25d9O3b1zz1rJ+fH4YMGWKu88kn\nn6BevXqoVq0aevXqhaCgIFy+fBk7duzAhQsXLKYDzazp+4cffoiff/7Z/H6Dg4Nx8eJFrFy5Etu2\nbYOnpyeGDBmCtWvXomXLlujWrRtCQ0MRFxeHf//9F6tXr8bp06ettljISmhoKEQEAwcORNOmTaHT\n6XLc4sDDwwNz5sxBly5dEBISgo4dO8LPzw9nz57Fhg0bULduXXNi737H7Lh9+zZmzJgBpRS2bdsG\nEcHMmTNRuHBhFC5c2DyY6JYtWzBo0CBERESgQoUKSEpKwtatW7FmzRrUrFkTr7zyinmfQ4YMwQ8/\n/IC6detiwIABKFKkCNatW4cff/wRvXr1shiQ9p133sHKlSsRHh6OwYMH486dO5g0aRKeeuopdOvW\nzSLW4OBghIeH49dffzWX5fQ6Boyfxmualu0n8hMnTsT+/ftRq1YtODg4YM2aNdi0aRPGjRtn8TC4\ndu1afPDBB4iIiEDZsmVx48YNLFu2DAcOHMD48eNRtGhRc13TdfDOO++gY8eOcHR0RKtWrfJsjAMf\nHx/UrVsX3bt3x3///Yfp06ejYsWKeO211wAYE37z589HixYt8MQTT6B79+4oVaoULly4gM2bN8PL\nywvfffddtseZMWMG6tWrh5CQELz++usoW7YsTp06hY0bN5rvwwkTJmDLli2oVasWevXqhSpVquDG\njRuIjo7Gr7/+el8Jj9DQUHzzzTd46623ULNmTbi7u6Nly5Y52jYoKAgffPAB3nnnHZw6dQovvfQS\nPDw8cPLkSXz77bfo3bs33nzzTQD3P2bH2bNn8cUXXwCAuaXSuHHjABhb9kVGRgIwDuQ5duxYDBgw\nAO3bt0fTpk2xdetWLFu2DB9++KHFdLYRERGYNm0aunfvjgMHDsDX1xezZ8+GwWDAqFGjzPXSDhBt\n4uXlBRFBzZo10apVK3P5uHHjsG3bNjRr1gxlypTBjRs3sGrVKuzatQuDBg2yaB0yc+ZMjBkzBlu2\nbEFYWBgA4xSz7dq1w/Dhw3H58mWUL18eixYtwpkzZ7Bw4cJcn7dMZTddy6P+AqeezRXXN2qIT9+X\n5cCRRFuHQkRENpbXU88+ro4fPy69e/eWoKAgcXZ2Fk9PT6ldu7bMnDlT7t69a66n1+tl7NixUq5c\nOXFycjJPz5d+SsuyZctKq1atMhwnPDxcGjZsaFG2YMECKV++vDg6OlpMyxoYGGh1H7mJw9rxatSo\nIaVKlcr2nIwcOVJCQkLE29tbnJycJDAwUAYMGCBXrlyxWn/btm1St25dcXV1lWLFismgQYMkNjY2\nQ73cvl9r7yE905SRkydPlqlTp0pAQIC4uLhIeHi47Nu3L0P9U6dOSbdu3aRkyZLi5OQk/v7+0qpV\nK1m9erW5jmnq1syu+3Pnzkm3bt2kWLFi4uLiIuXLl5dBgwZZTGkbFxcn7777rlSsWFGcnZ2laNGi\nUrduXZk6dap5eszTp0+LpmkyZcqUDMfQNE3GjBljXtbr9TJ48GApVqyY6HQ6iykp09dNP/WsyW+/\n/SbNmzcXb29vcXV1lQoVKkiPHj2ynEY3p0w/B03TMrzSTq954sQJ6datm5QvX17c3NzE1dVVqlWr\nJmPGjJH4+PgM+/3777/lhRdeMP+8KleuLBMmTBC9Xp+h7sGDB6VZs2bi7u4uPj4+0qVLF6vXrKZp\nVq+rnF7Hfn5+UqdOnWzPyYYNG+TZZ58VLy8vcXd3l9q1a1udwjM6Olpat24t/v7+5v+DwsLCMp3u\nc9y4ceLv7y8ODg4WP2dN02TQoEEZ6pctW1Z69OiRZaxbtmwRTdPk66+/lnfffVeKFy8ubm5u0qpV\nKzl37lyG+v/8849ERESIn5+fuLi4SNmyZaVjx46yefNmcx3T1K3Xr1+3esyDBw9K27ZtxcfHR1xd\nXSU4OFhGjRplUefq1asycOBACQgIECcnJylZsqQ0adJEFixYkCH29OfLdH+lnU42Li5OIiMjxcfH\nx+LatFZ31KhRotPpMsS9Zs0aCQsLEw8PD/Hw8JAqVarIoEGD5NixY1mc4ZwxTfVq7T6yNpX4/Pnz\nJTg4WJydnaVChQoyY8YMq/u9deuW9OrVS/z8/MTd3V0aNmyYo/s+s3O7adMmadWqlZQuXVqcnJzE\ny8tL6tWrJ0uWLMmwD9N1kH7a8bt378rQoUOlZMmS4uLiIrVq1cow3bo1uZl6VkkeDdhiK0qpEADR\n0dHRFhlSss4tqibKOIbg0EcZRzUmIqKCZffu3Qg1zhUbKiJW2+Tz96z9iI2NhY+PD2bMmIE+ffrY\nOpw8c+bMGZQtWxaTJk0yf6pKlF8OHjyIqlWrYuPGjRZjJTzufvvtNzRo0AArV67Eyy+/bOtwiDKV\nk79dTDhmBxEREVEBsHXrVpQuXdrcHJ2Icm/Lli2oXbu2XSU6iOwVkx1EREREBUCLFi1w8uRJi2lG\niSh3+vXrZ3VmCyJ69DDZUYAcOXcNSSkpuBpzB9di4m0dDhEREVGeUErl+5ScRPaO9xDZG6b2C4gm\nY8dhU9JYwBCE68nnUXRCabxVYR4+7tHW1qERERER3beAgID7msqViO6pX78+7yOyO2zZUQBEzf8G\nmwwj8IwMhrODK/zdg1DybkNMOt0R63cesnV4RERERERERHmKyY4C4PP9s+BzszF2jB0Pg9MNuDk7\n4eDYL6HdLYKR33JWFiIiIiIiIrIvTHYUAHGOpxDsWQthH/VFktspOIorPN2cUCQpBP8lnrZ1eERE\nRERERER5ismOAqBwchXsvbMJR/+7ANwpjvmvjMXF63dw1Xk7At2DbR0eERERERERUZ5isqMAiHru\nTcQV3onrd2LgFR+C307sRPDYFoCWgo879LZ1eERERERERER5ismOAuDdDk3Rs8hCGFQSYpKvYej+\nxkjQXcHMZ79H3aqBtg6PiIiIiIiIKE9x6tkCYv6Ablj7xgrcTRLMDd+DtnWfhIOOuS4iIiIiIiKy\nP3zaLUCUUnB1KoQO9asz0UFERPSQ6fV6aJqGDz/80NahPFJM5+XNN9+0dSiPpRMnTkDTNCxbtsxc\nNmLECDg6OtowKnrYTNfBjBkzbB3KY+mXX36BpmnYvn27uSwyMhIVKlSwYVT0oPjES0RERAVO69at\n4ebmhri4uEzrvPLKK3BycsLNmzfz7LhKKSil8mx/1nzyySf44osvHmgfI0eOhKZpGV6enp5W63/7\n7bcICQmBq6srAgMDMWbMGOj1eos627Ztw+jRoxEbG/tAsRVUX375JWbOnGl1Xfpr6mFcZ+nt3LkT\nffv2RWhoKAoVKoRChQpZrRcfH48ePXqgatWqKFy4MDw9PfH0009j1qxZSElJyVD/77//RosWLVC8\neHF4enqievXq+OSTT2AwGDLU/f3331GnTh24ubmhRIkSiIqKQkJCQo7fQ06u49yaPn06goOD4ezs\nDH9/fwwdOjRDTKaEX/qXTqfDlClTLOpu2LABY8eOfaCYCrKs/n98FO6jH3/80Xx/6HQ6VKxYMdO6\nIoLx48cjKCgILi4uqF69OlasWGG17vnz59GuXTt4e3vDy8sLbdq0wenTp7OM5fjx43BycoKmafj3\n338t1u3fvx/t2rVDUFAQ3Nzc4Ofnh/DwcHz//fc5fq+3bt3Ca6+9Bj8/P3h4eKBx48b4559/crx9\nTrAbCxERERU4r7zyCtavX481a9YgMjIyw/qEhASsXbsWLVq0gLe3d54cU6fTISEhId8/cZ81axb8\n/f3x6quvPtB+lFKYO3cuXFxczGXWYl+3bh3atm2Lxo0bo3///vjnn38wevRoXL9+HdOnTzfX++OP\nPzBmzBj06tUL7u7uDxRbQbR06VKcOHECAwcOtCgvV64cEhISMk0uPCzr16/HwoUL8dRTTyEoKAgn\nT560Wi8+Ph5HjhxBy5YtERgYCE3T8Mcff2Dw4MHYtWsXFi1aZK67c+dOhIWFITg4GMOHD4eLiws2\nbtyIgQMH4vTp0/j444/NdaOjo/H888/jySefxNSpU3H27FlMnjwZJ0+exHfffZdt/Dm9jnPjrbfe\nwtSpU9GxY0dERUVh//79mDp1Kg4dOoR169ZlqN+sWbMM/x+FhoZaLK9fvx4LFizAyJEj7yumgi6z\n/x8bNWr0SNxHS5cuxZo1axASEoKSJUtmWXfo0KGYPHky+vbti5CQEKxevRodOnSATqfDyy+/bK53\n584d1K9fH4mJieZE9uTJk9GgQQPs3bsXXl5eVvc/ePBgODk5WU1Cnj59GvHx8ejevTtKlSqF2NhY\nrFy5Ei+88AI+//xzdOvWLcvYDQYDmjVrhsOHD2PIkCHw8fHBrFmzEB4ejt27d6Ns2bLZn6ycEJHH\n+gUgBIBER0cLZa3oGy9I8Tda2zoMIiJ6RERHRwsAARAiBez3bEJCgnh6ekrz5s2trl+2bJlomiYr\nVqx44GMZDAZJTEx84P3kVOXKlaVJkyYPtI8RI0aIpmkSExOTbd2KFStKzZo1xWAwmMuGDRsmDg4O\ncvz4cXPZ+PHjRdM0uXDhgsX2KSkpopSSqKioB4o5v8XFxdn0+M2aNZMKFSrkqO6IESPE0dExnyOy\ndOXKFbl7966IiPTp0yfXx+/bt6/odDq5du2auax79+7i6uoqt2/ftqhbt25d8fX1tShr0qSJ+Pv7\nS3x8vLns008/FU3TZPPmzdkeP6fXcU6dO3dOHBwc5LXXXrMonzZtmmiaJj/88IO5LDf3QO/eva2e\n2+PHj4tSSqZPn57rWB+mtD8fW8jN/4+RkZE5vufyyqVLlyQlJUVEsr7nz549K46OjvLmm29alNeu\nXVvKli1rUTZu3DjR6XTyzz//mMsOHDggOp1O3n//fav7X7dunbi4uMjIkSNF0zSLbTNjMBikWrVq\nUq1atWzrfvnll6Jpmqxdu9ZcdvnyZfHy8pKuXbtmuW1O/nYxvdiNpYCY/cMmXCm8AcDDbYpFRET0\nKHJ2dsbLL7+MX375BdeuXcuwftmyZfDw8MCLL75oLps4cSLq1KmDIkWKwNXVFTVr1sS3335rsV3a\n8Se++OILPPHEE3B2dsYvv/xidcyO06dPo2/fvqhUqRJcXV3h6+uLjh074uzZsxb7nT9/PjRNw86d\nO/HGG2/Az88P7u7uiIiIwI0bN8z1/P39cfToUWzatMncFP7555+/7/NkMBhw586dTNfv27cPx44d\nQ+/evS2ae/fv3x96vR6rVq0CYOwW88477wAASpcubW6if/HiRYv9rV69GlWrVoWzszOqVauGTZs2\nZfQNjHgAACAASURBVBujqa/9qlWrMGzYMBQvXhzu7u5o06ZNhv0DwI4dO9C0aVN4eXnBzc0NDRo0\nwJ9//mlRZ8SIEdA0DUePHkWHDh3g7e2NBg0amNcfOnQI7dq1g5+fH1xdXREcHIz333/fYh8XLlxA\nt27dULx4cfP7Wbx4sdXY16xZg7Fjx6J06dJwdf1/e/cdH1WZ9n/8e00SCIReA1JFBCxItVNUiohr\nARQRFgSElbIqq67l56OPPMraFgERRVkVC1FWFFllQVFURJGlulhQyiJSRaVIEQjX74/JzM4kk0JJ\nBief9+t1Xjr33HPONXfOGWauc5fS6tixo9auXRuu16ZNG82ePVurVq0K/11D3dtjzdmRm8mTJ6tV\nq1YqXbq0KleurN69e8dsoyNRtWrVo7orXrduXbm7duzYES7btWuXSpUqpbJly0bVTU9Pj+pxtH37\nds2dO1f9+vWLKu/fv79KlSqlqVOn5nnsgp7Hh+PTTz/VoUOH1LNnz6jya665Ru6uV155Jebr9u3b\np19//TXmc7///e/19NNPRw17idXmEydOVIMGDVSqVCmdffbZWrp0ab7xhj5jPvnkEw0aNEiVK1dW\nhQoV1L9//6i/Scjbb7+tNm3aqEyZMipfvrwuu+wyff3111F1+vTpo4oVK2rVqlXq0qWLypUrp379\n+kW1UZcuXVSxYkWVKVMmPEQp0ldffaXu3buHP3fPPPNMzZw5M2bsR/P5GCqLnLMjFnfX6NGjw5/t\nNWrU0NChQ7Vz585827gg0tPTlZSUlG+9N954Q5mZmRoyZEhU+ZAhQ7Ru3TotXLgwXDZt2jSdffbZ\natq0abjslFNOUfv27WNeGwcOHNCf/vQn3XLLLapbt26BYzcz1apVS9u3b8+37rRp01SzZs2of2Or\nVaumHj16aPr06Uc9fCyEYSzFxMufzJWSpD+2uDXeoQAAcFzo3bu3Jk+erKlTp2ro0KHh8p9//lnv\nvPNOeM6OkHHjxql79+7q06eP9u/frylTpqh79+765z//mSOhMHv2bL3yyisaNmyYKlWqpDp16sSM\n4bPPPtO//vUv9e7dWyeccILWrl2rJ554QosXL9aKFSvCxw/9ABs6dKiqVKmikSNHas2aNRozZoxK\nlSoVHoM+fvx4DR06VJUrV9add94pd1eNGjWOqH3cXXXq1NEvv/wSTh48+uijqlq1arjO0qVLZWY5\nutrXqlVL6enp4R9ZV111lVatWqWpU6dq/PjxqlChgiSpUqVK4dd88MEH+vvf/66hQ4eqTJkyGjNm\njLp3767vvvsu127WkUaOHKnk5GTddddd2rRpk8aMGaNOnTppyZIl4R+E7777ri699FKdddZZGjly\npCTp2Wef1QUXXKBPPvlEzZs3j2rvbt26qXHjxnrooYfCx1m2bJnatWun1NRUDRkyRHXq1NGqVav0\n9ttv67777pMkbd68WWeeeaZKlCihG2+8UZUrV9bMmTPVv39/7d69O+p8k6T7779fKSkpuv322/XT\nTz/p4YcfVt++fTVv3jxJ0r333qtbb71VW7du1V//+le5e44EQH7uu+8+jRw5Utdee60GDRqkrVu3\nauzYsVq4cKGWLl0aHlq0d+9e7dmzJ9/9JSUlhf+OR2L//v3atWuX9uzZo4ULF+qxxx5TgwYNorqv\nt2/fXq+//rqGDBmim2++WampqXr77bc1Y8YMjRkzJlzv888/V2ZmZo7zsESJEmratGm+P/YLeh4f\njlDCIjL5IkmlS5eWFBx2k92kSZM0duxYubtOOeUU3XPPPbr66qvDzw8bNkybNm3Shx9+qBdeeCF4\n5zoQfe968uTJ2rNnj4YOHSp310MPPaTu3buHE2W5CZ3zQ4YMUeXKlTVy5Eh9/fXXmjBhgr7//nu9\n++674brPP/+8Bg4cqK5du+rhhx/W7t27NWHCBLVp00ZLly5VrVq1wvs8cOCAOnfurAsuuECjR49W\nWlqaJGnWrFm6/PLLVatWLd1yyy2qVq2avvzyS7399tsaNmyYpGASqk2bNqpbt67uvPNOlS5dWq++\n+qouu+wyTZ8+XZdeemlU7Efz+VjQ+TkGDBigjIwMDRgwQDfffLPWrFmjxx9/XMuXL9e8efPCbbx7\n927t27cv3/2VKFHisK9lKfg5VK5cOZ100klR5WeeeabcXUuXLtWZZ56pzMxMrVixIkdSJFT34Ycf\n1r59+5Samhouf/TRR7V7927deeedysjIyDOO0Pvcvn27pk+frnfeeUd9+/bNN/6lS5fmuN5CMT33\n3HNatWqVGjVqlO9+8pVf14/jfVOCdq891s79n7s8+Zb6+VcEABQbxXkYi7t7Zmam16xZ088777yo\n8lDX9zlz5kSVZx+KcuDAAT/llFP84osvDpeFuqOnpKT4t99+G1U/9NwDDzyQ6z7d3efPn+9m5q+8\n8kq4bNKkSW5mfskll0TVvfHGGz0lJSVqiMWxGMYyevRov+mmmzwjI8OnTZvmN910kycnJ3uTJk2i\njvXggw96IBDwzZs359hHixYtvG3btjnq5jaMpVSpUr5u3bpw+ZIlS9zMfOLEiXnGOmfOHDczr1ev\nXlQX+YyMDDczf/LJJ9092MW6QYMG/rvf/S7q9Xv27PF69ep5165dw2V33323m1nM7tTnnnuuV6xY\n0Tdu3JhrTP369fPatWv79u3bo8qvuuoqr1y5su/fvz8q9qZNm4a7rrsH2z8QCPjKlSvDZbl1aQ8N\nX3j55Zej4o8c6rB69WpPSkryRx99NOq1n3/+uScnJ/sjjzyS473nt+XVvb8gw1heeumlqP2dffbZ\n/uWXX0bVOXjwoA8dOtRTUlLC9UqUKOGTJk2KqvfKK694IBDwBQsW5DhOt27dvE6dOnnGcjjncUEt\nXLjQzcwfeuihqPK33nrLzcwrVaoULsvMzPQ2bdr4+PHj/a233vKJEyf66aef7maW473m1rah86B6\n9eq+a9eucPnrr7/ugUDAZ8+enWe8oc+Yc845xzMzM8PloeFn//znP93dfefOnV6+fHkfPnx41Os3\nb97s5cuX92HDhoXL+vTp44FAIMdQiYMHD3qdOnW8YcOGUbFm165dO2/ZsmXUteHuftZZZ/mpp56a\nI/aj+XycM2eOBwIBnz9/flT8kef53Llz3cz8tddei3rtzJkz3cyihj326dOnQNdRXp/VeQ1jufji\ni71x48Y5ynfu3Olm5vfcc4+7B/8uZuYPPvhgjrrjxo3zQCDga9asCZdt2LDBy5Yt688//7y7B9s2\nr2Es119/ffi9JCcn+zXXXJNj2FksqampfsMNN+QonzFjhgcCAX///fdzfe3hDGOhZwcAADgm9hzY\no6+3fZ1/xaPUuEpjlU4pfdT7CQQCuuaaazRmzBh999134d4XU6ZMUfXq1XXhhRdG1Y/s5bF9+3Yd\nPHhQ559/fo6hLFJwsrvsd9xiidzngQMHtGvXLp188skqW7aslixZEtUF3sz0hz/8Ier1bdq00fjx\n4/Xdd9+pcePGBXvjBTBixIiox926dVPLli3Vr18/PfXUU+FlYkOrSkS+j5DU1NTDWgnj4osvjuoB\n07x5c6WlpeU60WV21113XdRd9J49e+rmm2/WzJkzdcMNN2jRokVas2aN7r//fv3444/heu6uCy64\nIEd3bjPTDTfcEFW2ZcsWffrpp7rtttty7THj7nrjjTfUt29fHTx4MOpYnTp10rRp07Rs2TK1bt06\nXD5w4MCorutt2rSRu2vNmjV5rsZQUNOmTZOZqXv37lHx1KhRQyeeeKLmzp2rW28N9v4dMGBA1JCd\n3IR6KBypjh07as6cOdq+fbveffddffHFFzlW6klKSlKDBg3UpUsX9ezZUykpKZoyZYqGDh2q9PR0\nde3aVdLRn4fH8jwOad26tVq1aqVRo0YpPT1d7dq10xdffKGhQ4cqJSUlap+BQEAfffRR1Ov79++v\nZs2a6Y477lDfvn0LPLHxtddeGzUBcOS5lJ/QZ0xkD5Bhw4bp7rvv1syZM3XxxRdr1qxZ2rVrl665\n5pqocykpKUmtW7fW3Llzc+w3+3W0aNEirV+/Xk888USukxVv27ZNH330kR588MGoYRHurs6dO+v+\n++/XDz/8EO5pVhSfj6+99poqV66s9u3bR733Vq1aqVSpUpo7d6569OghSbrrrrvUv3//fPdZuXLl\nI4pl7969uZ6voecj/1uQupJ02223qVGjRlHDjfJy6623qlevXtq4caNeffVVZWZmat++ffn2Vvn1\n119zjcndj+iai4VkBwAAOCa+3va1Wj6ds1vqsbZ48GK1qNHimOyrd+/eeuyxxzRlyhTdcccd2rBh\ngz7++GPdfPPNObo0z5gxQ6NGjdLy5cujxtTHGjNfr169Ah1/7969euCBBzR58mRt3Lgx1JtGZhZz\nnHzt2rWjHodWijmWy+Pm5ve//71uueUWzZkzJ5zsCCUXYs0xsG/fvsNadSX7e5OkChUqFPi9ZU8u\nmZkaNGgQXl5x1apVkoI/BrMLdWHfvXt3uJu9pBwrAqxevVqSdOqpp+Yax+bNm7Vr1y5NmDAhx/wD\noWNt3bo1qqyw/66rVq1SZmamTjzxxJjxRC4pXL9+/WO3EkIeqlWrFk4oduvWTf/3f/+njh07atWq\nVapSpYqk4PCep59+Wt988034h9lVV12ltm3batiwYbrkkktkZvmeh9mHkmR3LM/jSG+88Yauvvpq\n9e/fP3iXOTlZf/7znzVr1qwc8/Jkl5KSomHDhumPf/xjeEhCQRztuZT9OipbtqyqV68edR25u9q0\naZPjtWYWNTRNCv7ITk9PjypbvXq1zCzP6+jbb7+VJN1555264447Yh5r69atUcPqCvs6+vbbb/Xj\njz9GHTN7PCFNmjRRkyZNjslxYylVqpS2bNmSozw0dCZ0Tud3bkfW+fjjj/Xqq6/mSLzlpVGjRuHh\nJn369FGHDh10xRVXaP78+Xm+rmTJkrnGFHlNHy2SHQAA4JhoXKWxFg/OOQ69MI5zrLRo0UKNGzdW\nRkaG7rjjjvAkj9l/EM+dO1dXXnmlLrzwQj311FNKT09XSkqKnnnmmZiTFxb0i9qQIUM0ZcoUjRgx\nQmeffbbKlSsnM1OPHj106NChHPVzm7gulCQpbLVr146a8C/Uu2HTpk2qXr16VN1NmzapXbt2Bd53\nYb+3UHuOGTNGp512Wsw62f9uR/KFO3Scfv36xVzWWJLOOOOMqMdF8d6Tk5M1a9asmM9H3oXdvXt3\njh4WsSQnJx/xXelYevTooXvvvVf/+Mc/wnfEn3zySXXq1ClqPgFJuuyyy3T77bdr/fr1qlOnjmrU\nqCF316ZNm3Lsd9OmTfku4Xksz+NIJ5xwgubPn69Vq1Zpy5YtOvnkk1W1alU988wzBeqxE/rxHnnN\n5acoziUzU0ZGRjgpFSl7D5Tsf7vDOY4k3X777erQoUPMOtmTckXx3mvWrKkXX3wx5j6rVasW/v+d\nO3cWqHdCyZIlj2jumxo1auSYWFlS+BoInfNVqlRRSkpKrteGmYXP/z//+c+64IILdMIJJ2jdunWS\nFJ7Ae8OGDapUqVJ4PpbcdO/eXcOHD9fatWvzTJrWqFEj15gi4z9aJDsAAMAxUTql9DHrcVGUevfu\nrXvuuUf//ve/lZGRoYYNG+aYOO31119XWlqaZs2aFfWFeuLEiUd17GnTpmngwIFRE2Du3bs3Zq+O\ngirIJHtHwt31n//8R+eee264rFmzZnJ3LVq0SM2aNQuXr1+/Xps3bw5P+FmYcYWE7gRHxrt69erw\nHfEGDRpIksqVK5djiFJBhfaxYsWKXOukp6crLS1Nhw4dOuLjxHI07degQYNwz478eh09+OCDeuCB\nB/Ld50knnaRvvvnmiGPKLvTDMPLc37p1a8xVGQ4cOCBJOnjwoCTp9NNPVyAQ0KJFi3TFFVeE6+3f\nv1/Lly/Pd8LEwzmPj8RJJ50U7jHx+eef64cffsgxSW0soZ5EkT0JiuI6Ou+888KPd+3apS1btoTP\nm9A1UK1aNbVv3/6IjtGgQQO5u1asWKG2bdvmWkcK9pw7nq6jefPm6fzzz893WNGwYcP08ssv57vP\nDh066J133jnsWJo1a6bJkydr1apVUb1xFixYIDMLn8dJSUk69dRTtWjRohz7+Oyzz9SwYcNwQmr9\n+vXauHFjjiSFmalr166qUqVKjl5p2cW6jnOL/1//+leO8gULFqhMmTIFGgZaECw9CwAAirXevXvL\n3XXPPfdo2bJlMe/GJyUlKRAIRP3wWrNmjf7xj38c1bGTkpJy9OAYM2bMUd2JTEtLK9DSf3mJHI8e\n8vjjj+vnn39Wly5dwmVNmzZVw4YN9fTTT0fFPGHCBCUlJalbt25RcUk66thyM3nyZO3evTv8+JVX\nXtHWrVt1ySWXSArO8l+vXj098sgjMVcbibUEcXbVq1fXueeeq0mTJmnDhg0x6yQlJenKK6/U1KlT\n9dVXX+V7nIL++Dqav2v37t1lZuHVYrKL7DkwYMAAzZkzJ9/thRdeOKJYYp1bkvTMM8/IzNSqVatw\nWcOGDTV79uyoH06ZmZmaOnWqKlSoEP5RVrFiRV144YV68cUXo/62zz33nPbt2xe1osnBgwe1cuXK\nqCEAh3MeH41Dhw7p9ttvV9myZTV48OBweaxzb+fOnRo7dqzS09OjEjBpaWnKzMws0Io5h8vdNXHi\nxKjPufHjx8vdw9dRly5dVKZMGT3wwAMxE1EFuY5at26tOnXq6LHHHst1ydb09HSdf/75evLJJ2P+\nwC7IcWI5muvo6quv1v79+3X//ffneO7gwYNR7+Wuu+4q0HX0yCOPHFEsV1xxhQKBgCZMmBBVPnHi\nRNWpU0dnnXVWuKxHjx5asGCBli9fHi778ssv9eGHH0ZdG88++6zeeOMNTZ8+PbyFVsYZM2ZMeFUb\nSfrhhx9yxLR//3698MILSktLi5ojZfPmzVq5cmXUv3U9evTQxo0bNWPGjHDZ1q1b9frrr+uKK64o\n0PK7BVEkPTvMbJikWyWlS1ou6Y/unjOVk/N150n6QNK/3f23d6sIAAAc9+rVq6dzzz1Xb775psws\n5pwOXbt21bhx49S5c2f16tVLmzZt0oQJE9SoUSN98cUXR3zsSy+9VM8995zKlCmjRo0a6ZNPPtGH\nH36YY9y7lHtX7OzlLVu21N/+9jeNGjVKDRo0CE+OKAWX0ixdunSed+QzMzNVt25d9ezZU6eddppS\nU1P14YcfaurUqWrdurWuv/76qPqPPPKIunXrps6dO6tnz55atmyZJkyYoKFDh0bdnWvZsqXcXXfe\neaeuuuoqpaSk6IorrlBy8rH5Olq+fHm1adNG1113nTZu3KixY8eqSZMmGjBggKTgJJCTJk3SpZde\nqtNOO03XXXedatasqQ0bNui9995T1apVYw5Jyu7xxx9Xu3bt1Lx5cw0ePFj16tXTmjVr9M4774Tv\nnj788MP66KOPdOaZZ2rQoEFq0qSJfvrpJy1atEjz5s3T5s2bw/sraGKrZcuWev3113XbbbepZcuW\nKleuXPgHaH4aNmyo++67T/fcc49Wr16tyy67TGXKlNGaNWv0xhtv6I9//KNuvPFGSUc+Z8e6dev0\n0ksvSZKWLFkidw/3EKlfv374unr++ef17LPP6vLLL1f9+vW1c+dOzZo1S++//766deum888/P7zP\nO+64Q/3791fr1q01ePBglSxZUi+//LI+//xzPfTQQ1GJolGjRqlNmzZq166dBg0apHXr1mn06NG6\n5JJLoiZcXbdunZo0aaLrr79eTz/9dLi8oOdxZmamUlJSCnRHfvjw4crMzNQZZ5yh/fv366WXXtLS\npUv10ksvRU1wO3bsWM2cOVNdu3ZVnTp1tGHDBj377LPauHGjMjIyon74hXqdDR8+XB06dFBKSoqu\nuuqqgv+h8rF371516NBBPXr00JdffqmnnnpK7du318UXXywpeJ2NHz9eAwYMUIsWLXTNNdeoSpUq\nWrdund5+++3wErN5Cf1Iv/LKK9WsWTP1799f6enp+vrrr7Vy5Uq99dZbkoLDmNq2bavTTjtNgwYN\nUv369bVlyxbNnz9fW7dujeqtcCw+H/O7Fi+88EINHDhQ999/v5YsWaIOHTooOTlZ33zzjV577TU9\n+eSTuuyyyyQd+Zwdy5cvD7//NWvW6Oeffw5fR82bNw9f83Xq1NGNN96osWPHat++fWrZsqWmTZum\nBQsW5Jhsefjw4Zo0aZK6dOmiW265RYFAQKNHj1atWrV08803h+t17NgxRzw//PCD3F3t27dX06ZN\nw+UDBw7U3r171bZtW9WsWVObNm3SSy+9pG+//Vbjxo2LGr506623asqUKfr+++/Dw1N69uypsWPH\nqm/fvrr11ltVsWJFTZgwQWame++997DbLVf5LddytJuknpL2SeorqbGkiZJ+klQln9eVl7RK0j8l\nLcmjXsIuiXesrNuy3XVXGU++pV68QwEAHEeK+9KzkSZMmOCBQMDPOeecXOtMmjTJTz75ZC9VqpSf\neuqp/uKLL+ZY4vPgwYMeCAT8T3/6U47Xh54bNWpUuGz79u3ev39/r1atmpcrV84vvfRSX7Vqldeu\nXdsHDx4cdexYy//FWi5x06ZN3rVrVy9XrpwHAoGopQ0rVqzo7dq1y7c9Bg0a5KeeeqqXL1/eU1NT\nvVGjRn733XdHLeEY6Y033vDmzZt7qVKlvE6dOn7fffdFLV8ZMnLkSK9Vq5YnJyeHl6HNq82yt0Ms\noTaYNm2a33HHHZ6enu5paWl+xRVX+Pfff5+j/tKlS71bt25epUoVL1WqlNevX9979erlH374YbjO\n3Xff7YFAwHfs2BHzmCtWrPArr7zSK1Wq5GlpaX7KKaf4yJEjo+ps3brVhw0b5nXr1vWSJUt6zZo1\nvVOnTuElHSNjf/PNN6Neu2rVKg8EAlHLye7atcuvvfZar1SpkgcCgfCSlLHq3n333V6iRIkccU+b\nNs3btGnjZcuW9bJly/opp5ziN910k69evTqvJi6Q0DK6gUAgxxZ5Di5cuNCvvvpqr1u3rqempnrZ\nsmW9devW/vjjj8c8Z2bNmuXt27f3qlWrempqqjdv3tz/9re/xYxh3rx5ft5553np0qU9PT3db775\n5qjliN3/216xzquCnMfbt2/PdVni7J599llv1qyZlylTxsuXL++dOnXyefPmxXyPnTp18ho1anjJ\nkiW9UqVKfskll/hHH32Uo25mZqYPHz7cq1Wr5klJSeHPn9D7GjduXFT9WJ87sYQ+Y+bPn++DBw/2\nSpUqefny5f26667LsYSye3AZ1s6dO3uFChU8LS3NTz75ZB84cKAvXbo0XKdPnz5RS+xm9/HHH3vH\njh29XLlyXrZsWW/evHmOpabXrFnjffv29Ro1anhqaqrXqVPHL7/8cp8+fXqO2I/m8zG3pWdPPvnk\nHHE//fTT3qpVK09LS/MKFSp4s2bN/K677vItW7bk+l4LKvReYm2DBg2Kqnvo0CEfNWqU16tXz1NT\nU71p06Y+derUmPtdv3699+jRwytUqODly5f3K6+80teuXVvgeLK3bUZGhnfs2NHT09O9RIkSXrly\nZe/cubPPnDkzxz769OnjSUlJOZYd//nnn33gwIFepUoVL1OmjHfo0MGXLVuWb0yHs/SseSFPaGVm\nCyR95u43ZT02SesljXP3h/N4XYakbyQdknS559Kzw8xaSFq8ePFitWhB549Y/vr6+7r13xfp9D3D\n9flDj8c7HADAcWLJkiWhu4Qt3X1JrDr8O5s4Pv/8czVr1kzvvPNOrhP+/Ra999576tixo6ZPnx6+\nqwoUlhkzZqh79+5asWJFeBWKRPC3v/1NgwcP1tKlS6Pu4APHm4J8dwkp1Dk7zCxFUktJ74XKPJhd\nmSPpnDxe119SfUmxBxbiiIy7dkS8QwAAAHHywQcfqG3btgmV6Agp7EkbgZAPPvhAffr0SahEB5Co\nCnvOjiqSkiRlXwR4i6SYnxBm1lDSKEnnu/sh/vECAAA4ejfeeGN4XoZEU9g9lYGQ/Oaj+C3jOkKi\nOa5WYzGzgKSXJd3r7qtDxXEMCQAAAMc5bo4BR4/rCImmsHt2bJOUKal6tvLqkjbnrK6yklpJamZm\nT2SVBRSc6mO/pE7u/kGsA40YMULly5ePKuvVq5d69ep15NEDAJAgMjIylJGREVUWuZwj8Ft10UUX\nxVwCE0DBDRw4UAMHDox3GMAxVajJDnc/YGaLJV0kaYYUnqD0IknjYrxkp6TTspUNk3SBpO6S/pPb\nsR577DEmTgMAIBexbgBETPIFAACQUAq7Z4ckjZb0fFbSY6GkEZJKS3pekszsL5Jqunu/rMlLv4x8\nsZltlbTP3b8qglgBAAAAAMBvXKEnO9x9qplVkTRSweEryyR1dvcfsqqkS6pd2HEAAAAAAIDioSh6\ndsjdJ0iakMtz/fN57X1iCdqjMm5uRnBdHAAAAAAAioEiSXYgvtanvS4dLKlWjWrGOxQAwG/UV18x\nmhQAAMTX4XwfIdlRLJg6l/xflUlNjXcgAIDfnm2BQGBfnz59+EcEAADEXSAQ2Hfo0KFt+dUj2QEA\nAHLl7t+ZWSMxIBIAABwHDh06tM3dv8uvHskOAACQp6wvFPl+qQAAADheBOIdAAAAAAAAwLFEsgMA\nAAAAACQUkh0AAAAAACChkOwAAAAAAAAJhWRHgrv3yeXyUj8qmaloAQAAAADFBMmOBDdn01RJ0p09\nO8U5EgAAAAAAigbJjmIgeVd9nVe/RbzDAAAAAACgSJDsAAAAAAAACYVkBwAAAAAASCgkOwAAAAAA\nQEIh2QEAAAAAABIKyQ4AAAAAAJBQSHYkuG++dUke7zAAAAAAACgyJDsS2Laf92tb478oKTnekQAA\nAAAAUHRIdiSwPQf2SpK61u4d50gAAAAAACg6JDuKgbqpZ8Q7BAAAAAAAigzJDgAAAAAAkFBI9jTj\nbwAAE7xJREFUdgAAAAAAgIRCsgMAAAAAACQUkh0AAAAAACChkOxIYN99F+8IAAAAAAAoeiQ7Etgj\n0+ZIkmqdYHGOBAAAAACAokOyI4FtDCyQJPVre0GcIwEAAAAAoOiQ7EhwKTtPVuXSleMdBgAAAAAA\nRYZkBwAAAAAASCgkOwAAAAAAQEIh2QEAAAAAABIKyY4E5ofiHQEAAAAAAEWPZEcCW/zVj/EOAQAA\nAACAIlckyQ4zG2Zma81sr5ktMLPWedS90szeMbOtZrbDzD4xs05FEWfCaf6c0socjHcUAAAAAAAU\nqUJPdphZT0l/lXSvpOaSlkuabWZVcnlJW0nvSOoiqYWkuZL+YWZnFHasCedQQBeU+0O8owAAAAAA\noEgVRc+OEZImuvsL7v61pBsk7ZE0IFZldx/h7o+6+2J3X+3u/0/St5J+VwSxJhYPKNXKxTsKAAAA\nAACKVKEmO8wsRVJLSe+FytzdJc2RdE4B92GSykr6qTBiBAAAAAAAiaWwe3ZUkZQkaUu28i2S0gu4\nj9skpUmaegzjAgAAAAAACSo53gHkxcyulfQ/ki5z923xjue3JDMz3hEAAAAAABAfhZ3s2CYpU1L1\nbOXVJW3O64Vmdo2kpyX1cPe5+R1oxIgRKl++fFRZr1691KtXr8MKOFG8+veDUtJBlS0b70gAAMeD\njIwMZWRkRJXt2LEjTtEAAAAUrkJNdrj7ATNbLOkiSTOk8BwcF0kal9vrzKyXpEmSerr7rIIc67HH\nHlOLFi2OPugEMe/H1yRJnc4t6GghAEAii3UDYMmSJWrZsmWcIgIAACg8RTGMZbSk57OSHgsVXJ2l\ntKTnJcnM/iKpprv3y3p8bdZzN0r6l5mFeoXsdfedRRBvQjjg+yRJXRteGudIAAAAAAAoWoWe7HD3\nqWZWRdJIBYevLJPU2d1/yKqSLql2xEsGKTip6RNZW8hk5bJcLQAAAAAAQEiRTFDq7hMkTcjluf7Z\nHl9QFDEBAAAAAIDEVNhLzyJOftkd7wgAAAAAAIgPkh0JatnyA5KkAH9hAAAAAEAxw0/hBLW1enB5\nwaSAxTkSAAAAAACKFsmOBGWepOq7L1JSICneoQAAAAAAUKRIdiSwEgerxDsEAAAAAACKHMkOAAAA\nAACQUEh2JKBDh6TtO+IdBQAAAAAA8UGyIwH98ovrUNVlKlMm3pEAAAAAAFD0kuMdAI69Xw78IqVt\nU92K1eIdCgAAAAAARY6eHQns5FLnxTsEAAAAAACKHMkOAAAAAACQUEh2JKCvvgr+1yy+cQAAAAAA\nEA8kOxLQsrXfSZIaN45zIAAAAAAAxAHJjgT0/a9fSJLOrNM8zpEAAAAAAFD0SHYksPS0GvEOAQAA\nAACAIkeyIwFt3RrvCAAAAAAAiB+SHQnou20/SpJKl4pzIAAAAAAAxAHJjgT0Y/IySVJaydQ4RwIA\nAAAAQNEj2ZGQTKW3t1RKUkq8AwEAAAAAoMiR7EhAv/wiSRbvMAAAAAAAiAuSHQlo+8FNSkqKdxQA\nAAAAAMQHyY4EtL/MapVLLR3vMAAAAAAAiAuSHQmqambzeIcAAAAAAEBckOxIMAcPSr/uP8QwFgAA\nAABAsUWyI8Hs3ntAqrxSVSuVjHcoAAAAAADEBcmOBHPg0AEpcEh1SzeJdygAAAAAAMQFyY4Es2VL\n8L8lk+nZAQAAAAAonkh2JJjla9dLkurVItkBAAAAACieSHYkmG17f5AkNazYOM6RAAAAAAAQHyQ7\nEszaDbslSVUq86cFAAAAABRP/CJOMFt/XSdJOr1O7ThHAgAAAABAfCTHOwAcWz9u/1WyJKWVSIt3\nKAAAAAAAxAXJjgSz7cB6WaB0vMMAAAAAACBuSHYkmJ/27FCJZIawAAAAAACKryKZs8PMhpnZWjPb\na2YLzKx1PvXbm9liM9tnZt+YWb+iiDMR7PJNKm2V4h0GAAAAAABxU+jJDjPrKemvku6V1FzSckmz\nzaxKLvXrSXpL0nuSzpA0VtIkM+tY2LEmgl9sk8oFqsc7DAAAAAAA4qYoenaMkDTR3V9w968l3SBp\nj6QBudQfImmNu//Z3Ve6+xOSXsvaD/Lxa8pmVSt1QrzDAAAAAAAgbgo12WFmKZJaKthLQ5Lk7i5p\njqRzcnnZ2VnPR5qdR31k2b5rv7zMBp1YqW68QwEAAAAAIG4Ku2dHFUlJkrZkK98iKT2X16TnUr+c\nmZU8tuElln8uWC2Z66wTT413KAAAAAAAxE3CrMYyYsQIlS9fPqqsV69e6tWrV5wiKnqrt25W0i91\ndG3bs+IdCgDgOJORkaGMjIyosh07dsQpGgAAgMJlwVElhbTz4DCWPZK6u/uMiPLnJZV39ytjvOZD\nSYvd/U8RZddJeszdK8ao30LS4sWLF6tFixbH/k0AAJCglixZopYtW0pSS3dfEu94AAAAjpVCHcbi\n7gckLZZ0UajMzCzr8Se5vOzTyPpZOmWVAwAAAAAA5KkoVmMZLWmQmfU1s8aSnpJUWtLzkmRmfzGz\nyRH1n5J0opk9ZGaNzGyopB5Z+wEAAAAAAMhToc/Z4e5TzayKpJGSqktaJqmzu/+QVSVdUu2I+v8x\ns66SHpN0o6TvJQ109+wrtAAAAAAAAORQJBOUuvsESRNyea5/jLKPFFyyFgAAAAAA4LAUxTAWAAAA\nAACAIkOyAwAAAAAAJBSSHQAAAAAAIKGQ7AAAAAAAAAmFZAcAAAAAAEgoJDsAAAAAAEBCIdkBAAAA\nAAASCskOAAAAAACQUEh2AAAAAACAhEKyAwAAAAAAJBSSHQAAAAAAIKGQ7AAAAAAAAAmFZAcAAAAA\nAEgoJDsAAAAAAEBCIdkBAAAAAAASCskOAAAAAACQUEh2AAAAAACAhEKyAwAAAAAAJBSSHQAAAAAA\nIKGQ7AAAAAAAAAmFZAcAAAAAAEgoJDsAAAAAAEBCIdkBAAAAAAASCskOAAAAAACQUEh2AAAAAACA\nhEKyAwAAAAAAJBSSHQAAAAAAIKGQ7AAAAAAAAAmFZAcAAAAAAEgoJDsAAAAAAEBCIdkBAAAAAAAS\nCskOAAAAAACQUEh2JJiMjIx4h3BcoB3+i7YIoh2CaIf/oi0AAAASV6ElO8ysopm9bGY7zOxnM5tk\nZml51E82s4fM7HMz+8XMNpjZZDOrUVgxJiK+vAfRDv9FWwTRDkG0w3/RFgAAAImrMHt2TJHURNJF\nkrpKaitpYh71S0tqJuk+Sc0lXSmpkaQ3CzFGAAAAAACQYJILY6dm1lhSZ0kt3X1pVtkfJb1tZre6\n++bsr3H3nVmvidzPcEmfmVktd/++MGIFAAAAAACJpbB6dpwj6edQoiPLHEku6azD2E+FrNdsP4ax\nAQAAAACABFYoPTskpUvaGlng7plm9lPWc/kys5KSHpQ0xd1/yaNqqiR99dVXRxhqYtmxY4eWLFkS\n7zDijnb4L9oiiHYIoh3+i7aI+rczNZ5xAAAAHGvm7gWvbPYXSbfnUcUVnKeju6S+7t4k2+u3SLrH\n3fOau0NmlizpdUk1JF2QV7LDzK6V9HLB3gEAAIiht7tPiXcQAAAAx8rh9ux4VNJz+dRZI2mzpGqR\nhWaWJKlS1nO5ykp0/F1SbUkX5tOrQ5JmS+ot6T+S9uVTFwAA/FeqpHoK/lsKAACQMA6rZ0eBdxqc\noPQLSa0iJijtJGmmpFqxJijNqhNKdJyoYI+On455cAAAAAAAIKEVSrJDksxspoK9O4ZIKiHpWUkL\n3f33EXW+lnS7u7+ZleiYpuDys5cqes6Pn9z9QKEECgAAAAAAEkphTVAqSddKGq/gKiyHJL0m6aZs\ndRpKKp/1/ycomOSQpGVZ/zUF5wG5QNJHhRgrAAAAAABIEIXWswMAAAAAACAeAvEOAAAAAAAA4Fgi\n2QEAAAAAABLKcZPsMLO7zGy+me02s5irsJhZbTN7O6vOZjN72MwC2eo0NbOPzGyvma0zs9ti7Ke9\nmS02s31m9o2Z9YtR5yoz+yprP8vNrMuxe7fHnpkNM7O1WfEuMLPW8Y7pcJhZGzObYWYbzOyQmV0W\no85IM9toZnvM7F0zOynb8yXN7Akz22Zmu8zsNTPLvgRyRTN72cx2mNnPZjbJzNKy1cn3PCsMZnan\nmS00s51mtsXM3jCzk2PUS+h2yDr2DVnX3Y6s7RMzuzhbnYRvh+zM7I6s62N0tvKEbwszuzfrvUdu\nX2ark/DtkHX8mmb2Ytb72JN1rbTIVqdYtAUAAEBujqcvJCmSpkp6MtaTWV+eZio4qerZkvpJuk7S\nyIg6ZSXNlrRWUgtJt0n6XzO7PqJOPUlvSXpP0hmSxkqaZGYdI+qcK2mKpGcUXB3mTUnTzeyUY/FG\njzUz6ynpr5LuldRc0nJJs82sSlwDOzxpCk5MO1TBSWmjmNntkoZLGizpTEm7FXyPJSKqjZHUVVJ3\nSW0l1VRwhZ9IUyQ1kXRRVt22kiZGHCff86wQtZH0uKSzJHVQ8Jp4x8xKRcRXHNpBktZLul3B67il\npPclvWlmTbLiKy7tEGbBBOZgBa/vyPLi1BYrJFWXlJ61nR8RX7FoBzOrIGm+pF8ldc6K9RZJP0fU\nKRZtAQAAkCd3P642Bb8s/RSjvIukA5KqRJT9QcEveMlZj4dI2hZ6nFX2F0lfRjx+SNLn2fadIWlm\nxONXJM3IVudTSRPi3T65tNkCSWMjHpuk7yX9Od6xHeH7OSTpsmxlGyWNiHhcTtJeSVdHPP5V0pUR\ndRpl7evMrMdNsh43j6jTWdJBSekFPc+KsB2qZMV7fnFuh4jj/yipf3FsB0llJK2UdKGkuZJGF7dz\nQsFk7pI8ni8u7fCgpA/zqVMs2oKNjY2NjY2NLa/teOrZkZ+zJf3b3bdFlM1WcOnaUyPqfOTuB7PV\naWRm5SPqzMm279mSzol4fE4B6hwXzCxFwTvf74XK3N0VjP+4i/dImFl9Be/iRr7HnZI+03/fYysF\n7y5G1lkp6buIOmdL+tndl0bsfo6CPUnOiqiT33lWVCpkxfaTVHzbwcwCZnaNpNKSPimm7fCEpH+4\n+/uRhcWwLRpacKjbajN7ycxqS8WuHX4naZGZTbXgcLcl2XovFqe2AAAAyNVvKdmRLmlLtrItEc8d\nbZ1yZlYynzrpOv5UkZSk3068RyJdwS/Yeb3H6pL2Z32pz61OuqStkU+6e6aCyYTDOYcKnZmZgt3M\nP3b30LwExaodzOw0M9ul4B3oCQrehV6p4tcO1yg4nO7OGE8Xp7ZYoOAQic6SbpBUX9JHWXNIFKd2\nOFHBXowrJXVScOjnODP7fUQMxaUtAAAAcpVcmDs3s78oOO4+Ny6pibt/U5hxKDisA/gtmSDpFEnn\nxTuQOPpawXl1ykvqIekFM2sb35CKlpnVUjDp1cHdD8Q7nnhy99kRD1eY2UJJ6yRdreC5UlwEJC10\n9//JerzczE5TMAH0YvzCAgAAOL4Uds+ORyU1zmNrImlNAfe1WcG7UZGqRzyXVx0vQJ2d7v5rPnU2\n6/izTVKmfjvxHonNCias8nqPmyWVMLNy+dTJvtpAkqRKyv/8kIqoPc1svKRLJLV3900RTxWrdnD3\ng+6+xt2Xuvv/U3BizptUvNqhpaSqkpaY2QEzOyCpnaSbzGy/gnfRi0tbRHH3HZK+kXSSitc5sUnS\nV9nKvpJUJyKG4tIWAAAAuSrUZIe7/+ju3+SzHcx/T5KCE4Senm2FkU6Sdkj6MqJO26wvZJF1VmZ9\nMQ7VuSjbvjtllSuPOh2z1TkuZN3tXayIeLOGQFwk6ZN4xXUsuftaBb84R77HcgqOGw+9x8UKTpwX\nWaeRgj8AQn+3TyVVMLPmEbu/SMEfBp9F1MnvPCs0WYmOyyVd4O7fRT5XnNohFwFJJYtZO8yRdLqC\nw1jOyNoWSXpJ0hnuvkbFpy2imFkZBRMdG4vZOTFfwclEIzVSsJcLnxMAAAAh8Z4hNbRJqq3gF/l7\nFPyiFPpin5b1fEDBO7v/lNRUwXHbWyT9X8Q+yik4C/1kBYcA9JT0i6SBEXXqSdql4KosjRRc6nS/\ngt3EQ3XOUXCegD9l1flfSfsknRLvdsql7a6WtEdSXwV7zExUcOWKqvGO7TDeQ1rW37uZgisA3Jz1\nuHbW83/Oek+/U/DH33RJ30oqEbGPCQouO9xewTvi8yXNy3acmQr+WGyt4BCRlZJejHg+3/OsENtg\ngoKrGLRR8O5oaEuNqJPw7ZB1/FFZ7VBX0mkKrqp0UNKFxakdcmmb7KuxFIu2kPSIgkuf1pV0rqR3\ns45fuZi1QysF/326U1IDSdcq+G/aNcXtnGBjY2NjY2Njy2uLewDhQKTnFByOkX1rG1GntqS3FExg\nbFEwYRHItp/TJH2o4I//7yTdGuNYbRW8s7U36wvg72PU6a7gOPC9kj6X1DnebZRP+w2V9J+seD+V\n1CreMR1m/O0UTHJk//s/G1HnfxVMZu1RcMb/k7Lto6SkxxUc2rNL0t8lVctWp4KCd8V3KJhYeEZS\n6Wx18j3PCqkNYr3/TEl9s9VL6HbIOvYkBYe47VXwLvU7ykp0FKd2yKVt3ldEsqO4tIWCS4R/n3VO\nfCdpiqT6xa0dso5/iYL/Lu2R9IWkATHqFIu2YGNjY2NjY2PLbTN3FwAAAAAAQKL4LS09CwAAAAAA\nkC+SHQAAAAAAIKGQ7AAAAAAAAAmFZAcAAAAAAEgoJDsAAAAAAEBCIdkBAAAAAAASCskOAAAAAACQ\nUEh2AAAAAACAhEKyAwAAAAAAJBSSHQAAAAAAIKGQ7AAAAAAAAAnl/wPwMhd6kBQ2XAAAAABJRU5E\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa47d1b4450>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "experiment_probes = {}\n",
    "experiment_probes[\"payload/processes/content/histograms\"] =  [\"TIME_TO_NON_BLANK_PAINT_NETOPT_MS\"]\n",
    "probe_names = list(itertools.chain(*experiment_probes.values()))\n",
    "\n",
    "df = exptPings(EXPERIMENT_SLUG, 0.1, \"20180220\", \"20180305\")\n",
    "\n",
    "for p in probe_names:\n",
    "    chart_pref(p, True, 1.0)\n",
    "    chart_ecdfs(p, 1.0, 50, 95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fetching 50628.56669MB in 517 files...\n",
      "len(branch_vals) = 3719787, len(control_vals) = 3890581\n",
      "TIME_TO_DOM_CONTENT_LOADED_START_ACTIVE_NETOPT_MS (logscale=True):\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlsAAAGUCAYAAAACk+eNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+Ng3/fwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa29a1ab310>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDsAAAFkCAYAAAAuUzCnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa447331ed0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "experiment_probes = {}\n",
    "experiment_probes[\"payload/processes/content/histograms\"] =  [\"TIME_TO_DOM_CONTENT_LOADED_START_ACTIVE_NETOPT_MS\"]\n",
    "probe_names = list(itertools.chain(*experiment_probes.values()))\n",
    "\n",
    "df = exptPings(EXPERIMENT_SLUG, 0.1, \"20180220\", \"20180305\")\n",
    "\n",
    "for p in probe_names:\n",
    "    chart_pref(p, True, 1.0)\n",
    "    chart_ecdfs(p, 1.0, 50, 95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fetching 50628.56669MB in 517 files...\n",
      "len(branch_vals) = 3592911, len(control_vals) = 3763422\n",
      "TIME_TO_LOAD_EVENT_START_ACTIVE_NETOPT_MS (logscale=True):\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmIAAAGUCAYAAACWdU5/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzt3X+UZVV95/33B1AUlUZD7MZoBvwJSQxaJSoxIyZECLAS\nnYTEFBoIogZFZdr4wPjECQQfB5UIgkpgZJQwQiUEk5igTkdIFDWCSZXiaBpRgSBCt4JYJPw03d/n\nj3Oa3C7qVld131Onq3m/1rqruft897777HVsv73PPvukqpAkSdLS26nvDkiSJD1cmYhJkiT1xERM\nkiSpJyZikiRJPTERkyRJ6omJmCRJUk9MxCRJknpiIiZJktQTEzFJkqSemIhJ0hJL8pkkf9d3PyT1\nz0RM0nYtyVOTnJ/k20nuTTKT5PNJ3pzkUR395n5JTknyk120D/huOUkA7NJ3ByRpmCRHAJcC9wEX\nAV8DHgn8PPAe4KeA4zv46Z8CTgH+Hri5g/YlCTARk7SdSrI3MAncCPxiVX1v4PAfJ/nvwBFd/TyL\nmLVK8qiquq+jvkjagXlrUtL26mTgMcBxs5IwAKrqhqp6P0CSnZP89yTfSnJfkhuTvDPJIwfrJLkp\nyV8neVGSa9pbnd9O8tsDMcfQzMIBfCbJxiQbkrx4VhuHJPnHJPcCr1tMPyRpk1S5VEHS9ifJd4D7\nquoZC4i9EDiaJoH6DPAC4BjgL6vq1wfibqS5zbkC+F/ArcCrgecCz66qtUn2AU4E3gS8E7iurf7p\nqvp+28aPgCcA5wM3Ad+oqqsW0Y+/B6qqfnGx4yJpx+KtSUnbnSSPA34C+KsFxP4sTfLzP6tq03qx\n85J8H/i9JAdV1WcHqjwT+M9V9Q9t/T8HvgMcC5xUVTcm+RxNInZFVV01x88+DTi0qq7Yhn5Ikrcm\nJW2Xdm///NcFxB5Os57rrFnl76VZ6zV7Hdk/b0rCAKrqduAbwFMX0b8bB5OwreyHJJmISdou3dX+\n+bgFxP4nYCPwrcHCqloP/LA9PmiupyDvBB6/iP7dOIJ+SJKJmKTtT1X9K836rZ9ZTLUFxm0YUp5F\n/Na9I+iHJJmISdpuXQ48LckLthD3LzR/l222qD/JE4E92uOLtTXJVBf9kLSDMxGTtL16D3APcEGb\nzGwmydOSvBn4JM1s1n+dFfJ7NAnVJ7bit+9u29xjEXW66IekHZxPTUraLlXVDUmOAv4UWJtkcGf9\nFwFHAh+uqnOS/AnwuiSPBz5Ls23E0cBfbOWTil+huYV5cpI9gPuBK9uF/cP6+9UO+iFpB2ciJmm7\nVVV/024L8f8Av0rzOqMHaBKytwL/sw09Dvg28DvAy4F1NHuAnTa7SYbfdnywvKrWJ/ld4G3ABcDO\nwC8AV82OnWWh/ZivDUkPI27oKkmS1JMlWSOW5IT2VR/3Jrk6yQFbiH9Jkqn2FSHXt68cmR2zIskH\nk9zaxl2X5Je7OwtJkqTR6jwRS/IKmg0NT6F5jci1wJokew6J35vmaakrgf2Bs2kW6750IOYRwBXA\nTwK/RrNT9muB73Z1HpIkSaPW+a3JJFcD11TVie330LxO5Jyqes8c8e8GDquqnx0omwRWVNXh7ffj\naZ5E2reqhu0JJEmStF3rdEasnbkap5ndApq33NLMZh04pNoL2+OD1syK/xXgi8C5SdYl+b9J3pbE\n7TgkSdKy0XXisifN00brZ5WvB1YNqbNqSPzuSXZtvz8V+A2a/h9G80TS7wG/P4I+S5IkLYnlun3F\nTjTJ2evaGbYvJ3kyzePs75gdnOTHgEOBm4D7lrCfkiRpx/AoYG9gTVXdMapGu07EbqfZFHHlrPKV\nNPvrzGXdkPi7qur+9vttwAO1+QK3tcCqJLtU1b/Pqn8ocPFiOy9JkjTLK4FLRtVYp4lYVf0oyRRw\nMPDX8OBi/YOBc4ZU+yLN7cZBh7Tlm3wBmJgV8yzgtjmSMGhmwvjoRz/Kfvvtt5hT0CKsXr2as846\nq+9u7LAc3+45xt1yfLvnGHdn7dq1vOpVr4I2pxiVpbg1eSZwYZuQfQlYDewGXAiQ5HTgSVW1aa+w\n84AT2qcnP0yTtB0JHD7Q5h+3MecA76fZvuJtwPuG9OE+gP3224+xsbHRnZk2s2LFCse3Q45v9xzj\nbjm+3XOMl8RIlzh1nohV1aXtnmGn0dxi/ApwaFV9vw1ZBTxlIP6mJEcAZwFvBm4BjquqKwZibkly\naBtzLc3+YWfRvCRYkiRpWViSxfpVdS5w7pBjx85RdhXNthfztXkN8HMj6aAkSVIP3HdLkiSpJyZi\nGpmJidnPT2iUHN/uOcbdcny75xgvP52/4mh7kGQMmJqamnIRoyRJWrTp6WnGx8cBxqtqelTtOiMm\nSZLUExMxSZKknpiISZIk9cRETJIkqScmYpIkST0xEZMkSeqJiZgkSVJPTMQkSZJ6YiImSZLUExMx\nSZKknpiISZIk9cRETJIkqSdLkoglOSHJjUnuTXJ1kgO2EP+SJFNJ7ktyfZJj5on9rSQbk/zF6Hsu\nSZLUnc4TsSSvAN4LnAI8F7gWWJNkzyHxewOXA1cC+wNnAxckeemQ2DOAq0bfc0mSpG4txYzYauD8\nqrqoqq4DjgfuAV49JP71wA1VdVJVfaOqPghc1rbzoCQ7AR8F/gC4sbPeS5IkdaTTRCzJI4Bxmtkt\nAKqqgCuAA4dUe2F7fNCaOeJPAdZX1UdG01tJkqSltUvH7e8J7Aysn1W+HnjWkDqrhsTvnmTXqro/\nyc8Dx9LcupQkSVqWlt1Tk0keC1wEvLaq7uy7P5IkSVur6xmx24ENwMpZ5SuBdUPqrBsSf1c7G7Yv\n8J+Av0mS9vhOAEkeAJ5VVXOuGVu9ejUrVqzYrGxiYoKJiYkFno4kSdrRTU5OMjk5uVnZzMxMJ7+V\nZslWd5JcDVxTVSe23wPcDJxTVWfMEf8u4LCq2n+g7BJgj6o6PMmuwNNmVXsn8FjgzcA3q+rfZ7U5\nBkxNTU0xNjY2wrOTJEkPB9PT04yPjwOMV9X0qNrtekYM4EzgwiRTwJdonn7cDbgQIMnpwJOqatNe\nYecBJyR5N/Bh4GDgSOBwgKq6H/jnwR9I8sPmUK3t/GwkSZJGpPNErKoubfcMO43mFuNXgEOr6vtt\nyCrgKQPxNyU5AjiLZobrFuC4qpr9JKUkSdKythQzYlTVucC5Q44dO0fZVTTbXiy0/Ye0IUmStL1b\ndk9NSpIk7ShMxCRJknpiIiZJktQTEzFJkqSemIhJkiT1xERMkiSpJyZikiRJPTERkyRJ6omJmCRJ\nUk9MxCRJknpiIiZJktQTEzFJkqSemIhJkiT1xERMkiSpJyZikiRJPVmSRCzJCUluTHJvkquTHLCF\n+JckmUpyX5Lrkxwz6/hrklyV5Aft59NbalOSJGl703kiluQVwHuBU4DnAtcCa5LsOSR+b+By4Epg\nf+Bs4IIkLx0IOwi4BHgJ8ELgO8DfJtmrk5OQJEnqwFLMiK0Gzq+qi6rqOuB44B7g1UPiXw/cUFUn\nVdU3quqDwGVtOwBU1W9X1XlV9dWquh54Dc25HNzpmUiSJI1Qp4lYkkcA4zSzWwBUVQFXAAcOqfbC\n9vigNfPEAzwGeATwg63urCRJ0hLrekZsT2BnYP2s8vXAqiF1Vg2J3z3JrkPqvBv4Lg9N4CRJkrZb\nu/TdgW2V5L8BvwkcVFUP9N0fSZKkheo6Ebsd2ACsnFW+Elg3pM66IfF3VdX9g4VJ3gqcBBxcVV/f\nUmdWr17NihUrNiubmJhgYmJiS1UlSdLDxOTkJJOTk5uVzczMdPJbaZZsdSfJ1cA1VXVi+z3AzcA5\nVXXGHPHvAg6rqv0Hyi4B9qiqwwfKTgLeBhxSVf+4hT6MAVNTU1OMjY2N4rQkSdLDyPT0NOPj4wDj\nVTU9qnaX4qnJM4HXJjk6yb7AecBuwIUASU5P8icD8ecBT03y7iTPSvIG4Mi2Hdo6JwOn0Tx5eXOS\nle3nMUtwPpIkSSPR+Rqxqrq03TPsNJpbjF8BDq2q77chq4CnDMTflOQI4CzgzcAtwHFVNbgQ/3ia\npyQvm/Vzf9j+jiRJ0nZvSRbrV9W5wLlDjh07R9lVNNteDGtvn9H1TpIkqR++a1KSJKknJmKSJEk9\nMRGTJEnqiYmYJElST0zEJEmSemIiJkmS1BMTMUmSpJ6YiEmSJPXEREySJKknJmKSJEk9MRGTJEnq\niYmYJElST0zEJEmSemIiJkmS1BMTMUmSpJ6YiEmSJPVkSRKxJCckuTHJvUmuTnLAFuJfkmQqyX1J\nrk9yzBwxv5FkbdvmtUkO6+4MJEmSRq/zRCzJK4D3AqcAzwWuBdYk2XNI/N7A5cCVwP7A2cAFSV46\nEPNzwCXAh4DnAB8H/irJT3V2IpIkSSO2FDNiq4Hzq+qiqroOOB64B3j1kPjXAzdU1UlV9Y2q+iBw\nWdvOJm8GPlVVZ7YxfwBMA2/s7jQkSZJGq9NELMkjgHGa2S0AqqqAK4ADh1R7YXt80JpZ8QcuIEaS\nJGm7tkvH7e8J7Aysn1W+HnjWkDqrhsTvnmTXqrp/nphV29bd5e2OO+7gu9/9bt/d2Cp77703u+++\ne9/dkCRpSXWdiGkJXXrppbzhDW/ouxtb5cILL+TZz352393YKjvttBMbN27suxtbZbn2fbn2G+x7\nX5Zr35drvwH22msv9tprr767sd3rOhG7HdgArJxVvhJYN6TOuiHxd7WzYfPFDGsTgNWrV7NixYrN\nyiYmJpiYmJiv2rLx67/+6zzvec/ruxtb5WMf+xi/8zu/03c3tspBBx3EZz/72b67sVWWa9+Xa7/B\nvvdlufZ9ufYb4JRTTuHUU0/tuxtbZXJyksnJyc3KZmZmOvmtNEu2upPkauCaqjqx/R7gZuCcqjpj\njvh3AYdV1f4DZZcAe1TV4e33PwUeXVUvG4j5AnBtVT1kSijJGDA1NTXF2NjYaE9QI3Hbbbdx2223\n9d2NrbKc/8W6XPu+XPsN9r0vy7Xvy7XfsOPNiE1PTzM+Pg4wXlXTo2p3KW5NnglcmGQK+BLN04+7\nARcCJDkdeFJVbdor7DzghCTvBj4MHAwcCRw+0ObZwGeSvAX4BDBB81DAazs/G3ViR/sfrCRJC9F5\nIlZVl7Z7hp1Gc/vwK8ChVfX9NmQV8JSB+JuSHAGcRbNNxS3AcVV1xUDMF5McBbyz/XwTeFlV/XPX\n5yNJkjQqS7JYv6rOBc4dcuzYOcquopnhmq/NjwEfG0kHJUmSeuC7JiVJknpiIiZJktQTEzFJkqSe\nmIhJkiT1xERMkiSpJyZikiRJPTERkyRJ6omJmCRJUk9MxCRJknpiIiZJktQTEzFJkqSemIhJkiT1\nxERMkiSpJyZikiRJPTERkyRJ6klniViSxye5OMlMkjuTXJDkMQuod1qSW5Pck+TTSZ4+q81zklzX\nHv+XJGcn2b2r85AkSepKlzNilwD7AQcDRwAvBs6fr0KSk4E3Aq8Dng/cDaxJ8sg25EnAXsBbgJ8G\njgF+Gbigg/5LkiR1apcuGk2yL3AoMF5VX27L3gR8Islbq2rdkKonAu+oqsvbOkcD64GXA5dW1deB\n3xiIvzHJ7wP/O8lOVbWxi/ORJEnqQlczYgcCd25KwlpXAAW8YK4KSfYBVgFXbiqrqruAa9r2htkD\nuMskTJIkLTddJWKrgO8NFlTVBuAH7bFhdYpmBmzQ+mF1kuwJvJ0t3PKUJEnaHi0qEUtyepKN83w2\nJHlmV52d1ZfHAZ8Avgb84VL8piRJ0igtdo3YHwEf2ULMDcA64ImDhUl2Bp7QHpvLOiDASjafFVsJ\nDN7iJMljgTXAD4Ffa2fbtmj16tWsWLFis7KJiQkmJiYWUl2SJD0MTE5OMjk5uVnZzMxMJ7+Vqhp9\no81i/a8DzxtYrH8I8EngycMW6ye5FTijqs5qv+9Ok5QdXVV/3pY9jiYJuxc4vKruX0B/xoCpqakp\nxsbGtvn8JEnSw8v09DTj4+PQPIg4Pap2O1kjVlXX0SRLH0pyQJIXAe8HJgeTsHY/sJcNVH0f8PYk\nv5Lk2cBFwC3Ax9v4xwGfBnYDXgPskWRl+3FzWkmStKx0sn1F6yjgAzRPS24ELqPZnmLQM4AH7xVW\n1XuS7Eaz+H4P4HPAYVX1QBsyBhzQ/ve32j9Ds8h/H+Dm0Z+GJElSNzpLxKrqh8CrthCz8xxlpwKn\nDon/LPCQOpIkScuRt/MkSZJ6YiImSZLUExMxSZKknpiISZIk9cRETJIkqScmYpIkST0xEZMkSeqJ\niZgkSVJPTMQkSZJ6YiImSZLUExMxSZKknpiISZIk9cRETJIkqScmYpIkST0xEZMkSepJZ4lYkscn\nuTjJTJI7k1yQ5DELqHdakluT3JPk00mePk/sp5JsTPKro+29JElS97qcEbsE2A84GDgCeDFw/nwV\nkpwMvBF4HfB84G5gTZJHzhG7GtgA1Gi7LUmStDQ6ScSS7AscChxXVf9UVf8AvAn4rSSr5ql6IvCO\nqrq8qr4GHA08CXj5rPafA6wGXg2ki3OQJEnqWlczYgcCd1bVlwfKrqCZvXrBXBWS7AOsAq7cVFZV\ndwHXtO1tins0cDHwhqr63ui7LkmStDS6SsRWAZslSVW1AfhBe2xYnQLWzypfP6vOWcDnq+ry0XRV\nkiSpH4tKxJKc3i6OH/bZkOSZXXW2XZT/izS3JSVJkpa1XRYZ/0fAR7YQcwOwDnjiYGGSnYEntMfm\nso5mvddKNp8VWwlsusX5C8BTgZlks6Vhf5Hkqqr6xfk6tnr1alasWLFZ2cTEBBMTE/NVkyRJDyOT\nk5NMTk5uVjYzM9PJb6Vq9A8dtov1vw48b9M6sSSHAJ8EnlxVcyZjSW4Fzqiqs9rvu9MkZUdX1Z8n\neSKw56xqX6N5EODyqvqXIe2OAVNTU1OMjY1t+wlKkqSHlenpacbHxwHGq2p6VO0udkZsQarquiRr\ngA8leT3wSOD9wORgEpbkOuDkqvp4W/Q+4O1JvgXcBLwDuAX4eNvu95i19qydGfvOsCRMkiRpe9VJ\nItY6CvgAzdOSG4HLaLanGPQM4MF7hVX1niS70ew3tgfwOeCwqnpgnt9xHzFJkrQsdZaIVdUPgVdt\nIWbnOcpOBU5dxO88pA1JkqTlwHdNSpIk9cRETJIkqScmYpIkST0xEZMkSeqJiZgkSVJPTMQkSZJ6\nYiImSZLUExMxSZKknpiISZIk9cRETJIkqScmYpIkST0xEZMkSeqJiZgkSVJPTMQkSZJ6YiImSZLU\nk84SsSSPT3Jxkpkkdya5IMljFlDvtCS3JrknyaeTPH2OmAOTXJnk39r2P5Nk127ORAs1OTnZdxd2\naI5v9xzjbjm+3XOMl58uZ8QuAfYDDgaOAF4MnD9fhSQnA28EXgc8H7gbWJPkkQMxBwKfAv4P8Lz2\n8wFg4+hPQYvhXwDdcny75xh3y/HtnmO8/OzSRaNJ9gUOBcar6stt2ZuATyR5a1WtG1L1ROAdVXV5\nW+doYD3wcuDSNuZM4H1VdcZAvW92cBqSJEmd6mpG7EDgzk1JWOsKoIAXzFUhyT7AKuDKTWVVdRdw\nTdseSX68rX97ki8kWdfelnxRN6chSZLUna4SsVXA9wYLqmoD8IP22LA6RTMDNmj9QJ2ntn+eQnOb\n81BgGrgyydO2vduSJElLZ1G3JpOcDpw8T0jRrAvryqbE8byquqj977ckORh4NfD7Q+o9CmDt2rUd\ndk0zMzNMT0/33Y0dluPbPce4W45v9xzj7gzkEI8aZbupqoUHJz8G/NgWwm4Afhv4o6p6MDbJzsB9\nwJFV9fE52t4H+DbwnKr66kD5Z4AvV9XqJHu37b+qqi4ZiPlT4EdV9dtD+n0UcPFCzlGSJGkerxzM\nQbbVombEquoO4I4txSX5IrBHkucOrBM7GAjNmq+52r4xybo27qttO7vTrAn7YBtzU5JbgWfNqv5M\n4JPzdGkN8ErgJppkUJIkaTEeBexNk1OMzKJmxBbVcPJJ4InA64FHAh8GvjQ4a5XkOuDkTTNkSU6i\nufX5OzRJ0zuAnwZ+uqoeaGNOBE4FXgN8pY19C/AzVXVjJycjSZLUgU62r2gdRbO/1xU0e3xdRrM9\nxaBnACs2famq9yTZjWYh/h7A54DDNiVhbczZ7eatZwJPAK4FfskkTJIkLTedzYhJkiRpfr5rUpIk\nqScmYpIkST0xEZMkSeqJiZikh70kOyfZmOT/7bsvkh5eTMQkbXeSfDzJ3UkeM0/MxUnuT/L4Ef1s\ntZ/OJDkhyZwbT0t6eDIRk7Q9uphm88T/MtfBJI8GfhX4ZFXdua0/1r4L99HAu7a1rS14I82bRyQJ\nMBGTtH36a+DfaPYjnMvLgd3YxleXpbErQFU9UO7nI2mJmYhJ2u5U1X3AXwAHJ9lzjpCjgH8F/gYg\nyclJvpDkjiT3JPnHJC8frDCwDuzMJL+d5Os0rzw7eK41Ykn2TvLHSb7Rtnl7kj9N8pOz2n1NW/cF\nSd6X5PtJ/i3JZUmeMBD3HZrXsf1SG78xyd+OaMgkLVNd7qwvSdviYuAY4DeBczcVtmvCDgEurqr7\n2+I3Ax8DPkrzSrWjgI8lOayqZic7hwK/RfMO2x8ANw/5/RcAB7T9+C6wD3ACMJ7kZwZ+e9Ms2rnA\n7cAfAE8F/itwL/9xK/KNbcwdwOk07969bYFjIWkH5c76krZLSXYCvgPcWFU/P1D+uzQJzSFVdWVb\ntutAYkSSXWhef3ZzVR3Wlu0M/Aj4d+CnqupbA/Gbjr29qv7HXG22ZT8HfB6YqKo/a8uOAz4EfKqq\njhiIPZvmXbt7VNU9bdla4DtVdcgoxkjS8uetSUnbparaCPwpcOCs24FHAeuBvxuIHUzC9qB5V+3n\ngbE5mr5yMAmb5/cH23xEe5vxeppborPbLZp35A76HLAz8JNI0hAmYpK2ZxfT3MI7CiDJTwA/D0wO\nLqxP8qtJrk5yL83txu8BrwVWzNHmTQv54SSPTvL/tWu77qO57fg94LFD2v3OrO+bnuYc1fYaknZA\nJmKStltVNQ1cB0y0RZueorxkU0ySXwD+kmam6njgMOCXgD9j7r/j7l3gz/8xcFL7W0cCL23bnRnS\n7oYh7WSBvyfpYcjF+pK2dxcDpyV5Nk1C9s2qmho4/mvA3cAvt/uBAQ+uJdsWvw78r6o6eaDNRzP3\nbNhCuShX0macEZO0vdt0e/I04Dk0T0YO2gBspFmPBUCSpwK/so2/u4GH/h35X9m2Ga67adavSRLg\njJik7VxV3ZTkH4CX0cwoXTIr5BM021esSTIJ7AW8AfgG8NPb8NOXA8cm+be2rZ8DDqJZgzbbsORs\ndvkUcFy7X9m3gXVV9dlt6KOkZc5ETNJycDFwIHBNVd0weKCqPp3ktTTrud4H3AD8HvAsHpqIzfc+\nydnHTgAeAF5F87qlq2jWiP39HG3M1+agU4EnAyfTLPq/EjARkx7G3EdMkiSpJ0uyRizJCUluTHJv\n+4j5AVuIf0mSqST3Jbk+yTFzxKxI8sEkt7Zx1yX55e7OQpIkabQ6T8SSvAJ4L3AK8Fya3a7XDHl/\nHEn2plmbcSWwP3A2cEGSlw7EPAK4gmajxF+jeX/ba2leQyJJkrQsdH5rMsnVNOs6Tmy/h2bjw3Oq\n6j1zxL8bOKyqfnagbBJYUVWHt9+Pp1kDsu/g4+qSJEnLSaczYu3M1TjN7BYA7W7YV9AsvJ3LC9vj\ng9bMiv8V4IvAuUnWJfm/Sd7WvptOkiRpWeg6cdmTZm+f9bPK1wOrhtRZNSR+9yS7tt+fCvwGTf8P\no9lf6PeA3x9BnyVJkpbEct2+Yiea5Ox17Qzbl5M8GXgr8I7ZwUl+DDiU5h1z9y1hPyVJ0o7hUcDe\nwJqqumNUjXadiN1Oszv1ylnlK4F1Q+qsGxJ/V1Xd336/DXigNl/gthZYlWSXqvr3WfUPpdmHSJIk\naVu8koduLL3VOk3EqupHSaaAg4G/hgcX6x8MnDOk2hdpbjcOOqQt3+QL/MdLgDd5FnDbHEkYNDNh\nfPSjH2W//fZbzCloEVavXs1ZZ53Vdzd2WI5v9xzjbjm+3XOMu7N27Vpe9apXQZtTjMpS3Jo8E7iw\nTci+BKwGdgMuBEhyOvCkqtq0V9h5wAnt05MfpknajgQOH2jzj9uYc4D302xf8TaaXbXnch/Afvvt\nx9jY2OjOTJtZsWKF49shx7d7jnG3HN/uOcZLYqRLnDpPxKrq0nbPsNNobjF+BTi0qr7fhqwCnjIQ\nf1OSI4CzaN4fdwtwXFVdMRBzS5JD25hrafYPOwt4yHYYkiRJ26slWaxfVecC5w45duwcZVfRbHsx\nX5vX0LyEV5IkaVly3y1JkqSemIhpZCYmZj8/oVFyfLvnGHfL8e2eY7z8dP6Ko+1BkjFgampqykWM\nkiRp0aanpxkfHwcYr6rpUbXrjJgkSVJPTMQkSZJ6YiImSZLUExMxSZKknpiISZIk9cRETJIkqScm\nYpIkST0xEZMkSeqJiZgkSVJPTMQkSZJ6YiImSZLUExMxSZKknixJIpbkhCQ3Jrk3ydVJDthC/EuS\nTCW5L8n1SY6ZJ/a3kmxM8hej77kkSVJ3Ok/EkrwCeC9wCvBc4FpgTZI9h8TvDVwOXAnsD5wNXJDk\npUNizwCuGn3PJUmSurUUM2KrgfOr6qKqug44HrgHePWQ+NcDN1TVSVX1jar6IHBZ286DkuwEfBT4\nA+DGznovSZLUkU4TsSSPAMZpZrcAqKoCrgAOHFLthe3xQWvmiD8FWF9VHxlNbyVJkpbWLh23vyew\nM7B+Vvl64FlD6qwaEr97kl2r6v4kPw8cS3PrUpIkaVladk9NJnkscBHw2qq6s+/+SJIkba2uZ8Ru\nBzYAK2eVrwTWDamzbkj8Xe1s2L7AfwL+Jkna4zsBJHkAeFZVzblmbPXq1axYsWKzsomJCSYmJhZ4\nOpIkaUc3OTnJ5OTkZmUzMzOd/FaaJVvdSXI1cE1Vndh+D3AzcE5VnTFH/LuAw6pq/4GyS4A9qurw\nJLsCT5tV7Z3AY4E3A9+sqn+f1eYYMDU1NcXY2NgIz06SJD0cTE9PMz4+DjBeVdOjarfrGTGAM4EL\nk0wBX6JHZfHEAAATyElEQVR5+nE34EKAJKcDT6qqTXuFnQeckOTdwIeBg4EjgcMBqup+4J8HfyDJ\nD5tDtbbzs5EkSRqRzhOxqrq03TPsNJpbjF8BDq2q77chq4CnDMTflOQI4CyaGa5bgOOqavaTlJIk\nScvaUsyIUVXnAucOOXbsHGVX0Wx7sdD2H9KGJEnS9m7ZPTUpSZK0ozARkyRJ6omJmCRJUk9MxCRJ\nknpiIiZJktQTEzFJkqSemIhJkiT1xERMkiSpJyZikiRJPTERkyRJ6omJmCRJUk9MxCRJknpiIiZJ\nktQTEzFJkqSemIhJkiT1xERMkiSpJ0uSiCU5IcmNSe5NcnWSA7YQ/5IkU0nuS3J9kmNmHX9NkquS\n/KD9fHpLbUqSJG1vOk/EkrwCeC9wCvBc4FpgTZI9h8TvDVwOXAnsD5wNXJDkpQNhBwGXAC8BXgh8\nB/jbJHt1chKSJEkdWIoZsdXA+VV1UVVdBxwP3AO8ekj864EbquqkqvpGVX0QuKxtB4Cq+u2qOq+q\nvlpV1wOvoTmXgzs9E0mSpBHqNBFL8ghgnGZ2C4CqKuAK4MAh1V7YHh+0Zp54gMcAjwB+sNWdlSRJ\nWmJdz4jtCewMrJ9Vvh5YNaTOqiHxuyfZdUiddwPf5aEJnCRJ0nZrl747sK2S/DfgN4GDquqBvvsj\nSZK0UF0nYrcDG4CVs8pXAuuG1Fk3JP6uqrp/sDDJW4GTgIOr6utb6szq1atZsWLFZmUTExNMTExs\nqaokSXqYmJycZHJycrOymZmZTn4rzZKt7iS5Grimqk5svwe4GTinqs6YI/5dwGFVtf9A2SXAHlV1\n+EDZScDbgEOq6h+30IcxYGpqaoqxsbFRnJYkSXoYmZ6eZnx8HGC8qqZH1e5SPDV5JvDaJEcn2Rc4\nD9gNuBAgyelJ/mQg/jzgqUneneRZSd4AHNm2Q1vnZOA0micvb06ysv08ZgnOR5IkaSQ6XyNWVZe2\ne4adRnOL8SvAoVX1/TZkFfCUgfibkhwBnAW8GbgFOK6qBhfiH0/zlORls37uD9vfkSRJ2u4tyWL9\nqjoXOHfIsWPnKLuKZtuLYe3tM7reSZIk9cN3TUqSJPXEREySJKknJmKSJEk9MRGTJEnqiYmYJElS\nT0zEJEmSemIiJkmS1BMTMUmSpJ6YiEmSJPXEREySJKknJmKSJEk9MRGTJEnqiYmYJElST0zEJEmS\nemIiJkmS1JMlScSSnJDkxiT3Jrk6yQFbiH9Jkqkk9yW5Pskxc8T8RpK1bZvXJjmsuzOQJEkavc4T\nsSSvAN4LnAI8F7gWWJNkzyHxewOXA1cC+wNnAxckeelAzM8BlwAfAp4DfBz4qyQ/1dmJSJIkjdhS\nzIitBs6vqouq6jrgeOAe4NVD4l8P3FBVJ1XVN6rqg8BlbTubvBn4VFWd2cb8ATANvLG705AkSRqt\nThOxJI8AxmlmtwCoqgKuAA4cUu2F7fFBa2bFH7iAGEmSpO1a1zNiewI7A+tnla8HVg2ps2pI/O5J\ndt1CzLA2JUmStju79N0Bjc6HP/xh3vrWt/bdja3ygQ98gH333bfvbmyVnXbaiY0bN/bdja2yXPu+\nXPsN9r0vy7Xvy7XfAHvttRd77bVX393Y7nWdiN0ObABWzipfCawbUmfdkPi7qur+LcQMaxOA1atX\ns2LFis3KJiYmmJiYmK/asjE2Nsbb3va2vruxVa6++mpe+cpX9t2NrXLQQQfx2c9+tu9ubJXl2vfl\n2m+w731Zrn1frv0GOOWUUzj11FP77sZWmZycZHJycrOymZmZTn4rzZKt7iS5Grimqk5svwe4GTin\nqs6YI/5dwGFVtf9A2SXAHlV1ePv9T4FHV9XLBmK+AFxbVW+Yo80xYGpqaoqxsbHRnqBG4rbbbuO2\n227ruxtbZTn/i3W59n259hvse1+Wa9+Xa79hx5sRm56eZnx8HGC8qqZH1e5S3Jo8E7gwyRTwJZqn\nH3cDLgRIcjrwpKratFfYecAJSd4NfBg4GDgSOHygzbOBzyR5C/AJYILmoYDXdn426sSO9j9YSZIW\novNErKoubfcMO43m9uFXgEOr6vttyCrgKQPxNyU5AjiLZpuKW4DjquqKgZgvJjkKeGf7+Sbwsqr6\n567PR5IkaVSWZLF+VZ0LnDvk2LFzlF1FM8M1X5sfAz42kg5KkiT1wHdNSpIk9cRETJIkqScmYpIk\nST0xEZMkSeqJiZgkSVJPTMQkSZJ6YiImSZLUExMxSZKknpiISZIk9cRETJIkqScmYpIkST0xEZMk\nSeqJiZgkSVJPTMQkSZJ6YiImSZLUk84SsSSPT3Jxkpkkdya5IMljFlDvtCS3JrknyaeTPH1Wm+ck\nua49/i9Jzk6ye1fnIUmS1JUuZ8QuAfYDDgaOAF4MnD9fhSQnA28EXgc8H7gbWJPkkW3Ik4C9gLcA\nPw0cA/wycEEH/ZckSerULl00mmRf4FBgvKq+3Ja9CfhEkrdW1bohVU8E3lFVl7d1jgbWAy8HLq2q\nrwO/MRB/Y5LfB/53kp2qamMX5yNJktSFrmbEDgTu3JSEta4ACnjBXBWS7AOsAq7cVFZVdwHXtO0N\nswdwl0mYJElabrpKxFYB3xssqKoNwA/aY8PqFM0M2KD1w+ok2RN4O1u45SlJkrQ9WlQiluT0JBvn\n+WxI8syuOjurL48DPgF8DfjDpfhNSZKkUVrsGrE/Aj6yhZgbgHXAEwcLk+wMPKE9Npd1QICVbD4r\nthIYvMVJkscCa4AfAr/WzrZt0erVq1mxYsVmZRMTE0xMTCykuiRJehiYnJxkcnJys7KZmZlOfitV\nNfpGm8X6XweeN7BY/xDgk8CThy3WT3IrcEZVndV+350mKTu6qv68LXscTRJ2L3B4Vd2/gP6MAVNT\nU1OMjY1t8/lJkqSHl+npacbHx6F5EHF6VO12skasqq6jSZY+lOSAJC8C3g9MDiZh7X5gLxuo+j7g\n7Ul+JcmzgYuAW4CPt/GPAz4N7Aa8Btgjycr24+a0kiRpWelk+4rWUcAHaJ6W3AhcRrM9xaBnAA/e\nK6yq9yTZjWbx/R7A54DDquqBNmQMOKD972+1f4Zmkf8+wM2jPw1JkqRudJaIVdUPgVdtIWbnOcpO\nBU4dEv9Z4CF1JEmSliNv50mSJPXEREySJKknJmKSJEk9MRGTJEnqiYmYJElST0zEJEmSemIiJkmS\n1BMTMUmSpJ6YiEmSJPXEREySJKknJmKSJEk9MRGTJEnqiYmYJElST0zEJEmSemIiJkmS1JPOErEk\nj09ycZKZJHcmuSDJYxZQ77Qktya5J8mnkzx9nthPJdmY5FdH23tJkqTudTkjdgmwH3AwcATwYuD8\n+SokORl4I/A64PnA3cCaJI+cI3Y1sAGo0XZbkiRpaXSSiCXZFzgUOK6q/qmq/gF4E/BbSVbNU/VE\n4B1VdXlVfQ04GngS8PJZ7T8HWA28GkgX5yBJktS1rmbEDgTurKovD5RdQTN79YK5KiTZB1gFXLmp\nrKruAq5p29sU92jgYuANVfW90XddkiRpaXSViK0CNkuSqmoD8IP22LA6BayfVb5+Vp2zgM9X1eWj\n6aokSVI/FpWIJTm9XRw/7LMhyTO76my7KP8XaW5LSpIkLWu7LDL+j4CPbCHmBmAd8MTBwiQ7A09o\nj81lHc16r5VsPiu2Eth0i/MXgKcCM8lmS8P+IslVVfWL83Vs9erVrFixYrOyiYkJJiYm5qsmSZIe\nRiYnJ5mcnNysbGZmppPfStXoHzpsF+t/HXjepnViSQ4BPgk8uarmTMaS3AqcUVVntd93p0nKjq6q\nP0/yRGDPWdW+RvMgwOVV9S9D2h0DpqamphgbG9v2E5QkSQ8r09PTjI+PA4xX1fSo2l3sjNiCVNV1\nSdYAH0ryeuCRwPuBycEkLMl1wMlV9fG26H3A25N8C7gJeAdwC/Dxtt3vMWvtWTsz9p1hSZgkSdL2\nqpNErHUU8AGapyU3ApfRbE8x6BnAg/cKq+o9SXaj2W9sD+BzwGFV9cA8v+M+YpIkaVnqLBGrqh8C\nr9pCzM5zlJ0KnLqI33lIG5IkScuB75qUJEnqiYmYJElST0zEJEmSemIiJkmS1BMTMUmSpJ6YiEmS\nJPXEREySJKknJmKSJEk9MRGTJEnqiYmYJElST0zEJEmSemIiJkmS1BMTMUmSpJ6YiEmSJPXEREyS\nJKknJmIamcnJyb67sENzfLvnGHfL8e2eY7z8dJaIJXl8kouTzCS5M8kFSR6zgHqnJbk1yT1JPp3k\n6XPEHJjkyiT/1rb/mSS7dnMmWij/AuiW49s9x7hbjm/3HOPlp8sZsUuA/YCDgSOAFwPnz1chycnA\nG4HXAc8H7gbWJHnkQMyBwKeA/wM8r/18ANg4+lOQJEnqzi5dNJpkX+BQYLyqvtyWvQn4RJK3VtW6\nIVVPBN5RVZe3dY4G1gMvBy5tY84E3ldVZwzU+2YHpyFJktSprmbEDgTu3JSEta4ACnjBXBWS7AOs\nAq7cVFZVdwHXtO2R5Mfb+rcn+UKSde1tyRd1cxqSJEnd6WRGjCah+t5gQVVtSPKD9tiwOkUzAzZo\n/UCdp7Z/ngL8HnAtcAxwZZKfrqpvD2n7UQBr165dzDlokWZmZpienu67Gzssx7d7jnG3HN/uOcbd\nGcghHjXShqtqwR/gdJq1WMM+G4BnAm8D1s5Rfz3wu0PaPrCtv3JW+Z8BkwMxG2luXw7GXAu8c55+\nH0WT5Pnx48ePHz9+/GzL56jF5E5b+ix2RuyPgI9sIeYGYB3wxMHCJDsDT2iPzWUdEGAlm8+KrQQ2\n3eK8rf1z9tTWWuAn5+nTGuCVwE3AffP2XpIk6aEeBexNk1OMzKISsaq6A7hjS3FJvgjskeS5A+vE\nDqZJtK4Z0vaNSda1cV9t29mdZk3YB9uYm5LcCjxrVvVnAp/cQr8v2VK/JUmS5vEPo26wk8X6VXUd\nTcb4oSQHtIvp309zi/HBGbEk1yV52UDV9wFvT/IrSZ4NXATcAnx8IOYM4M1Jfj3J05K8gyYx+19d\nnIskSVJXulqsD826rA/QPC25EbiMZnuKQc8AVmz6UlXvSbIbzX5jewCfAw6rqgcGYs5uN289k+ZW\n57XAL1XVjR2eiyRJ0silXcwuSZKkJea7JiVJknqy7BOxJG9L8qUkdyVZn+QvkzxzAfVekmQqyX1J\nrk9yzFL0dznamjFOclCSjbM+G5I8cb56D0dJjk9ybfve1Jkk/5Dkl7dQx+t3ERY7xl6/2ybJf2vH\n7MwtxHkdb4WFjK/X8OIkOWWO8frnLdQZyfW77BMx4D/TPAjwAuCXgEcAf5vk0cMqJNkbuJxmF//9\ngbOBC5K8tOvOLlOLHuNW0awDXNV+9qqq781f5WHpO8DJwBgwDvwd8PEk+80V7PW7VRY1xi2v362Q\n5ACa9wVfu4W4vfE6XrSFjm/La3hxvkazZdam8fr5YYGjvH53uDViSfak2dX/xVX1+SEx76Z5COBn\nB8omgRVVdfjS9HT5WuAYH0Tzf3aPb19VpUVIcgfw1qp6yL59Xr+jsYUx9vrdCkkeC0wBrwf+O/Dl\nqnrLkFiv40Va5Ph6DS9CklOAl1XV2ALjR3b97ggzYrPtQfOvgB/ME/NCmqc5B62hfaeltmghYwzN\nvnFfSXJrkr9N8nPdd215S7JTkt8CdgO+OCTM63cbLHCMwet3a3wQ+Juq+rsFxHodL95ixhe8hhfr\nGUm+m+TbST6a5CnzxI7s+u1y+4ollyQ0e5F9vqrmu7e7irnfabl7kl2r6v6u+rjcLWKMbwN+F/gn\nYFfgtcBnkjy/qr7SfU+XlyQ/Q5MUPAr4V+C/tPvxzcXrdysscoy9fhepTW6fAzxvgVW8jhdhK8bX\na3hxrgZ+B/gGsBdwKnBVkp+pqrvniB/Z9btDJWLAucBPAS/quyM7sAWNcVVdD1w/UHR1kqcBq2le\n1K7NXUezzmAFcCRwUZIXz5MoaPEWPMZev4uT5Mk0/0D7par6Ud/92dFszfh6DS9OVQ2+tuhrSb4E\n/Avwm2z51Y7bZIe5NZnkA8DhwEuq6rYthK+jWZA3aCVwl/8KG26RYzyXLwFPH22vdgxV9e9VdUNV\nfbmqfp9mIe7sDZA38frdCosc47l4/Q43Dvw4MJ3kR0l+BBwEnJjkgXYmfTav44XbmvGdi9fwAlXV\nDE0iO2y8Rnb97hAzYm2C8DLgoKq6eQFVvggcNqvsEOZfL/KwthVjPJfn8B8vbtf8dqK5nTAXr9/R\nmG+M5+L1O9wVwLNnlV0IrAXeVXM/FeZ1vHBbM75z8RpeoPbBiKfTvGpxLiO7fpd9IpbkXGAC+FXg\n7iSbMtSZqrqvjfkfwE9U1abp2POAE9qnHj5M86LxI2lmezTL1oxxkhOBG4Gv06zJeS3wC4CPps/S\njt2ngJuBxwGvpPnX7iHt8dOBJ3n9br3FjrHX7+K0a2g2WzOa5G7gjqpa23737+GttDXj6zW8OEnO\nAP6G5nbkTwB/CPwImGyPd3b9LvtEDDie5gm+z8wqP5b/yGT3Ah58+qGqbkpyBHAW8GaaF4sfV1Wz\nn4BQY9FjDDwSeC/wJOAe4KvAwVV1Vac9XZ6eCPwJzRjO0IzVIQNPRq3C63dbLWqM8fodhdmzNP49\nPFrzji9ew4v1ZOAS4MeA7wOfB15YVXe0xzu7fne4fcQkSZKWix1msb4kSdJyYyImSZLUExMxSZKk\nnpiISZIk9cRETJIkqScmYpIkST0xEZMkSeqJiZgkSVJPTMQkSZJ6YiImSZLUExMxSZKknpiISZIk\n9eT/B8SlR3VacvhJAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa29a1ac650>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDsAAAFkCAYAAAAuUzCnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfioAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa446d92610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "experiment_probes = {}\n",
    "experiment_probes[\"payload/processes/content/histograms\"] =  [\"TIME_TO_LOAD_EVENT_START_ACTIVE_NETOPT_MS\"]\n",
    "probe_names = list(itertools.chain(*experiment_probes.values()))\n",
    "\n",
    "df = exptPings(EXPERIMENT_SLUG, 0.1, \"20180220\", \"20180305\")\n",
    "\n",
    "for p in probe_names:\n",
    "    chart_pref(p, True, 1.0)\n",
    "    chart_ecdfs(p, 1.0, 50, 95)"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "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
}

## HTTP response throttling v2 (Pref Flip).py

# coding: utf-8

# In[1]:

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import plotly.plotly as py
from scipy.stats import chi2_contingency
from scipy.stats import ttest_ind
from collections import defaultdict as dd

import datetime as DT

from pyspark.sql import Row
from pyspark.sql import SQLContext
from pyspark.sql.types import *
from pyspark.sql.functions import *
import itertools

from moztelemetry.dataset import Dataset
from moztelemetry import get_pings_properties

from scipy.stats import mannwhitneyu
from __future__ import division

py_max = __builtin__.max
py_map = __builtin__.map

get_ipython().magic(u'pylab inline')

## these ones pull the data
#
def recursive_get(d, keys):
    if len(keys) == 1:
        return d.get(keys[0],{})
    return recursive_get(d.get(keys[0],{}), keys[1:])

def extract_probes(p):
    branch = p.get("environment",{}).get("experiments",{}).get(EXPERIMENT_SLUG, {}).get("branch", "warning")

    output = []
    for path, probe_names in experiment_probes.iteritems():
        for probe_name in probe_names:
            probe = recursive_get(p, path.split("/")).get(probe_name, {})
            #take all entries in histogram right now. this is probably problematic. inspect more
            for k,v in probe.get("values",{}).iteritems():
                if v <= sys.maxint:
                    output.extend([{"probe": probe_name, "branch": branch, "val": float(k)}] * int(v))
    return output

def exptPings(slug, samp, start= DT.date.today().strftime("%Y%m%d"), end= DT.date.today().strftime("%Y%m%d")):
    # returns a spark df with values from probe dictionary by branch and nothing else, default start/end dates are today
    cohorts = Dataset.from_source("telemetry-cohorts")
    main_pings = cohorts.where(submissionDate = lambda x: x >= start and x <= end).where(experimentId= slug).where(docType= "main").records(sc, sample= samp)
    #main_pings.cache() this was in ilana's ognb, mreid advised to remove it for the exception I was hitting

    probe_dicts = main_pings.flatMap(extract_probes)
    return sqlContext.createDataFrame(probe_dicts.map(lambda d: Row(**d)))

## these ones do the analysis
#

###### get the bins we use for the histogram for a probe by looking at all branches
def get_bins(probe_name, logscale=False):

    all_branches = [r.val for r in df.where(df.probe == probe_name).collect()]

    #remove top 0.5%, bottom 0.5% for easy outlier
    trim = int(len(all_branches)/200.0)
    all_branches_trimmed = sorted(all_branches)
    all_branches_trimmed = all_branches_trimmed[trim:-1*trim]

    if logscale:
        if all_branches_trimmed[0] < 1:
            all_branches_trimmed = py_map(lambda d: d+1, all_branches_trimmed)
        return list(np.linspace(np.log10(all_branches_trimmed[1]), np.log10(all_branches_trimmed[-1]), 10))

    n,b = np.histogram(all_branches_trimmed,10)
    return b

# get values for branch of experiment for pref, and trim off outliers
def get_vals(pref_name, branch, samp, seed=None):
    x_vals = [r.val for r in df.where(df.probe == pref_name)                               .where(df.branch == branch)                               .sample(False, samp, seed).collect()]

    trim = int(len(x_vals)/200.0)
    x_trimmed = sorted(x_vals)[trim:-1*trim]
    return x_trimmed

def median(lst):
    return lst[(len(lst))/2]

# return (pval, direction) if significant p value for mannwhitneyu vs control
def test_unequal(branch_vals, control_vals, p_threshold=.05):

    try:
        r = mannwhitneyu(branch_vals, control_vals)
    except:
        return None, None

    prefix = ""
    if r.pvalue < p_threshold:
        prefix = "***"

    if median(branch_vals) > median(control_vals):
        return (prefix + str(r.pvalue), "> control")

    return (prefix + str(r.pvalue), "< control")

# return (proceed bool, reason)
def can_chart_pref(pref_name):
    n = df.where(df.probe == pref_name).count()
    if n==0:
        return (False, "0 entries for pref %s"%pref_name)
    elif n>100000000:
        return (False, "%i values for pref %s"%(n,pref_name))
    return (True, None)

# chart histograms for all branches of a probe, log/std, and calculate if any branches vary from the mean
def chart_pref(pref_name, logscale, samp):

    sig_branches = []
    fig, axarr = plt.subplots(n_branches, 1, sharex=True, sharey= 'col')

    b = get_bins(pref_name, logscale)
    plt.tight_layout()

    control_vals = get_vals(pref_name, "Control", samp, 666)

    for i in range(n_branches):

        if branches[i] == "Control":
            x = control_vals

        else:
            x = get_vals(pref_name, branches[i], samp, 666)

        if logscale: #always assume 0 as lowest val for now
            x_trans = py_map(lambda d: d+1, x)
            ap,bp,cp = axarr[i].hist(np.log(x_trans), bins=b)

        else:
            axarr[i].hist(x, bins=b)

        axarr[i].set_title(branches[i])

        if branches[i] == "Control": continue

        print "len(branch_vals) = " + str(len(x)) + ", len(control_vals) = " + str(len(control_vals))
        if len(x) != 0 | len(control_vals) != 0:
#            p, direction = test_unequal(x, control_vals)
#            if p is not None and p.startswith("***"):
#                print branches[i], p #, direction
            print pref_name + " (logscale=" + str(logscale) + ")" + ":"
        else:
            print pref_name + " branch with no values"
            continue

    plt.show()

from statsmodels.distributions.empirical_distribution import ECDF

def chart_ecdfs(pref_name, samp, *percentiles):

    legend = []

    for i in range(n_branches):
        legend_entry = ""
        v = get_vals(pref_name, branches[i], samp, 666)
        if len(v) == 0: continue
        cdf = ECDF(v)
        curr_plot = plt.plot(cdf.x, cdf.y)
        curr_color = curr_plot[0].get_color()
        legend_entry += branches[i]
        for pct in percentiles:
            p = np.percentile(np.array(v), pct)
            plt.scatter(p, pct/100.0, facecolors = "none",                      edgecolors = curr_color, label="_nolegend_")
            legend_entry += ", " + str(pct) + "th percentile=" + str(p)

        legend.append(legend_entry)

    plt.legend(legend, bbox_to_anchor=[1, .5], loc='center left')
    plt.show()

EXPERIMENT_SLUG = "pref-flip-http-response-throttling-algo-v2-beta-1434388"
branches = ['Control', 'Variant']
n_branches = len(branches)


# In[2]:

experiment_probes = {}
experiment_probes["payload/processes/content/histograms"] =  ["TIME_TO_DOM_INTERACTIVE_MS"]
probe_names = list(itertools.chain(*experiment_probes.values()))

df = exptPings(EXPERIMENT_SLUG, 0.05, "20180220", "20180305")

for p in probe_names:
    chart_pref(p, True, 1.0)
    chart_ecdfs(p, 1.0, 50, 95)


# In[3]:

experiment_probes = {}
experiment_probes["payload/processes/content/histograms"] =  ["TIME_TO_NON_BLANK_PAINT_NETOPT_MS"]
probe_names = list(itertools.chain(*experiment_probes.values()))

df = exptPings(EXPERIMENT_SLUG, 0.1, "20180220", "20180305")

for p in probe_names:
    chart_pref(p, True, 1.0)
    chart_ecdfs(p, 1.0, 50, 95)


# In[4]:

experiment_probes = {}
experiment_probes["payload/processes/content/histograms"] =  ["TIME_TO_DOM_CONTENT_LOADED_START_ACTIVE_NETOPT_MS"]
probe_names = list(itertools.chain(*experiment_probes.values()))

df = exptPings(EXPERIMENT_SLUG, 0.1, "20180220", "20180305")

for p in probe_names:
    chart_pref(p, True, 1.0)
    chart_ecdfs(p, 1.0, 50, 95)


# In[8]:

experiment_probes = {}
experiment_probes["payload/processes/content/histograms"] =  ["TIME_TO_LOAD_EVENT_START_ACTIVE_NETOPT_MS"]
probe_names = list(itertools.chain(*experiment_probes.values()))

df = exptPings(EXPERIMENT_SLUG, 0.1, "20180220", "20180305")

for p in probe_names:
    chart_pref(p, True, 1.0)
    chart_ecdfs(p, 1.0, 50, 95)

	# coding: utf-8

	# In[1]:

	import matplotlib.pyplot as plt
	import pandas as pd
	import numpy as np
	import plotly.plotly as py
	from scipy.stats import chi2_contingency
	from scipy.stats import ttest_ind
	from collections import defaultdict as dd

	import datetime as DT

	from pyspark.sql import Row
	from pyspark.sql import SQLContext
	from pyspark.sql.types import *
	from pyspark.sql.functions import *
	import itertools

	from moztelemetry.dataset import Dataset
	from moztelemetry import get_pings_properties

	from scipy.stats import mannwhitneyu
	from __future__ import division

	py_max = __builtin__.max
	py_map = __builtin__.map

	get_ipython().magic(u'pylab inline')

	## these ones pull the data
	#
	def recursive_get(d, keys):
	if len(keys) == 1:
	return d.get(keys[0],{})
	return recursive_get(d.get(keys[0],{}), keys[1:])

	def extract_probes(p):
	branch = p.get("environment",{}).get("experiments",{}).get(EXPERIMENT_SLUG, {}).get("branch", "warning")

	output = []
	for path, probe_names in experiment_probes.iteritems():
	for probe_name in probe_names:
	probe = recursive_get(p, path.split("/")).get(probe_name, {})
	#take all entries in histogram right now. this is probably problematic. inspect more
	for k,v in probe.get("values",{}).iteritems():
	if v <= sys.maxint:
	output.extend([{"probe": probe_name, "branch": branch, "val": float(k)}] * int(v))
	return output

	def exptPings(slug, samp, start= DT.date.today().strftime("%Y%m%d"), end= DT.date.today().strftime("%Y%m%d")):
	# returns a spark df with values from probe dictionary by branch and nothing else, default start/end dates are today
	cohorts = Dataset.from_source("telemetry-cohorts")
	main_pings = cohorts.where(submissionDate = lambda x: x >= start and x <= end).where(experimentId= slug).where(docType= "main").records(sc, sample= samp)
	#main_pings.cache() this was in ilana's ognb, mreid advised to remove it for the exception I was hitting

	probe_dicts = main_pings.flatMap(extract_probes)
	return sqlContext.createDataFrame(probe_dicts.map(lambda d: Row(**d)))

	## these ones do the analysis
	#

	###### get the bins we use for the histogram for a probe by looking at all branches
	def get_bins(probe_name, logscale=False):

	all_branches = [r.val for r in df.where(df.probe == probe_name).collect()]

	#remove top 0.5%, bottom 0.5% for easy outlier
	trim = int(len(all_branches)/200.0)
	all_branches_trimmed = sorted(all_branches)
	all_branches_trimmed = all_branches_trimmed[trim:-1*trim]

	if logscale:
	if all_branches_trimmed[0] < 1:
	all_branches_trimmed = py_map(lambda d: d+1, all_branches_trimmed)
	return list(np.linspace(np.log10(all_branches_trimmed[1]), np.log10(all_branches_trimmed[-1]), 10))

	n,b = np.histogram(all_branches_trimmed,10)
	return b

	# get values for branch of experiment for pref, and trim off outliers
	def get_vals(pref_name, branch, samp, seed=None):
	x_vals = [r.val for r in df.where(df.probe == pref_name) .where(df.branch == branch) .sample(False, samp, seed).collect()]

	trim = int(len(x_vals)/200.0)
	x_trimmed = sorted(x_vals)[trim:-1*trim]
	return x_trimmed

	def median(lst):
	return lst[(len(lst))/2]

	# return (pval, direction) if significant p value for mannwhitneyu vs control
	def test_unequal(branch_vals, control_vals, p_threshold=.05):

	try:
	r = mannwhitneyu(branch_vals, control_vals)
	except:
	return None, None

	prefix = ""
	if r.pvalue < p_threshold:
	prefix = "***"

	if median(branch_vals) > median(control_vals):
	return (prefix + str(r.pvalue), "> control")

	return (prefix + str(r.pvalue), "< control")

	# return (proceed bool, reason)
	def can_chart_pref(pref_name):
	n = df.where(df.probe == pref_name).count()
	if n==0:
	return (False, "0 entries for pref %s"%pref_name)
	elif n>100000000:
	return (False, "%i values for pref %s"%(n,pref_name))
	return (True, None)

	# chart histograms for all branches of a probe, log/std, and calculate if any branches vary from the mean
	def chart_pref(pref_name, logscale, samp):

	sig_branches = []
	fig, axarr = plt.subplots(n_branches, 1, sharex=True, sharey= 'col')

	b = get_bins(pref_name, logscale)
	plt.tight_layout()

	control_vals = get_vals(pref_name, "Control", samp, 666)

	for i in range(n_branches):

	if branches[i] == "Control":
	x = control_vals

	else:
	x = get_vals(pref_name, branches[i], samp, 666)

	if logscale: #always assume 0 as lowest val for now
	x_trans = py_map(lambda d: d+1, x)
	ap,bp,cp = axarr[i].hist(np.log(x_trans), bins=b)

	else:
	axarr[i].hist(x, bins=b)

	axarr[i].set_title(branches[i])

	if branches[i] == "Control": continue

	print "len(branch_vals) = " + str(len(x)) + ", len(control_vals) = " + str(len(control_vals))
	if len(x) != 0 \| len(control_vals) != 0:
	# p, direction = test_unequal(x, control_vals)
	# if p is not None and p.startswith("***"):
	# print branches[i], p #, direction
	print pref_name + " (logscale=" + str(logscale) + ")" + ":"
	else:
	print pref_name + " branch with no values"
	continue

	plt.show()

	from statsmodels.distributions.empirical_distribution import ECDF

	def chart_ecdfs(pref_name, samp, *percentiles):

	legend = []

	for i in range(n_branches):
	legend_entry = ""
	v = get_vals(pref_name, branches[i], samp, 666)
	if len(v) == 0: continue
	cdf = ECDF(v)
	curr_plot = plt.plot(cdf.x, cdf.y)
	curr_color = curr_plot[0].get_color()
	legend_entry += branches[i]
	for pct in percentiles:
	p = np.percentile(np.array(v), pct)
	plt.scatter(p, pct/100.0, facecolors = "none", edgecolors = curr_color, label="_nolegend_")
	legend_entry += ", " + str(pct) + "th percentile=" + str(p)

	legend.append(legend_entry)

	plt.legend(legend, bbox_to_anchor=[1, .5], loc='center left')
	plt.show()

	EXPERIMENT_SLUG = "pref-flip-http-response-throttling-algo-v2-beta-1434388"
	branches = ['Control', 'Variant']
	n_branches = len(branches)


	# In[2]:

	experiment_probes = {}
	experiment_probes["payload/processes/content/histograms"] = ["TIME_TO_DOM_INTERACTIVE_MS"]
	probe_names = list(itertools.chain(*experiment_probes.values()))

	df = exptPings(EXPERIMENT_SLUG, 0.05, "20180220", "20180305")

	for p in probe_names:
	chart_pref(p, True, 1.0)
	chart_ecdfs(p, 1.0, 50, 95)


	# In[3]:

	experiment_probes = {}
	experiment_probes["payload/processes/content/histograms"] = ["TIME_TO_NON_BLANK_PAINT_NETOPT_MS"]
	probe_names = list(itertools.chain(*experiment_probes.values()))

	df = exptPings(EXPERIMENT_SLUG, 0.1, "20180220", "20180305")

	for p in probe_names:
	chart_pref(p, True, 1.0)
	chart_ecdfs(p, 1.0, 50, 95)


	# In[4]:

	experiment_probes = {}
	experiment_probes["payload/processes/content/histograms"] = ["TIME_TO_DOM_CONTENT_LOADED_START_ACTIVE_NETOPT_MS"]
	probe_names = list(itertools.chain(*experiment_probes.values()))

	df = exptPings(EXPERIMENT_SLUG, 0.1, "20180220", "20180305")

	for p in probe_names:
	chart_pref(p, True, 1.0)
	chart_ecdfs(p, 1.0, 50, 95)


	# In[8]:

	experiment_probes = {}
	experiment_probes["payload/processes/content/histograms"] = ["TIME_TO_LOAD_EVENT_START_ACTIVE_NETOPT_MS"]
	probe_names = list(itertools.chain(*experiment_probes.values()))

	df = exptPings(EXPERIMENT_SLUG, 0.1, "20180220", "20180305")

	for p in probe_names:
	chart_pref(p, True, 1.0)
	chart_ecdfs(p, 1.0, 50, 95)