Earthquake Tweets Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Earthquake Tweet Analysis Sample\n",
    "================================\n",
    "\n",
    "Examining Tweets about Earthquakes between ```April 13, 2016``` and ```May 3, 2016```\n",
    "\n",
    "    Total number of tweets:        4,296,188            \n",
    "    Total number of unique tweets: 4,195,261"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pymongo\n",
    "import json\n",
    "from pprint import pprint"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Step 1: Get all tweets from \"Earthquake\" Event\n",
    "- This was done using the events.rb functionality from epic_utils."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2: Filter & Import Tweets about the Kumamoto Earthquake\n",
    "- From the ```tweets.json``` file generated by events.rb, filter for the _kumamoto_ keyword.\n",
    "\n",
    "To do this, we will parse tweets.json and filter on the following:\n",
    "1. any part of the tweet includes _kumamoto_. This includes tweets from users whose profile may include \"kumamoto\", as well as any tweet where \"kumamoto\" was mentioned in the text\n",
    "\n",
    "1. Any Geo-tagged tweet that falls within the bounds of Kumamoto"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#Open the big one, iterate over it\n",
    "def parse_entire_file_for_geotagged(file=\"tweets.json\"):\n",
    "    geo_tagged_tweets = []\n",
    "    idx = 0\n",
    "    with open(file) as tweets_json:\n",
    "        for line in tweets_json:\n",
    "            idx+=1\n",
    "            tweet = json.loads(line)\n",
    "            if tweet['coordinates'] != None:\n",
    "                geo_tagged_tweets.append(tweet)\n",
    "            if (idx%100000==0):\n",
    "                print(str(idx) + \"...\", end=\"\")\n",
    "    return geo_tagged_tweets\n",
    "\n",
    "#geo_tagged_tweets = parse_entire_file_for_geotagged()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def parse_entire_file_to_array(file=\"tweets_kumamoto.json\"):\n",
    "    tweets = []\n",
    "    idx = 0\n",
    "    with open(file) as tweets_json:\n",
    "        for line in tweets_json:\n",
    "            idx+=1\n",
    "            tweets.append(json.loads(line))\n",
    "            if (idx%1000==0):\n",
    "                print(str(idx)+\" . \", end=\"\")\n",
    "    return tweets\n",
    "\n",
    "#all_kumamoto_tweets = parse_entire_file_to_array()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Geographic Filtering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import ogr, shapely"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def create_geojson(geo_tweets, outfile_name):\n",
    "    # Create structure for GeoJSON feature collection object\n",
    "    tweet_geometries = {\"type\": \"FeatureCollection\", \"features\": [] }\n",
    "    for tweet in geo_tweets:\n",
    "        # Add a feature object to the features array for each tweet\n",
    "        tweet_geometries['features'].append(\n",
    "            {'type':'Feature',\n",
    "             'geometry':tweet['coordinates'],\n",
    "             'properties':{'text':tweet['text'], 'time':tweet['created_at'], 'user':tweet['user']['screen_name']}\n",
    "            })\n",
    "    print(len(tweet_geometries['features']),\" tweets\")\n",
    "    with open(outfile_name, 'w') as outfile:\n",
    "        json.dump(tweet_geometries, outfile)\n",
    "    print(\"Geojson saved to\", outfile_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Identified 24554 Geo-Tagged Tweets in entire collection\n"
     ]
    }
   ],
   "source": [
    "print(\"Identified %d Geo-Tagged Tweets in entire collection\"%len(geo_tagged_tweets))\n",
    "\n",
    "# Write to GeoJSON:\n",
    "# create_geojson(geo_tagged_tweets,'all_geotagged_kumamoto_earthquakes.geojson')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## Filter those tweets to only those _in Kumamoto_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"http://research.yetilabs.co/nbserver/KumamotoEarthquakes2016/kumamoto_bbox.png\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image at 0x7ff0c46cf710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.display import Image, display\n",
    "kumamoto_geojson = '''{\"type\":\"Polygon\",\"coordinates\":[[[128.5290527344,30.9446357394],[128.5290527344,33.9331713056],[132.2644042969,33.9331713056],[132.2644042969,30.9446357394],[128.5290527344,30.9446357394]]]}'''\n",
    "kumamoto = ogr.CreateGeometryFromJson(kumamoto_geojson)\n",
    "display(Image(url='http://research.yetilabs.co/nbserver/KumamotoEarthquakes2016/kumamoto_bbox.png'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "in_kumamoto = []\n",
    "for t in geo_tagged_tweets:\n",
    "    p = ogr.CreateGeometryFromJson(str(t['coordinates']))\n",
    "    if p.Within(kumamoto):\n",
    "        in_kumamoto.append(t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Identified 4700 Geo-Tagged Tweets in Kumamoto bounding box\n"
     ]
    }
   ],
   "source": [
    "print(\"Identified %d Geo-Tagged Tweets in Kumamoto bounding box\"%len(in_kumamoto))\n",
    "\n",
    "# Write to GeoJSON:\n",
    "# create_geojson(in_kumamoto,'in_kumamoto.geojson')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## See these Tweets on a Map:\n",
    "\n",
    "### [All Geo-Tagged Tweets (24k over the world)](http://research.yetilabs.co/nbserver/yeti-sites/pages/maps/dynamic-heatmap.html?dataset=/nbserver/KumamotoEarthquakes2016/all_geotagged_kumamoto_earthquakes.geojson)\n",
    "### [Just Kumamoto Tweets](http://research.yetilabs.co/nbserver/yeti-sites/pages/maps/geojson-animator.html?dataset=/nbserver/KumamotoEarthquakes2016/in_kumamoto.geojson)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Visualize: When these were tweeted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline\n",
    "time_format = \"%a %b %d %H:%M:%S %z %Y\"# \"Wed Apr 13 09:58:03 +0000 2016\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4cAAAINCAYAAACNsQqcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff0df3af940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "times = [datetime.datetime.strptime(t['created_at'], time_format) for t in in_kumamoto]\n",
    "pylab.figure().set_size_inches(15,8)\n",
    "pylab.title(\"Tweets per day for tweets IN Kumamoto\", fontsize=30)\n",
    "pylab.xlabel(\"Date (Times in UTC)\", fontsize=18); pylab.ylabel(\"Number of Tweets\", fontsize=18)\n",
    "hist(times, bins=20)\n",
    "None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3AAAAH4CAYAAAALlg9LAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xd8FNX+//H3LElIo0hJAgkEkSKhlyhFNJRAKBZAI4oQ\nUBThIldsiOVrF7wCVlBUuGAXRJqAoEAUQSkCUgVEaYGEDoEQSjK/P/jtuEm2JCQQhvt6Ph55PDY7\nZ86cObt7Zj5zzpwxTNM0BQAAAAC47DmKuwAAAAAAgPwhgAMAAAAAmyCAAwAAAACbIIADAAAAAJsg\ngAMAAAAAmyCAAwAAAACb8CvuAgAAAMC7o0eP6p133lF2draqV6+u3r17F3eRgAty+vRpjRkzRmfO\nnFHFihU1aNCg4i6S7RDAAQAAXObKli0rPz8/Pf3003I4HCpXrpy6dOlS3MUCCqxkyZKKiopSUlKS\nDMOQn5+fHnjggeIulq0whBJAsfjtt9/kcDjkcDj0xBNPFHdxityZM2c0ZswYtWrVSuXKlZOfn5+1\nv3v37i3u4l1y9erVk8PhUFhYmNvl06ZNs+pn3Lhxl7h0+F9w8803y+FwqESJEsrIyCh0fg899JD1\nnd20aVMRlNC34cOH6/7771d2draSkpK0Z8+eS7JdoKj17t1bL7zwgkzT1MMPP6x169YVd5FshR64\ny8TOnTt19dVXF0lekyZNUp8+fYokL7uZN2+eVqxYIUn617/+pQoVKhRzieCLYRjFXYQid/bsWbVt\n21bLli2T9M8+GoZxRe5vfuR33wtbP7/88osWLFggSerVq5dq1KhRqPyuRPv27dMHH3wgSWrZsqXi\n4+OLuUSXh82bN2vKlCmSpC5duqhZs2b5Wu9S/6bHjRunXbt2af78+UpMTNSSJUtUokQJn+vVq1fP\nCjQ3bNigmJgYr+mfeOIJjRo1SpIUFBSkqVOnqnPnzoXfAVw2Tp48aX3G9erVU48ePS7p9p999ln9\n9ddfmjx5su644w799ttvCg0NvaRlsCsCuMvM/+rJXVGZO3euxo4dK8MwdMcddxDAoVj897//1bJl\ny2QYhmrXrq0HH3xQVapUsU6y/le/l6ZpXvRtLFu2TC+88IIMw1BsbCwBnBt79+616ujRRx/9nwrg\nvF1I2LRpk1UvFStWzHcAd6mVKFFCU6dOVevWrbV8+XI98cQTGj16tM/1XC8k+TJw4ECNHz9ehmGo\nVKlSmjVrlm688cZClx2XlxMnTuiFF16QJPXs2fOSB3CS9OGHH2r37t1atGiR+vfvry+//PKSl8GO\nCOAuE2FhYZoxY4bH5QsXLtQ777wjwzDUpk0bDRkyxGPaJk2aXIwiAsinefPmWa/nzJlTZL3rAC7c\n7Nmzi7sIRSY0NFRz5sxR8+bN9eabb+qmm27SLbfcUuh8s7Ky1LdvX3322WeSpKuuukrffffdZRvM\nomgUZ+eBn5+fvvnmG7Vq1UpTp07VTTfdpIEDBxZbeeyCAO4yERQU5LXxPXLkiPW6atWqRdJQA7g4\ndu/eLen8SRbBG4CLoXLlytq1a1eR5Xf27Fndeeed1sXkSpUqacGCBapbt26RbQNwp3Tp0lq/fn1x\nF8NWCOAAoIidPn1a0vmZtoDLzaUYympH/8v1curUKd122236/vvvZRiGoqOj9cMPP6h69erFXTRc\nRP/L33m7YxbKK9Bbb71lzYz1008/uU2zcOFCK43D4dDWrVvdpps8ebKVZvr06R63uW3bNj3xxBNq\n2rSpKlSooJIlS6py5crq2rWrJk+erOzs7HyVPSsrS5988ol69Oih6OhoBQcHq2zZsqpfv74efvhh\nbd++3e16gwcPlsPh0NixYyWdb5Scs965/iUmJuZZd8OGDXrooYfUqFEjlSlTRgEBAQoLC1NMTIw6\nduyoF154QX/88Ue+yu9O165dc8x8lp2drY8++khxcXGKiIhQcHCwatSoocGDB2vHjh35zrcwdX7o\n0KE8dbJ//349//zzatKkicqXL1/o2SFXrlyp3r17q2rVqgoKClJUVJRuvvlmzZo1K995ZGdna+HC\nhRo2bJhuuukmVapUSSVLllSpUqVUs2ZN9enTRwsXLvS4/tKlS6397N+/f762+dRTT1nrTJs2Ld9l\ndZ1FcePGjZKkgwcP5vkOupthMTs7W5MnT9att96qKlWqKDAwUOXLl1fTpk01fPhwnzPNuZvBcd26\ndRo4cKBq166tUqVKyeFwaO7cufneH6fDhw9r8uTJ6tu3rxo2bKiyZcsqICBAFSpU0PXXX6+nnnpK\nKSkpBc63qI0ePVoOh0OPP/64pPNtgPO35/p33XXXSTo/qsH53r333usx36ioKCvdU0895TbN6dOn\nFRgYKIfDoVtvvdVjXpmZmRo3bpw6d+6sqKgoBQYGqly5cmrWrJmefvpppaam5nt/f/vtNz300EOq\nX7++ypUrp6CgIFWtWlV33HGHx+H4zplfnXVgmqZGjRqVp44cDkeeWRpPnDihMWPGqG3btoqIiFDJ\nkiVVunRpVa9eXc2bN9fAgQM1e/bsCzoh7NatmxwOh4KDg3XmzBm3aZ555hmrbJGRkR7z6tu3rxwO\nh/z8/HT48OEcy3K3xU7O34+zLTRN0zqmuP5VrFjR574sWrRIPXr0UNWqVRUYGKhKlSqpR48e+vnn\nn/NTFcUiPT1dHTp0sIK32rVr6+eff/YavA0cONDn+UN+03711VfWcufkOmvXrtW9996r6tWrKzg4\nWNHR0UpMTNSaNWtyrGuapr788kvFx8crMjJSQUFBqlmzpp544gkdO3bMa7nOnDmjWbNmaciQIWrR\nooUqVqyogIAAlS1bVnXr1tWAAQPybM+dhIQE63vl/P5+/PHHateunSIiIhQaGqp69erp+eefz1Om\n/fv367nnnlPDhg1VpkwZlSlTRq1bt9bnn3/uc7tOixYtUt++fVWjRg2FhoaqdOnSuvbaazVgwACt\nXLnS7TppaWlyOByqXLmypH/q0V1b4Onz/e233zRw4EDVqVNHpUuXVkhIiK655hr16dNHP/zwQ77L\njwtkwhYmTZpkGoZhOhwOs1+/fl7Trl271kr7/PPPu03z9NNPW2kcDof5/vvvu02XlJRkGoZhlihR\nwjx06FCe5dnZ2eawYcNMf3//HPk5/wzDMA3DMBs3bmzu3r3ba7nXrVtn1qpVy1rHXV7+/v7mG2+8\nkWfdwYMH59lm7vUdDod5xx135FjvP//5j+nn5+cxvTO/+Ph4r2X3pmvXrlb+qampZtu2bT3uX0hI\niDlt2jSv+RVFnR88eNBa94477jB//PFHs2LFinnye/zxxy9on19++WWzRIkSHst37733mitWrLCW\ne9qOs+48fT7OZXfeeaeZmZnpNo969eqZDofDLFWqlHnixAmv5T537pxZqVIl0+FwmBEREea5c+fy\nvc9ff/11nrK5K/PYsWNzrPf333+b9evX9/q9Dw4ONsePH+912851x44da7711ltmQEBAnrzmzJmT\n7/0xTdNMTU31+D1zLV9QUJD52Wefec2rXr16pmEYZsWKFfO1DwU1atSofLUB1113nbVOo0aNTIfD\nYVarVs1tntu2bcuRT/Pmzd2mW7x4sZXOXftkmqa5aNEis3Llyl6/yyEhIeZXX33ldT8zMzPNvn37\nevyOOfe9ffv25tGjR3Osu2rVqnzVUYkSJcyTJ09a661bt86MjIz0+Vt0OBxmSkqK1/K78+abb1rr\nJycnu03TqlWrHNv5448/3KaLjo42HQ6H2ahRozzLXNti1/1z/e1628ewsLAc+Q0ePNhKu3HjRnPo\n0KFeP5N33nmnwHVTlJy/QWd5TdM0Dx06ZDZt2tR6v3Hjxub+/ft95vXggw9a62zZsqVQab/88ktr\n+fjx483x48e7bb+c5wDffPONaZqmmZ6ebnbu3DlPnTvru2bNmmZaWprHcl177bX5Or489thjXvcv\nISHByuPw4cNmp06dPJYpJibGqt9ly5aZYWFhHtM+9NBDXrebkZFh9ujRw+uxw+FwmAMHDsxzLEtN\nTbWW+2oLcn9mWVlZ5r/+9S+fbVC3bt1y/M5QtBhCeQVq0KCBypUrpyNHjmjx4sV67rnn8qRZvHhx\nnv8HDBiQJ11ycrIMw7Cu8ubWp08fffbZZzIMQ2FhYbrrrrvUuHFjhYSEaPfu3Zo6dap+/fVXrV27\nVm3bttXq1avdThG7Zs0axcXFKT09XYZhqF27durUqZOioqJ05swZLV++XB9//LHS09P16KOPKiAg\nQIMGDbLWHzBggOLj4/XBBx9o7ty5MgxDb731lqpWrZpjO65Xbp09O4ZhKDAwUHfeeadatmypChUq\n6PTp09q3b59Wr15tTUleFAYOHKjFixfr2muvVVJSkqpXr64DBw7oq6++0s8//6yMjAzdfffdWrhw\noVq1auU2j6Kqc6ddu3ape/fuOnr0qG6//XbFx8erbNmy2rVrl6666qoC7+N7772nZ5991prtrUeP\nHkpISFBISIg2bNigCRMmaNKkSTp69KjPvDIzMxUSEqK2bduqWbNmuvrqqxUUFKTU1FRt3rxZn376\nqdLT0zV16lSVLl3aunrrasCAARoyZIhOnjypzz//XPfff7/H7X377bdKTU2VYRjq27dvvqbmdmrZ\nsqXVSz1kyBDt2rVLpUqV0scff5wjXcOGDa3XBw4cUKtWraxtVqtWTf369VOtWrV0/PhxzZ07VzNn\nztSpU6esq9i+ehJnz56tBQsWKDQ0VElJSYqNjZW/v782bNjg9jfszblz55SVlaVq1aqpffv2qlev\nnvUct127dik5OVnz589XZmamkpKSVKVKFbVu3bpA2ygq3bp1U82aNTVjxgxNmjRJhmHo6aefzjP5\ngut3uk2bNvr999+1a9cu7dixQ9WqVcuR1rWdNE1Tv/32m06cOJHn9+Sark2bNnnKNm/ePN122206\nd+6c/Pz81LVrV+vqfHp6un766Sd98cUX1u8/JCTE7QOas7Ky1KlTJ6tdrlatmnr27Km6deuqZMmS\n2r59uz777DNt3LhRCxcuVOfOnfXTTz9Z3+MaNWpo+vTp2r59ux599FEZhqFbb71Vffv2zbOtoKAg\nSefvi+rWrZv27dsnwzDUvHlz3XbbbYqOjlaJEiV05MgRbd68WYsWLbrge1hc62zx4sW66aabcizP\nyMiwehKcky0sXrxYtWvXzpHu77//1q5du2QYhuLi4vK9/VatWmn69On65Zdf9Nprr8kwDN1///15\npsz3NiT65Zdf1pdffqkaNWqod+/eql27tjIyMjRr1izNnDlTkvTII4+odevWOdqA4rRv3z7Fx8dr\n06ZN1mc7d+5clSlTptjKtGDBAs2YMUPly5fX/fffr/r16yszM1MzZ87UzJkzde7cOfXu3Vvbt2/X\nAw88oO+++04tW7ZUYmKiKleurN27d2vcuHHavn27tm/frsGDB1uPhsgtMzNT5cqVU/v27dW4cWNV\nqVJFAQEBSklJ0dq1a/Xll19az/OsWLFivkakDBgwIE+Zdu3apffee09//fWX/vjjDw0cOFCjR49W\nQkKCTp8+rfvuu0+tW7dWYGCgli5dqg8++ECnT5/W2LFjlZCQ4PbRDaZp6uabb9aiRYtkGIZKly6t\nfv36qVmzZsrOztayZcs0efJkZWZm6v3331dGRoYmTZpkrX/VVVdpxowZOnr0qPr27SvDMHTDDTfo\n0UcfzbOtqKioHP/fd999mjx5sgzDUMmSJZWUlKRWrVrJz89Pq1ev1sSJE3X06FHNmDFDN998s9dR\nMiiE4o4gkT8F6YEzTdPs1q2baRiGGRgYmKdn4sSJE6a/v7/pcDjMVq1amYZhmOHh4Xny+Ouvv6xt\nPvzww3mWf/TRR9byxMREj70bI0aMsNK5u6KUmZlp1qhRwzQMwyxdurT5/fffu81n165dZs2aNa0r\n/u6u9Oa+IupNz549rbSLFy/2mO7cuXPmihUrvObljetVX8MwzB49ephnzpzJk+6ll16y0tWqVcvM\nysrKk6ao6ty1B87Zw/PDDz9c8D46paSkmKGhoabD4TD9/f3NGTNm5Elz+PBhs0mTJjmu+nnqgUtO\nTjYzMjI8bu/YsWNmhw4dvH7mR48eNUNCQkyHw2HGxsZ6LX+XLl2svLZt2+Zjbz3z1dvkdPvtt1vb\nu+WWW8xTp07lSTN16lTTz8/PdDgcZkhIiLljx448aVx7r5xXnnft2nXB5XdKT083f/rpJ69pfvnl\nF7Ns2bJ5erdyu9g9cE6jRo2y8vHV4zhz5kwr7cSJE/Msv+uuu0yHw2G2aNHCuro8d+7cPOluvPFG\n0zAMs3z58nmW7d+/36xQoYLpcDjMyMhIc82aNW7Lsm7dOrNixYqmw+Eww8PD3X4XnnnmGau8Q4YM\nMc+ePZsnTXZ2tjlo0CAr3ejRo/OkWbVqlc/fntN3332X72PP77//fsFX3MuXL286HA7zxhtvzLNs\n/vz5VhlatmxpGoZhJiYm5kk3ceJEK527tsdTD5xTQb+Drscb50gAd22764iXpKQkn/leLK49cN9+\n+615zTXXWP+3b9++QJ/dxeqBMwzDbNasmXn48OE86YYMGWKlc46seOmll/KkO3z4sFm1alXTMM6P\nHvLUFi5YsMDrKIuUlBSzQYMGpmEYZmhoaJ4ebSdnD5yzbO5GPh05csSsVq2a1atVt25ds0KFCuba\ntWvzpJ06daqVV7t27dxu07XXulatWuaePXvypNm4caMZHh5upXP2XLpy7Ym76667PNaF04wZM6z0\nYWFh5vr16/OkSUlJsUZTORwOc8yYMT7zRcERwNlEQQO4t99+20q/cOHCHMtcD8g//PCDx5PfCRMm\nWMtmzpyZY9m5c+esoSoNGjTwOdTMOcwhJCTEPH78eI5l48ePt7bja/iQ67C7p59+Os/yggRwzZo1\nMw3DMGvUqOE1XWG5DgOMjIz0Ooyvffv2Hhvboqxz1wDO4XCYI0eOvPAddOF6ojJs2DCP6bZt22aW\nLFnS2v6FDtU0TdPct2+fFeAMHz7cbZp7773XKtfvv//uNs2ePXusfNq0aXPB5THN/AVwf/75p7X/\nkZGRZnp6use0Tz75pFX+oUOH5lnuPPF0pvn1118LVf6Cev31132ezF2OAdyRI0fMEiVKmA6Hw+zd\nu3ee5c7htKNGjTIbN27s9rt66tQp67vcrVu3PHkMHz4835/LlClTrLQffvhhjmWHDx+2Lo506tTJ\naz7Z2dnWsNwqVarkWV6QAO7dd9+10v78889e0xZG9+7drYuOuYNX5/e/Xr161rEt93BG0zTN3r17\nWyftR44cybP8YgVwhmGY1atX9ziMOzMz0yxbtqxpGIYZERHhM9+LxTWACw0NtV7feuut5unTpwuU\n18UI4AzDMAMCAsy//vrLbT4HDhyw2miHw+H11oYxY8ZY2/zoo48KtG+ufv31VyufCRMmuE3jOoTS\nU8Blmqb5xhtv5Djufvrppx7TOgNHf3//PN+rrKwsK0D18/PzeFHINE3z22+/tbbp7gJmQQM45wUU\nd+eFrn7//Xfrs4qMjCzQ7QjIHyYxuUK5Dh9xN1xSkurVq6d27dqpfPnybtMlJydLkhwOR54hLT/+\n+KM1ffGQIUN8DjW75557JJ2f6WrJkiU5ln366aeSznfTu5tkxFVsbKxq1qwpSYUe2hgcHCzp/M28\nuW92vxgMw9ADDzygkJAQj2lchy/knjSmKOvcyTRN+fn5uR0+eyGckyc4HA49/PDDHtPVqFFDt956\na5HMgBUREWFN1b98+XK3aR588EHr9Ycffug2zYQJE5SVlSVJRVYf3syYMcPa/4EDB3od5vroo4/K\n399fkvTNN994TGcYhpo0aaLrr7++aAvrQ4sWLazXnj6Dy1HZsmXVsGFDmaaZp/37448/rElF2rRp\nozZt2rhNt2zZMmviAnfDJz/99FMZhqGWLVv6/Fy6detmtQ+527fp06fr5MmTkuR2mJMrwzB01113\nSZJSUlK0adMmr+m9cbaTkqzJeS4G5zHrzJkzWrp0aY5lixcvtp6B6kx38ODBPOVxDi11TrZzqTiH\nXHoaYlmyZEm1bNlS0vlJK1wfC3SpOYeguk7i8sgjjyggIKC4imQxDEMdO3b0+OiVChUqqFatWjna\nTU9uuOEG63Vhvv+xsbHWsTY/bZvrrR25OW+LME1TV111lfUbdcdZ/qysrDyTiKxcuVK7d++2bjdp\n1KiRx3y6dOmiunXrWkPAfU2I5U1aWpp++eUXGYahWrVqeX2cVYMGDZSQkCDTNLVv3z798ssvF7xd\nuMc9cFeo+vXrq3z58jp8+LDbwMx5MJSkm266Sd98840WL16sf/3rX3nSNWjQIM+YeNeA4Pjx49YY\nf09cn1WzefNma0z3mTNntGLFChmGoYiICJ/5SOdPKEzT1ObNm32m9SY+Pl5LlizRyZMn1bp1aw0f\nPlw333zzRR3/365dO6/Lb7rpJjkcDpmmmWf2qKKqc1eGYahu3bpFcrJz8uRJ/fHHHzIMQ9dee60i\nIiK8pm/Xrp2mTp3qM9+MjAx98cUXmj17ttavX68DBw7o5MmTboM/Twen2NhYNWrUSGvXrtVnn32m\nUaNG5TjZMk1T//3vfyVJ5cqVU/fu3X2Wq7BWrFhhvY6Pj/eatkKFCmrUqJF14E5LS1N4eLjbtBfj\nHrRt27Zp0qRJWrJkibZu3apjx45Zj0rIrTAnCMWhTZs2WrNmjfbu3as///xTNWrUkPTPBawyZcqo\ncePG2rt3r9544w2tXbtWx48fV+nSpSXlvPCV+76rnTt3as+ePTIMQ2XLls1X+1a6dGllZGTkad9c\nf//79u3zmdfBgwet15s3b1ZMTIzPbbvTtm1blShRQllZWRo6dKh2796te+65J8/9Z4WV+z44Z1t5\n4sQJrV692ipL7mOb8xll27dvt+q6IPe/FZXmzZt7Xe56//WRI0cu6P7iomCapgzDUL169bR+/Xrr\nXqr58+f73IdLwddFjvDwcOu34ZxN1VM6J28B86FDh/TJJ59o/vz52rhxow4fPpxnBlbp/LEyP22b\nt/I7y2QYhho3biyHw3Mfirfyux47OnTo4LNMHTp0sC52LF++PM89bflVkGOWc7tz5syxtusaVKPw\nCOCuYM7AbOXKlcrIyFBwcLDS09P122+/SfrnZKNNmzb65ptv9OOPP1rruh4M3V1Vdp3u3tfV4Nxc\nG6N9+/bpzJkzMgxDK1euVLdu3fKdT0ZGhs6ePWv1TBTU0KFDNXPmTK1evVqbN29Wnz595HA41LBh\nQ7Vs2VJxcXHq1KlTjivQheU8OfQkKChIlStX1p49e7R3794cy4qqznPzNiV3QaSmpio7O1uGYfjc\nT8l3XUjSL7/8osTERGuaeufV49yvpfMnJsePH/eY14MPPqgHH3xQx44d09SpU60eSkn6/vvvtXPn\nThmGoT59+lzwd6og9u3bZ72uVauWz/S1atWygvp9+/Z5DOCK6vN0ev755/Xqq6/q3Llzkv6pd3f1\nL8nrZ3A5iouL05gxYySdn47b+b10BmY33nijDMPQjTfeqBIlSig7O1vJycnW1WdnoFe+fHnVr18/\nR96uv9k5c+ZYJzP5kfs365pXnz598p2Pu7wKIjo6Wi+++KKeeeYZZWZm6pVXXtErr7yiypUrq1Wr\nVmrdurW6dOlS6AfW16tXTxUqVNChQ4e0aNEi6/0lS5bo3LlzcjgcuvHGGyWd/8ymTZumRYsWafDg\nwZL++Rwk9z2hF1uFChW8Lne9YJSZmXmxi+PTpEmT9Mwzz2jevHlKT09Xp06d9N13313y3vvcnCOC\nPHGtR29p81Pf06dPV//+/a3fh6e2zSk/bVt+y1SQ/cxd/gs5drhbt6CKa7twjyGUVzDnQezcuXPW\n1dslS5YoKytLhmFYwyKd6Q4fPqzff/9dku9Z1VyfZWL8/9kG8/MnKcdzfi40H+ff2bNnL7h+QkJC\ntGTJEr3wwguKjIyUYRgyTVNr167V2LFjdccddyg8PFxPPvlkkR1w8xMMOodQOYdLORVVnefmnG2u\nsE6cOGG9Lsh+erJnzx4lJCRo7969VlD40EMP6d1339UXX3yhb775RtOnT9f06dNVpUoVSbKGQLpz\n9913W8MUcw+jdP0/v8+LK6z09HTrta+6kJRjiKXrurkV1ecpnX+m5IsvvqisrCw5HA7Fx8frueee\n08SJE/XVV19Z9T9q1ChrHW+fweXIGZhJOdu93CMVSpcurcaNG+dId+rUKWsEQe5h5lLh2rfcbdvF\n+v3nx/DhwzVnzhy1atXKynffvn2aOnWqhgwZomuuuSbHVf4LddNNN1lDvZztn7OuXWdCdn4mrs85\ndaZzDfQuJW+9KZejkiVL6ptvvrF6cI4dO6aEhASPzw27VApSj4Wp8+XLl+vOO+/U0aNHZRiGGjVq\npMcee0zvv/++vvjiC6ttmz59uvz8zvd15Kdty2+ZClP2i3XsuFy3C/fogbuC5b4PrmPHjtZBrkGD\nBtYQjpiYGFWsWFEHDx7U4sWL1bBhQ+tqZokSJdweDF1/mOvXr7/g4Tmu+XTv3j1fQ+qKUmBgoJ55\n5hk988wzWr9+vZYuXaqlS5dq4cKFSktLU0ZGhv7zn//o119/te7DKIyMjAyVKlXKaxrniUvuBrKo\n6vxicS2fuyEoueUOUHP7z3/+Yz1WYujQoTmChNy83W/nWr5evXpp/Pjx+vnnn7V161bVqlVLBw4c\n0KxZs2QYhlq0aKE6der4zKsouH4PTp486XMYq2uA7Os7VBSysrL00ksvyTAMBQUFaeHChR6vzrue\nSNtN6dKl1ahRI/32229Wu7dx40YdOHAgzwiEuLg4rVq1ympHly5dqrNnz3ocqeD6m3jkkUf0+uuv\nX3A5c58MFeXIgPxISEhQQkKCDhw4oCVLlmjZsmVKTk7W2rVrZZqmfvjhBzVv3lw///zzBU+T7+xZ\nc150TEhIyHH/m2s66XzP4tq1a9WoUSMr4G7cuLE1vBXelSxZUjNnztStt96qBQsW6NixY+rYsaO+\n//57NW2/zfkqAAAgAElEQVTatMi2k52dXWR5FZUXXnhB586dk2EYeueddzzeu5adnW2NPrhc5D52\n+FJUx47i2i7cs9clIxRI3bp1VbFiRUn/XJ10PjMk98lGXFxcjhv0nQfDRo0auf3huQ7T2r179wWX\nMSIiwrr6Xdz3ztSvX18PPvigPvnkE+3du1dz5syxnnm1ZMmSIgku//zzT6/LMzMzrectVa5cOcey\noqrziyUiIsK6quhrP/OTxvnsmNKlS2vkyJEe0509ezbfwzNcJzP56KOPJJ0fSuTs7bgUk5c4VapU\nyXq9bds2n+ld0+T+blwMq1evtib3SUpK8jq0aufOnRe9PBeTsz3cv3+/Nm/ebLWD5cqVU4MGDfKk\n27BhQ577i93dd+X6my1s+3a5/P4rVqyo7t27a9SoUVq1apW2b9+url27Sjp/4Wb48OEXnHfu++CO\nHTumNWvW5FkWExNjtc2LFy/Wtm3brCHnxXH/m52VLFlSM2bMUPv27SVJR48eVYcOHaz7Dr2t5+Sr\nh9f1fszLhfPCQL169bxOPLJz584imWyrKBXXseNyP2b9ryGAu8I5h6SsWbNGO3futIZItm3bNkc6\n1yEpf/zxh3Uw9HQvgetwocLMBhkcHKymTZvKNM0cJ4wXynVYQmEb3YSEBL322mvW/z///HOh8pOU\n494Od5KTk61hGrGxsTmWFVWdXywhISGqU6eOTNPMMYOfJ74e7pmWlibDMFSlShVrCIs7ycnJHifU\nyK1hw4a67rrrZJqmJk+erDNnzmjChAmSzgeKd9xxR77yKQquN+B///33XtMePHhQa9asserDefJ6\nMaWlpVmvr7nmGq9p58+ff7GLk28X0ga4nvQvWrTICsxyBwOtW7eWn5+fdbHL2WNXsWJFtz3itWvX\nVnh4uEzTVHJycqGu5Bfl778o28no6Gh99dVXCgoKkmmahWonnaNBpPMn2D/99JOys7PdzoTsetGx\nqO5/K8p6sZPAwEDNmjXLOi84cuSIOnTooLVr13pcx3XEQO77tXNznfzicnD06FHrmOGrbfvuu+8u\nRZEKpCDHDilne5F74peCfOeLcrsoPAK4K5zzYJaVlaWXX35Z2dnZbodFOtMdP35co0ePzvN+bm3b\ntlVUVJRM09SECRMKdUU4KSlJ0vl79V588cULzkfKOcwoP138vlSrVs16XdhhFKZp6oMPPtCpU6c8\npnFOpiApz0yIRVnnF4tzEprs7Gy99dZbHtNt375dM2fO9Dok1Tnb6I4dO7weWF5++eUCldHZC3fw\n4EE98sgj2rp1qwzD0D333KPAwMAC5VUY3bp1sw6e7733ntdhp2PGjLF6CW+//fZLUj7XIXrbt2/3\nmG7z5s36+uuvCz28uKhcSBvQunVrayTAwoUL9dNPP7kdqRAaGmoNLZs9e7ZWrlzpc9ZD54Qj+/fv\n17vvvluQXcnh9ttvt34To0ePLtRkMUXdTgYGBlqT6hS2nXQGZmvWrLEeS+JuWKTzs1myZIl++OEH\nSeeH/BdmFtairhc7CQwM1OzZs3PcEx8fH29d9M3N9YKFtwuT06ZNsyahuly43ifsrW07deqURo8e\nfdm0bU6xsbGqUqWKTNPU999/r3Xr1nlMO3fuXG3YsEGGYSg2NjbPDJQF+c6Hh4erZcuWMk1TW7Zs\n8Top07p16zRv3jxJ53vuXB81g6JBAHeFcz2xmDx5sqTzB8PcwyJr165tTfvuTOftYBgQEKCXXnpJ\n0vn7MTp06ODzWSurVq3S0KFD87x/3333Wc92eeedd/TSSy95HTN/4sQJjRs3TnPnzs2zzHUmNF9D\nQAYPHuy14TNNU++//771/4Xe1+FkGIb27t2rpKQkt5OvvPLKK9aJiLtnrBRlnV8sDz74oEJCQmSa\npsaMGeN2qvPDhw8rMTFRZ8+e9RqYOXsgT548qVdffTXP8qysLA0ePFhLliwp0AG2Z8+e1tXjcePG\nWe9fqslLnKpXr64ePXrINE3t3btXvXr1cjtZzrRp06z7/4KCgvTQQw9dkvI1btxYJUqUkGma+uST\nT7Rly5Y8aXbu3KnbbrvN52d5KRWkDXAqVaqUmjRpItM09e233+rQoUOS3F/Acj4P7vPPP7eCFW8B\n3OOPP64KFSrINE0NGzbM43MInQ4dOqSRI0dq1apVOd6vUKGCHn/8cUnnh1B26tTJ54nx4sWL9dxz\nz+V5v2rVqlbA6quOPvjgA02ZMsXrhFHz5s2zZsksbDvprMvs7Gx98sknHu8vdL3o6HxmZpMmTbw+\nT9GXC/nuXEmCgoI0e/Zsq7fz0KFDio+Pd3ucbNeunfz9/WWapsaPH6+///47T5q1a9dq0KBBl10A\nVLJkSdWvX1+maWrDhg366quv8qQ5deqUevbsqb/++qsYSuidw+HQY489Jun878R1pmZXmzZt0n33\n3Wf9/+STT+ZJExQUpLCwMJmm6TFYdzVs2DBJ58+P+vfv73bior1796pnz57WhHmPPfaYz+fWouCY\nxOQKV6dOHYWHh2v//v3Wj8lTr1qbNm30xRdfWOmaNm3qdaahpKQkrVq1SuPGjdOWLVvUsGFDdenS\nRXFxcapcubJM09TBgwe1fv16LVq0SH/++aciIiL0xhtv5MgnICBAM2fOVOvWrXXo0CFrlrvbb79d\n9erVU2hoqE6cOKG///5bK1eu1KJFi3T69OkcwZWTcwiIaZp69tlnlZ6erjp16ljTwkdERFgnGJMn\nT9a4ceN07bXXqk2bNtbzhU6dOqUdO3boq6++sp43Ex0drV69ehX8A8jltttu09dff63169erX79+\nuvrqq3XgwAFNmTLFmgjC399fEyZMcDtLVVHV+cVSuXJlvf766xo0aJDOnj2r7t27q0ePHkpISFBI\nSIg2btyoDz74QAcOHFC3bt28PpR68ODBmj59uvVZLlu2TJ07d1bFihX1119/6eOPP9Yff/yhFi1a\n6ODBg/kaky+dv9Lcu3dvvfPOO9bMo82aNSv0ieeFGDt2rJYuXWo916tu3brq16+fatWqpePHj2vu\n3LlWL4RhGHrzzTcVHR19Scp21VVXqVevXvr444914sQJNWvWTA888IAaNmwowzD066+/6uOPP1ZG\nRob69OljXfgpbs2bN1dwcLAyMjL07rvvKjAwUE2aNLHu2SlTpozb5121adNGK1eutIYwh4WFuZ3Q\npk2bNho5cmSOGem8DdurUKGCpk2bpk6dOunUqVMaMGCA3nrrLXXr1k21a9dWUFCQjh07pj///FPL\nly+3hg02adIkT17/93//pzVr1mj27Nn65ZdfVKNGDXXv3l2tWrVSWFiYzp49q7S0NK1bt04LFizQ\n3r171bx5c73wwgs58nE+WHrJkiVauXKl7r33XnXq1ClHL1eHDh1kGIbWr1+vsWPHqkyZMurQoYOa\nNm2qqKgo+fn5KS0tTQsXLrSuxBuGoaefftrbx+OTa106J5lwV7+1atVSpUqVlJqa6vPYll81atRQ\ndHS0du7cqWnTpunRRx/VDTfcYPVGBwQEFMsjCi6l4OBgffvtt+rcubOWLFmigwcPKj4+XosWLbKe\nuSednwa/b9+++vDDD3Xs2DHFxsZq0KBBql+/vk6ePKkff/xRn3/+uUqXLq34+PjLbtj/Qw89pAce\neECmaeruu+/WrFmzdNNNN6lMmTLatGmTJk6cqL179+rOO+/UjBkzCj2Ta1EbPHiwZs2apUWLFmnr\n1q2qV6+e7r33XjVt2lTZ2dlatmyZJk2apMzMTOvxOLfddpvbvNq1a6cvvvhCe/bsUffu3ZWYmJjj\nOYU33nij1Wt58803q1+/fpo0aZLS0tLUrFkz9enTRzfccIP8/Py0Zs0affTRR9bsnm3atNG///3v\nS1In/3NM2MKkSZNMwzBMh8Nh9uvXr0Dr3nnnnaZhGNb63333ndt0H374YY50w4cPz1f+Y8aMMUND\nQ02Hw2Gtn/vP4XCYDofDvO666zzms2PHDrNFixZWWm/5BAYGmtOnT3ebz4ABAzzmcccdd1jpSpUq\nla9t1a1b1/zzzz/zVRfudO3a1covLS3NbN++vdttOhwOMzQ01Pzmm2985lnYOj948KCVxrVOisor\nr7xi+vn5edzPBx54wFy5cqX1/+OPP+4xH0+fkcPhMK+//npz3759Zr169UzDMMyKFSvmq3ybNm3K\nUUcfffRRUe5+gcqzY8cOs2HDhl73Mzg42Pzggw885vH1119baceOHVtk+3Hs2DHzuuuu81i2EiVK\nmE8//bS5atUqn5+lrzopyn149dVXPZY5NjbW7Trz5s3Lka5nz55u0508edIMCAiw0lWuXDlfZVq7\ndq1Zp06dfLU5pUqVMlesWOE2n6ysLHPYsGFmQEBAvn7/iYmJbvP58ccfzZIlS3r8zp08edI0TdN8\n6KGHvJbZmT4kJMT88MMP81UXvkRERFh5BwQEmCdOnHCbrlevXjnKMH/+fK/5urbFzv3L7bPPPjNL\nlCjhdn9zf3cHDx5s5bdx40av2y5I2ovF+RvMTxlOnDhhtm7d2tr3sLAwc8OGDTnSHDt2zLz++us9\nfj8qVapkLl261HzwwQet7W7ZsiXPtr788ktr+fjx472WKyEhwUp7+vRpj+lSU1OtdHfddZfbNP37\n9/fa7nbt2tXMyMgwAwMDTcMwzBYtWlz0MjmNHDnS5/f61KlT5u233+51HxwOhzlo0CAzKyvL47Y2\nbdpkli5d2uNvPPdnlp2dbQ4ePNjj78S53e7du5sZGRle9xMXjiGUNmL4eMikJ23atLGe3ePv7+9x\nWKRrOsPHfR2uhg4dqh07dujVV19Vu3btVKlSJZUsWVJBQUGKiopSu3bt9NRTT2nJkiVavny5x3yi\no6O1bNkyzZs3T/fee6+uvfZalS1bVv7+/ipbtqwaNGigXr16acKECUpJSfF4Nen999/XhAkT1K5d\nO4WHh8vf3z/Pc5Gk82PfJ02apPvuu0+xsbGqUKGC/P39FRQUpOjoaN1yyy2aPHmyfv/9d583OudX\naGioFixYoA8//FA33nijwsLCFBgYqOrVq2vQoEFav359vh5mXhR1fqHfp/x46qmn9Msvv+iuu+5S\nZGSkSpYsqcqVK6tLly6aOXOmxo8fn+fz8JTPokWLdMsttygsLEwBAQGqVKmS4uLi9N577+nnn3+2\nhv7mJz+nOnXqqFKlSjJNUyEhIerZs2eh9zm3/JYnOjpaa9as0aRJk9SlSxervq666io1btxYw4YN\n09atW3X//fcXyfYKonTp0lqyZIlGjRqlZs2aKTQ0VMHBwapevbr69OmjH3/80boHMT/b9lXGotqH\n4cOHa9q0aTnq010b4Kp169Y52orcEz05BQcH6/rrry9wO9mwYUNt3LhRX331le6++27VqFFDpUqV\nkr+/v8qVK6emTZvq3nvv1eeff659+/blmcTIyeFwaOTIkdq2bZueeeYZtWrVSuHh4QoICFBwcLCq\nVaumTp066cUXX9Tq1avdDg+Tzl9V//XXX9W7d2/VqFFDwcHBbutozJgxWrBggYYPH6727duratWq\nCgoKkr+/vypUqKBWrVrp2Wef1ZYtW4psGHJcXJxVjmbNmnkcCZL72HbDDTf4zNvXd+zuu+/WwoUL\ndfvttys6OlpBQUFevzsF+c5ejN9oQeW3DCEhIZo7d6717L+DBw+qffv21qgU6Xz78NNPP1ntQ6lS\npRQSEqKYmBg99dRTWrt2rVq2bJmv7V6MevR1jPvwww81ZcoUtW/fXuXKlVNAQICioqLUuXNnffHF\nF5o9e7bV81RU5S/IcddXnoGBgZo6dap++OEH3XPPPapevbqCg4MVGhqqWrVqqX///lq+fLnGjh3r\n9blzderU0erVqzVgwADVqVNHoaGh1rbdrWf8/0cvrFixQvfff79q166t0NBQBQUF6eqrr1avXr00\nf/58TZs2rUifS4qcDNP0fePCsWPH1L9/f23YsEEOh0MTJ05UrVq1dOedd2rnzp2qVq2apkyZojJl\nykiSRowYoYkTJ8rPz09vvfWW9aDI1atXq2/fvsrMzFTnzp315ptvXty9Ay4DN998s+bMmSPDMIrl\n+U3Ia/ny5WrRooUMw1D//v01fvz44i4SAABAvuSrB+7f//63OnfurM2bN+v333/Xtddeq5EjR6p9\n+/basmWL2rZtqxEjRkg6f9PklClTtHnzZs2bN0+DBg2ybm4fOHCgJkyYoK1bt2rr1q2X1dTTAP53\nuE5e8sADDxRjSQAAAArGZwB3/PhxLVmyRP369ZMk+fn5qUyZMpo5c6Y1/XtSUpJ1o/2sWbPUs2dP\n+fn5qVq1aqpZs6ZWrFih1NRUpaenW8NC+vTpY60DAJfKli1b9OWXX8owDLVo0cKaFh4AAMAOfM5C\n+ffff6tChQrq16+ffv/9dzVr1kxvvvmm0tLSrOe+REREaP/+/ZKklJSUHM97iIyMVEpKivz8/HI8\nfyIqKuqyezYIgCvTwoULdebMGW3fvl2jRo3S2bNnZRiGnn/++eIuGgAAQIH4DODOnTun1atXa+zY\nsWrWrJmGDh2qkSNH5rmxsihvzC3um3yBi8E5YQYuD6ZpKj4+vriLAQAA4JG76Up8DqGMiopSlSpV\n1KxZM0lSjx49tHr1aoWHhystLU2SlJqaqrCwMEnne9x2795trb9nzx5FRkZ6fN9bYYvq77nnnivS\n/Pijbgvy17VrVzkcDpUoUUInT56kXovhr1SpUnI4HAoICFDjxo31/vvvKzs7u9jLdaX88X2lbu32\nR71Sr3b6o17/d+vWE58BXHh4uKpUqaKtW7dKOj8UqW7durrllls0adIkSecfiHzrrbdKkm655RZ9\n+eWXOnPmjP7++2/9+eefuu666xQREaEyZcpoxYoVMk1TH3/8sbUOcCWbPXu2srKydO7cOWagLCbH\njx9XVlaWhg8fbk2XTE8/AACwI59DKCXp7bffVq9evXT27FlVr15d//3vf5WVlaXExERNnDhR0dHR\nmjJliiQpJiZGiYmJiomJkb+/v8aNG2edKI0dOzbHYwQSEhIu3p4BAAAAwBUmXwFcw4YNtXLlyjzv\n//DDD27TDx8+XMOHD8/zftOmTbV+/foCFrHw8vugVRQcdXtxUK8XB/V6cVCvFw91e3FQrxcH9Xpx\nUK8Xj13rNl8P8r7UDMPwOu4TAAAAAK5knmKifD3IGwAAAABQ/AjgAAAAAMAmCOAAAAAAwCYI4AAA\nAADAJgjgAAAAAMAmCOAAAAAAwCYI4AAAAADAJgjgCiEiopoMw7is/yIiqhV3NQEAAAAoIjzIuxAM\nw5B0uZfTHnUJAAAA4B88yBsAAAAAbI4ADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAO\nAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4A\nAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAA\nAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAA\nAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAA\nbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABs\nggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyC\nAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbIIADgAAAABsggAOAAAAAGyCAA4AAAAAbCJf\nAVy1atXUsGFDNW7cWNddd50k6ciRI+rQoYNq166tjh076tixY1b6ESNGqGbNmqpTp44WLFhgvb96\n9Wo1aNBAtWrV0sMPP1zEuwIAAAAAV7Z8BXAOh0PJyclas2aNVqxYIUkaOXKk2rdvry1btqht27Ya\nMWKEJGnTpk2aMmWKNm/erHnz5mnQoEEyTVOSNHDgQE2YMEFbt27V1q1bNX/+/Iu0WwAAAABw5clX\nAGeaprKzs3O8N3PmTCUlJUmSkpKSNGPGDEnSrFmz1LNnT/n5+alatWqqWbOmVqxYodTUVKWnpys2\nNlaS1KdPH2sdAAAAAIBv+QrgDMNQfHy8YmNj9dFHH0mS0tLSFB4eLkmKiIjQ/v37JUkpKSmqUqWK\ntW5kZKRSUlKUkpKiqKgo6/2oqCilpKQU2Y4AAAAAwJXOLz+Jli5dqkqVKunAgQPWfW+GYeRIk/t/\nAAAAAEDRylcAV6lSJUlSxYoVddttt2nFihUKDw+3euFSU1MVFhYm6XyP2+7du6119+zZo8jISI/v\ne/L8889br+Pi4hQXF1eQ/QIAAAAA20hOTlZycrLPdIbpnGHEg4yMDGVnZys0NFQnT55Uhw4d9Nxz\nz2nhwoUqV66chg0bptdee01HjhzRyJEjtWnTJvXq1UvLly9XSkqK4uPjtW3bNhmGoebNm+vtt99W\nbGysunTpoiFDhighISFvoQxDPop1WTjf63i5l9MedQkAAADgH55iIp89cGlpaerWrZsMw9C5c+fU\nq1cvdejQQc2aNVNiYqImTpyo6OhoTZkyRZIUExOjxMRExcTEyN/fX+PGjbOGV44dO1Z9+/ZVZmam\nOnfu7DZ4AwAAAAC457MHrjjQA1eU7FGXAAAAAP7hKSbK1yyUAAAAAIDiRwAHAAAAADZBAAcAAAAA\nNkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2\nQQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZB\nAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEA\nBwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAH\nAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcA\nAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAA\nAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAA\nADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAA\nNkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADaR7wAuOztbTZo00S233CJJ\nOnLkiDp06KDatWurY8eOOnbsmJV2xIgRqlmzpurUqaMFCxZY769evVoNGjRQrVq19PDDDxfhbgAA\nAADAlS/fAdxbb72lmJgY6/+RI0eqffv22rJli9q2basRI0ZIkjZt2qQpU6Zo8+bNmjdvngYNGiTT\nNCVJAwcO1IQJE7R161Zt3bpV8+fPL+LdAQAAAIArV74CuD179mju3Lnq37+/9d7MmTOVlJQkSUpK\nStKMGTMkSbNmzVLPnj3l5+enatWqqWbNmlqxYoVSU1OVnp6u2NhYSVKfPn2sdQAAAAAAvuUrgBs6\ndKhef/11GYZhvZeWlqbw8HBJUkREhPbv3y9JSklJUZUqVax0kZGRSklJUUpKiqKioqz3o6KilJKS\nUiQ7AQAAAAD/C3wGcHPmzFF4eLgaNWpkDYV0xzW4AwAAAAAUPT9fCZYuXapZs2Zp7ty5OnXqlNLT\n09W7d29FRERYvXCpqakKCwuTdL7Hbffu3db6e/bsUWRkpMf3PXn++eet13FxcYqLi7uA3QMAAACA\ny19ycrKSk5N9pjNMb91qufz4448aPXq0Zs2apSeeeELly5fXsGHD9Nprr+nIkSMaOXKkNm3apF69\nemn58uVKSUlRfHy8tm3bJsMw1Lx5c7399tuKjY1Vly5dNGTIECUkJOQtlGF47e27XJzvdbzcy2mP\nugQAAADwD08xkc8eOE+efPJJJSYmauLEiYqOjtaUKVMkSTExMUpMTFRMTIz8/f01btw4a3jl2LFj\n1bdvX2VmZqpz585ugzcAAAAAgHsF6oG7VOiBK0r2qEsAAAAA//AUE+X7OXAAAAAAgOJFAAcAAAAA\nNkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2\nQQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZB\nAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEA\nBwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAH\nAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcA\nAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAA\nAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAA\nADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcAAAAA\nNpEs+AQAAB45SURBVEEABwAAAAA2QQAHAAAAADZBAAcAAAAANkEABwAAAAA2QQAHAAAAADZBAAcA\nAAAANkEABwAAAAA24TOAO336tK6//no1btxYdevW1VNPPSVJOnLkiDp06KDatWurY8eOOnbsmLXO\niBEjVLNmTdWpU0cLFiyw3l+9erUaNGigWrVq6eGHH74IuwMAAAAAVy6fAVzJkiW1ePFirVmzRuvW\nrdOiRYu0dOlSjRw5Uu3bt9eWLVvUtm1bjRgxQpK0adMmTZkyRZs3b9a8efM0aNAgmaYpSRo4cKAm\nTJigrVu3auvWrZo/f/7F3TsAAAAAuILkawhlcHCwpPO9cdnZ2brqqqs0c+ZMJSUlSZKSkpI0Y8YM\nSdKsWbPUs2dP+fn5qVq1aqpZs6ZWrFih1NRUpaenKzY2VpLUp08fax0AAAAAgG/5CuCys7PVuHFj\nRUREKC4uTjExMUpLS1N4eLgkKSIiQvv375ckpaSkqEqVKta6kZGRSklJUUpKiqKioqz3o6KilJKS\nUpT7AgAAAABXNL/8JHI4HFqzZo2OHz+ujh07Kjk5WYZh5EiT+38AAAAAQNHKVwDnVLp0aXXu3Fmr\nVq1SeHi41QuXmpqqsLAwSed73Hbv3m2ts2fPHkVGRnp835Pnn3/eeh0XF6e4uLiCFBUAAAAAbCM5\nOVnJyck+0xmmc4YRDw4ePCh/f3+VKVNGp06dUseOHfXcc89pwYIFKleunIYNG6bXXntNR44c0ciR\nI7Vp0yb16tVLy5cvV0pKiuLj47Vt2zYZhqHmzZvr7bffVmxsrLp06aIhQ4YoISEhb6EMQz6KdVk4\n3+t4uZfTHnUJAAAA4B+eYiKfPXD79u1TUlKSTNNUdna2evfurXbt2qlx48ZKTEzUxIkTFR0drSlT\npkiSYmJilJiYqJiYGPn7+2vcuHHW8MqxY8eqb9++yszMVOfOnd0GbwAAAAAA93z2wBUHeuCKkj3q\nEgAAAMA/PMVE+ZqFEgAAAABQ/AjgAAAAAMAmCOAAAAAAwCYI4AAAAADAJgjgAAAAAMAmCOAAAAAA\nwCYI4AAAAADAJgjgAAAAAMAmCOAAAAAAwCYI4AAAAADAJgjgAAD4f+3df0zU9x3H8dcJtLbTqrX1\n6ACxQ34pvw4H/qitlfirtrXLtA5dp07NEtnmuqWrLlkz3TLRNTWxnfuVaru6idqtFRMr2qyea42K\nExetaJFUUFjB2VqnTvn52R+Mi5RDEI+7+xzPR0LCfe/7/d773sD3ey8+3x8AAFiCAAcAAAAAliDA\nAQAAAIAlCHAAAAAAYInwQBeAnnanHA5HoIvolNMZq5qaikCXAQAAAAQ1hzHGBLqIL3I4HArCstpp\nCUbBXqcNNUqSHT9zAAAAwB86ykQcQgkAAAAAliDAAQAAAIAlgvYcuBMnTgS6hJsaMGBAoEsAAAAA\n0MsE7Tlw99wzItBl3FRdXaXq6q4q+M8v4xw4AAAAwDYdnQMXtAEu2ENHRER/NTRcUbDXSYADAAAA\n7MNFTAAAAADAcgQ4AAAAALAEAQ4AAAAALEGAAwAAAABLEOAAAAAAwBIEOAAAAACwBAEOAAAAACxB\ngAMAAAAASxDgAAAAAMASBDgAAAAAsAQBDgAAAAAsQYADAAAAAEsQ4AAAAADAEgQ4AAAAALAEAQ4A\nAAAALEGAAwAAAABLEOAAAAAAwBIEOAAAAACwBAEOAAAAACxBgAMAAAAASxDgAAAAAMASBDgAAAAA\nsAQBDgAAAAAsQYADAAAAAEsQ4AAAAADAEgQ4AAAAALAEAQ4AAAAALEGAAwAAAABLEOAAAAAAwBIE\nOAAAAACwBAEOAAAAACxBgAMAAAAASxDgAAAAAMASBDgAAAAAsAQBDgAAAAAsQYADAAAAAEsQ4AAA\nAADAEgQ4AAAAALAEAQ4AAAAALEGAAwAAAABLEOAAAAAAwBIEOAAAAACwBAEOAAAAACxBgAMAAAAA\nS3Qa4KqqqpSTk6ORI0cqNTVVL7/8siTp4sWLmjJlihITEzV16lRdunTJs0x+fr7i4+OVnJysPXv2\neKaXlJQoLS1NCQkJevbZZ3vg7QAAAABA6Oo0wIWHh2vt2rU6ceKEDhw4oPXr1+vUqVNavXq1Jk2a\npI8++kg5OTnKz8+XJJWWlmrbtm06efKkdu3apby8PBljJElLlizRhg0bVFZWprKyMu3evbtn3x0A\nAAAAhJBOA1xkZKQyMjIkSf369VNycrKqqqpUWFio+fPnS5Lmz5+v7du3S5J27Nih3NxchYeHa9iw\nYYqPj1dxcbFqamp0+fJlZWVlSZLmzZvnWQYAAAAA0LlbOgeuoqJC//znPzVmzBjV1tbK6XRKagl5\n58+flyRVV1crJibGs0xUVJSqq6tVXV2t6Ohoz/To6GhVV1f74j0AAAAAQK/Q5QB35coVzZo1S+vW\nrVO/fv3kcDjaPP/FxwAAAAAA3wrvykyNjY2aNWuWvvWtb+mpp56SJDmdTs8oXE1NjYYMGSKpZcTt\n3LlznmWrqqoUFRXV4fSOrbjh+0f//wUAAAAAocftdsvtdnc6n8O0XmHkJubNm6f77rtPa9eu9Uxb\ntmyZ7r33Xi1btkxr1qzRxYsXtXr1apWWluqb3/ymDh06pOrqak2ePFmnT5+Ww+HQmDFj9PLLLysr\nK0uPP/64li5dqmnTprUvyuGQ1GlZARUR0V8NDVcU7HVKwd/LFg514VcRAAAA6BUcDu+fjzsNcPv3\n79cjjzyi1NRUORwOORwOrVq1StnZ2Zo9e7bOnTun2NhYbdu2TQMHDpTUchuBDRs2KCIiQuvWrdOU\nKVMkSUeOHNGCBQt0/fp1TZ8+XevWreuw2GAPHQQ4XyPAAQAAAK26HeACgQDnS8HfyxYEOAAAAKBV\nRwHulq5CCQAAAAAIHAIcAAAAAFiCAAcAAAAAliDAAQAAAIAlCHAAAAAAYAkCHAAAAABYggAHAAAA\nAJYgwAEAAACAJQhwAAAAAGAJAhwAAAAAWIIABwAAAACWIMABAAAAgCUIcAAAAABgCQIcAAAAAFiC\nAAcAAAAAliDAAQAAAIAlCHAAAAAAYAkCHAAAAABYggAHAAAAAJYgwAEAAACAJQhwAAAAAGAJAhwA\nAAAAWIIABwAAAACWIMABAAAAgCUIcAAAAABgifBAFwC0uFMOhyPQRXTK6YxVTU1FoMsAAABAL0WA\nQ5Cok2QCXUSnamuDP2QCAAAgdHEIJQAAAABYggAHAAAAAJYgwAEAAACAJQhwAAAAAGAJAhwAAAAA\nWIIABwAAAACWIMABAAAAgCUIcAAAAABgCQIcAAAAAFiCAAcAAAAAliDAAQAAAIAlCHAAAAAAYAkC\nHAAAAABYggAHAAAAAJYgwAEAAACAJQhwAAAAAGAJAhwAAAAAWIIABwAAAACWIMABAAAAgCUIcAAA\nAABgCQIcAAAAAFiCAAcAAAAAliDAAQAAAIAlCHAAAAAAYAkCHAAAAABYggAHAAAAAJYgwAEAAACA\nJQhwAAAAAGCJ8EAXANjlTjkcjkAX0SmnM1Y1NRWBLgMAAAA+RoADbkmdJBPoIjpVWxv8IRMAAAC3\njkMoAQAAAMASBDgAAAAAsAQBDgAAAAAsQYADAAAAAEsQ4AAAAADAEgQ4AAAAALAEAQ4AAAAALEGA\nAwAAAABLEOAAAAAAwBKdBrhFixbJ6XQqLS3NM+3ixYuaMmWKEhMTNXXqVF26dMnzXH5+vuLj45Wc\nnKw9e/Z4ppeUlCgtLU0JCQl69tlnffw2AAAAACD0dRrgvv3tb2v37t1tpq1evVqTJk3SRx99pJyc\nHOXn50uSSktLtW3bNp08eVK7du1SXl6ejDGSpCVLlmjDhg0qKytTWVlZu3UCAAAAAG6u0wA3fvx4\nDRo0qM20wsJCzZ8/X5I0f/58bd++XZK0Y8cO5ebmKjw8XMOGDVN8fLyKi4tVU1Ojy5cvKysrS5I0\nb948zzIAAAAAgK7p1jlw58+fl9PplCRFRkbq/PnzkqTq6mrFxMR45ouKilJ1dbWqq6sVHR3tmR4d\nHa3q6urbqRsAAAAAep1wX6zE4XD4YjVfsOKG7x/9/xcAAAAAhB632y23293pfN0KcE6nU7W1tXI6\nnaqpqdGQIUMktYy4nTt3zjNfVVWVoqKiOpx+cyu6UxoAAAAAWOfRRx/Vo48+6nm8cuVKr/N16RBK\nY4znYiSSNGPGDL3++uuSpD/+8Y966qmnPNO3bNmi+vp6nTlzRuXl5crOzlZkZKQGDBig4uJiGWP0\nxhtveJYBAAAAAHRNpyNwc+fOldvt1qeffqqhQ4dq5cqVWr58uZ5++mlt3LhRsbGx2rZtmyRpxIgR\nmj17tkaMGKGIiAj95je/8RxeuX79ei1YsEDXr1/X9OnTNW3atJ59ZwAAAAAQYhzmxqG1INES+oKu\nrDYiIvqroeGKgr1OKfh72YI6fcuhIPzTBgAAQBc5HN4/z3XrKpQAAAAAAP8jwAEAAACAJQhwAAAA\nAGAJAhwAAAAAWIIABwAAAACWIMABAAAAgCUIcAAAAABgCQIcAAAAAFiCAAcAAAAAliDAAQAAAIAl\nCHAAAAAAYAkCHAAAAABYggAHAAAAAJYgwAEAAACAJQhwAAAAAGAJAhwAAAAAWIIABwAAAACWIMAB\nAAAAgCUIcAAAAABgCQIcAAAAAFiCAAcAAAAAliDAAQAAAIAlCHAAAAAAYAkCHAAAAABYIjzQBQDo\nCXfK4XAEuohOOZ2xqqmpCHQZAAAA1iDAASGpTpIJdBGdqq0N/pAJAAAQTDiEEgAAAAAswQgcgADi\nUE8AAIBbQYADEEAc6gkAAHArOIQSAAAAACxBgAMAAAAASxDgAAAAAMASBDgAAAAAsAQBDgDgN5GR\nw+RwOIL+KzJyWKBbBQCAVw5jTNBdAq7lsuJBV1YbERH91dBwRcFepxT8vWxBnb5Fnb7lUBBuKq1k\nw/a9BT9zAEBgORze90WMwAEAAACAJQhwAAAAAGAJAhwAAAAAWIIABwAAAACWIMABAAAAgCUIcAAA\nAABgCQIcAAAAAFiCAAcAgKW4MToA9D7cyLubuJG3r1Gnb1Gnb3FTZ1+xYfvewo6fOf0EgNDV0Y28\nwwNQCwAAQe7O/4cjAACCCwEOAIB26mTLyBYAoHchwAFAp4J/NMbpjFVNTUWgywAAAD2MAAcAnQr+\n0Zja2uAOmAAAwDe4CiUAAAAAWIIABwAAAACW4BBKAAgJwX+eHgAAuH0EOAAICcF/nl4LQiYAALeD\nQygBAAAAwBIEOAAAAACwBAEOAADAIpGRw+RwOIL6KzJyWKDbBIQshzEm6E6aaDkRP+jKaiMior8a\nGq4o2OtsOd8k2GuUqNPXqNO3bKjThhol6vQ1W+rsq5bzNIOb0xmrmpqKQJfRKRs+J0kOBeFHTMAq\nDof3vyMuYgIAAHqYHRfZqa3lIjsAgh8BDgAAAD5mx61NbBl1BW5EgAMAAJBkS+iwA6OuQE8hwAEA\nAEiyJXRwP0Wgd+MqlAAAAABgCUbgAAAA0EvZcdgs5+rhRgQ4AAAA9FJ2HDbLuXq4EQEOAAAACGp2\njBT26XO3mpv/G+gyOmX7iCYBDgAAAAhqdowUNjfbcJN5qba2rxWBuCN+v4hJUVGRkpKSlJCQoDVr\n1vj75QEAAAD0aq2BONi/vPNrgGtubtb3vvc97d69WydOnFBBQYFOnTrlh1d2++E1eit3oAsIUe5A\nFxCi3IEuIES5A11ACHMHuoAQ5Q50ASHKHegCQpQ70AWEMHegC+gWvwa44uJixcfHKzY2VhEREcrN\nzVVhYaEfXtnth9fordyBLiBEuQNdQIhyB7qAEOUOdAEhzB3oAkKUO9AFhCh3oAsIUe5AFxDC3IEu\noFv8GuCqq6sVExPjeRwdHa3q6mp/lgAAAAAA1grai5jcc8+TPlvX9esfqW/fIz5bnyRdvVrn0/UB\nAAAAQGccxhi/XSrm4MGDWrFihYqKiiRJq1evlsPh0LJly9oWZfFVYQAAAADAF7xFNb8GuKamJiUm\nJupvf/ubHnjgAWVnZ6ugoEDJycn+KgEAAAAArOXXQyjDwsL061//WlOmTFFzc7MWLVpEeAMAAACA\nLvLrCBwAAAAAoPv8fiNvSdq+fbv69OmjsrKy217XX/7yF6WkpCgsLEwlJSWe6ZWVlbr77ruVmZmp\nzMxM5eXleV3+mWeeUVJSktLS0rR48WI1NTV5nlu6dKni4+OVkZGho0ePeqYvWrRITqdTaWlp7db3\nyiuvKDk5WampqVq+fPltv79b4Y++bt68WS6XS5mZmXK5XAoLC9OxY8faLU9fveuorw0NDVq4cKHS\n0tLkcrm0b98+r8vTV++ef/55JScnKyMjQzNnztR//vMfSdJnn32mnJwc9e/fX0uXLu1w+VDqq+Sf\n3h4+fFgul0sul0vp6enaunWr1+VDqbf+6Cv7rtvTUV/Zd92ejvram/Zdffr00bx58zyPm5qadP/9\n92vGjBk+WX9+fr7i4+OVnJysPXv2eKb/9Kc/1dChQ3XPPfd0uOzmzZuVnp6u9PR0jR8/vs3vdlFR\nkZKSkpSQkKA1a9Z4pnf0eUSSjh07pnHjxiklJUXp6emqr6/3yXv0JlB9feyxx+RyuZSSkqLFixer\nsbGx3bJB3VcTAN/4xjfMk08+aVasWHHLyzY1NbV5fOrUKVNWVmYmTpxojhw54pleUVFhUlNTO13f\nrl27PN/PmTPH/O53vzPGGPPOO++Y6dOnG2OMOXjwoBk9erRnvvfff98cPXq03fr37t1rJk+ebBoa\nGowxxvz73/++xXd3e/zR1xsdP37cDB8+3Otz9LVFV/u6fv16s3DhQmOMMefPnzejRo3yuj762uKL\nfX333Xc905YtW2aWL19ujDHm6tWrZv/+/eb3v/+9+f73v9/h+kKpr8b4p7fXrl3zTP/kk0/M4MGD\nTWNjY7v1hVJv/dFX9l23pqt9vRH7rs51ta+9ad/Vr18/43K5zPXr140xLe/J5XKZJ5988rbXXVpa\najIyMkxDQ4M5c+aMiYuLM83NzcYYYw4dOmRqampM//79O1z+wIED5vPPP/fU1drHpqYmExcXZyoq\nKkx9fb1JT083J0+eNMZ0/HmksbHRpKWlmePHjxtjjPnss888tfSEQPX18uXLnvlmzpxp/vSnP7Vb\nPpj76vcRuKtXr+rQoUNav369tmzZ4pm+b98+TZgwQU888YSSkpLa/Nexf//+eu655+RyuXTw4ME2\n60tMTFR8fLzXK7R4m/ZF06ZN83yfnZ2tqqoqSVJhYaHnPwKjR4/WpUuXVFtbK0kaP368Bg0a1G5d\nv/3tb7V8+XKFh7ecWnjfffd1+vq+4s++tiooKFBubq7X5+jrrfW1tLRUOTk5kqT7779fAwcO1D/+\n8Y929dBX732dNGmS+vRp2ZyNGTPG05e7775b48aN05133nnTekKlr5L/etu3b1/P9GvXrmnAgAEK\nCwtrV0+o9NZffZXYd0k909dW7Lt819fetu+aPn26du7cKanl92jOnDme5w4fPqxx48Zp1KhRGj9+\nvE6fPi1JmjBhQpuRm4cffljHjx9vs97CwkLl5uYqPDxcw4YNU3x8vIqLiyW19MvpdN60rjFjxmjA\ngAGe71vvsVxcXKz4+HjFxsYqIiJCubm5KiwslNTx55E9e/YoPT1dKSkpkqRBgwb1+NXhA9HXfv36\nSWoZRa6vr9fgwYPb1RXMffV7gCssLNTUqVMVExOjIUOGtBkuP3z4sNavX6+TJ0+qvLxcb731lqSW\nDdHYsWN19OhRjRs3rsuvVVFRoczMTE2cOFEffPDBTedtbGzUpk2b9Nhjj0lqf9PxqKioTm86XlZW\npr///e8aM2aMJk6c6HUj1lP82ddWW7dubfNH5g197Vpf09PTtWPHDjU1NenMmTM6cuSIzp071+H8\n9LXjvm7cuNHTl1tle18l//a2uLhYKSkpSklJ0dq1a29al+299Wdf2Xe16KltAfsu3/W1N+27HA6H\ncnNzVVBQoLq6Oh07dkyjR4/2PJ+cnKwPPvhAR44c0cqVK/WTn/xEkrR48WK99tprkqTTp0+rrq5O\nqampbdbdnT505NVXX+2wv9HR0V3qr9QSur/61a/qxRdf7FYdXRXIvk6bNk2RkZG666672vyTwZtg\n66vfA1xBQYFmz54tSXr66ae1efNmz3PZ2dmKjY2Vw+HQnDlzPDuusLAwff3rX7+l1/nyl7+ss2fP\nqqSkRC+99JLmzp2rK1eudDh/Xl6eJkyY0K0g06qxsVEXL17UwYMH9atf/crzPv3BX31tVVxcrC99\n6UsaMWLETeejr12zcOFCRUVFKSsrSz/60Y/00EMPeR3NaEVfvfvlL3+piIgIzZ07t1t12d5Xyb+9\nzc7O1ocffqiSkhL94Ac/8JwX443tvfVXX9l39ey2gH2Xb/va2/ZdKSkpqqioUEFBgR5//PE2oyyf\nf/65Zs2apdTUVP3whz9UaWmpJGnWrFnauXOnmpqatHHjRi1YsKDH6tu7d69ee+21Nudk3arGxkbt\n379fBQUFev/99/X2229r7969PqyyvUD1taioSJ988onq6ur0xhtvdDhfMPbVr7cRuHjxot577z19\n+OGHcjgcampqksPh8KTQLw4ltj6+6667bnmYMSIiwjP0npmZqbi4OJWVlSkzM7PdvD//+c914cIF\n/eEPf/BMi4qKavNfpKqqKkVFRd30NWNiYjwbv6ysLPXp00effvqp12FZX/JnX1tt2bKl0/9g0teu\nCwsLazOC8dBDDykhIcHrvPTVu9dff13vvPOO3nvvvW7VZXtfpcD1NjExUXFxcTp9+rRGjRrV7nnb\ne+vPvrLv6tnfV/Zdvu1rb9p3tZoxY4Z+/OMfy+1268KFC57pL7zwgnJycvTWW2+psrJSEydOlNTS\n18mTJ2v79u168803deTIkXbr7E4fvujYsWP6zne+o6KiIs82JCoqSmfPnr2l9UZHR+uRRx7xrGP6\n9OkqKSnxvJ+eEqi+3nHHHZo5c6aKi4vbXEylVbD21a8jcG+++abmzZunM2fO6OOPP1ZlZaUefPBB\nz39/iouLVVlZqebmZm3dulUPP/ywpK6dD/DF+S5cuKDm5mZJ0scff6zy8nJ95StfabfMq6++qt27\nd6ugoKDN9BkzZnjS+MGDBzVw4MA2xyAbY9rV9bWvfc2zYSsrK1NDQ4NfNij+7Gvr423btnV4DoFE\nX7vixvmuXbum//73v5Kkd999VxEREUpKSmq3DH31rqioSC+++KJ27NjR4fluN1s+FPoq+be3FRUV\nnivJVVZWqry8XPHx8e2WD4Xe+rOv7Lt6blvAvsv3fe1N+67WOhYuXKif/exnGjlyZJvnL1265PkQ\n33poX6tFixZp6dKlys7O9pxTdaMZM2Zoy5Ytqq+v15kzZ1ReXq7s7Gyvr+/N2bNnNXPmTG3atElx\ncXGe6VlZWSovL1dlZaXq6+u1ZcsWr1d3vHHdU6dO1fHjx3X9+nU1NjZq3759nY5Y345A9PXq1auq\nqamR1DIytnPnTmVkZLRbPqj72u3Ln3RDTk6O2b17d5tpr7zyisnLyzNut9s88sgj5oknnjBJSUkm\nLy/PM8/Nrrzz9ttvm+joaNO3b18TGRlppk2bZowx5q9//asZOXKkcblcZtSoUWbnzp1elw8PDzfD\nhw83GRkZxuVymV/84hee57773e+auLg4k5aW1uZKMnPmzDEPPPCAueOOO0xMTIzZuHGjMcaY+vp6\n88wzz5iUlBQzatQo43a7b71J3eDPvhpjjNvtNmPHjr1pTfTVu476WlFRYRITE82IESPM5MmTzdmz\nZ70uT1+9Gz58uBk6dKhxuVzG5XKZJUuWeJ4bNmyYGTx4sOnfv7+JiYnxXCnqRqHQV2P829tNmzZ5\ntrHZ2dmmqKjI6/Kh0Ft/9pV9V89tC9h3+b6vvWnf5a0/brfbc7XEAwcOmISEBJOZmWleeOEF8+CD\nD7aZNykpyezZs6fD9a9atcrExcWZpKSkNj+/559/3kRHR5uwsDATExNjVq5c2W7ZxYsXm3vvvde4\nXC6TkZFhsrKyPM/t2rXLJCQkmOHDh5v8/HzP9Jt9zvvzn/9sRo4caVJTU71eydWXAtHX2tpak5WV\nZdLT001aWpp57rnnvF4RMpj7GjQ38t63b59eeukl7dixI9ClhBT62jPoa8+grz2H3vYM+toz6GvP\noK+B8a9//Us5OTk6depUoEsJKb25rwG5kTcAAAAQ6jZt2qSxY8dq1apVgS4lpPT2vgbNCBwAAAAA\n4OYYgQMAAAAASxDgAAAAAMASBDgAAAAAsAQBDgAAAAAsQYADAAAAAEsQ4AAAAADAEv8DzlNbcttQ\nLhkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff0c47bf8d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "times = [datetime.datetime.strptime(t['created_at'], time_format) for t in all_kumamoto_tweets]\n",
    "pylab.figure().set_size_inches(15,8)\n",
    "pylab.title(\"Tweets per day for all tweets with `Kumamoto`\", fontsize=30)\n",
    "hist(times, bins=20)\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Geo Tagged Tweets in Kumamoto WITH images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 337 tweets with images in them from the geotagged tweets in Kumamoto\n",
      "Unique Users with images in geo-tagged tweets: \n",
      "['HyesungTama47' 'LastQuake' 'QuakesZone' 'fulchin0000' 'japankatamichi'\n",
      " 'kyosin_bot']\n"
     ]
    }
   ],
   "source": [
    "counter = 0\n",
    "media_tweets = []\n",
    "media_tweets_full = []\n",
    "for t in in_kumamoto:\n",
    "    if 'media' in t['entities'].keys():\n",
    "        counter+=1\n",
    "        media_tweets_full.append(t)\n",
    "        media_tweets.append({ \"user\" : t['user']['screen_name'], \"url\":t['entities']['media'][0]['media_url']} )\n",
    "print(\"Found %d tweets with images in them from the geotagged tweets in Kumamoto\"%counter)\n",
    "print(\"Unique Users with images in geo-tagged tweets: \\n%s\" %str(np.unique( [u['user'] for u in media_tweets])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "337  tweets\n",
      "Geojson saved to tweets_with_images.geojson\n"
     ]
    }
   ],
   "source": [
    "# Write these files out as GeoJSON:\n",
    "create_geojson(media_tweets_full, 'tweets_with_images.geojson')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unfortunately, most of these tweets (all but 3) are from automated bots and sensors that tweet earthquake updates. You can see the results on this map:\n",
    "### [View these on a Map](http://research.yetilabs.co/nbserver/yeti-sites/pages/maps/geojson-animator.html?dataset=/nbserver/KumamotoEarthquakes2016/tweets_with_images.geojson)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}