1. Yelp Reviews Explored in Python.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analysis of Yelp Reviews using Python\n",
    "\n",
    "You need to create your environment and activate it first.  Use the environment.yml file in the repo and \n",
    "```` \n",
    ">conda env create -f environment.yml\n",
    "... agree to all installs...\n",
    ">source activate esnlp\n",
    "````\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import gensim\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import string\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import nltk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the data and put into Pandas Dataframe\n",
    "\n",
    "You will need to gunzip the yelp data first if you load this from the repo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "!gunzip data/yelp_academic_dataset_reviews.json.gz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data = []\n",
    "with open(\"data/yelp_academic_dataset_reviews.json\") as handle:\n",
    "    for line in handle.readlines():\n",
    "        yelp = json.loads(line)\n",
    "        data.append(yelp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "229907"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = pd.DataFrame.from_dict(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>type</th>\n",
       "      <th>user_id</th>\n",
       "      <th>votes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9yKzy9PApeiPPOUJEtnvkg</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>fWKvX83p0-ka4JS3dc6E5A</td>\n",
       "      <td>5</td>\n",
       "      <td>My wife took me here on my birthday for breakf...</td>\n",
       "      <td>review</td>\n",
       "      <td>rLtl8ZkDX5vH5nAx9C3q5Q</td>\n",
       "      <td>{u'funny': 0, u'useful': 5, u'cool': 2}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ZRJwVLyzEJq1VAihDhYiow</td>\n",
       "      <td>2011-07-27</td>\n",
       "      <td>IjZ33sJrzXqU-0X6U8NwyA</td>\n",
       "      <td>5</td>\n",
       "      <td>I have no idea why some people give bad review...</td>\n",
       "      <td>review</td>\n",
       "      <td>0a2KyEL0d3Yb1V6aivbIuQ</td>\n",
       "      <td>{u'funny': 0, u'useful': 0, u'cool': 0}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6oRAC4uyJCsJl1X0WZpVSA</td>\n",
       "      <td>2012-06-14</td>\n",
       "      <td>IESLBzqUCLdSzSqm0eCSxQ</td>\n",
       "      <td>4</td>\n",
       "      <td>love the gyro plate. Rice is so good and I als...</td>\n",
       "      <td>review</td>\n",
       "      <td>0hT2KtfLiobPvh6cDC8JQg</td>\n",
       "      <td>{u'funny': 0, u'useful': 1, u'cool': 0}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>_1QQZuf4zZOyFCvXc0o6Vg</td>\n",
       "      <td>2010-05-27</td>\n",
       "      <td>G-WvGaISbqqaMHlNnByodA</td>\n",
       "      <td>5</td>\n",
       "      <td>Rosie, Dakota, and I LOVE Chaparral Dog Park!!...</td>\n",
       "      <td>review</td>\n",
       "      <td>uZetl9T0NcROGOyFfughhg</td>\n",
       "      <td>{u'funny': 0, u'useful': 2, u'cool': 1}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6ozycU1RpktNG2-1BroVtw</td>\n",
       "      <td>2012-01-05</td>\n",
       "      <td>1uJFq2r5QfJG_6ExMRCaGw</td>\n",
       "      <td>5</td>\n",
       "      <td>General Manager Scott Petello is a good egg!!!...</td>\n",
       "      <td>review</td>\n",
       "      <td>vYmM4KTsC8ZfQBg-j5MWkw</td>\n",
       "      <td>{u'funny': 0, u'useful': 0, u'cool': 0}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id        date               review_id  stars  \\\n",
       "0  9yKzy9PApeiPPOUJEtnvkg  2011-01-26  fWKvX83p0-ka4JS3dc6E5A      5   \n",
       "1  ZRJwVLyzEJq1VAihDhYiow  2011-07-27  IjZ33sJrzXqU-0X6U8NwyA      5   \n",
       "2  6oRAC4uyJCsJl1X0WZpVSA  2012-06-14  IESLBzqUCLdSzSqm0eCSxQ      4   \n",
       "3  _1QQZuf4zZOyFCvXc0o6Vg  2010-05-27  G-WvGaISbqqaMHlNnByodA      5   \n",
       "4  6ozycU1RpktNG2-1BroVtw  2012-01-05  1uJFq2r5QfJG_6ExMRCaGw      5   \n",
       "\n",
       "                                                text    type  \\\n",
       "0  My wife took me here on my birthday for breakf...  review   \n",
       "1  I have no idea why some people give bad review...  review   \n",
       "2  love the gyro plate. Rice is so good and I als...  review   \n",
       "3  Rosie, Dakota, and I LOVE Chaparral Dog Park!!...  review   \n",
       "4  General Manager Scott Petello is a good egg!!!...  review   \n",
       "\n",
       "                  user_id                                    votes  \n",
       "0  rLtl8ZkDX5vH5nAx9C3q5Q  {u'funny': 0, u'useful': 5, u'cool': 2}  \n",
       "1  0a2KyEL0d3Yb1V6aivbIuQ  {u'funny': 0, u'useful': 0, u'cool': 0}  \n",
       "2  0hT2KtfLiobPvh6cDC8JQg  {u'funny': 0, u'useful': 1, u'cool': 0}  \n",
       "3  uZetl9T0NcROGOyFfughhg  {u'funny': 0, u'useful': 2, u'cool': 1}  \n",
       "4  vYmM4KTsC8ZfQBg-j5MWkw  {u'funny': 0, u'useful': 0, u'cool': 0}  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# I'm deleting votes because it's out of scope for this talk...\n",
    "del df['votes']\n",
    "del df['type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9yKzy9PApeiPPOUJEtnvkg</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>fWKvX83p0-ka4JS3dc6E5A</td>\n",
       "      <td>5</td>\n",
       "      <td>My wife took me here on my birthday for breakf...</td>\n",
       "      <td>rLtl8ZkDX5vH5nAx9C3q5Q</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ZRJwVLyzEJq1VAihDhYiow</td>\n",
       "      <td>2011-07-27</td>\n",
       "      <td>IjZ33sJrzXqU-0X6U8NwyA</td>\n",
       "      <td>5</td>\n",
       "      <td>I have no idea why some people give bad review...</td>\n",
       "      <td>0a2KyEL0d3Yb1V6aivbIuQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6oRAC4uyJCsJl1X0WZpVSA</td>\n",
       "      <td>2012-06-14</td>\n",
       "      <td>IESLBzqUCLdSzSqm0eCSxQ</td>\n",
       "      <td>4</td>\n",
       "      <td>love the gyro plate. Rice is so good and I als...</td>\n",
       "      <td>0hT2KtfLiobPvh6cDC8JQg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>_1QQZuf4zZOyFCvXc0o6Vg</td>\n",
       "      <td>2010-05-27</td>\n",
       "      <td>G-WvGaISbqqaMHlNnByodA</td>\n",
       "      <td>5</td>\n",
       "      <td>Rosie, Dakota, and I LOVE Chaparral Dog Park!!...</td>\n",
       "      <td>uZetl9T0NcROGOyFfughhg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6ozycU1RpktNG2-1BroVtw</td>\n",
       "      <td>2012-01-05</td>\n",
       "      <td>1uJFq2r5QfJG_6ExMRCaGw</td>\n",
       "      <td>5</td>\n",
       "      <td>General Manager Scott Petello is a good egg!!!...</td>\n",
       "      <td>vYmM4KTsC8ZfQBg-j5MWkw</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id        date               review_id  stars  \\\n",
       "0  9yKzy9PApeiPPOUJEtnvkg  2011-01-26  fWKvX83p0-ka4JS3dc6E5A      5   \n",
       "1  ZRJwVLyzEJq1VAihDhYiow  2011-07-27  IjZ33sJrzXqU-0X6U8NwyA      5   \n",
       "2  6oRAC4uyJCsJl1X0WZpVSA  2012-06-14  IESLBzqUCLdSzSqm0eCSxQ      4   \n",
       "3  _1QQZuf4zZOyFCvXc0o6Vg  2010-05-27  G-WvGaISbqqaMHlNnByodA      5   \n",
       "4  6ozycU1RpktNG2-1BroVtw  2012-01-05  1uJFq2r5QfJG_6ExMRCaGw      5   \n",
       "\n",
       "                                                text                 user_id  \n",
       "0  My wife took me here on my birthday for breakf...  rLtl8ZkDX5vH5nAx9C3q5Q  \n",
       "1  I have no idea why some people give bad review...  0a2KyEL0d3Yb1V6aivbIuQ  \n",
       "2  love the gyro plate. Rice is so good and I als...  0hT2KtfLiobPvh6cDC8JQg  \n",
       "3  Rosie, Dakota, and I LOVE Chaparral Dog Park!!...  uZetl9T0NcROGOyFfughhg  \n",
       "4  General Manager Scott Petello is a good egg!!!...  vYmM4KTsC8ZfQBg-j5MWkw  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Save to a file for re-use\n",
    "df.to_msgpack('data/yelp_full_df.msg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load from file:\n",
    "df = pd.read_msgpack('data/yelp_full_df.msg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A few quick exploratory plots..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11349b510>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAFkCAYAAADxHkghAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt0VPWhL/Dv3jOZR2Ymk2QyeQASMIASH7ysFD1CQPAF\nl6NtYgkQPEtOXXrqOqgtreBZy6KnR6R3eT1asHC8vVRtpUVpsQhqaRSfQEQJGN6BhJD3c96TyWT2\n/WOYgUCSmSQzmZ2Z72etFjJ7sufHz5189++3fw9BkiQJREREFFNirAtAREREDGQiIiJZYCATERHJ\nAAOZiIhIBhjIREREMsBAJiIikoGwArm1tRUFBQU4d+5cj9e3bt2KRYsWYcWKFVixYgWqqqqiUUYi\nIqK4pwz1Bq/Xi2effRYajeaqYxUVFdiwYQPy8/OjUjgiIqJEEbKF/OKLL6K4uBiZmZlXHauoqMDm\nzZuxdOlSbNmyJSoFJCIiSgT9BvKOHTtgMplw++23o7cFvRYuXIh169bhjTfewKFDh7Bv376oFZSI\niCieCf0tnbl8+XIIggAAOHHiBMaPH4/XXnsNJpMJAGC326HX6wEAf/zjH2GxWPDYY4/1+4GSJAXP\nSURERH79PkN+6623gn8vKSnBc8891yOMFy1ahD179kCj0WD//v0oLCwM+YGCIKC52TbEYicGs9nA\nugoD6yk8rKfwsa7Cw3oKn9lsCPmekIO6AgKt2l27dsHlcqGoqAhPPfUUSkpKoFarMWvWLMyePXvw\npSUiIkpg/XZZRwvvqMLDu8/wsJ7Cw3oKH+sqPKyn8IXTQubCIERERDLAQCYiIpIBBjIREZEMMJCJ\niIhkgIFMREQkAwxkIiIiGQh7HjIRUbySJAk2mzX4tUrlg9Xqn85jMKRwdUEaFgxkIkp4NpsVfz9w\nBtpkHQBAr2uD3dEJl9OBBTMnICXFGOMSUiJgIBMRAdAm65Cs8y/eoNNr4IM7xiWiRMNnyERERDLA\nQCYiIpIBBjIREZEMMJCJiIhkgIFMREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDLAQCYiIpIB\nLp1JRER9unLjjctlZOiHuTTxjYFMRER9unLjjQCX04HiDAPY0Ro5DGQiIurX5RtvUPTw1oaIiEgG\nGMhEREQywEAmIiKSAQYyERGRDIQVyK2trSgoKMC5c+d6vF5aWorCwkIsWbIE27dvj0oBiYiIEkHI\nUdZerxfPPvssNBrNVa+vX78eO3bsgFqtRnFxMe68806kp6dHrbBERETxKmQL+cUXX0RxcTEyMzN7\nvF5ZWYnc3Fzo9XokJSVhxowZKCsri1pBiYiI4lm/gbxjxw6YTCbcfvvtkCSpxzG73Q6D4dK8NJ1O\nB5vNFp1SEhERxbl+u6x37NgBQRDwxRdf4MSJE/jFL36B1157DSaTCXq9Hna7Pfheh8OBlJSUsD7U\nbOYE83CxrsLDegoP66l3KpUPel0bdPpLj+YMeg1EeJCRYYDRmLj11lvdAIAIDwBeU5HUbyC/9dZb\nwb+XlJTgueeeg8lkAgDk5eWhuroaVqsVGo0GZWVlWLlyZVgf2tzMlnQ4zGYD6yoMrKfwsJ76ZrXa\nYHd0wgc3AH8Y2+xuOB2daGmxweNJ3AkpV9ZNgNPRCYC/z8MVzo1L2EtnCoIAANi1axdcLheKioqw\nZs0aPPzww5AkCUVFRVc9ZyYiIqLwhB3Ib7zxBgBg/PjxwdcKCgpQUFAQ8UIRERElmsTthyEiIpIR\nBjIREZEMMJCJiIhkgIFMREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDLAQCYiIpIBBjIREZEM\nMJCJiIhkgIFMREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDLAQCYiIpIBBjIREZEMMJCJiIhk\ngIFMREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDLAQCYiIpIBBjIREZEMKEO9wefz4T/+4z9w\n7tw5iKKIdevWYcKECcHjW7duxTvvvIP09HQAwHPPPYdx48ZFrcBERETxKGQgl5aWQhAEvP322zh4\n8CBeeuklbNq0KXi8oqICGzZsQH5+flQLSkREFM9CBvL8+fMxb948AEBtbS2MRmOP4xUVFdi8eTOa\nm5tRUFCARx55JDolJSIiimMhAxkARFHE008/jb179+KVV17pcWzhwoVYtmwZ9Ho9fvKTn2Dfvn2Y\nM2dOVApLREQUrwRJkqRw39za2oqioiLs3r0bGo0GAGC326HX6wEAf/zjH2GxWPDYY49Fp7RERFFg\nsViw92A1dPqUHq877FbMvzX3qp7BRMK6GT4hW8g7d+5EY2MjHnnkEajVaoiiCFH0D8622+1YtGgR\n9uzZA41Gg/3796OwsDDkhzY324Ze8gRgNhtYV2FgPYWH9dQ3q9UGu6MTPrgBAAa9Bja7G05HJ1pa\nbPB4EndCypV1E+B0dALg7/Nwmc2GkO8JGch33XUX1qxZg+XLl8Pr9WLt2rX46KOP4HK5UFRUhKee\negolJSVQq9WYNWsWZs+eHZHCExERJZKQgazVavHyyy/3eXzx4sVYvHhxRAtFRESUaBK3H4aIiEhG\nGMhEREQywEAmIiKSAQYyERGRDDCQiYiIZICBTEREJAMMZCIiIhlgIBMREckAA5mIiEgGGMhEREQy\nwEAmIiKSAQYyERGRDDCQiYiIZICBTEREJAMMZCIiIhlgIBMREckAA5mIiEgGGMhEREQywEAmIiKS\nAQYyERGRDDCQiYiIZICBTEREJAMMZCIiIhlgIBMREckAA5mIiEgGGMhEREQywEAmIiKSgZCB7PP5\nsHbtWhQXF2PZsmU4c+ZMj+OlpaUoLCzEkiVLsH379qgVlIiIKJ6FDOTS0lIIgoC3334bq1atwksv\nvRQ85vV6sX79emzduhVvvvkm/vSnP6GtrS2qBSYiIopHIQN5/vz5eP755wEAtbW1MBqNwWOVlZXI\nzc2FXq9HUlISZsyYgbKysuiVloiIKE4pw3mTKIp4+umnsXfvXrzyyivB1+12OwwGQ/BrnU4Hm80W\n8nxmsyHke8iPdRUe1lN4WE+9U6l80OvaoNNrgq8Z9BqI8CAjwwCjMXHrrbe6AQARHgC8piIprEAG\ngPXr16O1tRVFRUXYvXs3NBoN9Ho97HZ78D0OhwMpKSkhz9XcHDq0yX+hs65CYz2Fh/XUN6vVBruj\nEz64AfjD2GZ3w+noREuLDR5P4o5/vbJuApyOTgD8fR6ucG5cQl5lO3fuxJYtWwAAarUaoihCFP3f\nlpeXh+rqalitVng8HpSVlWHq1KlDLDYREVHiCdlCvuuuu7BmzRosX74cXq8Xa9euxUcffQSXy4Wi\noiKsWbMGDz/8MCRJQlFRETIzM4ej3ERERHElZCBrtVq8/PLLfR4vKChAQUFBJMtERESUcBL3wQgR\nEZGMMJCJiIhkgIFMREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDLAQCYiIpIBBjIREZEMMJCJ\niIhkgIFMREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDLAQCYiIpIBBjIREZEMMJCJiIhkgIFM\nREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDLAQCYiIpIBBjIREZEMMJCJiIhkQNnfQa/Xi7Vr\n16K2thZdXV149NFHMW/evODxrVu34p133kF6ejoA4LnnnsO4ceOiWmAiIqJ41G8gv/fee0hLS8OG\nDRtgsVhw//339wjkiooKbNiwAfn5+VEvKBERUTzrN5Dvvfde3HPPPQAAn88HpbLn2ysqKrB582Y0\nNzejoKAAjzzySPRKSkREFMf6DWStVgsAsNvtWLVqFZ588skexxcuXIhly5ZBr9fjJz/5Cfbt24c5\nc+aE/FCz2TCEIicW1lV4WE/hYT31TqXyQa9rg06vCb5m0GsgwoOMDAOMxsStt97qBgBEeADwmoqk\nfgMZAOrr6/H4449j+fLluO+++3oce+ihh6DX6wEAc+bMwbFjx8IK5OZm2yCLm1jMZgPrKgysp/Cw\nnvpmtdpgd3TCBzcAfxjb7G44HZ1oabHB40nc8a9X1k2A09EJgL/PwxXOjUu/V1lLSwtWrlyJ1atX\n44EHHuhxzG63Y9GiRXC5XJAkCfv378cNN9wwtBITERElqH5byJs3b4bVasWmTZuwceNGCIKABx98\nEC6XC0VFRXjqqadQUlICtVqNWbNmYfbs2cNVbiIiorjSbyA/88wzeOaZZ/o8vnjxYixevDjihSIi\nIko0iftghIiISEYYyERERDLAQCYiIpIBBjIREZEMMJCJiIhkgIFMREQkAwxkIiIiGWAgExERyQAD\nmYiISAYYyERERDLAQCYiIpIBBjIREZEMMJCJiIhkgIFMREQkAwxkIiIiGWAgExERyQADmYiISAYY\nyERERDLAQCYiIpIBBjIREZEMMJCJiIhkgIFMREQkAwxkIiIiGWAgExERyQADmYiISAYYyERERDKg\n7O+g1+vF2rVrUVtbi66uLjz66KOYN29e8HhpaSk2bdoEpVKJH/7whygqKop6gYmIiOJRv4H83nvv\nIS0tDRs2bIDFYsH9998fDGSv14v169djx44dUKvVKC4uxp133on09PRhKTgREVE86bfL+t5778Wq\nVasAAD6fD0rlpfyurKxEbm4u9Ho9kpKSMGPGDJSVlUW3tERERHGq3xayVqsFANjtdqxatQpPPvlk\n8JjdbofBYAh+rdPpYLPZwvpQs9kQ+k0EgHUVLtZTeFhPvVOpfNDr2qDTa4KvGfQaiPAgI8MAozFx\n6623ugEAER4AvKYiqd9ABoD6+no8/vjjWL58Oe67777g63q9Hna7Pfi1w+FASkpKWB/a3BxecCc6\ns9nAugoD6yk8rKe+Wa022B2d8MENwB/GNrsbTkcnWlps8HgSd/zrlXUT4HR0AuDv83CFc+PS71XW\n0tKClStXYvXq1XjggQd6HMvLy0N1dTWsVis8Hg/KysowderUoZWYiIgoQfXbQt68eTOsVis2bdqE\njRs3QhAEPPjgg3C5XCgqKsKaNWvw8MMPQ5IkFBUVITMzc7jKTUREFFf6DeRnnnkGzzzzTJ/HCwoK\nUFBQEOkyERERJZyQz5CJaOSTJAkWiwVW69XP+wyGFAiCEINSEdHlGMhECcBms+LDr2rgk3r+yLuc\nDiyYOQEpKcYYlYyIAhjIRAkiOVkHH1SxLgYR9SFxx/ITERHJCAOZiIhIBhjIREREMsBAJiIikgEG\nMhERkQwwkImIiGSAgUxERCQDDGQiIiIZYCATERHJAAOZiIhIBhjIREREMsBAJiIikgEGMhERkQww\nkImIiGSAgUxERCQDDGQiIiIZYCATERHJAAOZiIhIBhjIREREMsBAJiIikgEGMhERkQwwkImIiGSA\ngUxERCQDYQVyeXk5SkpKrnp969atWLRoEVasWIEVK1agqqoq0uUjIiJKCMpQb3j99dexc+dO6HS6\nq45VVFRgw4YNyM/Pj0rhiIiIEkXIFnJubi42btzY67GKigps3rwZS5cuxZYtWyJeOCIiokQRMpAX\nLFgAhULR67GFCxdi3bp1eOONN3Do0CHs27cv4gUkIiJKBCG7rPvz0EMPQa/XAwDmzJmDY8eOYc6c\nOSG/z2w2DOVjEwrrKjysp/6pVD7gbBsMek2P10V4kJFhgNGY2PWnUvmg17VBd1n9GPQa1g96rxvA\nf+0A/NmLpLADWZKkHl/b7XYsWrQIe/bsgUajwf79+1FYWBjWuZqbbQMrZYIymw2sqzCwnkKzWv31\nY7O7e7zudHSipcUGjyexJ1xYrTbYHZ3wwV8/Br0GNrub9YOr6ybA6egEwN/n4QrnxiXsQBYEAQCw\na9cuuFwuFBUV4amnnkJJSQnUajVmzZqF2bNnD760RERECSysQB49ejS2bdsGAFi0aFHw9cWLF2Px\n4sXRKRkREVECSdx+GCIiIhlhIBMREckAA5mIiEgGGMhEREQywEAmIiKSAQYyERGRDDCQiYiIZGBI\nS2cSEcWLs/UOnKpthlGvwqgMPQxaJVK1UuhvJIoQBjIREYBzjU7YXV2wu7pQ2+wAANw4zoDZU2Jc\nMEoYDGQiSnhdXh867F0wGTWYN3007O5u7PmqCi1WT6yLRgmEgUxECe9CsxOSBJhTNdCqlcg06ZGs\nUaLd1nXVxjrk32yI9RJ5DGQiSnhVjf4uarNRG3wtw6jB+UY7OuxdMBpjVTL58XR146+fnYNCBMQk\nDWZOzkF6iib0N1JIHGVNRAmvOhDIqZcC2WT0h8z5JkdMyiRXje0uuD3dcLi7sf3jc1j92pfYuucE\nW8wRwEAmooRX1WCHOkmETnup0zCDgdyrpnYnAGDW5DSsXDQJGUYNPi2vQ3OHK8YlG/kYyESU0Nqs\nbnQ4umBKUQX3fQcA08Vu2JpmZ6yKJkuNbS4IApCVpsbc6aMwb/oYAMDZemuMSzbyMZCJKKGdrfMH\nSbohqcfrqiQF9FoFzjc54WN3LAD/aPRWqxumFA2UCn98jM9JAQCcq7PFsmhxgYFMRAmtss4CADAZ\nVFcdS9er4PZ0o6md3bEA0GJxQZKAzLRLz9pzswwQBQHn2EIeMgYyESW0yjorRAFIu6KFDFx6jWHj\nF7gxuTyQ1SoFRpt1qG60wdvti1XR4gIDmYgSlrfbh6p6G0aZtMEu2MulXWw1M5D9GoOBnNzj9fE5\nKejy+oIrnNHgMJBpxJEkCVarJfg/i+XS3zn1ggaipskOb7cPuVn6Xo+n6pQQBaCqns9HfT4JLR0u\npOpV0KgUPY5dO+ric2TeuAwJFwahEcdms+LvB85Am6wDAOh1bbA7OuFyOrBg5gSkpHAVBwpPZa3/\n+fG4bB06PVcvk6lUiMhO1+J8ow3dPh8UYuK2YVqtbni7pR7d1QHBgV31VhRMGz3cRYsbiXt10Yim\nTdYhWWdAss4AnT4FyTpDMKCJwlV5cYT1uKy+r52xmTp42B172fPj5KuOjcpIhipJZAt5iBjIRJSw\nKmst0GuTkGFU9/mesZn+AKpqSOxu68Dz46xeWsgKUcS4LANqWxxwe7zDXbS4wUAmooTkdHehxeLG\nuBxDjwVBrjQ20996TuTWnyRJaGp3QqdRQqe9ejQ6AIwflQJJAqoT/MZlKBjIRJSQAi2+7PSru2Av\nl5OuRZIysbtjrU4vPF0+ZPVTV5eeIzOQB4uBTEQJqfHimsxZvTwTvZxCIWCUSYeG1sRdsSuwL3Rv\nA7oCrr0YyFxCc/DCCuTy8nKUlJRc9XppaSkKCwuxZMkSbN++PeKFIyKKlqa2vp+JXikzTQuP14cO\nW2e0iyVLFkcXgEsbbvTGZNTAkJyEc3UM5MEKOe3p9ddfx86dO6HT9RyF6PV6sX79euzYsQNqtRrF\nxcW48847kZ6eHrXCEhFFSmOPVae6+n1vVro/tJvaXQm596/d5R+oZUi+ennRAEEQMD4nBUcqW2Fx\neGDU9f1e6l3IFnJubi42btx41euVlZXIzc2FXq9HUlISZsyYgbKysqgUkogo0pranVCIQnDf4/5k\npvq7tZsSdItBm9OLZI0SScr+I+PaHC4QMhQhA3nBggVQKBRXvW6322EwGIJf63Q62Gx8mE9EI0Nj\nuwsZqdqwFvsIPDsNPHdOJJ1d3XB5fEgJo8U7NtufCTVN9mgXKy4NeqUuvV4Pu/1SpTscDqSkpIT1\nvWazIfSbCADrqjcqlQ96XRt0+kstG4NeAxEeZGQYYDSyzq6kUvmAs20w6Hu2BhO1zuxOD+yuLlw/\nLh1msyHkNZVu9s9T7nB0JdzPZH2bfzWzDKO2x/Ujwj/Q6/L6uAH+6WMWZ+LVUySEHchXrhGcl5eH\n6upqWK1WaDQalJWVYeXKlWGdq7mZLelwmM0G1lUvrFYb7I5O+OAG4P/FabO74XR0oqXFBo+Hkweu\nZLX6ryOb3d3j9USts0CXaqouCc3NtpDXlMEgQK1SoKbBlnA/kyfPtQEAtCpFj+vH6fAPcLu8PoRu\nHxSigOp6S8LVUyjh3KCEHciBifO7du2Cy+VCUVER1qxZg4cffhiSJKGoqAiZmZmDLy310NrWhpa2\nRrS19ewiSzEkIzsrK0alIooP4U55ChAEAVmpWjS0OyFJUr8LicSbZos/eMPpslYqRGSkatHYlpjP\n2ocqrEAePXo0tm3bBgBYtGhR8PWCggIUFBREpWCJ7lxNEwSdGTZ7zx+CpvYWBjLREA1kylNAZpoW\n55vssDg8SNX3vdRmvGlq97eKU3S9r9B1pew0LcrbnLC7uqDvY1Uv6l1i9VMREeGyKU8hVum6XGBT\nhca2xBrY1WxxQxTQ55KZVwqs5pWIA+CGioFMRAknOOUpJfyWbqA1Hdj1KBFIkoSmDjf0WiXEMLvp\ng4GcYDcukcBAJqKEM5ApTwGBqU+JNBfZ6vDA7fFBrw1/Qk72xXpq4HPkAWMgE1FCcbi7YHd1Dej5\nMXBZl3UCtZAbLrZyDQMIZLaQB4+BTEQJpanHkpnhS9WroEoS0ZRAQTOYQE41qKFKEhnIg8BAJqKE\nMtApTwGCICAzNRmNHa6r1mWIV4HpS/rk8ANZFARkpSUHp4hR+BjIRJRQBjPlKSArTYtOTzeszv43\no4gXl1rIVy+f3J+s9GR4unzosHuiUay4xUAmooQSaCEPZMpTQHBgV4JM6alvcyJZrYA6aWCBnH1x\ndyx2Ww8MA5mIEkpTu2vAU54CgptMJMAIYm+3Dy0dLphTB77dZOBxQEOC3LhECgOZiBLKYKY8BQSC\npqkj/oOmxeJGt09C5iACOZsjrQeFgUxECWOwU54CMhNocZDA8+PM1IH3JFya+hT/9RRJDGQiShiD\nnfIUkGpQI0kpJkTQNLQGAnngLWS9Ngk6jTIY6hQeBjIRJYxAF+pApzwFiIKAzFQtmjrif0rPpRby\nwAMZ8HdbN3e40O3zRbJYcY2BTEQJIxAyWemDayED/ta1q7MbNld8T31qbHNCAGAyDm5nq6z0ZHT7\nJLRY3KHfTAAYyESUQAKBnD2IKU8BwYFdcd5t3dDmhMmogUo5uJjgEpoDx0AmooTR2OaCSikiPWVw\n3bDApdZ1PD8fdbq7YHF4kG0a/I1L4KaHm0yEj4FMRAlBkiQ0tDuRmZYc9laCvQkETX2bI1JFk536\niwO6Rpl0gz5HYCQ790UOHwOZiBJCh92DTk/3kFp9AJBzMaQCo5DjUSCQh1JXwcVB4rieIo2BTEQJ\noaHV36LNHsKALgAwJMf/lJ76i3U1lBayWqVAeoo6rusp0hjIRJQQIjGgC/Dv+pSdnoymdhe83fE5\npScSLWTAH+jttk64Or2RKFbcYyATUUKoDwby4Ft9Admm+J7SU9/mhE6jhEGbNKTzjMrw13Vda/w+\nb48kBjIRJYRItZAvP0d9HAaNt9uH5nYXcjJ0EIYw+A24LJBb4q+eooGBTEQJoaHViRSdCska5ZDP\nFc8DuxrbXfBJEnIicOMSeAZd3xJ/9RQNQ78yiUi2PF3deOujU2i1OGB1dgEQYTJqMHViRqyLNqy6\nvN1otbgx8ZrUiJwvxxSY+hR/QVN/sTWbM4QBXQE5Gf56Ypd1eBjIRHFsX3kdPj9a3+O12hYHRmUk\nI3OQ6zmPRE3tLkiITHc1AJhTtRAFIS5byIGbjJwhDugCAJ0mCUa9il3WYWKXNVGc6vL68MGB81Al\nifjlipuwfEEuFnxvDADgaGVbjEs3vCL5/BgAlAoR5jRtXE7pCUwPy8kYegsZ8Hdbt1jc6PR0R+R8\n8SxkIEuShGeffRZLlizBihUrUFNT0+P41q1bsWjRIqxYsQIrVqxAVVVVtMpKRAPw5Xf1aLd1omDq\naKTqVVAqROSYdMhK06K2xYHWOB0h3JtgIEeg1ReQk54Mu6sLNqcnYueUg7pWJ5QKERlDWF70coGB\nXfG8slmkhOyy3rt3LzweD7Zt24by8nK88MIL2LRpU/B4RUUFNmzYgPz8/KgWlIjC1+3zYff+aigV\nIu6+dSwgXQrfm/JMaPz6Ao6ebcWtk1JiWMrhE+hajsRApYBsUzJwxj9n15Csith5Y0mSJDS0OpGd\nroUoDm2EdcDlI63HZSfG9TZYIVvIhw4dwh133AEAmDJlCr777rsexysqKrB582YsXboUW7ZsiU4p\niWhADh5vQnOHG3fcnIM0Q8/t83JMyTClaHC+0Q6rI763EAxoaHNCIQowGSPT6gMuhXs8dVu32zrR\n2dUdkQFdAaMu9krUcaR1SCED2W63w2AwBL9WKpXwXbbh9MKFC7Fu3Tq88cYbOHToEPbt2xedkhJR\nWHyShPe/qoYoCLh35tirjguCgJvy0gEAJy7Yh7t4w06SJDS0OWFO1UKpiNywmXic+hRYoSsSA7oC\nOBc5fCG7rPV6PRyOSxXp8/kgipcu6oceegh6vR4AMGfOHBw7dgxz5szp95xms6Hf4wSkpibD0gUY\n9D3v6JN82oSvP5XKB72uDbrL6sag10CEBxkZBhiNiV0/Xx2tQ12LA/NuuQaTJ2YC8NcZzrYFr6d8\nnRpHKttQ0+yCT6GM62vKYu+Ew+3FjXkZff47B3NNqZP9PQ+t9s64qT/riSYAwHXjTcF/U291AwAi\n/M/OQ/3bzQCMehUaO1xxU0/REjKQp0+fjo8//hj33HMPDh8+jEmTJgWP2e12LFq0CHv27IFGo8H+\n/ftRWFgY8kObm21DK3UC6OhwQtDpYLP3HHij7HIlfP1ZrTbYHZ3wwV83Br0GNrsbTkcnWlps8HgS\ne/LArs/OAgDmTh0VvFasVv+fl19Pk3NT8cXRBnz41Xmk6iLXRSk3p2o6AABpelWfPzuDvab02iSc\nr7fGzc/kmep2AIAuSexx7VxeNwFORyeA8H6fZ6cl49SFDtTWdUCVpIhwqUeGcG5GQgbyggUL8MUX\nX2DJkiUAgBdeeAG7du2Cy+VCUVERnnrqKZSUlECtVmPWrFmYPXv20EtORIPi6vTiWFUbxpj1GB1i\n2srYLAO++q4BR8524EfDVL5YaIzwlKfLZZuScbbWii6vD0nKkX8jWN/qgAAgK8J1NSpDh5M1HWho\nc2JsFlv1Z4JTAAAZI0lEQVTJfQkZyIIgYN26dT1eGz9+fPDvixcvxuLFiyNfMiIasCOVrfB2S5hx\nnTnke5OUIrLS1Khvc6GxzRnxX8JyEek5yJfLSU/GmQsWNHW4Qt4AjQT1rU6YjBqoI9yKvXyTCQZy\n30b+LR0RBR061QwAmDEpdCADwGiTf2/gby5+XzyKxhzkgHga2OV0d8Hi8ESlnjjSOjwMZKI44enq\nxtHKVmSlaTHaHF5rLcekgSgAX5+M30C+0GyHXps05K0Ee5MdnPo08kcQB0ZYj4rglKeA4OIgHGnd\nLwYyUZyoONeGzq5uTL/OHPa2eeokERNGG3Cu3oo2a/yt3GV1eNDc4ca1o1KGvJVgb3KCLb+RHzTn\n6q0AgGsy9RE/d4pOBZ1GyU0mQmAgE8WJS93VmQP6vpuv9e+AFI/d1pV1FgDAtaOis0KUOVULrVqB\ns/Ujf5T1mVp/XU0YY4z4uQVBQE6GDo1tLni7faG/IUExkInigLfbh8OnW5BmUGNczsAGzdw0Po4D\nudbf6ssbHfmQAQBRFJA3yojGNiesI3xN68paCwzJSchM1Ubl/KNMOvgkKdg1TldjINOIZHN6UX6m\nBV+faMLHh2qwv6IBdpc31sWKmZPnO+Ds9GLGJDPEAXbNGnUqTBhtxMmajhEfKlc6W2eBAODanOit\noRxoUVZesETtM6KtzepGq7UTE0Ybo9K1D1zqpTh9oSMq548HDGQacTrsHnxypAXlZ1pxrKodx861\n4VSNBZ9915qwoXzopH+FpXCmO/Vm+iQzJAk4fLolksWKqW6fD2frrRhl1kGrjt7W7xMutr5P147c\nQA52V0epJwEArrvG3xNz8jwDuS8MZBpRurw+/O6DSnR2+TBlggn3zRqLpXddhxvHp8Ph7sb/+7Ay\n4Z5R+XwSvjndAkNyEiaOSR3UOaZfDPKvLwZ7PKhtdsDT5UNelJ4fB1w7KgWiIODMCG4hR/P5cUBm\nmhZGvQonazogSVLUPmckYyDTiPLHvadwvsmJsZla3JxnQoZRi7QUDaZNysDoDA0q6+x444OTCfUD\nf+J8O6wOj7+7epBb5mWmapGbbcDxqva42d+38mLI5I2KXsgAgEalxDWZelQ1WNHl7Y7qZ0VLZa0F\nClHAuOzoLdohCAKuuyYVVocnrnbIiiQGsox4urqx68sqvL7rGP5W1oo/7T2FL482JFyLry+fltdh\n3+E6jM7QYvqEns+6BEHA9yal4hpzMj4/Wo8PD9bEsKTD68CxRgDAzPysIZ1n5uQsdPukuJmTfObi\ngK5ro9gNGzBhjBHebgnVDSNv96zOrm6cb7RjXLYBScrorjN93dg0AMDJGnZb94aBLBPdPh9+u7MC\nOz49iy+/a0BNSyfaLG6cqbWg9FAturyJHcrNHS689dEp6DRKPHxPXq/b6CkVIv71vjwYdSrs+PRs\nXM6rvVKX14dDJ5uRqlcNurs64NbJ/ulSgYAf6c7WWaBVKyO6lWBfJo4JPEceeUFTVW9Ft0+K2kj0\ny026+Bz5FJ8j94qBLAOSJGHrnhM4fKYF+ePSsP7RWfjxgmw8cv+NGJulR0ObE3u/vgBP18jsDouE\nv3x6Ft5uH5YumARTirrP9xl1KvxwTh683T785eKuR/Gs4lwbnJ1e3Do5a9Dd1QHpKRpMGmPE6ZqO\nEX8zY3N60NjuCj7fjbbAYKiR+Bx5OAZ0BYwyJcOQnMTnyH1gIMvA9k8q8cXRBozPMeDxH9yEzFQt\nkpQiFAoRs6eMwrgcA5o7XPh72QV4ErClXN1gw/5jjcjNMoTVLXvbjdkYY9bhy6MNqGkaeV2IA3Hg\neGS6qwNm5mdBAlB2YmQP7jpbd3H+cZQHdAWkp2iQnqLGmVrLiAuaaM/VvpwgCJh0TSrabZ1o7nBF\n/fNGGgZyjP3j0AV8cOA8stOT8UTRFGhUPadniKKAf7o5B3mjU9BqdePg6fgOmN68s68SAFA4Ny+s\n1o4oCigsmAAJwDufVEa5dLHT6enGt6ebkZmqjdhgnBnXZ0IUhBHfbR1YoWs4QiZgwmgjbM4uNLaP\nnKCRJAlnai3IMGqQZui75ymSOP2pbwzkGGpqd2L7x2dgSE7CT380FYZkVa/vEwUB378hG2kGNU7V\nuXCsqm2YSxo7FVVtqDjXhhvGpeGGcelhf99N16Zjcm4ajp5txfE4ra/yyhZ4uny4NT8zYos5pCSr\nkD8uDVUNNjS2j9yRsIFWX7SWzOxN4Bn+SFr4orHdBbura1i6qwM4sKtvDOQYkSQJv//gJDxeH5bO\nnwSTUdPv+xWigNtuzIYgAFv3nIDbE/8LYPgkKdjCLSyYMKDvFQQBRXPzAAB//qQSvhHWjRiO4Ojq\nyZHprg4IdH8fHKGtZJ9Pwtl6K3JMydBpIr/DU19G4nPkwM3DcPYkjDbroNMo2ULuBQM5Rj4/Uo/j\n1e2YkmcKjm4NxWTU4KaxyWixuLFjX/wPWDp4vBHVDTZ8Pz8LuYPokh2XnYKZ+VmobrDhq+8aolDC\n2HG6u3D0bCtGm3UYbY7s7jzTJpqhVIg4cLxpxD0PBYBzDVZ0erqHtXUMAGMydVCrFMFBUiNBYLTz\nxCguCHIl8eJz5FarGy2WkdO9PxwYyDHQYe/En0rPQKNSoOTu6wbU3Th1vB7Z6cn4x6ELOBXHXT6d\nnm5s/7gSSoWI+2dfO+jz/HDOtUhSinhnX2Vc9Sp8frQB3m4p4q1jAEjWKHFzngl1LQ6cHkGtvYB9\nh+sAAN+7PvJ10x+FKGLCqBTUtzrRNAIGLDndXSg72QRTihpjInxTFwqfI/eOgTzMJEnCH/5+Cs5O\nL4oK8pCe0n9X9ZWUCgEP3zcZAPC73cfR6YnPqVC7vqpCu60T98wcO6TdZzKMWtw7cywsdg/e/6o6\ncgWMoc6ubuzeXw21SoE5U0dF5TPuvvUaAMDOz89F5fzR4nB34cCxRphTNbjx2vDHHETKbTflAAD+\n8fWFYf/sgfr8aAM8XT7MnT5myFPmBur6XP9z5HjcYWwoGMjD7KuKBhw62YyJY4yYM230oM4xYYwR\nd986Fk3tLvz5kzMRLmHsNbY78eHB80hPUWPhrNwhn+/emblIM6jx4cGaEdFyCeXjb2phdXiw4JYx\nfQ4EHKqJY1Jx4/h0HK9ux4nq9qh8RjR8fqQeXV4f5k4bMyzzj6/0veszkapX4bMjdXB1yrdHxidJ\nKP3mApKU/qmVw+2aTD3G56Tg29MtqG1xDPvnyxUDeRg1XVxtSqNS4F8X5Q/pF8YDs8djdIYOH39T\ni+/OtkawlLG3be9peLsl/GjeRKiThr6Un1qlQNFc/2Ih20tH9g2M2+PF7v3V0KoVuOt7Y6P6Wf98\nx3gAwF8/OzsiniX7JAkff1uLJKWIf7o5JyZlUCpE3DljDNyebnxWXheTMoTju7NtaGp3YebkLOi1\nwzfwLUAQBCy6eLO9O056riKBgTxMun0+/M97FXB7ulFy13UwD3ET8CSlP9QVooDf7T4Oh7srQiWN\nrSOVLSivbMX1Y1NxyyC3EuzNzMlZmDDGiEOnmkf0tLF/HLoAu6sLC265Juq/SPNGGXFzngmnLlhw\nbAS0ko+d84fMrZMzYxIyAXOmjoZKKWLvoQvw+eR5I1P6jb9L/c4ZY2JWhikTMzA6Q4cDxxq5SMhF\nDORh8rcvqlBZZ8XM/Cx8/4bIDDbJzTZg8T+NR4fdgzc/HPk7HFnsnXjzw5MQBQFLF0yK6EbpgiBg\n2fxJEATgf/52bEQuDenq9OKDA+eRrFZGvXUc8MAd/gF1I6GVXPpNLQBg3vTYhQwA6LVJuO2mHLRY\n3LJ8RtrY7sTRylZMGG0c1OyFSBEFAffNyoVPkvDBgfMxK4ecMJCHwZHKVvztyyqYUjQouSuyQXPf\n98cib1QKDh5vwjufVMr+l2Zf3B4vXt5+BK3WTvzzHeOjMuozN9uAJfMmwuLw4L/fOTLiRl3vOXAe\nDrcXd88ci2SNMvQ3REButgHTJmagstaKb0+3DMtnDkaLxYXyyhaMzzFgfM7wTnfqzYJb/DcFH30t\nv13HPv6mFhKAeTMGN4Ylkm6dnIkMowafHalHh70z1sWJOQZylH19ogmvvnsEClHEI4vzkRzhhQoU\nooif/OAmZKUnY8+B89jxqfxbMlcK7HRV3WjD7Ck5wWdL0TD/ljEomDoKNU12/M/fjo2YBUM+La/D\nri+rYNSrMH+YuxkfuONaKEQB//O3Y7Jc9MInSXh331lIEjB3WmxbxwE5Jh1uzjPhzAWLrKYn1rU4\n8NmROhh1KtxyXXjrH0STQhRx3/dz4e324aMy+d28DDcGchR9Wl6H13Z+hySliKcenDLk7fH6kqpX\n4+fF05CZpsX7X1Xjr5+NnKkqPknCWx+dwpHKVtx4bTqW3zWwedkDJVzsDp+cm4ZvT7dg2z9Oy/Y5\nX8BXFQ34/Z4T0Gv9S6xq1cPTOg4Yk6nHo/98I7q8Pvyf7Ydxrt46rJ/fH58k4fd7TuDAsUaMyzZg\nZn7sQybg3pn+xwqvvHMElTJYLKSuxYENb38LV2c3fjDn2l63MI2F22/KhlGvGvHrp0dCyP8ikiTh\n2WefxZIlS7BixQrU1PS8iyktLUVhYSGWLFmC7du3R62gI4nT3YV3PqnE1j0noNMkYXXxtOC8u2hJ\nM1wM5VQt/vZlFV599wjqW+U9naC6wYb1b32DfYfrMDZTj8f++cZh+SWhVIj4twduRHZ6MvZ+fQH/\n+cbXqG6wRf1zB+PrE014fdcxaNVK/PRHU4d9AYeAGdeZ8cjifLg93XjpT4dxvjH29eWTJLz54Ul8\ndqQeuVkG/HTJVCQphz4qP1KuG5uGf100GW5PN/73tsOoiOFgwvpWfxhbHR4sWzAJd9w8/FOd+pKk\nVODJoil4eOHkWBcl5kLeau/duxcejwfbtm1DeXk5XnjhBWzatAkA4PV6sX79euzYsQNqtRrFxcW4\n8847kZ4+/BPy5cDq9ODvZTX4x6ELcHu6kWZQ46kfTcXoDN2wfH56igY/XzoNv91ZgW9Pt6D8TCtm\nT8nBvd/PHfKo7kiRJAktFjc+OHAenxyuhSQBt1xnxrK7rhvWlp9Ok4Q1y6dj2z/O4KuKBjz/+68x\n/5YxmDN1FLLTk6PaSg/FJ0k4UtmKf3xdg4qqdmhUCjz5oykxHYADALdOzkKX14f/+/5xPP/7r/G9\nyZlYcMs1w/7MVpIkVDXY8FFZDQ4ca8TYTD1+umTqsK5bHa7bbsyBVqXEazu/w39vL0fxnRNxa37W\nsJXV1enFt6ebsf3jymAYx3JkdV/GZsX22paLkL8BDx06hDvuuAMAMGXKFHz33XfBY5WVlcjNzYVe\n779rnzFjBsrKynD33XdHqbjy5XR78cyW/XC4vUjRqfC/bh+Hgqmjh717MT1FgzXLp+ObUy14d18l\nPjlch08O1yHNoMbEMUaMz0mBUa+CQauCXpsEVZJ/32WlKEAhCoAgQAAAAbg8ki4PqMufUUsX/8//\npwSfBHi7fej2SfB0dcPu6oLN2QWr04OqeitO1nSgzeofvJFjSsbS+ZNww/jY3MAZklX48f/Kx203\nZePND07io7IafFRWg/QUNfJz0zEmU48UXRKMySrok1VQKgQoFSKUChGC4K+T/uoqUE/B2pL8r/ku\n/unx+uDp6oany4cOeyca251obHPhVE1HcAGT68emomjuBFkMVAKA22/KgTpJgb9+fg77Kxqxv6IR\nudn+gVTZaVpkpvs3oFcpFVAliUhSiBAEAeLF+gpU1JW3O1deR90+//+83T64Pd2wOjywODxobHfi\n0InmYP1ck6nHz4qnxXSaUyjTJpnxRNEUvPruUbz50Sn84e+ncd3YVNx0rQlpBjVSkpNgSFZBlSRC\nFAUoRP+fgWvrsj96XkvwX0fd3RK8Ph+83RLsTn89ddg9OHm+HUfPtsHb7YMAYOn8ibIMY7okZFrY\n7XYYDJfuXpRKJXw+H0RRvOqYTqeDzRb7rqxYUCWJ+H5+NrJNybjj5hyohrighQAfnJYmOO09p+fo\nVYDVGvp51MQcFVY/eD3KTrbiWJUFZxvsOHi8CQePx3bjeZ1GgZvHp+L6sSm49XoTlAoxrH/P5Ww2\nK1zOS93xIjxwOjrhcjpgsw38+eY16Qr87MHrcehUG07WWHG61orPj9YP+DyRkqQQMPN6E2bfnInR\nGckApAHX0ZVsNiucTgd8Us+RrIOps0mj1FhddB1OXbDh0yNNOFZtGdYuf5VSxPSJaZg2IR2Tx6bA\n1+WEdYjT8CN9TV1pTLoCP//RZHxzug1Hz3XgeHU7jg/D3O7sdA2mTUjH9AlpMKdqBnUdXVk3Ab29\nRkMTMpD1ej0cjksVHwjjwDG73R485nA4kJIS+k7ebI7P7oknls2I2LnumT8zIueZNPGaiJxHbqZO\nzY/4OSdfNzxze2Nl6tTInm/CBOC+gsj/d4iVaFxTl8sDcOu0vKh+RrREu27IL+QImunTp2Pfvn0A\ngMOHD2PSpEnBY3l5eaiurobVaoXH40FZWRmmRvqnnoiIKAEIUohJq5Ik4Ze//CVOnjwJAHjhhRdQ\nUVEBl8uFoqIifPLJJ/jNb34DSZJQWFiI4uLiYSk4ERFRPAkZyERERBR98pgZTkRElOAYyERERDLA\nQCYiIpIBBjIREZEMDFsgh1oTm3oqLy9HSUlJrIshW16vFz//+c+xbNkyPPjggygtLY11kWTL5/Nh\n7dq1KC4uxrJly3DmzJlYF0nWWltbUVBQgHPnRs4mLbHwgx/8ACtWrMCKFSuwdu3aWBdHtrZs2YIl\nS5bghz/8Id59991+3zts6zr2tyY29fT6669j586d0OmGZw3skei9995DWloaNmzYAIvFgvvvvx/z\n5s2LdbFkqbS0FIIg4O2338bBgwfx0ksv8WevD16vF88++yw0Gk2siyJrHo8HAPDGG2/EuCTydvDg\nQXz77bfYtm0bnE4nfve73/X7/mFrIfe3Jjb1lJubi40bN8a6GLJ27733YtWqVQD8LUClcnjXDB9J\n5s+fj+effx4AUFtbC6PRGOMSydeLL76I4uJiZGbKZxtHOTpx4gScTidWrlyJf/mXf0F5eXmsiyRL\nn3/+OSZNmoR/+7d/w2OPPYa5c+f2+/5h+y3W35rY1NOCBQtQW1sb62LImlbr373Kbrdj1apVePLJ\nJ2NcInkTRRFPP/009u7di1deeSXWxZGlHTt2wGQy4fbbb8dvf/vbWBdH1jQaDVauXImioiJUVVXh\nxz/+MT788EP+Pr9Ce3s76urqsHnzZtTU1OCxxx7DBx980Of7hy2Q+1sTm2gw6uvr8fjjj2P58uW4\n7777Yl0c2Vu/fj1aW1tRVFSE3bt3s1v2Cjt27IAgCPjiiy9w4sQJ/OIXv8Brr70Gk8kU66LJzrhx\n45Cbmxv8e2pqKpqbm5GVlRXjkslLamoq8vLyoFQqMX78eKjVarS1tfW5RfGwJWJ/a2JT77iIWt9a\nWlqwcuVKrF69Gg888ECsiyNrO3fuxJYtWwAAarUaoijyZrgXb731Ft588028+eabuP766/Hiiy8y\njPvw7rvvYv369QCAxsZGOBwOmM3mGJdKfmbMmIHPPvsMgL+e3G430tLS+nz/sLWQFyxYgC+++AJL\nliwB4F8Tm/p3+R7E1NPmzZthtVqxadMmbNy4EYIg4PXXX4dKpYp10WTnrrvuwpo1a7B8+XJ4vV48\n88wzrKcQ+LPXv8LCQqxZswZLly6FKIr4r//6L97k9aKgoABff/01CgsLgzON+ru2uJY1ERGRDPCW\nhoiISAYYyERERDLAQCYiIpIBBjIREZEMMJCJiIhkgIFMREQkAwxkojjz6quv4tChQ7EuBhENEAOZ\nKM4cPHgQPp8v1sUgogHiwiBEI1hjYyN+9rOfweVyQRRFzJkzB6+//joyMzPxm9/8Bu3t7Xj55Zfh\ndrthtVqxevVq3H333VizZg3a29tRU1OD1atX48CBA/jqq68giiLmzZuHxx9/PNb/NKKEwxYy0Qi2\nfft2zJ07F++88w5Wr16N5ORk3HTTTfjVr36FiRMn4g9/+AN+9atfYceOHfjP//zPHtt6pqWl4f33\n38ekSZPw2Wef4a9//Su2bduG8+fPB/e7JaLhw01kiUaw2267Df/+7/+OiooKzJ07F8uWLUNpaWlw\nY5Jf//rX+Pjjj7Fnzx6Ul5fD6XQGv3fKlCkAgKysLGg0GhQXF2Pu3Ll44oknuNY1UQywhUw0gk2f\nPh3vv/8+7rjjDuzevRuPPvpoj8Xri4uLcfToUdx444149NFHe+wgFth+UaFQ4M9//jOeeOIJdHR0\n4MEHH0R1dfWw/1uIEh1byEQj2K9//WtkZmbioYcewsyZM3H//ffjhhtugNfrhcViwfnz5/H2229D\npVLh1Vdf7XWw1/Hjx/H888/jrbfewsyZM3Hs2DGcO3cuuN8tEQ0PBjLRCFZSUoKf/vSn+Mtf/gKF\nQoF169ahrq4Ov/zlL/Hiiy+isLAQCxcuhMFgwNSpU+F2u+F2u3ucY/LkyZg2bRoWLlwIrVaL/Px8\nzJ49O0b/IqLExVHWREREMsBnyERERDLAQCYiIpIBBjIREZEMMJCJiIhkgIFMREQkAwxkIiIiGWAg\nExERycD/B08snUTqOh2cAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x160998ad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df.stars)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Group by Business"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "biz = df.groupby(\"business_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11537"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(biz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th colspan=\"3\" halign=\"left\">stars</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>len</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>hW0Ne_HTHEAgGF1rAdmR-g</th>\n",
       "      <td>844</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.236967</td>\n",
       "      <td>1.040466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VVeogjZya58oiTxK7qUjAQ</th>\n",
       "      <td>794</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.964736</td>\n",
       "      <td>1.217264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JokKtdXU7zXHcr20Lrk29A</th>\n",
       "      <td>731</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.310534</td>\n",
       "      <td>0.794251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ntN85eu27C04nwyPa8IHtw</th>\n",
       "      <td>679</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.998527</td>\n",
       "      <td>1.109757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWMwV5V9BxNs_U6nNVMeqw</th>\n",
       "      <td>645</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.130233</td>\n",
       "      <td>0.948242</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date  stars                    \n",
       "                        len median      mean       std\n",
       "business_id                                           \n",
       "hW0Ne_HTHEAgGF1rAdmR-g  844    3.0  3.236967  1.040466\n",
       "VVeogjZya58oiTxK7qUjAQ  794    4.0  3.964736  1.217264\n",
       "JokKtdXU7zXHcr20Lrk29A  731    4.0  4.310534  0.794251\n",
       "ntN85eu27C04nwyPa8IHtw  679    4.0  3.998527  1.109757\n",
       "EWMwV5V9BxNs_U6nNVMeqw  645    4.0  4.130233  0.948242"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_summary = biz.agg({'date': len, 'stars': [np.median, np.mean, np.std]})\n",
    "\n",
    "biz_summary.sort_values([('date','len')], ascending=False).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th colspan=\"3\" halign=\"left\">stars</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>len</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>-1bOb2izeJBZjHC7NWxiPA</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.32</td>\n",
       "      <td>1.096190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-O7H5LhBkE13VQZB0QZI0Q</th>\n",
       "      <td>50</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.92</td>\n",
       "      <td>1.139996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>679pJn1KyKr2X8NcQNyMzA</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.08</td>\n",
       "      <td>1.065986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74qDPs2dR6gSzDp6N1U-Og</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.94</td>\n",
       "      <td>1.150155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>863nEWkPZ7uD-ZRTmmJokw</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.66</td>\n",
       "      <td>1.318781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8qrICL2tS2Rq7b5gxUdQwQ</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.76</td>\n",
       "      <td>1.270706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BJfjd8oEMW_zq8KTkixyig</th>\n",
       "      <td>50</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.30</td>\n",
       "      <td>1.092647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>G5SASWuL_CVxpgwXXGC4DA</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.46</td>\n",
       "      <td>1.459801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hw8-13eEQKdtTnUxwUkkHw</th>\n",
       "      <td>50</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.46</td>\n",
       "      <td>0.734291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K4Oo7A5Kp5eSgBV7MgOu8Q</th>\n",
       "      <td>50</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.40</td>\n",
       "      <td>0.857143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Li5LOL873Ep8HoPRML18sw</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.84</td>\n",
       "      <td>1.131371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NA3tQYxR6Fq5O8nV6u41Tw</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.66</td>\n",
       "      <td>1.349376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Q8OMwgekiTtgClSiRcQcSw</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.22</td>\n",
       "      <td>1.329262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>QiNZaRZ5RmCfYiBgk3ApEQ</th>\n",
       "      <td>50</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.98</td>\n",
       "      <td>1.186489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>To5CjntVFawXM_hpAhvVdQ</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.08</td>\n",
       "      <td>0.876915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V7Da92RVtaWGTL6GNkPT8A</th>\n",
       "      <td>50</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.18</td>\n",
       "      <td>1.082137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XYhVyCbzJCrYclR6J4LYSw</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.18</td>\n",
       "      <td>0.628896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZJxs1L4oMhX9GW00ty0aAg</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.28</td>\n",
       "      <td>1.212856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dyd81TK9jY-WLt6UbMFWng</th>\n",
       "      <td>50</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.32</td>\n",
       "      <td>0.978128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jEkaxdgLxUCf4jtUnCPVWg</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.30</td>\n",
       "      <td>0.814411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lut0g04VsYRyblR1Ad-8HA</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.34</td>\n",
       "      <td>1.318781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ntJAKLAq0cXzzdFgPoo0Iw</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.52</td>\n",
       "      <td>1.265556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>upzSExkcvYMX9dJMgNARXg</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.06</td>\n",
       "      <td>1.095631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vvA3fbps4F9nGlAEYKk_sA</th>\n",
       "      <td>50</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.86</td>\n",
       "      <td>1.324957</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date  stars                \n",
       "                        len median  mean       std\n",
       "business_id                                       \n",
       "-1bOb2izeJBZjHC7NWxiPA   50    4.0  3.32  1.096190\n",
       "-O7H5LhBkE13VQZB0QZI0Q   50    3.0  2.92  1.139996\n",
       "679pJn1KyKr2X8NcQNyMzA   50    4.0  4.08  1.065986\n",
       "74qDPs2dR6gSzDp6N1U-Og   50    4.0  3.94  1.150155\n",
       "863nEWkPZ7uD-ZRTmmJokw   50    4.0  3.66  1.318781\n",
       "8qrICL2tS2Rq7b5gxUdQwQ   50    4.0  3.76  1.270706\n",
       "BJfjd8oEMW_zq8KTkixyig   50    3.0  3.30  1.092647\n",
       "G5SASWuL_CVxpgwXXGC4DA   50    4.0  3.46  1.459801\n",
       "Hw8-13eEQKdtTnUxwUkkHw   50    5.0  4.46  0.734291\n",
       "K4Oo7A5Kp5eSgBV7MgOu8Q   50    5.0  4.40  0.857143\n",
       "Li5LOL873Ep8HoPRML18sw   50    4.0  3.84  1.131371\n",
       "NA3tQYxR6Fq5O8nV6u41Tw   50    4.0  3.66  1.349376\n",
       "Q8OMwgekiTtgClSiRcQcSw   50    4.0  3.22  1.329262\n",
       "QiNZaRZ5RmCfYiBgk3ApEQ   50    3.0  2.98  1.186489\n",
       "To5CjntVFawXM_hpAhvVdQ   50    4.0  4.08  0.876915\n",
       "V7Da92RVtaWGTL6GNkPT8A   50    3.0  3.18  1.082137\n",
       "XYhVyCbzJCrYclR6J4LYSw   50    4.0  4.18  0.628896\n",
       "ZJxs1L4oMhX9GW00ty0aAg   50    4.0  3.28  1.212856\n",
       "dyd81TK9jY-WLt6UbMFWng   50    5.0  4.32  0.978128\n",
       "jEkaxdgLxUCf4jtUnCPVWg   50    4.0  4.30  0.814411\n",
       "lut0g04VsYRyblR1Ad-8HA   50    4.0  3.34  1.318781\n",
       "ntJAKLAq0cXzzdFgPoo0Iw   50    4.0  3.52  1.265556\n",
       "upzSExkcvYMX9dJMgNARXg   50    4.0  4.06  1.095631\n",
       "vvA3fbps4F9nGlAEYKk_sA   50    4.0  3.86  1.324957"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_summary[biz_summary.date.len == 50]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>34871</th>\n",
       "      <td>fXtPvVb8vbWvAa_mRjq2ig</td>\n",
       "      <td>2010-01-27</td>\n",
       "      <td>9Qvjh0Qnba8c_0MK32h4Iw</td>\n",
       "      <td>1</td>\n",
       "      <td>Will not be returning there.</td>\n",
       "      <td>NLCFdOpNwk2CzR5iQhIg2w</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116076</th>\n",
       "      <td>fXtPvVb8vbWvAa_mRjq2ig</td>\n",
       "      <td>2009-03-02</td>\n",
       "      <td>mdKpC4FTDRm6H-yfTBdU3g</td>\n",
       "      <td>4</td>\n",
       "      <td>So downtown its hard to find a decent Chinese ...</td>\n",
       "      <td>heGVdrtlVTbmxASUTVNlaw</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>224391</th>\n",
       "      <td>fXtPvVb8vbWvAa_mRjq2ig</td>\n",
       "      <td>2009-06-26</td>\n",
       "      <td>PvNGOvQ7xfggy55hxKMSfQ</td>\n",
       "      <td>4</td>\n",
       "      <td>Well, I was actually a Weekly Boofay regular o...</td>\n",
       "      <td>n_lAaNGeWgspbbCJc0DqRg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   business_id        date               review_id  stars  \\\n",
       "34871   fXtPvVb8vbWvAa_mRjq2ig  2010-01-27  9Qvjh0Qnba8c_0MK32h4Iw      1   \n",
       "116076  fXtPvVb8vbWvAa_mRjq2ig  2009-03-02  mdKpC4FTDRm6H-yfTBdU3g      4   \n",
       "224391  fXtPvVb8vbWvAa_mRjq2ig  2009-06-26  PvNGOvQ7xfggy55hxKMSfQ      4   \n",
       "\n",
       "                                                     text  \\\n",
       "34871                        Will not be returning there.   \n",
       "116076  So downtown its hard to find a decent Chinese ...   \n",
       "224391  Well, I was actually a Weekly Boofay regular o...   \n",
       "\n",
       "                       user_id  \n",
       "34871   NLCFdOpNwk2CzR5iQhIg2w  \n",
       "116076  heGVdrtlVTbmxASUTVNlaw  \n",
       "224391  n_lAaNGeWgspbbCJc0DqRg  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.query(\"business_id == 'fXtPvVb8vbWvAa_mRjq2ig'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x17cab0a50>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAFkCAYAAADxHkghAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4yWgF+O7QFcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x16989f050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# All the stars ratings, means\n",
    "sns.distplot(biz['stars'].aggregate(np.mean))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Grouping by Reviewers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "revs = df.groupby(\"user_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "45981"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(revs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "rev_summary = revs.agg({'date': len, 'stars': [np.median, np.mean, np.std]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th colspan=\"3\" halign=\"left\">stars</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>len</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>fczQCSmaWF78toLEmb0Zsw</th>\n",
       "      <td>588</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.930272</td>\n",
       "      <td>0.791932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90a6z--_CUrl84aCzZyPsg</th>\n",
       "      <td>506</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.944664</td>\n",
       "      <td>0.784062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0CMz8YaO3f8xu4KqQgKb9Q</th>\n",
       "      <td>473</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.002114</td>\n",
       "      <td>0.714555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4ozupHULqGyO42s3zNUzOQ</th>\n",
       "      <td>442</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.260181</td>\n",
       "      <td>0.654791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>joIzw_aUiNvBTuGoytrH7g</th>\n",
       "      <td>392</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.135204</td>\n",
       "      <td>0.985618</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date  stars                    \n",
       "                        len median      mean       std\n",
       "user_id                                               \n",
       "fczQCSmaWF78toLEmb0Zsw  588    4.0  3.930272  0.791932\n",
       "90a6z--_CUrl84aCzZyPsg  506    4.0  3.944664  0.784062\n",
       "0CMz8YaO3f8xu4KqQgKb9Q  473    4.0  4.002114  0.714555\n",
       "4ozupHULqGyO42s3zNUzOQ  442    4.0  4.260181  0.654791\n",
       "joIzw_aUiNvBTuGoytrH7g  392    3.0  3.135204  0.985618"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rev_summary.sort_values([('date','len')], ascending=False).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x117e2f290>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAFkCAYAAADxHkghAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nvPpwIPTkZO4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1614f7e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(revs['stars'].aggregate(np.median))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "user_id\n",
       "fczQCSmaWF78toLEmb0Zsw    588\n",
       "90a6z--_CUrl84aCzZyPsg    506\n",
       "0CMz8YaO3f8xu4KqQgKb9Q    473\n",
       "4ozupHULqGyO42s3zNUzOQ    442\n",
       "joIzw_aUiNvBTuGoytrH7g    392\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "revs.size().sort_values(ascending=False).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# multiple reviewers - good, there are a lot\n",
    "multiples = revs['user_id'].filter(lambda x: len(x) > 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "192214"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(multiples)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Are There Review Trends Over Time?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "one_business = df.query('business_id == \"VVeogjZya58oiTxK7qUjAQ\"')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1689a3bd0>"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeYAAAFkCAYAAAD165gcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nP3nznmCWPDgTAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x121ea9f90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "one_business.rolling(window=15).mean().plot('date', 'stars')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAekAAAFkCAYAAAAE4MORAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17d856d50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeYAAAFkCAYAAAD165gcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsvXmgHFWZ//2trW/f/SY3NwtJSEICBAhr2EEIywgioqPE\nERUXGEdfHZ2ReR1k5v0NMzg6ODMu42/cZxxE1FE0bgiiCAgEgbBDIISE7Ou9yd1671reP6pP1anT\ntZzqru7b3ZzPP7m5fbu6uurUefbnkSzLsiAQCAQCgaAlkGf6BAQCgUAgELgIwSwQCAQCQQshBLNA\nIBAIBC2EEMwCgUAgELQQQjALBAKBQNBCCMEsEAgEAkELodb6xre//e3o6+sDACxatAif+9znnNfu\nv/9+fO1rX4OqqnjHO96BtWvX1n+mAoFAIBC8DqhJMJdKJQDA7bffXvWaruu49dZbsW7dOnR1deGa\na67BJZdcgtmzZ9d3pgKBQCAQvA6oyZW9adMm5HI5XH/99fjABz6A5557znlt69atWLJkCfr6+qBp\nGlavXo0NGzYkdsICgUAgEHQyNVnM6XQa119/PdauXYvt27fjQx/6EO69917IsoxMJoP+/n7nb3t7\nezE9PZ3YCQsEAoFA0MnUJJiXLl2KJUuWOD8PDQ1hdHQU8+bNQ19fHzKZjPO32WwWAwMDocezLAuS\nJNVyKgKBQCAQdBQ1Ceaf/vSn2Lx5M26++WYcOHAA2WwWIyMjAIDly5djx44dmJqaQjqdxoYNG3D9\n9deHHk+SJIyOdq5VPTLS39HfD+j87yi+X/vT6d9RfL/2YmSkP/C1mgTz1VdfjZtuugnvfve7Icsy\nPve5z+Huu+9GPp/H2rVrcdNNN+G6666DZVlYu3Yt5s6dW/PJCwQCgUDwekJqlelSnaQJsXSapudH\np39H8f3an07/juL7tRdhFrNoMCIQCAQCQQshBLNAIBAIBC2EEMwCgUAgELQQQjALBAKBQNBCCMEs\nEAgEAkELIQSzQCAQCAQthBDMAoFAIBBQ/PSnP57Rz6957KNAIBAIBEnx4/u3YMOmg4GvK4oEw4jX\nduOMlXPxzotXxD6X22//b7zjHe+M/b6kEIJZIBAIBK9bdu3aic997p+gqiosy8Lq1WdgamoKX/zi\n5/GRj/wlbr31n5HJZHDo0Cj+9E/X4m1vewc+/vEPY9as2ZiensInP/m3+Jd/ucV5/803/zNGRurr\ndikEs0AgEAhmnHdevCLUum1U568NGx7H8cevwkc/+gk899wzGBqahZ///Ke44YYbsXnzJlx66WW4\n4II1GBsbw8c//hd429veAQB44xsvx/nnX4h16+70vD+TydQtmEWMWSAQCASvW6688q3o6+vDDTd8\nHOvW/RiqqjivzZ49jIceegCf+cw/4Lvf/W/ouuG8tnjxEt/3K4pS9RlxEYJZIBAIBK9bHn74Dzj5\n5FPxH//xNaxZcwnuuOO7AOxY9g9/eAdWrToJ/+f/3IKLLrrE+T0AZ1Qx+/7vf/+7dZ+TcGULBAKB\n4HXLypXH4bOf/UdomgbTNPGJT9yAAwf24zOf+QdceeVb8aUv/St+//vfoq+vD4qiolwuO0I56P31\nIqZLNYFOm4riR6d/R/H92p9O/47i+7UXYrqUQCAQCARtghDMAoFAIBC0EEIwCwQCgUDQQgjBLBAI\nBAJBCyEEs0AgEAgELYQQzAKBQCAQtBBCMAsEAoFA0EIIwSwQCAQCQQshBLNAIBAIBC2EEMwCgUAg\nELQQQjALBAKBQNBCCMEsEAgEAkELIQSzQCAQCAQthBDMAoFA0AR0w8TYRH6mT0PQBgjBLBAIBE3g\nm7/YiL/9xh+x+2Bmpk9F0OIIwSwQCARN4KnNowCAHQc6Z6awoDEIwSwQCARNRJGlmT4FQYsjBLNA\nIBA0EVkIZkEEQjALBAJBE5ElIZgF4QjBLBAIBE1EEoJZEIEQzAKBQNBELMua6VMQtDhCMHcQpnjg\nZxTLspAv6jBMc6ZPRdDCGGZrPaeWZYm9o8UQgrlDKOsm/vzzD+A7v355pk/ldct37n4ZH/vSQ/jb\nr/8RZd2Y6dMRtCi60VqK29d+/iK+8L/PzvRpCCiEYO4QDk8XAACPvLBvhs/k9cvWPVMAgPHpIsan\nizN8NoJWpdUs5l0HMnht79RMn4aAQgjmDqFYEhbaTJMrlJ2fswV9Bs9E0Mq0msVsmBaKZaPlzuv1\njBDMHUJBCOYZxbIsjzDOUkJaIKAxjNaymElOhFAmWwchmDsEIQhmlmLZ8Lgoc2KTEwSgt1hyoFlZ\nt9m82ENaBSGYO4RsXgiCmYRcf9I8QmxyAhqTUtr0lrOY7fMRymTrIARzhyAs5pmFXP+RWd0AgIzY\n5AQUZd21ko0Wi+USwZwRe0jLIARzhyDiQzMLuf5zh2zBnBObnICiRJXPtZrFbDoWs1izrYIQzB1A\nsWzgrke3x3rPXY9uFzXPCUJc10Qw3/vELuEaFAAApnMl3PydJ5z/t1r2s+HEmMV6bRWEYO4A9h/K\nOT/3plWu96x76DU88sI+0fEnIYpl2yI6YqTX+d2ug2LursCevzyRKTn/b6WsbMuyXMEsLOaWQQjm\nDoDOBo7bIL9QFFpyEpB70KXJeNclRwMAMsICEcCNL19w8gIArZWVTSvmIhzWOgjB3AHQrrG4fZpF\nklIykIQeVZEdr4WwQASAK5i7u+x10UoWM50tLtZr6yAEcwdAZ3nGTSwRCR/JQK67IsvoTWsARPmJ\nwKZU9grmVrKY6f1CxJhbB76ApKCl0T01kvEeevEwJgPZbFVFQo+wmAUUZKBJd6oimFvJYrbopjhi\nvbYKwmLuAGhhbFle91QUQngkg2MxKxJ6KhaziNkJAKBU5cpuHYuZzk8RYa3WQQjmDoCNWUVZzfTr\nQngkgxNjlmX0EYtZdP8SgI4xKwBay2Km9w5hMbcOQjB3AKwgjnrw6S5ERHjkizq+d+8rODCeC3pb\nTTzwzB5885cbcWiykOhxWw1yzVVFdizmQ1OFhlxTwis7x3HnA1tgiZK3loa1mHOFMn5w32ZMZGZ+\nNKgn+Suvi7XUIgjB3AGwgjgqM5sWzCRB6e7HduCBZ/bgaz97MdFz+969r+Dxlw7gqc2jiR631SDK\nkaJI0FQZqiLjtb1TeOCZPfjKT55vyGd+/gfP4J7Hd2LHAVEv3cqQGHM6pUKRJWzdO4X7ntyd+LNW\nC/ReYVqWmFLXItQsmA8dOoQ1a9Zg27Ztnt/fdtttuPLKK/G+970P73vf+7B9+/Z6z1EQASuIoyxm\nuj0g6Y87nbP/ncyWfN9TL53u1iWxOlWxH6mU6j5aoxP5hn52XoQjWhpiMadU2UkMBIDx6Zn3IhlM\nPorIOWkNasrK1nUdN998M9LpdNVrGzduxL/+67/i+OOPr/vkBHwQQZzSZJTKZmRyiZ8rm2RnyvH6\nk3DT6aVDbh2zfQE1TQYqnspGxxRLeuskEwmqIc+bptmldEQJbgWqBHNex5zBGToZgUNNFvPnP/95\nXHPNNZg7d27Vaxs3bsQ3v/lNvPvd78a3vvWtuk9QEA0RCmmtklwSkZVN6ioBV2CS2JLcIMnc6Zq4\nm5VdbTE3GtIOVNCalB2LWeFumdss2AqOTn9O24XYq2TdunUYHh7Geeedh2984xtVr7/5zW/Ge97z\nHvT19eFjH/sY/vCHP+DCCy+MPO7ISH/cU2krGvn9urpTAICebg1TuTIGBrpDP28s4z58Rd3EyEg/\nUpUaS1VVaj7XsPeVTavt73HY+asVpWjeSD/mDHWjO60BcF2Vjfzuch33jKbd7w8PM/EdpYqyu2D+\nAGYNdgN7pwAAsiInfj5xjzdRUcxVRYZumFBTWkuvg1Y+tySpSTBLkoT169dj06ZNuPHGG/H1r38d\nw8PDAID3v//96OvrAwBceOGFeOmll7gE8+ho5yawjIz0N/T7TU7ZMUy1sgGMjmXQowZbvgfH3HOZ\nzBQxOjqNPIkBm1ZN5xr1HcenCm19j6O+XyZnx+YnJnKwyjrYq9/I735gLFP38Ru9RluBmfqOmUre\nxtREDhrlSDENM9HzqeX7jR3KAgD6ezSMTxex9+AURkdbU/h12hoNVfTjHuyOO+5wfr722mtxyy23\nOEI5k8ngyiuvxD333IN0Oo3HHnsMV199dQ2nLIgDcaN2EVd2RIyZjkkS1xWJNcWcgRGKtxSjs11k\nbIy5ma5s0b2ttSHPm6q6pXQAEKMPUMMgz+hATwrj08WOzwVpF+oKeJBJRnfddRfy+TzWrl2LG264\nAddeey26urpwzjnn4IILLkjkRAXBEKHQleITzDolmEtlE+seeg0bNh0EkGyMmU4sOTCexy8e2Ya+\nbg2L5/ahrJs4YdnsxD6r2WzeNYFXd09AkWWcsXKuW8cs+8eYH3puLy44+YiGnMvvntyFqy9aDjlJ\nrUpQN5PZEh7buB+bd01AVWTIkuSJMY9PF3F4qoDZA9VJtM2C7B39PZVudQkr0JZl4eHn9+HEo4Yx\nq78r0WN3MnUJ5ttvvx0AsGzZMud3V111Fa666qr6zkoQC3fkIF/yF0lGkSS7heddj253XktSMLOJ\nJb94xFta951PX5zYZzWbL935HIqVms/fPbkLC4Z7ANh1zACgqYrn72+7ZxNWLpmFuUPdiZ2DLEkw\nK/N0t++bxlFHDCR2bEH9fO1nL+DV3ZMAXGV5hLn/9z25G++8eEXTz41gVJI+iWDOJTwG9tlXx3Db\nPZswb3YP/uUvzk702J1Ma6UICmqCPPREMEf1yiYDF977J8dg4UgfXt4x7gjNJK2uuCMo24ki1Yhh\nfLrobLhKRbFJUcHEdEpBoWQkXm8syxLMiqUuMrNbj237ppyfP/CmlQCAc1bNx4LhXmTyJXz5zudR\nmOH7RlpypitdyeiKjSQYm7ITIA8cbkz3u05FdP7qANw6Zlsws7WJLERwp7tUHLN4CMsXupZWsoK5\nBYJoTcIwTKiK5IR3NMqVferRcwB468frxbQs7xzuFhqMILChvSZHDPcCsJ+vo44YwPzZtocl7jS4\npHH2gkoYrKwnqyi00uzpdkII5g6AbMrESot6GIjAJNZdL5WQkmSYspMFM3uZdMNyapgB76ZMrm8p\nwU2PFfKtNBhBYEPnGfQw9cukQ9xMK1SGI5grFnPCzWo62WvWSIRg7gB0JsZsRjSiJw8jsY57u13B\n3MgYcydBEu0Iumk65WqAd1Mm1zdJi7laMIsNsNXQfNYAgShxM61Qkb2g27GYk11HnbwHNBIhmDsA\nsikTYRClpZKHhSQq0ZmizbSYoxSIdkGS/CzmamupoYJZWCYth0cwMxYz8VbNtEJF9oKUpkCWpMQF\ncyd7zRqJEMwdgMHGmGO6ssk4OiBZDZc31t2O0N9NU2UnxkygLWbiyUjSlU2ORT5RxPJajxQVzlAV\n71ZL1spMCy6i0MmSBE2TE12jgKt8i0q+eAjB3AHEzcp2BbN9++mEryRda1GbDv36ZLaEZ7eMtU12\nMX2NS2UTY5MFp4YZ8MaYieUUJ36XK5SxM2ScY7mSPUuyaWfa8popDk7k8dQrB7Gv0sGqWZR1A1v3\nTGLbvils3TPpmzSlhTSZaZUYs0kp6ZoiI1/UsXXPZGJzmYnCqDRqOk6TIH0L6Ez7RiLKpToAw3FH\nyZ7/B/59ZTPwiycnqcFHbTqGYQGV0Bup+Xzr+cvw1vOXhb5vprEqtcMstOeBuC5lSXKs53KMUpRb\nbnsSByfy+MLHzvNtzFCsCIKeLgX5oj7jscqZ4ks/fg4HDufQ163hK3/1hqZ97v/cvQmPvXTA+f95\nJ87H9W/2TtQLe5ZcV3ZrxJgVRUJKkzE6UcBnv/cUPvq2VTh9ZfWQopqPL7evDbhnNINbv/+08/+b\n3nsajl401NDPbN+rJXDQDROS5FppvLFdWov9xw+e4RwrKaJiyPTre8dsi+fQ1MzPqI2CnDfbLOS6\nNx/n/Hzm8fPwgTetxN+/b7VjOZVjXNuDlRnOUwHzsUlN9ECvPcBkpi2vmWK6cn0y+TJKTfS2kE55\nhCc3jVb9DbGiP/uhs6pekyQJiizNeG6AM5JSkT1enh0h3po4OImmbWwxs2M69x1qfE22EMwdgG5Y\nTss/II4r231YjpzXjzmD6UQ3eF7LHXCznNuhpza5vnNnuYL54tMWYvHcPuf/XZqCC04+AssWDDgb\nXpKCI1PpcT7Ya1vTMx2rnCno751tYp9nks9B8HNbl3UTg30pLKjUMLPYE51m9r6VqFnRfnkR9WL6\n7DXtBqs8NaMPvhDMHQBJPCKLP1og+muxSsIbBW8SGuBa++3QRF/3uX50LTiL48quIePVgv81zDkW\ns1Y5p9enxUx/72YqdelUtGAu6WboJq4q0ox7OohVn1IVr2BOJSOY/YyAdoPdE9lEvkYgBHMHoJsW\nFFl2yp+4y6WYh0WVpWRd2TGysq3Kz+0wqN0vFMCWw9DUkvxFCGqRSIQQcWXPtOU1E7Cx/mauHdai\n9BPAZd2s6plOo8jSjN83sr40VfYoF0lZzGH5LO0Cqzw1I/wgBHMHoBsmFEVyFj9vNjQrmBVFihyA\nEQc2KS3odcBd7M10R9aKX6Yp69qkIa/V0u4wyMom12mgpyKYX4d1zOw6b+baYQWXnwAu6UZoZrbt\noZphi9mgBbP7HZLKyibKaCdZzHGSOGtFCOYOwDBMqLLsdJ7ijTGzWqyqyA2JMacCrAaPYDbax2L2\nu35hG7Cm1GExBwhzcp0G+yox5tehxcwKtaZazIyrl1U+LctCuczhyp7h3AAiZFKqN8aclCVf7gjB\n7F1ncZI4a6VtBLNlWRibzKNYNtq6MUUYhmnWZFXppgU1hsXsdv5imh5UXGsHxnOYzvlnA8eBfE4X\nh8VMFIJS2URZN6EbJg6O51AotZ4F7YYC3O8VZjFrWu0xZvY9hZKOg+M5TEwXAQADPW6MuRWvVdJM\nZIoYr3x3sn7Ilp/NN+/7V8WYFdlz/Q3Tzg4IF8zJKsK1QBQ/TVWcdQokl0xI9jN2r2kn2GuR9AQu\nP9qmjvmH972K+57aDQBYONKLz1xfXYLQ7vzt1/+I8eli7DnFhmFBTcuOoIh2ZVe0WKYdj1YRLjd9\n8zFIEnDLdWdi4Uhf1fu5z6vyOUHJEqbHle3+nCuUcds9m/Dc1kMY6NHw7x87rykJF7w414+yAvq7\nG5P8RW8CpmXh77/9uCOYZElCf8WV/crOCXz0iw9h7ZrleNPZS2J/Tjvw3JYx/MdPngcAfPiqE7By\nySwAwEBfCpOZUlMtZnYK2yu77Ov/rx85B3OGuqnYbViMeeazssmaTGmyx7OVlMJQoma/tyvstUh6\nApcfrbPbRfAa1XFlz2hzu/w0C7LhxoXEmJWYrmxF8T4tV56zBG84aQGWLRiAZbm1tLVi+FjmH6zM\npQW8cVF68Zd1063jzZVbrhsYff3+6boz8acXHIVjjwxuOJCqo1yKdpsVSwbGp4uYPdCF809agPde\ndowj9PdU6sDvfmxH7M9oF+iZvgcO55w1M1hJgGtmRj9Zu5eevsjze7JP5Yq2kkA3nWFRldaqY750\n9SKnNj+p2HeuaN+TdvZyEuWpniTOuLSNYBYE49QxE1c2Z5kSG2M+9shZ+OAVx+HcVfMB1D90gZ1i\nBQBvOPkIvPkc26IjD6tlWR7LQTctz3eYaauChb5+i+f24S3nLnXmMPshy7bSxHs96cSbMiXMyfuP\nWjCA6644DmtOWVjlIgxzqbc79IZYqoQ7ADczvZkWM1mf77r4aPRR3hKihBG3eli2vu3KbpE6ZlXG\norl9+OAVtuKc1DNHlKWZ/p71QJSnKyqeqKQHffghBHObY1kWDMOEIlN1zFEdtyJqC+txvfJ8DmvZ\ns6533TB948+tgvO9YvjnUprMrWnTmyL9Hse6oVyOrNeDjX12EmXmWpDrNNhDBHMTLWbDhARb6aIT\n/8j9yFWUBHbcI40i28lfMzllrVzJHCeKJQkZJWXJk7K+mU5yqweyzoj3QwhminbWuOISp1TBtOwk\nE1WRKYEX0aM6QjCTJJB6XTZBljn5P4krs/fWMLyb1UyXlLAEhQLC0BSZ+4GmY1heK7HSDIJK0lHl\n14/F7BXMhrMu0l0qNFVuaoMResynJ5tZ95b9hVvMfB6uRsI2QVESPKeybjjrt51HvBLDoLsr+Slx\nQbSNYG7GxWgV4miXRJujY8xxxz6yaEql7rbO2G6QAGMtZlY71w3TYyW3siubF01VuJNGSowAcn92\n44EElbWYO1gwlxiFhdwHVZHQm1ab7Mp2x3zSHgxy70jL1LCOcESwz6TiaTdBqVb0kjgn2oPRal6v\nOJD9p0dYzNU042LMJHS3rjiJEmTBqzIVY+Z2Zfvf/lRCFjP5HLXKlU1G3lUEMyN4WVd2q1nMUdfP\njziubDaWyv5Ml7Ww2eqd7MoOijGriozebq2pyV+GaTnXnhZsRHkg59LbHR5jJseaKVjBrCQ4J5r2\nYLS3K5tYzPa9bEa5VNsIZnZTa2fXiB/0zY5lMVNWg6J4BV4QRJgHhUhTCWUfBrXjY3t6V7e887qy\nW+2hJucbp2mCpvILZk/CF7UuyO/pshZZljz3Makex62Ix5VdNjwd2Hq7VOQKetP2BVIJEXSORCj1\nhFjMrit75hTPUtnwrCc1QSveYzG32DMcB3Luboy58d7blq9j/uYvN+LZV8eqSmamc2XcesdTuOLs\nJXjDyUfUdOzv/fYVPPL8PpR1E+mUgjedvQRvOXdpAmcdH7osJs4iJpuTN8YcIZgtC4osBWYSE9fc\n+FQBN37jUbtkqWRAkoCzjpuHv7jqBL5zs/w7f7mNUOzvzLYBNQzTo1y0Wn4B+V5xXNkpVeFu5ffz\nh7c5PxML7FePbsfPHnqtciyvPm3Xw9rH7jB91YNHMBumEwJRFRk9aQ0WgN88vtPJnm0kumE5bl9P\nFr0TYyau7OAtlnhcbvzmH3HByUfg3ZceU/d5ff93m3FgPIcb3nkK199XubIrgnn9C/vx5CujkCUg\nXzRw1vHzsHnXBD54xUpYFvCdX7+MT7/3NMyb1RN4bDq00M7lUkRxIt6ojdvHsXcsiyPm+E8NS4KW\nt5hHhtK+dazPbxnDgfE8/ueeTTUf+4Gn9zgPUqFk4IXXDtV8rHqhN+04i5hsyIoiOWVJPJ2/wqw9\nsvFv3TuF0YmCU39rWfAMh488t8q1fcNJC3Dq0XPw6fec5pwrOQ/Ax2I2rLZwZbMu+jA0VYZpWZED\nRgBgbNKdSU2UEiKUybFoLl29CHMG0wBa71olCRtj1iml9JSj5wAAtu2d8n1v0uimG2O+7MwjPecF\n2PsJAKRTwYL5zOPmYtmCfhiGhY3bDidyXr9/ajdefI3vWLphoqSbnlpr2gtQKhvIF+3v8fhLBzA+\nXcS3fvkSvvXLjZjMlnDfht2hxy+W3PvVzhazm8cjO532dh3MNPQzW14wv/2C5Thp+XBTPmsmZwHT\nm0685C/iVpW540OGYYVae2Tjn8jYDU/mMlox7+ZPNqnebg0ff8dJOGax3YRDYRQIsvCJAV8VY26x\nAQ1BYzPDcJoTcFjNJd0IdSmymdfvvHiF0wmvnTfAKJy+zpqMctn0hBTOOn4egOY0fwAq3fYq9+is\n4+c5Sidxc9IdtYI4ecUc/J/3n4HBvlTiOTQ8yj1p/tFDWfV0lv/wQLrqPXGmTtH3wjCtxAZjNBvH\nMyNLePuFywFET/Crl5YXzIB/CUhUglMtzOQs4LJnEfPfdNeVLfGXS1nhFjMRIkTrJ92ACOSBjiJo\nc2J7ehPh0+VMYaq2oFuJWmbMxqkNL+umU5rhN+1L82lPSpSyzraYTaiK3TqypBuUxezWEjcj/gdU\nYsye6WJexctJ1ONoJaupSiIKBS34eKpYiCFCZ47TFrNfDTadXBg0K5zArvU2lcsezwzvzPt6aQ/B\n7NMIvhFCNFsoz5hWx2qXvNBxNu6xj0aEK5tRhFjBzOtZCNqcql3Z9r8kcYndpFqt1MJvHnMUJG7P\ns2GWdBPdFReo33f3m2RFzqXVlJgkIc0wNFV2Bp0AtotRliSoitQ0i5muYwbcNU6EEUnUC5s6Rkip\nciIKBe065rkOWZ/Mcbr8rs8nPh6nTp5d6422MhuFU/lCDQpqdMy8bQVzvTWLfgJYN6ympML7QWuX\n8WLMbq2wLNkZujwxZh5XNmFkFmMxcypFZLPRmIeZHbbhNIqo/B29wdivt5awqcWVnYoxYaqsm0hp\nCmRJ8v3ufu5RSbI9Jq2mxCRJudIMI1URzHQdM0BqxZvlyjY9QkxjvD2lSlJVWKtWAvk+9TKdc/dE\nneN4OZ9aa7p9bpTFHAUJPbRCWVg90DFmYTFT+E1ooYVDLVYu+yCMDNnxlJmaB1yuMcZM1zEDttDj\nGWIRVoPLCuYqi5nzGjmu7KosYqZcymQt5tbWtF1XNv/jo8VyZRtIaXJlXq+fxey/OaqK7Ov67hSI\nsCOuX9eScTtwNcNiNk232x7BLTGsxJiN8FnMNJpqT5mq1wrL5N1RrVwWs08/b1qR8Cv1omPMEsKV\njjKTzdyumdnulDxJCGYaPwuBdqfWNoDe+x4ifJrZb5fGU8ccw0LUGatBkaXIzdkwzVA3rO0WdK85\nazHzXqNS2V8ws+6gKou53OIWs8/YxyicCVMRa9U07YEeKVWGoviPBQza8FXldWAxa4qd/EXFmImC\npCXkEo6CroQgsDkE5bLJ5cYGXEWrXqs5Q1nMPJPMorqT+ZV6xamTJ88/EcztqjTqhmX3RZck7kFB\n9dLydcyAf5zmyVdGnZ+z+TJ3tqBlWbjtnk14abu3pGBkqBvAOMani/j1H7djz2gW3WkV55+4ABfE\nrJO+57EdeHbLGPJFHYvn9uPGD5wZ+R6PKzuGB4C1GmRZitRMTdOK7PNsa/H2sUcGvdmZ3/7VS8jk\ny3h19yTed9mxnuk69z25C/mSgbecu9R1ZTMWHikz+v1Tu/H0q6MYqkwH6kp5O+tIACy0Vox5y55J\nfPc3rwCI2SubbNwRGyZZB6pqW8x+yVxBG76iyNg9msW//fAZ9HVreP/lKz0Zt7mCjv+66yUM9aVw\n7WXHQpJOuBxdAAAgAElEQVQkPPHyAewdy0JWFSyd24uTls/h/k7NZOueSWTyZQwPpKFVFBZS3ui6\nsmWPO7dROMlAlMeE3JNnXh3Dbx7fiVIlHs4DbW3HEXzTuRK+99vNeOt5S7FwpA/TOddi/tavXsI7\nL1oeej/d5C9/MeAnsOOMVSbPPykZa1uL2TChKHIlXGTfq0Y3smkLi/n8Exegu0vFqmWzfV+Pkwi2\n/3AODz+/D4em3NnHyxb0VwQz8OAze/DEywexZyyLLbsn8cAze2Kf750PbsWruyexezSLP27cjwOH\noudHexqMxNDGis5A9oo7T5Mj5xeXI5K/AFfL7e5SkNIUHLdkFk492n3If3jfq3hy00Hc/Ufv/N8f\n3PeqU3Nb0u0JPGw/5wXDvRjo0TA+XcSW3ZOOkuU0iScdrirKVitZzE9uOuj8vHCkj/t9RHEslMLv\njTOoQlWcsYD0hjY8kHbWKgu5zi/vGMeGTQfxy/XbPK9v2z+FZ7eM4cFn92K6sil/4xcb8cv12/Hz\nP2zFl+98nvv7NJvHK/XzcwbTWDBsl++9WKn9JYIt1aQYc5lKBiKoiowVCwcBAPdu2FmJh/MJWS1G\n/gHNr/+4A09uOoj/+9MXAHj3wb1j2cj7SfaJLqbWevWxI0inFCxfOICPvm2V57U4Wf/EO9RXSS5r\ntbnqvBTLhrPGmuXKbguLefZAGl/95AXO///ljqfw6u5J5/9x4sKsQLrotIW49o3H4vdP2cXyeaYU\nKAmXRSZfxqyQnrn259RWLuVovRWrtTetYSpbCntLpdtP+KbRm1YxPl10tOZPXXMqADth5C+//LDz\nd5mQa18OSIAZHkzjy594AwDg419+yHGNk7m6RUcw20pGK9Uxk43pluvPxMIYnX9I5mtUGICOy9th\nCdP53YlHDeOT7zw58L3s2mY3Qnot5wo6BirjEtsB8oy/65Kjke5S8OCze53XyNrXtGSSqKLw8wRJ\nkoS/u3Y1PvPdDdg9moUEvoxsoPYxq+R+kmsTd9BPUNnfx/70RM//v/Ppi5Er6PjLLz/kUZJ5y6WG\n+rsAzGw5aj1kC7rz/LqubFHHXAXrYsnk+W84q+i42rb/HNIkXBa0iykIesHHcflkmTgRmbITdN6m\nZUHnSEwhx6O7AgH2iD36MQ7aTCzLqmr35/s5lBt8sCKYiSu7qwUtZrqmMQ59lesZpUS6M5fdGLNr\nRYd/ZtQ50etqJpvp1AJRaHrSKrqZNUjWaqrSXa3RtdxhzUN60hrKulk1TjEMjTP/gIUtj4xbUeIM\nY+EIyRDvQBwl2RHMfbZgbrc1R8jmy84ac3pFCFd2NWxMJBfDYmYfWlIeQNxJ1bOBYy52nxvGE/ei\nF3wcNwlbi9ib1mBZQKHorz07G39IRyLATfJgH1lZkjxxy6AkE7vdnxFZ90grWcRiducO2+9tpRiz\nmwUfJ9rmdleKsphLjsWsOMlcvPcsKkuc9sRkmzjwIQmyhTIUWUI6pVStQbL2UwklUUVRYsJHNPTe\nxJYJBsFmdPPC9sbnSfiiidMoxyl5iqEkk+/jCOY2tJjJTGlyX3lH69ZLewpmpr4uzg1nL6g747Yi\nBNiBCjFjCX6bQpbDYvYMbYgjmPPVFjMQbJmVqY2/VujrH7QJlipWQ6TFTG1kg732A+xazKQtZesI\nEHfGdLxHh1yzKIu55LhJZaiyXf7E20WKjeWznhd6XWULZRR8Ori1atvEbF5Hb1p1wiK0BUM8K2pC\nU9GiCHuG6GeDp+sXQCcG1msx1yaYeerxyRSzOEoyyTEZ6OVb+62Ia/hU1luCYzHDaEvB3MNYzHFu\nOOuKKTFuKdaijmtV+AmqaQ4Xjm7UaTFXrglZQEHxnFKMjkRB0MI0aBMslU2ukhF/V7Z3vGEr1THr\nPok/PBBBEuXOK1PWGLGY2esRBOvKrsqXYFzZGZ81EpWcNlNkC2VPXS3ZA2hhneLMfK+XsNAC/WyE\n9cmm0WpUKNhEpLjJVfTYTB7YOvmofapcNqFpslO10Y4WMzu+k3hYRecvH9gYc30Ws/chYwVrXJeF\nn9bKE2OmF3mcm07c+GSjIv8GJWWRjNIogRn2qNLXP8hiLhsmV5OFXh+XZLGFs7JrjTGT7xmVAEPu\nT0pzY8y8rmwW9rmg11WuoPuGgFrRqrEsCzkqAQegEr4oZcV5hpsUY/Z3ZVMWM3fyV20ueDY2HPf9\nbmtZvvNUZG/5XlQs337+FW6ltBVhDR/RYCSEJGPM7kNmPxxhVgYP61/cX/W7u9dvx8btwaPYHnh6\nN7ZQWea8nzmVLWHTzgl0dynOw0UegiABUA5o+hEH2mNR1g3sPDCNXz38mmfj37l/miv7m7aCiOvv\nwHgegOvKbqUYs27WZjGnNLv86cVth5EJ2KAeem4vXt4xDsBej+QzvnevXTcddc9Yi4kVsvS6yhTK\nTucnmg1UORjN7oMZ/OB3m7FtX2PHKm7ZPYk7fvsK7n5sh+NW37xrAoZpeYQe2QPohH+y1p7f2tjx\nrWGCmX424pZL3fHbV/Dy9sO4+7EdXJ461tJtZIyZ/N3OAxlHWK1/YT+mAoyOiUwRe8ey0FQ5MrwW\nxmSmiF+t3zZjpVZ/qGT/O6GTJrUXbUvBvGDYLlMhD0YcTYy1vs48zh4XRza9PJM0FecGlHXDMzfX\n/UwTX/jfZ33fMz5dxPd+uxmv7JqgPpNPEJHZyHR/W1ILzCoYhBJnjPnCUxcCAN545uKq19gY8z/+\nzwZ86+cvYB9Vr/3rSn1zd0TDhCMqNanLFgxUbXQkAa04Q/3L/TCMeFYGDRE0T71SLfwKJR233bMJ\nv3l8JwB7PRKrfGdl9mtQ/TJhybx+z//DlMxcQffdKB/1USwB4JePbsd9T+3GnQ9sCT2HevnF+m24\n/+k9+MmDW7H/cA4AnFLGWZWyGwCYP9teN0cvGnR+N7fSoe6RF/Y19BzZ5EQasjcB0feL/bvJbAn/\n9r/P4icPbsUzm8ci38fGhlnhFaU8EoWXt+e7n2fy+7/d7Pu3pO7cMK3I8FoY/3PPJvzs4W2469Ht\nsd+bBE9vtnsszK/sU25cv7F7UlvUMbMsmd+PL3zsPHRpMm746nrfWFkQZDG+7fxluHj1Iif+QYQC\nq6nGK11yz+PS0xfhrecvw8epml8//LRcXmWA9Mb98FUnOL9jB0SwlHW+GPMpK+bgq5+8oKpcCvC6\n6+jPoV2I49MFAMD737Qy9HPOPmE+jlo4iKHelJO8QyB1tnE8Io1GN0xIUrwBFoQPXrES/3XXy76x\nRFZh1FTZY8m877Jjce6q+aHHf/+bjsUV5yzBrP4u/N23Hqtau2y5FFmv77rkaFx5wXK89+bfBK53\novw2OuO5UHKfIeJZIGvsHZVZuABw1fnLcPYJ850e9wBw8WkL8f3fba4agpI0pRCLecXCQfz7R89F\nSTcxbxafYD5h6Wx89G2r8LWfv+j8jsu6ZG4V2Us+c/2Z+L/rXsBhqomSH2ZMi9mPw5XnnIWcywcu\nX+mUt9Xiyh6btI+//1Cu5nOsFdO0UCwbmDOYxikr7OZKqtycGHNbCmbA1Z5701o8V3blgvZ2a55W\nkkGCqpYMaQDoTqno8RFqPMfn/UziihyiLAklogA+bFNh8RPKgDeUQJ8rLVymKiVifT4TaljYIRmE\nnrQKWZKQ5Zz/3Ax0w4odXyYQt71f3gKroJE6ZsLSBf2Rk4oUWXYsyZQqo1DVYMRbLkXW64LhHgz2\ndWGoLxWYU0GsMZ5pSfVAZyaT9c1OkQJsLxH5rgRJkrBopNfT1a+R5xgUWpg9kPb9fRjEIosDu0+Q\naoY5Q92Y1deFg+N5mJbl8aj5vb8ewRzUCpnsBd1dbnlbLc+x0zFvBlzZZO784rluhz/e0br10pau\nbJretOobKwuCbE5s4kRQvW2sLlyUxZzS+Ea++d1gXm2MbS4CVNc2sgRNfIoD/Xn057CutHRKqVmI\nEXrSaksljbDj/uLgNGnwUZrYhKUUFWMGggcNBOHXM92gvEHZQtlxLZJjq4ocuN7J2m54jbBOKw/2\nfY9j2fWmNeSLekMtGjeBsvaSQ5ZanscqwUx5w0jcOmz8Y5xyqSCCBDNbVtib1mp6jtNOOKv5gtnZ\nXynjQghmTnpiPoh+DegBfwtSliRYFn/JVC3JDX4bHbfFzGQMAtF1dqWAGclxCLKYWRdiXGHCUigZ\n6O3WWqqVX9TIzDDImvMVzEwcXdNkzxoNGjQQhCLLPs1yaMGsU9OFKhmnAdOsANcbFLcJRlzKuteq\nB+KN2STJV7kGelnc8rXkts9ahHy1xWxAVWTIksQ1zSwJV3Y65b8u2bLCnrRaU7lUV8CM9mbgNxaz\nWa7sthfMvWkVFvgfxKCsWj/B3BVzjihtufNaFn5j6nhLtHKFMjRV9lj7URpdElnZQQ1GqgRzRH/w\nKAolw2kx2iqNL/S6LObg+D+7gaYqdcyENEdYhEaRJY+FDLDlUuWqPutB06xIuRLQeIuZfh6yTIyZ\nx4vO28ilHsKysmuFt+aZhvVulMpueSLP/G/DtN3c9YQn0gHJnQZjAPV2V1qVxrR8u2bQYs75eCSF\nxcyJWx7E9yAaAXWo9gxi7wJVY3Z5oTcD3mYBfn/HbaHn9apmK0pEAXycGHMQ9GfSDwwbB6rfYtbR\nm9bsftEtkpldT4zZ8Wb4KF6sgsbGmIPihEH4urIr/+/SFOiGhfFpOxZLciFIpzGWQslw1mSju2qV\ndNMRwLQrW5H5BIjTkzxGeKuWcwSSFcxslzCeEkH2/hbL7qhJt2lJsEAzTKsuNzYQXM9PDCCy5t2S\nqXj3hZRMzoTFzHqUgOjk2qRof8FcscoeeWEfXtsbXWPpN+ScwLqT5JhuC3rR8VoWfgKH56HMFso4\nOJF3NiIC2cyDhpInoe0HCdxqV3adFnPRqKsGkmXrnkmMTtg10lPZUtVMbh5004zdjpNANjHWKj00\nWcDG7eOe32lMjDkuiiJVu7Ira6K/x75/2/dPo7tLddY56TQGAC9tP4ypbAm7D2Y8o0+nsiVPWVzS\nlHXTSezMOa5sk9vd6riyK+tl864JbNh0MNG+4OWQcqlaYZ9HHgWIFg5jk3nsq9QNA3QXtDCL2Yw1\nUzzoGATTsvDkpoN46Lm9jtKnUjFmALj/6d146pXRwHJOFnLbimUDT206ENgDoBE8+6pdskZ7CKNy\neJKibbOyCbP77QzIux7dgfuf2oP/pMZD+uEIZp94VUqVkacSOmf3d2EyU+LWjmirfen8/qrXLcuq\n0vrLRm3lUvc+Yde7siVGUcoEb7lUGKyVTqBLXQBgVn/87NQzVs51mlwctXAAhyftG5It6Jg9EPtw\nDqZp4bPfewqAPcbulu9uwOGpIv75z8/CETHGNxqGFXuABYE81KzS9KmvP1r1tynVbWW4aIT//JzP\nkiSYluVZc2RNjAx1O2UoXmtAgm5Y2Hcoi3//32cxq78LXZri1BMT/v7bj+M7n7449jlFYZgmDNPC\nYG8XDk8VPTFmXsuOtszKuolbv/80AOBT7zoF8+bWsYAoSgmEg1gkSYKmumMreRR7ep/49DceA+AK\nMqebWIiSb5qW42HjIaXJVYYEnZOwZfekp+QLcAXz7AFb2SL9DU4/dgQfZcZL+kGOXygZ+MdvP4Y3\nnX0k1q5ZwX3OtVLWDTzxsr0P0fXz5HI1uulR2wvmC085ArP6u/CrR7dj18EMSuXwiUauK9vPYnYf\ntL94y/F45tUxbNs3zS2YiZb7gTetxDk+NaemVf0g+Gm0PA9lvmAL2LdfcJTn90pEAbxTelJjAhNg\nu0Jv/sAZ+NavNmIfVV9I3NpL5/fj8rOOxKpls2Mf+/o3H4c/OX0xyoaJYxYP4lfrtwOov5aZzUEg\nNZ4TmWIswawbtVsZYVnZLClVxqWrF2PerB4sWxBfoNBJgOzIvqvOW+p0GPNk9Fc20YOVzmvj00V0\npRTMGUxj7UUrcFflGWsUZN0ThYQokcSVzQN59ku64bGuJjLRbXF5yTLZ7EmRogQzj8VMK9/EI0D2\nA4VjrRmmFWstf/4j52KiYgnvOpjBd+5+2VuC52PNkuNfunoxdo9mncYjrIco+By95z8xndx9DIPc\nhy5NwYqFbhMbSZJ88zeSpu1d2SlNwekr52JBpQ4wKobhJn/5WMyUQD960VBkTTALPdDeLybol/Hq\n9wDyubHsvxlmaiajerkmUSIB2E1eaE0ScDt0nXL0HJx53DxPu01eUpqCFYsGcdySWVBk2TlGnJnb\nfgQJ9riPVz0xZnd0XvT91VQFXSl7bQ8Pxvc8+HlOyM+9aQ2DfXbzFjpBj5xfiUnoGxnqxhkr59bl\nWueh5GyGduIb+b8RQzDTSU/0PU8yGSxXGUFZS8JWGLRh4JcUyuKXq7Bgjr0PumGT4BUeN8Y82JvC\nkvn9WDK/HydUlG7a++O3bxEDoCul4IKTj3B+z+uxY8+/Wc2GyPdaddTsKi+nIleHiZKm7QUzgTcb\n02mp6GcxUxuuqkix4wlRk5v8k378LGaOhzJgyHlU1mASJRLsZxFIjLne2mWapGLMSU22MUyzdlc2\nx2ZJqDexiHhm6HXgt2Zo5YkIXjZzltyDRifg0ZO1NNV1m9qWXcxJTWXTc8+TnGyUKejo7dYSb7ZC\n3/O4MWb3GJURmEq45wyI54lgcRMZ3eP7JZp5a/HpPuJ895NVYps1oSpsn1SU6sTKpOkcwcw5wSSo\njhnwTu9RFNkVcpxui6jEKnbkJOC/mOO4sVjXeJQyESTQa4G9hsSVXavg8iNqKAcvQesiThmWaVqw\nrPizmAlxXNn1Wqd+zfb9vCVpyktE3sOWpvQ6ruVGZ2QTxVaBpiqOgmoYwd2rWNxJTYZHmUuySU2u\nUK47sdEPun89z2xmv4Q2svcoMp/FXKtgVn2O77c+FI9gjj95i83HaNb0szDBLEst7Mo+dOgQ1qxZ\ng23btnl+f//99+Pqq6/Gu971Ltx55511nyAvvOn4YdOBaC2Otph53RZRgjnKYuapPXSOFeCS5nZl\nJ6DtV1vM9rWvVXD5QVytjbKY4wgbt2lCfQ1GeHIW6rXGwlzZiiS5Yz2pj3EtZu81Icl+zWouklJl\nO95aud6mFd+VXdJNT8lUUpaWZVnI5vXE48uAV4jxjK70C4mQPUz1sWir3m9akGttlsPkLQDVz5IE\n7z5Dh014J2/NlMWsB+yvQHNc2TWpfbqu4+abb0Y6na76/a233op169ahq6sL11xzDS655BLMnh0/\nCSgujsUc6coOiTFTi0VVZEfr5HZl66bTeccP345P1GLu0hSUdTOWYGaFoFNnF7BwzID31QK7WZI6\n5iRjke59rdNiDppPHUswBycO8sCTkJMUqo+CFrRm3PcQN7BXAJOSvGa149Q025U9nXMbjKRTvMlf\nrnKbbUCMmdR0B1Um1AOlLnE14vBT8FzBHG0xm2btFQZuvkR4jJlWMOn2nbyzxdnzz+bLvtUtSeNa\nzNXnqShya7qyP//5z+Oaa67B3LlzPb/funUrlixZgr6+PmiahtWrV2PDhg2JnGgUjsUckSREsoj9\n65gpV7Ysxe7yUtaN0NiJ34Zc8rGYD47nIzfvoAb0jqUU4GohMackYszsMbbumar8vgEx5lijPU2n\njhKwN+m9Y27tLV2Hu2XPZFWZF4tlWTg4nqNK7eorl2q0tg34j6eLSvwjCgfryiZCqPFdv4jFrCCl\nKs7/DcPkTlLSHFe2N8acVFtXsq4aYTFbVCri7tHo7PewGLNTmhdqMfNfVxZZttUI+vgk9DDYaycW\nsmdHC1NZkjzPaBDs+Rum22zowHgu8BiZfNlTK21ZFnYfzHAlj5XKBl7bOwkg2JV9cCLPFQabypVq\nUsRj76Dr1q3D8PAwzjvvvKoTy2Qy6O9363d7e3sxPT0d+6Rqwe2RG3zhD47n8Opu+4L7lVTRglmS\npMi+0ywl3axyY9OTSXxd2dQmSCzt/Ydz+MpPng/9rKAYSFQmeVJZ2WHH6IqYwRyHnpid3QDgSz9+\nDn/z1fXOQ/svdzyF+592m2T8/bcfd36+78nd+MKP/GdlE9a/sB/X//Pv3NrxGr0NpNSCdv/5PdxJ\nJM/5KZUGpVgsnW+XYNGjCYlCVQjoeb54nruWG9Eila6x11QZJd2AZVk1urKNhmRl33bPJgBuk5Yk\noSexjU4UsJma0e4Ha7VJkqtchbV/JdTb+YvtrU4E5lBfV9BbHDbvmsDffHV9ZLMaP4s/Wyhj92gG\nN33zMfy/X1vv6134xH88jL/6yiPO/zdsOoh/+M4TuOW7T0ae2zd+sRH/ddfLAPwFMzF61r/gP7uc\nMDqRx19/5RH8969fjvxMltj+mHXr1kGSJKxfvx6bNm3CjTfeiK9//esYHh5GX18fMhlX08tmsxgY\n4KvBHBmpbsgRh/GKpaxqauCxDlZq4EZmdWP5kuGq1weohhgjI/3oryywgYFurvMzTAvpLu/nf+Yj\n5+L6f/4dDNNCv89xJGoT7ulWcajSvOzFbYdDP1OpbEBz5/Z7Gsl399rfUdUU3/drlb+dO6cfI5zz\nYoPorcxLprng1IW46MwlgSMj42K7rQDIMvcaITW6ed3CMSP92L4/XDncumcq9Ngbd7wEALi/0gGr\nv6+r5vWqqfbUMfJ+dlO5aPUiXHLGkXU/D3299todHOxxjqVWlNF5c/tx0wfPxINP78ZlZy91Np/+\nSvmbxbgJ5wz3YmSkH//w5+fg/f90LwBg1uy+RFtSAkB6n32fZg11o7dHg2XZn2OYQFcq+LmmUbps\n4SYrivNsKbKEfNG+zvVeVxL7/bM3rsTI7PijGsP4f64+BQ8/uwf3PbEDh6eKsJTwNa+o7vczTAsp\nTcHcShOVWbPsjSTdnQo8hmmiar+KA7uWlcr6mjvcgx0H7HvJHvvfPvEGfOor7oz66aKJk8I+30dv\nSHWncDhn7/eWBajplOdeEIVFN0z3Odto108fHM9Hft9XKIWoz+dZv/j0xfjpA1ug+3w/mlcr6/nx\nlw7g/7v+7NDPZIm9e95xxx3Oz9deey1uueUWDA/bQm758uXYsWMHpqamkE6nsWHDBlx//fVcxx0d\nrc+ynp6ymyJMZ4qBxzp0yFYazls13/dvDGqTHB2dRqHiPj10OIPRvmgNuVgy0NWrVB37bRcux08f\n2IKxQxkMpr3WZCbrFsyz8Z6wa5KvWADjh7MeC4u4b3L5su/7czn78yYmsoBen3uv7OMCftt5S5GZ\nyiPJNhSKLCNf8P8+YYxP5LjfE/Z3csUpR8rBDN2oeb0qsoRCUXfez1pybzlnCYb6uup+HkoVz9HY\noQzSleWRr6znw4ez6NIUnHH0HByuPBMjI/0oVdbOxFTBc6ws9UydtHwYz289hAMHphL1jADAoYr1\nVCqUgcrmum//JAzDhGmaXNeEWMnTmaITbx7q68JEpgjLsjA2Vt/KnMqWMDKUhmTUvgaC6FUlXH76\nIqQVCbff+wrGx8PXb77intdUGUbJQEp1955cttJAZzIfeAzDMGEafNfVD0WWUCi5a3mauPmpdcEe\ne7hHQ1+35jR/KeSD92sAKPn0yN69dxKjk3nn/7v2TECmOijSnhJy7Mkp9+/37Z8M9UoVKBd40Wff\nWVbxHE1NFULPfYr6TL+/CxPqdZk1JGZw1113IZ/PY+3atbjppptw3XXXwbIsrF27tioO3Sh4XDck\n0y7oprANA+K7sg1fK4K0zfRP/nIXVNBsUz/MAJc0cWVFlkslEmOu/q6NKCMJmnoUBW9/5Khrwd4X\n3oxS389i3H9s3DapNo9+SYBR956sHbZLGp3s5rYVNdGFZAUzyfpOqYozlrSkm7FGbWpUuRQx/If6\nUzg0Vahy0ddCNl/Gghid4mohqrKCQOqQ7f3M8OxfUXXMpmVHtOvZBxSqtzrg3r+h/nBXNr1HRrnS\n/UpMs4Uyk3HvVW4zPvkE9HOQK+oY8PH2Eeir7peLxHt/6hn4UtcuevvttwMAli1b5vxuzZo1WLNm\nTT2HrQmeGlGn3CVgMZCNnFx43hsA2C7XMjV2jYY0LvEbLEHfvKDZpn7opu3iZTPAoxLWnAYrDYox\nJzk8nqAqck3TXHgzJ6NqulnLsB4XLqtksA9vUtfPqWenlJOo5jIkW5sdMEBncfNk+9YKPbWJPEdO\nCR7nelUVOymprJtOY2MS88zk6oszl3UDJd1EXwOUTxrepFOSvEXWb8qnJj3oPiWxD6iyv5IZFWOm\n98ioem3/GLPuySViE/v88lFoAZ/NlwMFM7tn+FXY8E6YqifhsGMajPC0O3S7fvl/bbrzEBBvxJdu\n2Bqor8Uccm70xhynxZ8ZYEXIPh2fPO+zkkv+SkK4c32OItUkCHgt5rJuhiYzsRZzXYJZ9ioZZSbG\nnFSpmbO5MxZz2Pxd8tmsYKbPiac+tlZ0SjCTa0ysXN61RoZBlHQT5bIBCcBQJd6eydfXZ5lkedfS\najYOSoTXi0AahDhtL6l16pTLBdwn1+NWp5JpViuZUYI5Toczv/NnLeYMI4j9KnPo/SNMYLLeIj9Z\nwXt/6mlq0zGC2S0PCHFlOzXM/g85SewgGl2csY9lyg3H4rqywxuMBA0d9yOoaw/J/I12ZSdQx9zg\n3skEVZZqEgS8ZUmWFa58sYKsHnezEmExJ1Wf6TfMJCoLl2zwVRYztVZcSyx5wVzS3eePFcxxFEky\npYlUSZDGFtO5ZAQzPQawEfAaBK4rm1jMfvepcSEtRZE9zxhRMof6gt3EgNcrFNW0xtdizuse9zUr\naP0y8On1Gpahz1rbQeVSQPT9ocv14o4d7RzBzLFh6Ea4UCo5FrO3FpDHYiYbSJjFHB1j5neRGUbw\nJivLUvB0qTprcWcCOy4bXxDEqbsN+1v2s7U65vAqAe6/pKHXrmVZ0A0ThhE+11jhiDGrjAKcZNmU\n05JTUxwFtxDTlQ3YLt2SbqBMBHPFwp3O1tk9rmIBNSKPgoZ34y+ViSvb3l9oV3ZUaC+Jfgaq7FUy\nC0FecI0AACAASURBVCU7x6YvQnGhhRTxkrDryDBNlHXD9/xzhbJH6NEKl2laTmMagr3+3eNn8zp0\nw/aSsZ/LNjLyd2WHx+/9zosou2Xd4DL02n7sI0HlSNQKa8cJuLFEMjuUd7rUwYk8Pv2NPwLwt6b8\nuuQQaItpoJdfEw+r7STlE77vMy27VV4CgrmWuG8tqIpcU7wmTgvJkm6iO8ADx963eizm6hhzY9pc\nEiH7kwe3esrFekLK2Mg6ZWUt7c6jFeB7n9iJH92/BV/42HlVk8bisnH7YdzzmF0nnqJizHFd2YCt\nHJMmKSlNcSzmW2/fUNcc6W//aiOAxjQXoeFxlb68/TAOTuQx2JtyhIuncyFnB8D665jttbx51wR2\nHsygp0uNLJWkP7Kk282A/uar6/HOi1bg8rOOxNhEHv/wnScCk/UyBd3jJr73iV244uwlkCQJf/et\nxzzjPu/47SueHgYA8ItHtuHbd9klkCuPHMLfvvs05zXWmvZN/uK4PwfGc3isMuISAH70+y1QFQkP\nPrsXwwNd+OyHwsunOkgwR1vM7ixm/431becvg66beMt5SwHwW8y7qRm1b6BGmxE0NVh7LZdNSADO\nO3EBLl29GIvn9uHLdz4fuoHa3yXY+gkTzPU2FfCeg/0ZXSkFl529BEfXMDeYBzXEAxCGG/NKYSJT\nwoeuPB5PbR7FMYuHsO9QFvsP5bBnLItMvlwV66WpspjrEMyaKkOvxLQlSWqYxUzuMS2UVywcxGnH\njAS+58Sjqmv7AW+yJK0A/+j+LQCAF1875Lvu4/CD3212ftZU2WnZSCyNOGs2pcpOoleXqmDVMvd7\n1dPOkayn044NvoZJwGORkbnYi+f2Od0M6TahTvvXgGPU214WsENvumGhrJvYts+um162oB/dXSqu\nPHcplszzLwe66rxl+OrPXoBuWCjpJp7fOgYA+PEDW3D5WUdi12gGhZKBBcM9GB5IY6ivC4+8sM95\nf65QRq6gY7A3hclKuenBiTwkSB6hDKBKKJO/JWza6W3iUmSUAV9XtlOZECwX6Dn1ADA2mXcs+UNT\nRYxPF7Ew8N2d5MrmiDE7btyAxTjQm8J1bz4OI0N24w2eEizAdXFdd8VxOGbxkM+5BR+npJtYumAA\n1735OHSlFJy0fA6WLeiPbGIfNuRcjogxJ+XGpocifOitJzozWpOGLTHihQg907SwYLgH56yaj798\n+4l44xmL8f7LV+LG95yG0yubbFgSCnvf6hXMFtx12jBXto/w+Yu3HI/Lzzoy8D2z+rtw0vJq4az6\nZGU3sq2opspOJYNrMfNfc60yAEPXTWiajIHelLM267neJd3EkXP7MHeovsY8UfAYBOS1S09f5HgH\n+qlM46j75Liy6+gyR2LttmvZ3gOvPHcpAODtFxyF1QEKzMkr5uBv/uwUAKSszbtWSfLW5WceiRv+\n7BScfcK8qtezhTL6ujW8pfJ55bLJNS43CnYf8B37yJF7RF571yVHQ5FtBZxOUmMT1lg6RjCTpKcw\nLTOqjpmFd+iAmxTib+UG1TGbldgf6xrVKn2Cw+J3phU8Ci/KlZ1U0pb7cDc2Xl1rHTPpqqUbwXWw\n5N7EiTHXU8econo52+fYIMHss8Z5son94qe+dcwNHMSRUhUnXlqbK9t+fopU+SL5t57a0nLZ5B6+\nUA88G7/Tt11x3fa0xex4ECMt5tq/Tx810S9uYhy5v6WyCXYbIwlY5Fjs85bJ2xZzb1p1Et5KlWS/\nemH3Af/pUvZn8ghmRZaQ0uwqgVyM3u0dI5gBe5FxZWVz10Ty1WwSbTEo9qQFHMcpD2Ee9hSHsDCM\niBhzSDZmEiMfybHI5zUSVZFhWfx1yQTiddBNM9BlxwpKP9j7Vq/FbH+e4fk3afw2lO6uaIXCbw37\n1jHXEFrgxVsuVZsrG7Cfd5LIGWekqh+GacK0rLqUMl546pjdufJuOCTd5ePKDtgH4u6FfvRQE/3c\nxDg+wezcD8P0TNUC3AYhRElkn7dDUwVYsAU3PbQkCe8Te4wwiznMlU2eD1mWoKkKcoWy59hRpVQd\nJpjDy2qcrGxOLZG3ZjPLLKTq4xBXtvc4bnmIf51slHs16HsoshwyXSr4fXGJaliRFLWOSyR16YZh\nBXd7o4YeBMHetzj15sGfZ3r+TZqgUroo0j7C21vH7KNkJnz7U5pP8lcMrwy9kbsWs9sRrBZKTI+D\nRsJTLuUmsrrnQ5dbuslf4cNs6rGY3VnpeuQeyOIYHz4WM2vo0M8b/R170iplxBiJJFKyx/DtFRHD\nla3IElKqjImMHQsn3zVqjG1HCWZFDm9EERVjrj4eX80m63phUVV/7ZVoUOzDTtw8oRZziOUrh9T9\nGmZ4yUwcjASblYShyj7CgIOSbls5YXF1TYtWgqot5tqtJrrVZNTn1kPt9zi4bpM+Lr2+kqiYoo+h\nyLJzjUnyl1/MPAi6bIg8Wzz3OQx3JGXzXNlhcXy3WZJ7XejOgbwWcz1hKGdWer6MXKFsd2zjLCUk\n97dEtU4l5Bi3OL0/0h27etOax4iJ6iLGA3uMMIuZq5mVbDe8IYoQyU+ImnbWWYI5ot7Vdf/Es5gj\nXdkR9Y1BGeMl3b/2mbi+wzTAsFhxVIw5MVe2UX/3IB7cnuUxLWbdoDLx63FlJ5iVrbCu7MZmZdeL\nBK+l7fdMJONC9K71qs5fcSxmygp0BLNz3Ws716BntRHEiTHTe5nHYo7wMulG/d6uXjrGnNc9Me4o\n6NAC68om+2mP48p2vxc9brM3rXqOk4SSGyf5ywjRSMlrRDAT5s6yp2D5dSej6ZhyKYAkCdVex8yi\ncMbTsgW90q3IX1v0y5A8PFXALbdtAFCthRPtPspirrVcqh5XLA3ZOOqJU/FQa3/mkm7il+u3AQgO\nXzgad0i5FKsZ12M1kWv/xR89h1OOnoNHnt8X8Y7aqHXDZd/Gur/JdfzJg1uc37EK5A9+txn9vSkn\nYzaIbKGMr657ASsWDeHQlHfgvePKrqFcis7ZIIqXkyRE3eeHntuLl7YfxoevOiHSze96t5oYYw7Z\n+HU/i5mKMcuSLe4CG4w4XRDrcGVTFnO2UOaaw0wg9/eZV8ewYtGg8/t9h7J4cdthqIpUlbgHeDPP\ne7s15/4eGM/hvid31/xdCOxa9lt3cVzZsix5QpWvS4tZjbCY4y5GN8YcnfwVlo1INn9awL+2d8qZ\nEcuWGWkBFjYNaWDvR0pTUCr7Z3UnWS519ZrlOHJeH66/8rhEjhdEVN9fGvphKZQM/PqPOwAECyo6\nCSUIouQsnNOLlUcOhU6miYJ8XiZf9gjl7i4FH3nrCTUfl4Vd4x+8YiXX+y49fTFWLBp06uhZebX8\nCLtWne6uxCqQ9z21Gz976LXIz3rwmT3YtHMCdz263fndmlPt6k7HYi7Hz8qmN3LSNMhJEqLu8233\nbMITLx+s6nTmx8y4ssOff8C+zx++6gSsWDiIU491J/lJkoRUSnEytlniVqj4QRqJ5Eu6kyXNC21F\nbts75fxMZiH396QcZYn+25NXDGN4II25s7pxzKIhRwmjhTJPkmMQehyLmSfUwFjMQ/323sHWS7N0\nlMWsyHLsuEwYUd1zCNm8jlkDwdqi340kQvd9lx+LU48e8f/7kEEUlhVsqfakVZiWhULJqOrCYyYo\nmOfN7sE/fvDMRI4Vhuu5iLaYaWWG7nsbpN3SSSjBx7Sv2Wf+/Cyu8w0jKKv35g+c4bi5koAWIOmU\ngjecxNcApK9bw9+9dzX+c90LeHrzaJVgXjjSh6MXDWLL7knnd7W6h9nOTp/90FlYMGyPVKwul4pT\nx+xe4z6n5Cb4PvPEyJ3JVy1SLkU3Szrr+Hk46/h5VcNW+tJqYFlO3HwbP4jAmcqWYCHecA9JkvC+\ny4/F7b95xVG+APf+vPvSY5zf0crDiUcN48JT3NYc+d3eBiEA8FdXn4yHn9uL9S/ur3otyqvK48qW\nJAmyJIW7sqmZBPSzSPbjKLd7h1nMUqjbueY65pBjmqaFfFEPLRPwq2MOi3f7TQZiP5P+OxbaxcSi\nJ9j5q1nwjPQksD1xCUGWA52EEnxM07mH9RIUo0zaRUp/Ti2KmLtGqt+bqjRJIdC12HHyAFiB7pdN\nXYsrm94ISZwyrI6ZZ12RznCtUy4VLVh70lqgyzSJOmZyvyam7TBEUB+HINjMe4CayR2gALFr2e9+\npDQ5cL34PWd0BUvV+gjqFaEEl6QC7nMgy5Knt353JUEvqjqgwwRzuMWsx3ZlR7uUc0XdrqkLceP4\nxUjD4t1RGrMRJZipMgaWJC3mZhGn25Rn7iq1KQUJZp6acT2k3CouwYI52UeR3oBqud/EG+P3VnZz\nozeZOA1TwmZRq8ymHSePgb6WfRUlVXUEc/U64BHMpYAKikZAPER8dczB59ObVpEvGr7KUhJ1zOTZ\nIaVAcXuIk/tdoEIJTiw/4Hlj9zw/hVlT5JBS0urvS++zVa15A6zisO6K9DEVRapKRlQV6fVlMZOk\np8Aa3piZiDyxhFxEcxHAXWSeEXwhGmtU1p+bdOV/+3qpwn+/9yYx8rGZxOk2Rd8r2o1XDBAYPGU0\nhmkGbhRxCYpRJh27pI9Xi4eEvMcvKYoVTrRSE2uiF7MJpnwsZnJfarWYiZLKZt/Ta4lH4QsqbWwE\nXFnZHImsZB/wc2cblOCoFSJYxzMViznm1C028x6IDhmwe5ffc6NpSqzyOvr+s/tA0GpWpOAEW4C6\nvpLkHcdZaTgSpcC21w4dQVSyVtQ85urjRVvMUe04Af95zI4rKiTrL6x7F/13LOQBYR9Iq1LT226u\n7DgzgOm/oa8e6SDFwlNGoxtmYm1Hgx7lxC1mZjOIi7NGfN7KboalAMEc1amNTbjzurK9Vnm8ecx0\nEwomxlw5P9qbxOXKbmLyF8/YR558mTDPWVzvoR8ppnok7pxqP1e2e539QwbVruzq80+pwa5sP+jr\nXDZMTzlpUFtkRQlv/0zv0Z6wkmLHnKNmIXSUYI7awF1hGC8rOyxZwK25C4kx+5wX3euWJaoXa1Qr\nTKIps43STSuex6BVcO5DzOQvmqAsSLeZS1iMOTlXdpACUOvEoyBSHld2/HMnLk6/s2LnUdPfiXYV\nR1nPtNUgS5LnGsuy5FmncVyutIXitHVkPCN0/gVPGZ4T+2xCjNmt2w8Py0nwnxdM6AnxnEXNpuc6\nT1nyhGDj1DED7rWkleayc50DLGZGEfGLGadUOZYiTQvYqn7oAbcgjivbM45TsQV1VIy5o7Kyo6ZB\n6ZXmGrzaFE8d8z2P7wTgNnTnPa+wxhdRY9+MEGsbcDXlnz30GtZQGYxxXfmtghtjjufKpomKMYe6\nsg0TakIzeBvVUIRFS8yVXf0au2k+u2UMuUIZPWmtSkh3pYIFWZ6KLfq5LlOa7JQUxrOYaVe2dxDC\nzx56DZOZoid0yNP3u6kx5sp3feG1QxibzOOZzWNYMr/fM7lON+zWumEKndMAhGlm8fTmUfx2g71v\n1TP2UZJsIUMUrL7YMeZqT2LUdWbXgV+SmKYqsfY4WsCWdAOaIqMIe90FiV41pFcE4DWe6OdFVew+\n8HS5oR8dZTE7m2zAJlwqG7HKHXjqmLfssctGFs3t4zgOZTGHZIhHZWVGubKPqJScTOfKmMqVuN/X\nqrhNQDgEc8A1W7tmReixw8qlDNNKbENe5TMac7C39rroILzus/j3e+Ecew2tpmpj/Y5NeHHbYQDB\nbm0/PILZ5zmgfxfHsps/uweyJGFWf5cjLOhzvv/pPXjgGXdOL1dSIXH9NsOVTT2fX/zRc/jh71/F\nrd9/2vM3RshgFgIpzWHDOP+57gUcrjR0qdcTRD+T84fjlfv5raNyQE/ylUfaSgnrIfATwKoi+f5+\n5ZFDOO2YOVW/97iydRMpTcYlqxcBQOBMaZlbMMtVz2KqMv0sjI6ymHuoFnGzB6pfj1sET7rnhFlq\npmlh4ZxeHL2oeg4zgYykpDXDsDpC8rsoV3bQQzVnqBurjx3BU6+MepQU8r6kEpmahdv6L1zLBNwN\nVJLchMqVRw7hT85Y7Pv3KZ/GE9XHtOqyLGjmze7B2jXLceeDWwHYTVouO9P/3OrBsxnU4Ca/6LRF\nOPP4eU6jkaBjD/RomMqVnTgmndAVlXlKv+5n+Sgewcz/HRYM9+I/P/kGqIobawxTrOI0rml0lzvA\nK3zGJvO+fxM2mIXAU82Q1Lo+78T5mDMYb061n7s6aLDPp645FbOH+zB+OOv5vSRJuO6K4/Cdu18G\nANz6kXPsOmPqPq1aNhufuPokZw2tvWgF/vorjzj7oWF6LfbuLhXv+ZNj8GcXrwi8xooso1gObkxj\nBsSYVVmGpsmRAzfaa4eOIKx+F6h06IpZBK8ocmBskyRT8Qh7hZkpHFbuoEQkf/AksZHGCrRmlkTj\n+pnAzTKP7tBEviMtUNjGCzQ8LTl1w0wsxgx4hyx0aUpDsuTpmG2tHpLetObrKqU3TdIQhDxztIIT\ndk0B79r0ixXS6zvud0inVM89C0va4sldcPvCN/fZCdsDop5jnvr/pCbNzepPx34Pm6sAuDFm1rMp\nMTkININ9rseJ9AunFTny/EqSBEmSqmSAyVrMlbUS9swrETFmZ1a9LHmed9JqNKqpTYcJ5uAsRMM0\nkS8asVP6WYFKQy4uz+JWZe+s6NA65ojkj7BSK4Kf+zeJTMyZIKz8i4VssnTHs7DvS5KMglxLRPlK\n0oVJ3/NGxizJBpO0Ikaf88gsb+9fer1FZZ7SCTB+glOt0WL2I6yBC1cZXpNGnLIEbeC6YUUO43Gm\n44UIkKS+T1iOTRB+97xYtl30cQbt0IKWHJPeO4PySwi6RzAbXJn3/K5sxmJWZK4EwvbaoSMgiR5+\nG7gzSixmgoIqB3d4obu7RB6HSa83QrKyo2LMYaVWBL+pSe2a/OWUfURMZAHc6+o3Ai+IlCYHul2d\nsEGCVi1tITdSMJNj1+LKDoPeuJym/JV746lpjsgJ8Liyfa4DvU7rXbNhg1t4Ysx0zHCmoPsz6GYy\nFnNSSnrcUinAf+0XSnrsLni0seXURhddYcy2fmUhlq9pWtANvnwSu2dGdLkUm/zFCuogOkswhxTU\n89Qb+xE2GCOOFs2OpHRd2X5Z2eHlUmGlVgR3Tqm7KNvdYs5xxZiJxUy7j8K/r6YqgYK5EUk/tMXc\nyLpY8jlJu1+9Y+wYi1nnjzF7XdnhMeZ6v0NYXgWPxRzVBrcZ0B2yeGLMShNjzH65CFGw5VaA3Ukx\nrrJKKwVk76QT3qIsZiJg40wQi3JlB8WYFUV+/QnmnpAkIfK7OI3WgfCm52YMwcweJ7yOOcJidpK/\nwizm6jIgv1Fx7UBPSIiChVxXrys7wmJWZegByRh6SFlbrdAbaiPHCJJNKqk4IoE+Z5LwQ+6NNys7\nZJSmaXrWd1B2rfN6ndc/bDPkqWOeKVc2DdsUJSoRrZkx5lRIHkcQkiRVuXULRT5XMo2fUkBbyVGT\nnMi9LUXUUNMoIZ5U+pik0xeBHmcZRkcJZseV7ZP8RVxt8WPMsqfOcfOuCadESo+hRauKzF3H7Lqy\ng+qY+WPM9BizqGzuVkVVZHSllMCkPhryHemHNWrz0VSvK3vL7klsroyfa8Q1U5pkMZPPSVqY0G7h\nvh4NPV2qo/jqnjrmYIFQFdP3cbfT4YN63fFh9b48gzfoZJ6Z4ucPb0OhpGMiU0ShZESua3L9Ql3Z\nCX2fWkMy7PuKZSP2sfz2X7oUL8qVvXmn/aw7FjNHSa0sS7AA3/bP+aKOp14ZBRBUx/w6izGTzdhv\nvipPT2s/NEX2xMpu/f7T+Nz3ngIQz2K2R1LyCUnysASWS3E0oCcarL8ru70sZsC+tzxzc30t5oi4\nICuYP3fHU07dqNEQVzaVLVyDpcELWZdxEml4mNXnjjgd7EmhhxovSLsNw2o1WaE9Z7A6q9drMSe/\nVZENk8dipjs5zRR/3LgfT28e9cweDiOq4RJQ/7peOt+u8x0ZilcqRfDb44b6gkfo8rLqqGHn5/NO\nnB/6t6R0MU7bVXLd/Nb4nQ9scX5m65hVRXJmMoceP/Iv2oiwbGY3xhxPMPekVewdy8K0rKoNLk4y\nlRLgyg61mIN6fjubRIjFrFS7sl2B3n76GBlQEgW5Zukur/sojJSqBCYqkfuUZO03rVDF9eDEgXTd\nioqxxWXR3D7c/IEzkE4p6Eop0FQZhUonI9rdGiaYyfVefcwILlm9CMuOqG48kGSM2Y90l4qSXuKq\nY47jHUuCT7/nNEc5PGHpLMyd3YMHnt6DYslwFNRrLjk69BiKjyub7f1cS2yY5sb3nIaJ6SJm9dcm\nTK+97Fh885cbAQDXv/k4DPV1Ycl8/6YeYXzp4+d7UtgvOm0hjq00JZk/O7rxSVk3qK5j0cqyYwQW\n9KpyzN1jbq21zJRLKbKMPzl9MY5a4NNog6KzBHOIpen2tI73lXvTGizYiRdsfJpMf+LNyvbtle1X\nx8yZlR0mcPw6WuktoPXXSlhJE41fHXPU99VUGaZl+br8kphby0ILnFqyWXkhmelBAzzqgd48U6qC\nkl4A4M3vCGuiQF7r7dawcsks379JslzKj+4uFVPZElcds2MxJ+x9CIIWdEsXDGDp/AE88PQe6Ibl\nNHEZ6Au3vPzGzbJ7Sr3ruktTMI9D8AWxgOoWdsKy2TVby2z3PFmSsGgkuBsjS7agO2uSx5VOnttc\noVyllKRTtCD2jn2UZQkyJBx7pP+ad/6O+8zbgDCBRjT5uP1ciUWTKehV2qbryuarYzZMyzmGblqQ\nJH+h7mRlBxQx8mRXp5ym/e2flQ3YwozHYnazsvld2fTkIdZd3pisbPdY9VosYXSnyJCAZC1mFk2z\nwz2WZXkqIsLKpXjchkmWS/nRU/GqxMnKboRL3Q/2u7uDXEynPjzK5eq3H/J812bS58montnEOj2G\nKzusZ0Y3Vaopy1KsNtDO+2K/o4UJF8wkxhzTYqYSyliLLWqYBI3KuNmNkG5SkWMfOVzoWgfVMQO2\nBszlyjarY8yRFrPmXiu21K4R2bj0sRrpGnUt5sYK5pQqw4J9regEPZ7kr7BNq57OXzyQ68MTY262\nK9szWUuRKbe05TRxiRIgflnZPN+1mdAezBkVzPlyrEElYV0m08zglloSPDtMMAfX//KMZ/SDnm3M\nbjRx65gB9yEJ67/M78oOsZh9khPa22IOrxskOBZzii6X4rOYS7pR9aA5MeYG1TE3krRjMSfvyqYh\nJS+lsumxIMJd2f49kWlq7ZXNC/FWxOmV3SzhQSsAiiI5Xh/DMN22lRGxUL+pbDzftZl0MfHXmSJb\nKFMKD0eMOcRippsb8R6PpaNizCT847f4skUdsiR5Gk/wQM80pYUcadUI8JdLAa6w1Q0zcCFGjn3k\niBW7DUaqM8HbNcYcp6wlXoMRV4lha+B5StPi0izFiCTA0V2QGoGboWp4rp8eajFH14w2OsacdlzZ\nPEmFzS2XovcGVZY9+0dJt2cxRyl4CkeMeaahy9ia2byFjRLmCrpzb3lcz30hbYLZLnO1fK8OE8x2\n32PDJzabzZfRk1ZjD6QnncJe3jHu+b1hWrEbjADuQ2KYwRYzuZFBFuK2fVOVY0bXMZd1A5ZlYeeB\njJOd244Wc1RvWgK5vrQ1wZP8BdhzammPimlZ2D+eA5DsNWvW1kg090Z/HhGuT786hulcGUN9KUxk\nSqGubGKdhMXu1Qa7/Em4g0vhi5HomQReV7bkybAu6SY0LXwWM3kfec+O/dMAENswaSYz6cqeyBSx\n80AGAF8FBrGYp7IlbNx+GLIk4djFQ5BlKTA3KA4dJZiB4FZpcUc+Eohm9Idn9+IPz+51fm8YViwL\nVJG9bqWyHhxjDiv7emn7YTzx8kEA4XXMxEVUKpt4/KUD+NavXnI+rx0FsypLsCz4lq3R+GWsd6fC\n7ztxNf30D695fp8r6Lj9N69UjpfcNSOCrNHb0BGVmcq1lJ/EgXyf791rX6uhvq5IwVzmCBF4Leb6\nrz/bfY+4smMlf82IK9u1mHXDqkxAihawxP09OlHAP922AQDwN+86xXl92YLGrou4NNNiPvbIIY+x\nRT/7PL0uSO7RPY/vxD2P7wQAfOStJ+DM4+Zxhdyi6DjBLPu0SrMsC9lCGcM+TQyiCIpJ61RLQZ4G\nDmwiRq6oY15AUX7Y2McdB6bdvwsRFt1UsxUyxN4RWu2Y/EV5EeQQRchJcFNk3PjuU7H3UA6nHl09\nHJ0myJ06mS05P6tqctds9kAan3jHSVg00pvYMf04efkwPvLWE7AyojSjXthYZ19PZeRoSP20EyII\nEbi0wpuEQPzmpy/FfY9txw9//yoAYLBSmsNThtfsxEn6u6uKm5VNYsxcgxYq75nIFJ3fjU7Y850H\nejT89dqTkzzlmvnMn5+FSeocm8FH/3QVnt48in2HcvhNRbACwKWrF2HVUbMj3+8nvKcrtfxkrbz3\njcc4r/3de1fH2kPaz3SKwM+VXSqb0A0rdg0zEDz0QjesWCUUdLxHN0wUS0bg+ZC/9dO8aDdLmBWX\nTimQJakqNs57vq2G43GI0EZpi/nYI2fholMXRnbXCoopZXKuYE6ywQgAnHL0HMypsVsSL5Ik4czj\n5mGgN7rTUD2wMTWyaYVZzMTdF+Zt8ljMCeRFzJ3dg9NXznX+TzqYRQ3bAGxXtlyZ59sMaGXfzsp2\n94+SbnJPQAK8DWamKsrmuasWoL+nseuCl4VzenH80mhhmCS9aQ1vOOkIXLp6kef3l56+iMs75ud9\nJfs1kT/HLh5yXluxaBBL54c3FaFpvx06Ar/m4iRAH7eGGQh2axiGGat/LrFSdcMtyQlqLkGsQ7/G\nB94uMsGfK0kSertVZPO6j2BuP4vZSYiLSNTRa7BsgoTuVM5N7GhHZaZZsNePKIVh85h5kqk8ZWUJ\nCUT6mKQ1Iq/FPFOTpRRZdvcP00S5bHKV4JCcG/r7ES9QO+4BjYDdg3mHyqQ0pUo5ckth689HG/Ni\nFAAAIABJREFU6LjdRvaJMZOU9los5qAGELoR15XtWnzZiL7dYR3M6E+KysrsTWt2GQBTttLWruyI\nxAqiLMWJCQdZ1FO0K1tsZIGwG5Cm2v2BwxqM8Dw7jcjKpu/jYG/FYuZoWWqa1owlJ9nJX6zFzCdA\n2OeArOl27GXQCFKq7LlGccoiWXlC9p4k8hE6TjArcnWHKFKbGneABRCs9Xhd2Tx1zK7F7PTtDnJl\nhwhm2gqJEj69leECrKuuHZO/XIs53LqppYVmkHCYplzZSSR0dCqsxZlSlcpgkLCxj9HPjifGnJBi\nRHs++ns026LkSP4yZlQwy26Oim5CN/gsZvu93nMmgrkd94BGIEmSZx+O0wyE9cA6ruwYHSGD6Li7\n41fvGiUIa8EwqazsOHXMhtsdKdqVXb1hlCgrJGqz6u3WYJiWk5TAnks7EdV0hVBLvWnQ3GDalZ1v\ncC1wO8MKtpQqI6XJoS5iHquCXqdJubJVT1JVtGVPMExzxlzZKtVghDSL4W3zyIZghGCuht6H67OY\niWC215NwZVOwHaIKJd3JRExyYADtyo5Xx2xGtgclx2MFKuAVIlEaGTn+BJPx2I7xpTAvAsEwTYxO\n2sMU4mw8QQLk8FTB+bnR3bPaGVawaZo9c7akm4FKD49VQYdckkq6YgV8ihn5GUSxbMzYc6NQLTlJ\nL3feZETWYp4QMeYqyD6pKtG14d73MUONKms6iRGhnVkuVbkwmXwZn/rao05WYpKC2TCsWEF+sgF9\n8cfPOb+LijHv2D+N7funPNl8PJsIgZR6FZleyUm2l2wWMofF/MUfPYe9Y1lIiPdQdKX843XPbz0U\n+TeCastBU2SkVBkHDpfw4X//A75+w4VV14+na14j1im78WqqEqg8EH58/xYcnmpuOQ+NIkvOnkCa\nYPDO8WbL0che0I57QKMg+3DcdqWsYWVUubKFYHZQqGEHE9NFFMsG5s3uwapls3FcjfWcf/fe1fjN\nEzvx9OZR53e6YbolHzEsZpogi1mSJCxb0I9t+6axZzTrEczEult97EjkDFS/4yuyhLkNLtNpBDyu\n7H2H7Dmo77r06Fiuz3NXLcDB8TxWLBzEM6+O4dXdEzgwnndeP+/E+bjs7KWYmsjVePadzeVnHYmy\nYTr1oGzG6nimWDUTl2fzWnXUMC46dSEWz+Uf38fDtZcd6zw7mir7tlWk+c0T/397dx4fVXX3D/wz\ne5YJWSAR2WWpLC4IboCFuNQHBTcEBBGsIIiFp1YEWWq1IkiKL1HgQV9g+RWl4r7hWlSUIkhFrFYR\nUTAQpKgxQMhMQmY7vz+Se3Nn5s5kJnMzcyb5vP9SJjO5J3Pu/Z79Wxb19eam15PLjDErmfZktVPb\nZqFr+xw4bBac0zP63v7WRHlOxruK5LJzOwMmoGM7J57/YF/YHHMi0y8tLzBrhrKVOdqze7TFuEaS\nikfTs1MufhPoFByYA4G4Jvn1tttE68FfeWFXrHrlq7BsR0pgvvai0xr9nXqfP2po96TtxTRSLPuY\nfX6BDu2y8ZtzO8f12TarGWMu7gkAOOdXhXj2/e+waech9fWbh/cOS4ZODTIdVoy9uKcamG1Wc6OL\naGLZaujMtGHi/5xu3IXWu/icjup/261mHI9jFCoV4mnUh9I2kEYO6oZBZ7Q37LpaiqaOpHZtn4Mp\nI/ri0M8u4IPwoexEdnK0uPEM7VC2uqfVgPmU0KEjn1/El/ZR52eibd/KinBIuprMO4ZAoXfzNmXL\nmAwaOz8cqHvYG7FyNnR4lQtl4mO3moO28+gNEQZiGMpOBlsji9RkoNuoj3GHibaBxHllfYk+E0On\n2eJJbhTxMxO6Igkp26WEaAic0Y79i1XYZnJ/IM4kFvHdXGoi7hr9HnNsybzDP78pW8ZkEO38cEW0\nVJpx/S7u8UyIzWoJqp966yJkyQ1ut1rqd1hEDs6pHmDSa9TH2mPWfg9sYOpL9JloVTsNdXXIiKHs\nFvdNqat3hVBPzjLiYR0aCH3++NI+hrZWzSZTWEJtLWf98Iq7NqTHHGOSdCBCYDZwAVwyNZYKE6h7\n2BtxQleqg0W6s5hNQdt59Hqk/jjWZzQnNT1qlC1Tofl1k023UR/jfawdueAhOfoM6zGrmQPrRu4S\nmTJscYFZO+Sp9pgNeFiHnrTjD8S7XSr82MJoX1xWxB5z7Ksq9c75NnIvdzIprc9IQ9mBgEBACENO\nNUv18GraMwVv59E7aESaoWxNLu5IHDHuGW4uRs0x81hZfYmuH1E7DaJhjjnRet3ivint6l2fJtNQ\nokIDYdDJX01Yld1YY8phs8BiNqE6ZI7ZqyZJ51C2VlOO4oz4uxiYE6ZNS6jbYzbw3kyEPZbAnOIe\ns97fKFLWu1BBQ9ms17pCk7DEK/SMBSNOiUvP7lMUyh/kd8v+qQbD5hjKXvf2N+p/xzaUHfz+xnq8\nylFx+/97At/9cBy9OtVlKlEyy8QyTKI3RJOui78a2y7VlKM4IzHqlKnWymyKYShbadSm+G+tjIRF\nOz7UkeI9v3oP+Vj3IWsXiXKOWV+iPWbl+f/Jnp9RfnwnPN7EF6E26SkdCARwzz33oLS0FGazGfff\nfz969uypvr5u3Tq8+OKLKCioS+W1cOFCdOvWLaELbQojH9Y2qxkXnXkq3Ce9+Pd3vwS91rmo8YTj\nXU/JQe8uefih3I2CHAeKNVs2ImnfNhsnqo9j3w+VamD2+WNL+QbUlbu4fwf88Isbfbvm4+iJ2qjz\n2jJrbLuUku7RiJWn2t8xZ/w5CX9ea/G/15+Jj/5zBL065WJvWUMSer35W3W7VIrnPZV7yRcla5my\nF/i3V/ROyjUpbh5+OkqPnFCfX+Mu6YlNnx7CGacVoCg/trMIuPircd07tMHA0wtxYd9TmvR+7XbZ\n0iNVAOryXSeiSYF58+bNMJlMeOaZZ/DJJ59g2bJleOyxx9TXd+/ejaVLl6Jv374JXVxTnPSEt3yN\nGJo0mUyYPKIP9v1QGRSYLxvYKaZeqDPThrtvHBDX77xmSDc89OznQataff74hkkmDU/uw6S5hC6w\nCGVkI0wbmPt0bdqhNK3ROb0KcU6vQgDB95zeyVqyzDEr1+mLmqJSwGGzYOjZHZJ1WQCAYf07Ylj/\nhgb85ed3weXnd4nrM4LmmDmUrctiNmPGdWcm8P7wv2ui9bpJgfmyyy7DJZdcAgA4fPgwcnNzg17f\nvXs3Vq9ejfLychQXF2PatGkJXWQ8dAOzga3y0CDcnEPDesNsPn8g5fNyqaBdba+nYWucsT1mSpzu\ndikDji00gja5TCS+QCBtVzTbuI+52ekF4ZTNMZvNZsybNw/vvfceVqxYEfTaiBEjMGHCBDidTsyY\nMQNbtmzBsGHDErrQWOklGzBiH7MidJtCc24/0lsx6g8Ys1c33TS2XUrZGmdEo8WIfKrUIOocc8oD\nc+M9Zp9B2/BSwW7lHHNz02vwJJLyEUhw8VdJSQkqKiowZswYvPXWW8jIyAAA3HzzzXA66863HTZs\nGL7++utGA3NhYePztLHQewi0Lcg27PPz8oM/v31hTkyf3ZTff7L+V1msFvX9QtStEjWqPEZqzmvK\nza2bU8vOduj+nur6Hk+OU//1eDjqV7xareagz5Lxb24kI8uX7Ww4x92mU1+t9QHjlKI2MSdkMELo\ndbRpU/fMcuZkRC2/3WZJi+8/9BoLNOfin1KUg7a56XdOvpas34HJVPdsVtgSrC9NCsyvvfYafvrp\nJ0ybNg0OhwNmsxnm+haCy+XCyJEj8fbbbyMjIwM7duzA6NGjG/3M8vKqplxKGPfJ8B6z23XSsM8P\nFfD6Gv3swsKcJv1+V1VdIoUTVbXq+z1ePyBszVaepmpqGWNV7a7L7nP8eI3u7ykvr8u646lt/Pto\nTJWrLt2j2dRQL5u7fKlmdPncmlSjx0+Ef2c19dsAjx11J22eWa+MtfXXUXHUHbH8Hq8fNotZ+u8/\nWvkA4PjxagTSOH2pzPegxWwKWUAoYooLkTQpMF9++eWYP38+brrpJvh8PixYsACbNm1CTU0NxowZ\ng1mzZmHixIlwOBwYNGgQhg4d2pRf0yQna8MrXnMOQ8Wa5aUplGEo7eKZdJ7vSkSj26XUfczGzTFz\n25Qxos0xp/pPrExzRVuV7fMH0nY3gy1oHzOHspuL2WwCNHUo0W2ATYoqmZmZePTRRyO+fvXVV+Pq\nq69u8kUlQu/2as5A1pytffW4QO0cs1+0yrmixvIxP7d5HwDOMcvIG2G7VKLHFhrBaom+dgFQzmBP\nz3uOR3ImR+izgid/hbhFZ69hohPxoZS0cW3bOAzPFasVuvgrIIQhp8qkI3uUgyD8gQD2/VAJwJiH\nzyUDOgFAs6QcbC0u7NeQXtAbIbuUDPVYachF3y5lTNayVOhUmI1MhwWdi5xBuZnJWKExJk+zxqIp\n0vMYqCh+fXYHbNp5CId/cav/ZnRLceL/nJ6Uh7bVYobZZFIDs1/dq5ueD4lEKOd+h+anBoIPsDCi\nZ9O5yIm1cy9OeW8unRXmZWLFHb/G75dvrVsXEcLvT/w8YSM07GNuLGtZega1Uwqy8H9/qJtKZH1u\nPqF1+cbLeiX0eS0uMOtJ160OQF2+WKWX2HC6VfqWp6nU/NQ13rDXtCvxjToPmA+xxEU7h9ov5Ogx\nN+xj1u8xB0R9cpQ0bgyzLjc/EXK+QqLbaFvFEz6dbyq7tSGRuzK/mq6t90Q46w9ycen1mDXD262x\n0SIrq84aCYUsQ9nqPuZIyVEkSbZBcgs9oCbRxYKtoral82pEm9WsDtU2pLFM/QMt2ZQec2i2LSC4\nRxZtrpCSy2wywWox6x7JKc1QdiMnf/kMPFGOWi5/SI850VGK9I1Y0YT8TdI5kNmsFnXxjJrGMo0b\nGk1ls5pht5nD8lMDwXPMtTrzmZQ62hEfrbpFjKmvx42d/NWaR6kodpHyxDdVq6ht6TwMVfdgq59j\nNnCvbjrKzrDBrddj1jxUa3XOSqfUqVsjEXm7VKo17GOOcNSrgVnLqOViYI5FyN8onQOZXTOU7Wvl\n8111gbmhx7znwFHs2vszvJpeMnvMconUYw4EhBTBTukJv/nxQd0hd3Uou5XecxQboxPftMjapk3P\nZreaY85fLCOb1Qx/oG5lqJEZlNJRdoYVNbU+tXX60LOfY9UrXwX1yM7s3jZVl0c6bFaL/napgCxz\nzA0H17y/63DY6w3TR6m/VpJXv24N6WF7d8lL+PNa5Hap35zXGef3PQVerx8Ou0WKuaymUh5egYAw\nNOdwOlK2IFTX+uDUbEdQemQjB3fF2T3bpeTaSJ8tQo/Z5xdSLMrUBtyqak/Y60qjQpuliSjU70ef\nhZ+P1SAn245sA1IBt8jADAC52fZUX4IhlEaFPyBa/XyXkvvaXeMNqvzKdql2aZ45pyVShrKFEEEr\nVf2SnPmubeTqbXFR1i/YbKlvRJC8bFYLOhYadwoka5vk1OQNftHqV4g667dMuU56g1ZiK0PZ6Txl\n0VLZrWYIBJ+sJYSQJsextnHg0Ek/qZzzbWfdoiRibZOcEpiD5pgl6GmkgvZYTu3qbA8fntKy6WRI\na2hgpr4eaxu5enmh2eijVGBtk5yaVckfaNX7mIHgYzm1q7Nr63PM8uEpn0gZ0gA5Rn60vfbQYxWB\nhgYF55gpmVJ/Z1BU2jzErX2OWZlXdp/0BZ2ZXVO/d9nGh6d07DqBWdmPL8NKZ22vXS+RBXvMlAqs\nbZKzaFdlB1r3nkplVfZn35bjm7Jj6r9/XXoUAIeyZWSrHx4+9FOVmvFNpv342saBrz596ElPw2iM\nl4GZUqDFrspuKcyaHrPykJBhbi4V8utznO45eAx7DjYE5rKfXQD48JSR0lha9cpXAID/N+8SqdZK\naBu5Xx84hhc+2I9+pxXgrhv6A2gIzHrzz0TNhU8yyamH7AcEaurnVbMciaUUS1cd2mXj1LZZEV/n\nw1M+eo2lhsQQqX/8WC1mjLm4BwDg4I9VAIDd9SMwQMNWPDb6KJlY2yRnMTX0mJWUh87M1jvQcV7v\nooiv2SQYGqVgoQHN6wtItSobAHp1rDupyR8IPwiF26UoFVjbJKc9+UtJeZid0Tp7zED0BOQ8BEI+\noauZq096pZpjBhoWU+qdUKYs/uKqbEomOe4Mikh7lq+yRSjLgCPf0pUzSqOEvRr5hPaYXSd9msQQ\ncvSYlXlmvVXZXPxFqdB6n/BpomG7VEDdIhSt19jSRWuUsFcjn9DGkrvGC3P99IwsuwuiNRC8nGOm\nFGBtk5x2u5T7pA9Wi7lV9wy1jZLQB7sM2YooWOj0QrWmxyzDPmYg+nU0DGW33nuOko+1TXJKsPEF\nBA78eALZGdagZACtjTZ5RU5W6x05SBc2S/Aohvukt+GAEWl6zOHXofSU1QNGuOKfkkiOO4MiUlrz\nH3x2GELoH7TfmmjTPeY5W0YGsZbMHtJjdp/0adKXytHA1GsgKOs5vGraRz4qKXk4xyw55Vzsn49V\nAwCGnNk+lZeTcjlZdkwafjqqqr2oqfWh9EhVqi+JoggNaLVef8NZ2RLsYwb0Gwjukz7kOR0NaR8Z\nmCmJWNskpwxln6w/D/qsHu1SeTlSKO7fEVcN7sZ9y2kg9Pzyun3Mkq3K1mkgKAstPb4ATCZ55sOp\ndeCTTXLKA6Gmtm5oLXRosDVrrck80kloj9nr82uSschRl/XqUbU6lB2A3Wpp1es6KPnkuDMoIktI\nj5m9xAbsxcgvdAjY4wvIN8esU4+UfN8en5/D2JR0rHGS0yaxALg6VIvbo+QXdiSnN6BJYiHH48dk\nMoUFZ3Xxly/AUSpKOtY4yYU+MLg6tIEJDMyyC00s4vH5G47klKhhFdpIUOaYvb4A83xT0vEpL7nQ\nhxeH1RoIEX6EIslFL4mFbHPMQMPoi7Idz3WyYfEXp48o2VjjJKd9eJlNJmmG/2TAsCy/0MB80uPH\nCx/uByDPHDPQ0MhT9sZ/8NlhHKuqhdfn51A2JR1rnOTMmtWg7C0H0/aY54zrn8IroUjMJhNGDu6K\nc3rVbfM7VlWrvtajQ26qLiuMMjJVmJep/tt/K9zw+QWnjyjpWOMkpx3KZmAOph3J7tOtIHUXQlGN\nGtoD/3v9WbBZzTjmqgvMxf07INMhz/lGyr3lsFkwurgHgIYtU5xjpmTjk15y2j2WHFILxqHs9GK3\nmlFbv+1Ptgxp2sxkSpBWtkyxx0zJxhonOXNQj5ktdy0u/kov2hGf7Ch5tVNByYLl8QXUQKz2mNkg\npiRjjZOcVROY2XIPwbicVrS90mh5tVNBubc8Pr96ncqWKa7KpmRjjZOcmXPMEQXYY04rUveY64Ov\n1xvQGcrmSBUlF5/0kjOzx0wthDYwOzPl6jErJ+p5fJrAXMOhbEoN1jjJaTPfcI45mNJhlmc3LEWj\nbVhmSdZjVq7Nq5lj3vVtedBrRMnCGic57XYpmQ5kkEHxOR1xSn4m7hhzdqovhWJwVs92cGba0OUU\nJ4ryMxt/QxKNLu6BU9tm4bdX9A47j95hZ4OYkkuu8SQKo90uxTnmYPk5Diy5bVCqL4NidOWFXXHl\nhV1TfRm6Tm2bjcVTLwQAlP1UFfSabPPh1PLxSS857RGcFp2E7kRkrNAGcLZkK8ip5eOTXnLawMyh\nbKLmFx6Y2WOm5GJglpx2jlmmbDxELVXo9ijZTimjlo9Peslpe8lWifLXErVUHMqmVGNglpwlaCib\nXxdRcws9k55D2ZRsfNJLTttjtnCOmajZhS6yZPIYSjbWOMlp8zGzx0yUHMpdl51hhcnEBjElFydP\nJGcy8YARomRbPacYZpMJjMmUCgzMaYT7mImSg6NTlEqsfWmEc8xERC0fA3MaYSueiKjla9KTPhAI\nYMGCBRg/fjwmTJiAffv2Bb2+efNmjB49GuPGjcMLL7xgyIUS9zETEbUGTQrMmzdvhslkwjPPPIM7\n7rgDy5YtU1/z+XwoKSnBunXrsH79ejz33HM4evSoYRfcmrHHTETU8jXpSX/ZZZfhgQceAAAcPnwY\nubm56mv79+9H165d4XQ6YbPZMHDgQOzcudOYq23lOMdMRNTyNXlVttlsxrx58/Dee+9hxYoV6r+7\nXC7k5OSo/5+dnY2qqiq9jwhSWJjT6M+kMyPKV5CfLfXfSeZrMwLLl/5aehlZvpYhoe1SJSUlqKio\nwJgxY/DWW28hIyMDTqcTLpdL/Rm32402bdo0+lnl5Y0H73RVWJhjSPmq3Sel/TsZVUZZsXzpr6WX\nkeVLL9EaGU0ayn7ttdewZs0aAIDD4YDZbIa5fo9tjx49cPDgQZw4cQIejwc7d+5E//79m/JrKASz\nSxERtXxN6jFffvnlmD9/Pm666Sb4fD4sWLAAmzZtQk1NDcaMGYP58+dj8uTJEEJgzJgxKCoqMvq6\nWyWGZSKilq9JgTkzMxOPPvpoxNeLi4tRXFzc1GsiIiJqtdgJSyc8uJeIqMVjYCYiIpIIAzMREZFE\nGJjTgJKoPSfTluIrISKi5sa0j2lg0a0X4Pv/nkCnImeqL4WIiJoZA3MaaJebiXa5mam+DCIiSgIO\nZRMREUmEgZmIiEgiDMxEREQSYWAmIiKSCAMzERGRRBiYiYiIJMLATEREJBEGZiIiIokwMBMREUmE\ngZmIiEgiDMxEREQSYWAmIiKSCAMzERGRRBiYiYiIJMLATEREJBEGZiIiIokwMBMREUmEgZmIiEgi\nDMxEREQSYWAmIiKSCAMzERGRRBiYiYiIJMLATEREJBEGZiIiIokwMBMREUmEgZmIiEgiDMxEREQS\nYWAmIiKSCAMzERGRRBiYiYiIJMLATEREJBEGZiIiIokwMBMREUmEgZmIiEgiDMxEREQSYWAmIiKS\nCAMzERGRRBiYiYiIJMLATEREJBEGZiIiIokwMBMREUmEgZmIiEgiDMxEREQSYWAmIiKSCAMzERGR\nRBiYiYiIJMLATEREJBFrvG/w+XxYsGABDh8+DK/Xi+nTp+OSSy5RX1+3bh1efPFFFBQUAAAWLlyI\nbt26GXbBRERELVncgXnjxo3Iz8/H0qVLUVlZiWuvvTYoMO/evRtLly5F3759Db1QIiKi1iDuwHzF\nFVdg+PDhAIBAIACrNfgjdu/ejdWrV6O8vBzFxcWYNm2aMVdKRETUCsQdmDMzMwEALpcLd9xxB+68\n886g10eMGIEJEybA6XRixowZ2LJlC4YNG2bM1RIREbVwJiGEiPdNR44cwcyZM3HTTTfhuuuuC3rN\n5XLB6XQCADZs2IDKykrcfvvtxlwtERFRCxf3quxffvkFU6ZMwZw5c3SD8siRI1FTUwMhBHbs2IF+\n/foZdrFEREQtXdw95sWLF+Ptt99G9+7dIYSAyWTC2LFjUVNTgzFjxmDjxo146qmn4HA4MGjQIMyc\nObO5rp2IiKjFadJQNhERETUPHjBCREQkEQZmIiIiiTAwExERSYSBmYiISCJRDxjROxe7Z8+emDdv\nHsxmM3r16oX77rtP/fmjR49i/PjxeP3112G329V/f/fdd/HOO+/g4YcfDvsdtbW1mDNnDioqKuB0\nOlFSUoL8/HwAgN/vx5133omxY8fioosuCnvvxx9/jOXLl8Nms6GgoABLly6Fw+HA7373Oxw/fhxW\nqxUZGRlYs2ZNs5XR5XJh9uzZcLvd8Hq9mDdvHvr37x9TGbdv346HH34YNpsNgwYNwh133BF2fV98\n8QUWL14Mq9WKIUOGYMaMGdi6dSvWrFkDk8kEIQR27dqFN954A927d0+ofM8//zyee+452Gw2TJ8+\nHcXFxQCAoUOHquedn3POOWGHykQq36effoqlS5fCbDbjvPPOw1133RW1fIMHD1ZX8ZeUlGDXrl2w\nWCy4++67MWDAgIS/P73yuVwu3HnnnaiurobD4cBDDz2Etm3bGla+SHX05ZdfxrPPPotAIIBLL700\n6l7/VNbRsrIy3HffffD5fLDb7Vi2bBlyc3PDrjH0Xm2uOmp0+WK5B6N9zwcPHsTMmTPx+uuvG/L9\naevo7bffjmHDhmHNmjXYunUrTCYTTpw4gV9++QUfffRRTOWbOHGi+h2UlpZi1KhRmDVrVtB7I92D\nixcvxmeffYbs7GzMnj0bZ511liFl1PsOFdFiRaSfiXT9WpHuw3jKmFQiipdeekk8+OCDQgghKisr\nRXFxsZg+fbrYuXOnEEKIe++9V7z77rtCCCG2bt0qrr32WjFw4EBRW1urfsaiRYvEFVdcIWbNmqX7\nO/72t7+JlStXCiGEePPNN8WiRYuEEEKUlZWJcePGiYsvvlhs3bpV973Dhw8XFRUVQgghHn74YbF+\n/XohhBBXXnlltGIZWsYVK1aIJ598UgghxPfffy+uu+66mMt47bXXiv379wshhBg/frz49ttvw957\nzTXXiEOHDgkhhJg6darYs2dP0Ot//etfxSOPPJJw+crLy8XIkSOF1+sVVVVVYuTIkcLj8YiDBw+K\n6dOnR/0bRirfqFGjxOHDh4UQQkycODHs2iOVb8+ePeKGG24QQghx4MAB3b+pUeV78sknxUMPPSSE\nEOL5558XJSUlhpZPr46WlZWJsWPHitraWhEIBMTy5cuFz+dLuIxCGF9HJ02aJL744gshhBCbNm0S\nn3/+edh7G7tXjaqjzVG+WO7BSN/zq6++KkaNGiWGDBkSsWzxlC9SHdW67bbbxPbt22Mun6KsrExc\nf/31orq6Ouy9evfgBx98IG699VYhhBBHjx4Vo0aNMqSMQjQ9VkT6mcaekULo34fxljGZog5lX3HF\nFWoL0u/3w2Kx4Ouvv8a5554LoK4n9fHHHwMALBYL1q1bF9aaHjBgAP785z9H/B27du3C0KFDwz7P\n7XZj8eLFuOCCCyK+d/369WoWK5/PB4fDgYqKCpw4cQLTp0/HhAkT8OGHH0YrYsJlvOWWWzBu3Lig\na2isjDt27AAA9O3bF8eOHYPH40FtbS3M5uCvw+Vywev1olOnTgCAiy66CNu3b1df//HVgt8+AAAJ\nzElEQVTHH7Fx40bMmDEjofJt374d//nPfzBw4EBYrVY4nU5069YNe/fuxVdffYWffvoJkyZNwm23\n3YbS0tJGy6f8vV544QV06NABbrcbLpcLWVlZMZXvlFNOQUZGBjweD6qqqoJa1EaX71e/+hVcLpd6\nPTabzbDyAfp1dPv27ejXrx/uvvtuTJw4EQMGDIDFYkmojM1RR2tra3H06FG8//77mDhxIv7973/r\n9iiqq6sj3qtG1dFU3YNA5O85Ly8PTz/9dMRyxVO+aHVUsWnTJuTm5mLQoEGNlk/5eykefPBBzJ49\nWz1SWaF3D27btg379u1TRynz8/NhsVhQUVGRUBkTjRV6P9PYM1Khdx/GW8ZkihqYMzMzkZWVFXQu\nttBse87OzkZVVRUAYNCgQcjNzQ16Haj7wqLRHuGZnZ2tPiR79+6tO+yl1a5dOwB1FfaTTz7BNddc\nA6/XiylTpuCxxx7DypUrsWTJEhw9erTZyuh0OmG321FeXo67775bdzgztIzK5/Xq1QvTp0/HyJEj\n0aFDB/To0SPofW63W31f6HuBuhSbv/3tb3WDSTzlc7lccLvdyMnJUf89KysLVVVVKCoqwm233Yan\nnnoK06ZNw5w5cxotn/Idms1mfPHFF7jqqqtQWFiI9u3bx1Q+q9UKk8mE4cOHY8qUKZg8eXKzlS8v\nLw/btm3DiBEjsHbtWowePdqw8gH6dfTYsWP49NNPsWTJEqxYsQIPPPCA+plNLWNz1NHjx4/ju+++\nw0UXXYT169ejsrISr7zySth7Tz/99Ij3qlF1NFX3IBD5ex42bBgyMjIiliue8kWro4o1a9ZEPLAp\nUh0FgL1798LtduPCCy8Me5/ePehyudC3b19s3boVPp8Phw4dwr59+1BdXZ1QGRONFXo/09gzUqF3\nH/bp0yeuMiZTo4u/jhw5gptvvhnXXXcdRowYEdSidLvdaNOmTdDPm0ymqJ9XVlaGiRMnYtKkSXjx\nxReRk5MDt9utfp62YoZ6+umn1ff+/PPPAOpu/HXr1mHt2rWw2+1o164dbrjhBpjNZhQUFKBPnz66\nvTwjy7h3715MnjwZd911F84999yYylhVVYU1a9bgrbfewqZNm9ClSxesXbs2qIx+vz/oBtNeixAC\nH3zwAUaMGBG1bLGWz+l06v6uM844Q03rOXDgQJSXl8f1HZ599tnYvHkz+vTpg9WrV8dUvldffRWF\nhYXYvHkz3n//faxcuRI//fRTs5Rv1apVmDp1Kt58802sXbsWM2fOxKFDhwwpX6Q6mpeXh/PPPx+Z\nmZkoKChAjx49pKyjeXl5yM7OxnnnnQcAuPjii/Hll1/qllGP0XXU6PLFcg8q5dN+z9HWrCRSvkh1\nFAD279+P3NxcdO7cGUB8z9GNGzdi7Nix6v/Hcg8OHjwY5557LiZNmoQnnngC/fr1U9f+JFJGrXhi\nxUsvvaT7M6GNEOX3xHIfDhkyJO4yJkvUxV/Kudj33nuv2trq06cPdu7cifPOOw///Oc/w1phoa2g\nUF26dMH69evV/6+qqsKWLVtw5plnYsuWLerQh54JEyZgwoQJ6v8//vjj2LNnD9atW6cOd27fvh1/\n//vfsWbNGrjdbuzbt0+3FWxUGfft24c//OEPePTRR3H66afHXEaHw4Hs7Gx1aKmwsBDHjh3DlClT\ngspot9tx6NAhdOrUCR999JHaYv7222/Ro0ePqMO88ZTvzDPPxCOPPKIO6X3//ffo1asXVqxYgby8\nPNx666345ptvcOqpp8b8HU6YMAGPP/442rRpg+zsbHg8nrDvUK98paWl6nBhZmYm7HY7ampqmqV8\nubm5aou7oKAAbrcbnTt3Nqx8enV0wIAB2LBhAzweD3w+H/bv34+uXbsm/B1qGVVHTzvtNOzatQsD\nBw7Ezp070atXL9x4441BZYzE6DraHOWL5R7U+57jkWgdBeqea7/+9a/Vz4znOfrxxx8Hpd+N5R48\ncOAA2rdvjw0bNuDHH3/E3Llzg3qmTS2jVryxQo8yWhJ6/WeddVaj92G8ZUymqIF59erVOHHiBB57\n7DGsWrUKJpMJf/zjH7Fo0SJ4vV706NFDzc2saKwVFGr8+PGYO3cubrzxRtjt9qir8bQqKiqwatUq\nnHHGGZgyZQpMJhOuvPJKjBs3Dtu2bVN7zbNmzUJeXl6zlXHZsmXweDxYvHgxhBBqL6yxMtrtdsyd\nOxeTJ0+Gw+FAmzZtUFJSEnZ9999/P2bPno1AIIAhQ4aoc3ylpaVq6zmaWMtnMpkwceJE3HjjjRBC\nYNasWbDb7erw9ZYtW2C1WrFkyZKw3xHpO5wyZQqmTp0Ku92OoqIiLFq0KKbynXHGGfjss88wbtw4\nCCFw1VVXqavCjS7f73//e9xzzz3YsGEDfD6f7jU2tXzR6ujo0aPVedEZM2aE9SaaUkYtI+ooACxa\ntAgLFy5EIBBAx44ddacyIjG6jhpdvljvwVjqsRHli1RHgbogMnjw4Ii/I9pztKKiQnclvULvHvR4\nPFi2bBmeeeYZOByOoBXViZRRK95YEc/1a0W6D0eNGhVXGZOJZ2UTERFJhAeMEBERSYSBmYiISCIM\nzERERBJhYCYiIpIIAzMREZFEGJiJiIgkwsBM1ArMnz8fr776atTXjxw5ksQrIqJIGJiJCP/6178a\nPYmJiJKDB4wQtVBLlizBhx9+iKKiIgghMHr0aJSWlmLHjh2orKxEfn4+Vq5ciZdffhkrVqxAt27d\n8PTTT+PgwYMoKSnByZMnkZ+fj4ULF6Jjx46pLg5Rq8EeM1EL9I9//APffPMN3n77bSxfvhwHDx6E\nz+dDaWkpnnvuObzzzjvo0qUL3njjDUybNg1FRUV44oknkJWVhT/96U9YtmwZXn75Zdxyyy245557\nUl0colYl6lnZRJSePvnkE1x++eVqlrWhQ4fCarVi7ty5eP7551FaWorPP/8cXbp0Ud8jhMCBAwdQ\nVlaG22+/HUIImEwmNWsRESUHAzNRC2QymRAIBNT/t1gsOHbsGCZPnozJkydj+PDhMJvNYfPKfr8f\nXbp0UfMuCyFQXl6e1Gsnau04lE3UAg0aNAjvvPMOPB4PKisr8dFHH8FkMuGCCy7ADTfcgO7du2Pb\ntm1q8LZarfD7/ejevTsqKyvx6aefAgBeeOEFzJ49O5VFIWp12GMmaoEuvfRSfPnll7jqqqtQWFiI\nnj17ora2Fnv37sXVV18Nm82G3r1744cffgAAFBcXY+rUqVi7di2WL1+ORYsWwePxwOl04i9/+UuK\nS0PUunBVNhERkUQ4lE1ERCQRBmYiIiKJMDATERFJhIGZiIhIIgzMREREEmFgJiIikggDMxERkUT+\nP3nznmCWPDgTAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17dec9710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe4AAAFkCAYAAADmPNjoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17de64390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeYAAAFkCAYAAAD165gcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17de1c4d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeYAAAFkCAYAAAD165gcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17d67b410>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# show reviews over time\n",
    "topmost = biz_summary.sort_values([('date','len')], ascending=False).head(n=5).index\n",
    "for bizid in topmost:\n",
    "    subset = df.query('business_id == \"' + str(bizid) + '\"')\n",
    "    subset.rolling(window=15).mean().plot('date', 'stars')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Might be useful to subset them..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "bad = df[df.stars < 3]\n",
    "good = df[df.stars > 3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "38473"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(bad)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "156071"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(good)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Do Some Text Analysis Now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "english_stops = stopwords.words('english')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What are the most common words in this corpus?\n",
    "\n",
    "Using nltk's Text collection functions, we can report... (or you can do by 'hand' with a counter.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def tokenize(string):\n",
    "    stoplist = nltk.corpus.stopwords.words('english')\n",
    "    tokens = nltk.word_tokenize(string)\n",
    "    return clean_tokens(tokens, stoplist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def clean_tokens(tokens, stoplist):\n",
    "    # Get rid of short words, stop words, punct, apostrophe-containers, and numeric strings\n",
    "    import re\n",
    "    import string\n",
    "    \"\"\" Lowercases, takes out punct and stopwords and short strings \"\"\"\n",
    "    return [token.lower() for token in tokens if (token not in string.punctuation) and \n",
    "                (token.lower() not in stoplist) and \n",
    "                len(token) > 2 and re.search('[a-zA-Z]', token) and not re.search(\"'\", token)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for row,vals in df.iterrows():\n",
    "    df.set_value(row, 'tokens', ' '.join(tokenize(vals['text'])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Save that out since it took so long:\n",
    "df.to_msgpack('data/yelp_df_tokens.msg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_msgpack('data/yelp_df_tokens.msg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "      <th>tokens</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9yKzy9PApeiPPOUJEtnvkg</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>fWKvX83p0-ka4JS3dc6E5A</td>\n",
       "      <td>5</td>\n",
       "      <td>My wife took me here on my birthday for breakf...</td>\n",
       "      <td>rLtl8ZkDX5vH5nAx9C3q5Q</td>\n",
       "      <td>wife took birthday breakfast excellent weather...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ZRJwVLyzEJq1VAihDhYiow</td>\n",
       "      <td>2011-07-27</td>\n",
       "      <td>IjZ33sJrzXqU-0X6U8NwyA</td>\n",
       "      <td>5</td>\n",
       "      <td>I have no idea why some people give bad review...</td>\n",
       "      <td>0a2KyEL0d3Yb1V6aivbIuQ</td>\n",
       "      <td>idea people give bad reviews place goes show p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6oRAC4uyJCsJl1X0WZpVSA</td>\n",
       "      <td>2012-06-14</td>\n",
       "      <td>IESLBzqUCLdSzSqm0eCSxQ</td>\n",
       "      <td>4</td>\n",
       "      <td>love the gyro plate. Rice is so good and I als...</td>\n",
       "      <td>0hT2KtfLiobPvh6cDC8JQg</td>\n",
       "      <td>love gyro plate rice good also dig candy selec...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>_1QQZuf4zZOyFCvXc0o6Vg</td>\n",
       "      <td>2010-05-27</td>\n",
       "      <td>G-WvGaISbqqaMHlNnByodA</td>\n",
       "      <td>5</td>\n",
       "      <td>Rosie, Dakota, and I LOVE Chaparral Dog Park!!...</td>\n",
       "      <td>uZetl9T0NcROGOyFfughhg</td>\n",
       "      <td>rosie dakota love chaparral dog park convenien...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6ozycU1RpktNG2-1BroVtw</td>\n",
       "      <td>2012-01-05</td>\n",
       "      <td>1uJFq2r5QfJG_6ExMRCaGw</td>\n",
       "      <td>5</td>\n",
       "      <td>General Manager Scott Petello is a good egg!!!...</td>\n",
       "      <td>vYmM4KTsC8ZfQBg-j5MWkw</td>\n",
       "      <td>general manager scott petello good egg detail ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id        date               review_id  stars  \\\n",
       "0  9yKzy9PApeiPPOUJEtnvkg  2011-01-26  fWKvX83p0-ka4JS3dc6E5A      5   \n",
       "1  ZRJwVLyzEJq1VAihDhYiow  2011-07-27  IjZ33sJrzXqU-0X6U8NwyA      5   \n",
       "2  6oRAC4uyJCsJl1X0WZpVSA  2012-06-14  IESLBzqUCLdSzSqm0eCSxQ      4   \n",
       "3  _1QQZuf4zZOyFCvXc0o6Vg  2010-05-27  G-WvGaISbqqaMHlNnByodA      5   \n",
       "4  6ozycU1RpktNG2-1BroVtw  2012-01-05  1uJFq2r5QfJG_6ExMRCaGw      5   \n",
       "\n",
       "                                                text                 user_id  \\\n",
       "0  My wife took me here on my birthday for breakf...  rLtl8ZkDX5vH5nAx9C3q5Q   \n",
       "1  I have no idea why some people give bad review...  0a2KyEL0d3Yb1V6aivbIuQ   \n",
       "2  love the gyro plate. Rice is so good and I als...  0hT2KtfLiobPvh6cDC8JQg   \n",
       "3  Rosie, Dakota, and I LOVE Chaparral Dog Park!!...  uZetl9T0NcROGOyFfughhg   \n",
       "4  General Manager Scott Petello is a good egg!!!...  vYmM4KTsC8ZfQBg-j5MWkw   \n",
       "\n",
       "                                              tokens  \n",
       "0  wife took birthday breakfast excellent weather...  \n",
       "1  idea people give bad reviews place goes show p...  \n",
       "2  love gyro plate rice good also dig candy selec...  \n",
       "3  rosie dakota love chaparral dog park convenien...  \n",
       "4  general manager scott petello good egg detail ...  "
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Most common words in the whole dataset of clean tokens?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# item for sublist in l for item in sublist\n",
    "allclean = [token for token in df.tokens.values for token in tokens.split(' ')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17243025"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(allclean)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from nltk import Text\n",
    "cleantext = Text(allclean)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "mostcommon = cleantext.vocab().most_common()[0:1500]\n",
    "mostcommon_words = [word[0] for word in mostcommon]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[u'excellent',\n",
       " u'best',\n",
       " u'get',\n",
       " u'amazing',\n",
       " u'quickly',\n",
       " u'pretty',\n",
       " u'ever',\n",
       " u'made',\n",
       " u'griddled',\n",
       " u'anyway',\n",
       " u'bloody',\n",
       " u'weather',\n",
       " u'simply',\n",
       " u'looks',\n",
       " u'delicious']"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mostcommon_words[0:15]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    My wife took me here on my birthday for breakf...\n",
       "1    I have no idea why some people give bad review...\n",
       "2    love the gyro plate. Rice is so good and I als...\n",
       "3    Rosie, Dakota, and I LOVE Chaparral Dog Park!!...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[0:4]['text']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def word_count(row):\n",
    "    # doesn't strip stops but ok if it applies to all\n",
    "    return len(row['tokens'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df['text_length'] = df.apply(lambda row: word_count(row),axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "      <th>tokens</th>\n",
       "      <th>text_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>63676</th>\n",
       "      <td>XTtNevBI5G5x_GjFqSBMpA</td>\n",
       "      <td>2012-11-20</td>\n",
       "      <td>4ij2LU0iZ4dJDXk8fnFtEQ</td>\n",
       "      <td>2</td>\n",
       "      <td>Like J.B. - I'll keep it simple and to the poi...</td>\n",
       "      <td>Id-8-NMEKxeXBR44eUdDeA</td>\n",
       "      <td>like j.b. keep simple point least try said try...</td>\n",
       "      <td>3603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141338</th>\n",
       "      <td>1M4oczf2lmkdgbrJ3J7OqA</td>\n",
       "      <td>2012-08-18</td>\n",
       "      <td>gqHC4DfRl_LE-9SDUwqijQ</td>\n",
       "      <td>5</td>\n",
       "      <td>Full review blogged, high points here.\\n\\nThe ...</td>\n",
       "      <td>5lq4LkrviYgQ4LJNsBYHcA</td>\n",
       "      <td>full review blogged high points space heard co...</td>\n",
       "      <td>3530</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50817</th>\n",
       "      <td>gFJtzwGmDTABRwL2F8GsSw</td>\n",
       "      <td>2012-11-06</td>\n",
       "      <td>YOCJYcjAj33y11K_ucHjKQ</td>\n",
       "      <td>4</td>\n",
       "      <td>Comp disclosure:  Generally not one to visit c...</td>\n",
       "      <td>5lq4LkrviYgQ4LJNsBYHcA</td>\n",
       "      <td>comp disclosure generally one visit chains eve...</td>\n",
       "      <td>3513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86251</th>\n",
       "      <td>Gy7ERh2GhmAYMZMO48ztBg</td>\n",
       "      <td>2010-02-18</td>\n",
       "      <td>mzif_3SuuGMyKHxiEY5ncQ</td>\n",
       "      <td>4</td>\n",
       "      <td>Back in the day, resorts and hotels were some ...</td>\n",
       "      <td>lPaYMDmJbAnv_3pmZH_inw</td>\n",
       "      <td>back day resorts hotels best culinary destinat...</td>\n",
       "      <td>3481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36257</th>\n",
       "      <td>8m08a9xJKmANwmeuR-0bPA</td>\n",
       "      <td>2012-10-01</td>\n",
       "      <td>x2-_KKM8qu54Yjj7GIRqtQ</td>\n",
       "      <td>4</td>\n",
       "      <td>Full review blogged:\\n\\nTrio of Duck Fat Fries...</td>\n",
       "      <td>5lq4LkrviYgQ4LJNsBYHcA</td>\n",
       "      <td>full review blogged trio duck fat fries includ...</td>\n",
       "      <td>3446</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   business_id        date               review_id  stars  \\\n",
       "63676   XTtNevBI5G5x_GjFqSBMpA  2012-11-20  4ij2LU0iZ4dJDXk8fnFtEQ      2   \n",
       "141338  1M4oczf2lmkdgbrJ3J7OqA  2012-08-18  gqHC4DfRl_LE-9SDUwqijQ      5   \n",
       "50817   gFJtzwGmDTABRwL2F8GsSw  2012-11-06  YOCJYcjAj33y11K_ucHjKQ      4   \n",
       "86251   Gy7ERh2GhmAYMZMO48ztBg  2010-02-18  mzif_3SuuGMyKHxiEY5ncQ      4   \n",
       "36257   8m08a9xJKmANwmeuR-0bPA  2012-10-01  x2-_KKM8qu54Yjj7GIRqtQ      4   \n",
       "\n",
       "                                                     text  \\\n",
       "63676   Like J.B. - I'll keep it simple and to the poi...   \n",
       "141338  Full review blogged, high points here.\\n\\nThe ...   \n",
       "50817   Comp disclosure:  Generally not one to visit c...   \n",
       "86251   Back in the day, resorts and hotels were some ...   \n",
       "36257   Full review blogged:\\n\\nTrio of Duck Fat Fries...   \n",
       "\n",
       "                       user_id  \\\n",
       "63676   Id-8-NMEKxeXBR44eUdDeA   \n",
       "141338  5lq4LkrviYgQ4LJNsBYHcA   \n",
       "50817   5lq4LkrviYgQ4LJNsBYHcA   \n",
       "86251   lPaYMDmJbAnv_3pmZH_inw   \n",
       "36257   5lq4LkrviYgQ4LJNsBYHcA   \n",
       "\n",
       "                                                   tokens  text_length  \n",
       "63676   like j.b. keep simple point least try said try...         3603  \n",
       "141338  full review blogged high points space heard co...         3530  \n",
       "50817   comp disclosure generally one visit chains eve...         3513  \n",
       "86251   back day resorts hotels best culinary destinat...         3481  \n",
       "36257   full review blogged trio duck fat fries includ...         3446  "
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sort_values('text_length', ascending=False).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(aka. SKETCHY TEMPE with BONNIE G, Part Three of Three)\n",
      "\n",
      "SET-UP: Bonnie G and Mr K walk up to the Palo Verde Lounge, a tiny run-down box nestled in the armpit of Broadway, gleaming in the diviest of glories. The outside light may as well have been flickering (but it wasn't). \n",
      "\n",
      "BONNIE G:  \"So, Mr K, what do you think? Isn't it all you ever dreamed about? You know it is.\" \n",
      "\n",
      "ACTION: Mr K stands there for a second and takes in the aura. It is what he always dreamed about. \n",
      "\n",
      "MR: \"Hell yeah. Let's do it.\" \n",
      "\n",
      "ACTION: Bonnie G smiles. They walk in.\n",
      "\n",
      "SCENE: Small place. Like a shoebox. A very divey shoebox which may have seen better days. Or perhaps each of its days has always been this way. Pool table to the left. Bar to the right. Graffiti this way and that. Some palpable decrepitation. Behind the bar, a microwave. Stale scent of lingering smoke of many years still permeates. Faint scent of other odors, too. Some hipsters milling about, doing whatever it is hipsters do. Others of a more miscreant yet harmless variety. Bartender with long hair and glasses hanging out behind the counter. Mr K stands there at the door.\n",
      "\n",
      "MR K: \"Niiiice.\"\n",
      "\n",
      "ACTION: Bonnie G and Mr K walk to the bar and sit down.\n",
      "\n",
      "MR K: \"Corn nuts. Yo, Bonnie, they've got Corn Nuts!\"\n",
      "\n",
      "BONNIE G: \"Get it!\"\n",
      "\n",
      "MR K: \"No wait. They've got Beef Jerky.\"\n",
      "\n",
      "BONNIE G: \"Get it!\"\n",
      "\n",
      "MR K: \"No wait. Slim Jims!\"\n",
      "\n",
      "BONNIE G: \"Get it!\"\n",
      "\n",
      "MR K: \"But yo, hold on, Ramen noodles. That's dope.\"\n",
      "\n",
      "BONNIE G: \"Get it!\"\n",
      "\n",
      "MR K: \"I'm hungry now.\"\n",
      "\n",
      "BONNIE G: \"Get it!\"\n",
      "\n",
      "BARTENDER WITH LONG HAIR AND GLASSES: \"So, what can I get you guys?\"\n",
      "\n",
      "MR K: \"Two g & ts, thanks.\"\n",
      "\n",
      "OVERHEAD SOUNDSYSTEM: \"Wheel in the sky keeps on tuuuuuuurniiiiin...I don't know where I'll be tomorrooow....\"\n",
      "\n",
      "MR K: \"GAH. I hate this song. I really hate this song.  No, I REALLY fucking hate this song.\"\n",
      "\n",
      "BONNIE G: \"Um, I can tell. Your body's contorting.\"\n",
      "\n",
      "ACTION: Mr K decides to ignore Journey and continue conversing.\n",
      "\n",
      "For the next long time, Mr K & BONNIE G: \".....yada yada yada, Lactose Intolerant War Stories, yada yada yada x10.....\"\n",
      "\n",
      "ACTION: Bonnie G and Mr K step away from the bar for a couple minutes as Mr K checks the rest of the place out. Upon their return to the bar:\n",
      "\n",
      "MR K: \"Uh, where'd our drinks go? They took away our drinks. Just like that. Gone. Doh.\"\n",
      "\n",
      "BARTENDER WITH LONG HAIR AND GLASSES: \"Um, uh, sorry, dude. I thought you guys left. I can make you some new drinks, though.\"\n",
      "\n",
      "ACTION: Bonnie G and Mr K don't really debate about more drinks. It's late. Mr K lives far away. And he's, well, old. They make motion to leave, except:\n",
      "\n",
      "MR K: \"What the hell is this?\"\n",
      "\n",
      "BONNIE G: \"It's a grip machine. It tests your gripping power. I kick all the guy's asses. Well, the weak guys who do nothing but read, anyway.\"\n",
      "\n",
      "MR K: \"Ok, let's try it out.\"\n",
      "\n",
      "BONNIE G: \"Go ahead.\"\n",
      "\n",
      "ACTION: MR K squeezes the grip gun as hard as he can. The machine indicates MR K has a grip of a 20 year old. He tries to figure out if that means he's strong or weak. He's convinced himself that it means he's strong, although he probably knows it means otherwise. Bonnie G gets up next. Bonnie G squeezes. But there's apparently an issue. Grip Machine reads that Bonnie G has the grip strength of something like a 10 year old. Bonnie G then uses both hands. No dice. The machine's not budging.\n",
      "\n",
      "BONNIE G: \"Hey, wait! It's stuck! It's not working!\"\n",
      "\n",
      "MR K: \"Dude, Bonnie. I just did it and it worked. You're weak.\"\n",
      "\n",
      "BONNIE G: \"No, wait! You try! It's stuck!\"\n",
      "\n",
      "ACTION: MR K tries to squeeze the grip machine, but it doesn't squeeze any further. \n",
      "\n",
      "MR K: \"OK, it's stuck. Or maybe that's really the most you could do.\"\n",
      "\n",
      "BONNIE G: \"I'm way stronger than this. Whatever. Let's go.\"\n",
      "\n",
      "ACTION: Bonnie G and Mr K walk outside to the parking lot. Mr K takes in his experience. \n",
      "\n",
      "BONNIE G: \"Well, Mr K? What ya think of Sketchy Tempe?\"\n",
      "\n",
      "MR K: \"Man, what a night. This kicked ass\"\n",
      "\n",
      "BONNIE G: \"There's some more sinister and scary places I could have taken you to, but you really didn't need to go there.\"\n",
      "\n",
      "MR K: \"That's cool. I mean, this rocked. Check it...from Fresh & Easy, to Plaid, to Gold Bar, to experiencing Monkey Pants, then exploring Time Out Lounge, and then forgetting to buy Corn Nuts & cringing to bad Journey songs at Palo Verde Lounge. All within the span of 6 hours. This was a Sunday for the ages. And this here dark underbelly of Tempe? Not bad. Not bad at all. Although some parts were better than others. But now we gotta hit Sketchy Phoenix. Dive bars abound. Abound, I tell you.\"\n",
      "\n",
      "BONNIE G: \"You're on.\"\n",
      "\n",
      "MR K: \"Awesome.\"\n",
      "\n",
      "BONNIE G: \"Bye Mr K!\"\n",
      "\n",
      "MR K: \"Later Bonnie G. Thanks again.\"\n",
      "\n",
      "ACTION: MR K walks to his car. He finds himself whistling to \"Wheel In The Sky Keeps On Turnin'.\"\n",
      "\n",
      "MR K: \"GAH!\"\n",
      "\n",
      "(end of Part Three)\n"
     ]
    }
   ],
   "source": [
    "print df.query('review_id == \"M7vzqBAS0RoF3MdO-5SvHA\"')['text'].values[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "I'm not a fan of strange phenomenons. Whether its $5 cupcakes, 25 Random Things About Me, or re-imagined frozen yogurt.\n",
      "\n",
      "BACKGROUND\n",
      "----------------------\n",
      "I bought way too many vegetables at the Wednesday night farmers market downtown. And I was determined to eat them all. Did you know that boiled beets suck? They make your whole damn place smell like feet. No one told me this. \n",
      "\n",
      "Anyway, 2 full days of eating nothing but healthy produce. Which is supposed to make one feel alive and organic and fibrous and whatever else. But it made me downright irate. And glum. And after my 10th bowl of salad, I had more than enough of a good thing. \n",
      "\n",
      "I needed junk. Immediately.\n",
      "\n",
      "Enter MoJo at the Biltmore, as I recall a friend who recently said, \"Have you been to MoJo? It's awesome!!\" And it happened to be open late on Friday night (till 11pm) and was just minutes away. Perfect.\n",
      "\n",
      "But I obviously hadn't been to the Biltmore in a while because I had no idea where Mojo was. I parked by the Borders and walked into the inner sanctum where the lawn is. There I found dozens of teenagers wandering around with blankets, throwing Cap'N Crunch at each other. Eh? I soon learned that the mall was playing \"The Breakfast Club\" just hours before and these kids were still not bored with their Stridex-ridden lives. \n",
      "\n",
      "Cap N' Crunch. In honor of whatshername's sandwich with the pixie stix. I all of a sudden felt ancient, as the Breakfast Club apparently had transformed into the Rocky Horror for this generation.\n",
      "\n",
      "I then see the blanketed Arcadia High scruffians slowly migrating somewhere, shuffling along with their flip-flops and bunny slippers, and I hear one say, \"ohmygodtheoriginaltartissogood! Let's get some more!\"\n",
      "\n",
      "Bingo.\n",
      "\n",
      "I walk behind them and they lead me back towards the south parking lot, where Mojo is situated right next to....Haagen Dazs? How does that make any sense? Anyway. \n",
      "\n",
      "SETTING\n",
      "-------------\n",
      "Brightly colored inside with some seating area in the front, as well as a few outdoor tables. Two teenage girls at the register. A messy section of toppings. And the fro-yo dispensers in the back. \n",
      "\n",
      "Crowd: It's 10:30 pm on a Friday night. One middle aged couple - woman with bleachy hair, sun-damaged skin and fat gold necklace, man in shorts, braided belt, and loafers. And then teens. Lots and lots of teens. Talking about absolutely nothing of consequence. Mojo has become their new domain.\n",
      "\n",
      "PROCESS\n",
      "---------------\n",
      "I go to the back, and all I see available is a stack of big, sundae-sized paper cups. Nothing in the size for. say, a single-serving. I take it Mojo must be pretty cheap since there's no prices posted, no scale to weigh as you go, and only these large cups available.\n",
      "\n",
      "A kid in front of me is going to town with the yogurt - pouring in at least 5 flavors and piling high. And so, when in Rome, make a Neapolitan - chocolate, vanilla, and strawberry. Just like those sandwiches from my childhood ice cream truck.\n",
      "\n",
      "But the chocolate isn't coming out like soft serve. It's glopping haphazardly into my cup, as if it was a cross between Hershey's syrup and molasses. The vanilla's doing the same thing, but the strawberry's fine. \n",
      "\n",
      "I tell one of the girls working there about the trouble with the chocolate and vanilla. She...shrugs. It's the end of the night, apparently. They'll fix it tomorrow. \n",
      "\n",
      "I proceed to then add some toppings - some crumbled thin mints basically and then I hand my cup over to the girl to pay.\n",
      "\n",
      "JAW DROPS\n",
      "------------------\n",
      "$8.17. \n",
      "\n",
      "QUE?!\n",
      "\n",
      "$8.17 for frozen yogurt?!? Are you kidding me?\n",
      "\n",
      "What's worse is I can't just say, \"Ha ha! Just kidding!\" and walk out of there. \n",
      "\n",
      "I am now an owner of an $8.17 cup of half-gloppy Mojo yogurt.\n",
      "\n",
      "BUT THE TASTE! IT'S AWESOME! BECAUSE FRO-YO IS AWESOME!\n",
      "---------------------------------------------------------------------------------------------\n",
      "No.\n",
      "\n",
      "NO?\n",
      "------\n",
      "I was bored after my 5th bite. Dessert wasn't meant to taste boring, and this stuff was boring. \n",
      "\n",
      "Maybe I don't like frozen yogurt. Or maybe theirs is just bad. I walked across the street, through the trip-tastic underpass, and sat on a planter of the Esplanade, trying to eat through the cup. In depressed silence.\n",
      "\n",
      "I gave up halfway through and threw the rest away.\n",
      "\n",
      "YOU\"RE PISSED, AREN'T YOU\n",
      "-------------------------------------------\n",
      "Of course, I am. I just spent a meal's worth of money on frozen yogurt. With no way of me checking how much it was going to cost, coming in as a first-timer. But Mojo's business process is not alone in angering me, for my quiet rage shakes its fist at the salad bars of AJ's and Whole Foods, too. *grits teeth*\n",
      "\n",
      "WHAT WOULD YOU CHANGE?\n",
      "-------------------------------------------\n",
      "Small cups would be nice. Plus a small scale and a sign simply saying, \"Mojo costs $.xx/ounce. Be wise.\"\n",
      "\n",
      "Also? Non gloppy chocolate and non-shruggy employees who charge me 8 McDonald's cheeseburgers for a cup of dubiousness.\n",
      "\n",
      "IN SUMMATION\n",
      "---------------------\n",
      "2 stars, for making me angry at dessert.\n",
      "\n",
      "Bah.\n"
     ]
    }
   ],
   "source": [
    "print df.query('review_id == \"XGKwd36kg4_mt5XoJI0L_Q\"')['text'].values[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1689a3ed0>"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAFdCAYAAACgiL63AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x168bcb510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# some evidence longer = worse review?  Doesn't match conclusions on the following more detailed data though.\n",
    "sns.lmplot(data=df, x='text_length', y='stars')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# a fast way to save a big df --to msgpack\n",
    "df.to_msgpack(\"data/yelp_df_tokens_length.msg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load it\n",
    "df = pd.read_msgpack(\"data/yelp_df_tokens_length.msg\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Let's do some simple dictionary-based sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Simple sentiment analysis\n",
    "from collections import defaultdict\n",
    "sentiment = defaultdict(int)\n",
    "with open('data/AFINN-111.txt') as handle:\n",
    "    for line in handle.readlines():\n",
    "        word = line.split('\\t')[0]\n",
    "        polarity = line.split('\\t')[1]\n",
    "        sentiment[word] = int(polarity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sentiment['pho']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sentiment['good']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-3"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sentiment['awful']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_net_sentiment(string):\n",
    "    tokens = string.split(' ')\n",
    "    total = 0\n",
    "    for word in tokens:\n",
    "        total += sentiment[word]\n",
    "    return total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_net_sentiment(\"the pizza is terrible but the cocktails are fair\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for row,vals in df.iterrows():\n",
    "    df.set_value(row, 'net_sentiment', get_net_sentiment(vals['tokens']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "      <th>tokens</th>\n",
       "      <th>text_length</th>\n",
       "      <th>net_sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9yKzy9PApeiPPOUJEtnvkg</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>fWKvX83p0-ka4JS3dc6E5A</td>\n",
       "      <td>5</td>\n",
       "      <td>My wife took me here on my birthday for breakf...</td>\n",
       "      <td>rLtl8ZkDX5vH5nAx9C3q5Q</td>\n",
       "      <td>wife took birthday breakfast excellent weather...</td>\n",
       "      <td>542</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ZRJwVLyzEJq1VAihDhYiow</td>\n",
       "      <td>2011-07-27</td>\n",
       "      <td>IjZ33sJrzXqU-0X6U8NwyA</td>\n",
       "      <td>5</td>\n",
       "      <td>I have no idea why some people give bad review...</td>\n",
       "      <td>0a2KyEL0d3Yb1V6aivbIuQ</td>\n",
       "      <td>idea people give bad reviews place goes show p...</td>\n",
       "      <td>703</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6oRAC4uyJCsJl1X0WZpVSA</td>\n",
       "      <td>2012-06-14</td>\n",
       "      <td>IESLBzqUCLdSzSqm0eCSxQ</td>\n",
       "      <td>4</td>\n",
       "      <td>love the gyro plate. Rice is so good and I als...</td>\n",
       "      <td>0hT2KtfLiobPvh6cDC8JQg</td>\n",
       "      <td>love gyro plate rice good also dig candy selec...</td>\n",
       "      <td>50</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>_1QQZuf4zZOyFCvXc0o6Vg</td>\n",
       "      <td>2010-05-27</td>\n",
       "      <td>G-WvGaISbqqaMHlNnByodA</td>\n",
       "      <td>5</td>\n",
       "      <td>Rosie, Dakota, and I LOVE Chaparral Dog Park!!...</td>\n",
       "      <td>uZetl9T0NcROGOyFfughhg</td>\n",
       "      <td>rosie dakota love chaparral dog park convenien...</td>\n",
       "      <td>280</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6ozycU1RpktNG2-1BroVtw</td>\n",
       "      <td>2012-01-05</td>\n",
       "      <td>1uJFq2r5QfJG_6ExMRCaGw</td>\n",
       "      <td>5</td>\n",
       "      <td>General Manager Scott Petello is a good egg!!!...</td>\n",
       "      <td>vYmM4KTsC8ZfQBg-j5MWkw</td>\n",
       "      <td>general manager scott petello good egg detail ...</td>\n",
       "      <td>248</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id        date               review_id  stars  \\\n",
       "0  9yKzy9PApeiPPOUJEtnvkg  2011-01-26  fWKvX83p0-ka4JS3dc6E5A      5   \n",
       "1  ZRJwVLyzEJq1VAihDhYiow  2011-07-27  IjZ33sJrzXqU-0X6U8NwyA      5   \n",
       "2  6oRAC4uyJCsJl1X0WZpVSA  2012-06-14  IESLBzqUCLdSzSqm0eCSxQ      4   \n",
       "3  _1QQZuf4zZOyFCvXc0o6Vg  2010-05-27  G-WvGaISbqqaMHlNnByodA      5   \n",
       "4  6ozycU1RpktNG2-1BroVtw  2012-01-05  1uJFq2r5QfJG_6ExMRCaGw      5   \n",
       "\n",
       "                                                text                 user_id  \\\n",
       "0  My wife took me here on my birthday for breakf...  rLtl8ZkDX5vH5nAx9C3q5Q   \n",
       "1  I have no idea why some people give bad review...  0a2KyEL0d3Yb1V6aivbIuQ   \n",
       "2  love the gyro plate. Rice is so good and I als...  0hT2KtfLiobPvh6cDC8JQg   \n",
       "3  Rosie, Dakota, and I LOVE Chaparral Dog Park!!...  uZetl9T0NcROGOyFfughhg   \n",
       "4  General Manager Scott Petello is a good egg!!!...  vYmM4KTsC8ZfQBg-j5MWkw   \n",
       "\n",
       "                                              tokens  text_length  \\\n",
       "0  wife took birthday breakfast excellent weather...          542   \n",
       "1  idea people give bad reviews place goes show p...          703   \n",
       "2  love gyro plate rice good also dig candy selec...           50   \n",
       "3  rosie dakota love chaparral dog park convenien...          280   \n",
       "4  general manager scott petello good egg detail ...          248   \n",
       "\n",
       "   net_sentiment  \n",
       "0           35.0  \n",
       "1           22.0  \n",
       "2            6.0  \n",
       "3           10.0  \n",
       "4           13.0  "
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x17d9cf050>"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAFdCAYAAACgiL63AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsvXmUXdV95/s5051vDZKqSlJVaUQgYSOEAYEtQ2OM03Kb\ndtvBGIMbxwnPbvqt5LXNStvujhOcbvdyOvF67k7beas7Xu04JM+Gjh8xHuJ0BMYIAZIioASoNICG\nGlTzeOcz7ffHGereqluDpBpUVfuzlrHurTPsc+693/Pbv/0bFCGEQCKRSCSLgrrUA5BIJJLVhBRd\niUQiWUSk6EokEskiIkVXIpFIFhEpuhKJRLKISNGVSCSSRURf7BPats2///f/nu7ubizL4tFHH+Wa\na67hK1/5CqqqsmPHDh5//HEAnnrqKZ588kkMw+DRRx/lrrvuWuzhSiQSybyy6KL7zDPPUF9fzx//\n8R8zPj7Ov/gX/4KdO3fy2GOPccstt/D4449z4MAB9uzZwxNPPMHTTz9NsVjkwQcfZN++fRiGsdhD\nlkgkknlj0UX3wx/+MPv37wfAcRw0TePEiRPccsstANx5550cOnQIVVW5+eab0XWdVCrFli1bOHXq\nFO9+97sXe8gSiUQybyy6Tzcej5NIJMhms/ybf/Nv+OIXv0h5UlwymSSbzZLL5Uin0+H7iUSCTCaz\n2MOVSCSSeWVJFtJ6enr4jd/4DT7+8Y/zkY98BFWdGEYul6OmpoZUKkU2m53y/mzIrGaJRHI1s+ju\nhcHBQR555BH+4A/+gNtvvx2AXbt2cfToUW699VZeeOEFbr/9dm644Qa+9a1vYZompVKJs2fPsmPH\njlmPrygKAwNLYxE3NKTlueW55blX8Lnng0UX3f/+3/874+Pj/Nmf/Rnf+c53UBSF3/u93+PrX/86\nlmWxfft29u/fj6IoPPzwwzz00EMIIXjssceIRCKLPVyJRCKZV5SVWGVstT6F5bnlueW5F/bc84FM\njpBIJJJFRIquRCKRLCJSdCUSiWQRkaIrkUgki4gUXYlEIllEpOhKJBLJIiJFVyKRSBYRKboSiUSy\niEjRlUgkkkVEiq5EIpEsIlJ0JRKJZBGRoiuRSCSLiBRdiUQiWUSk6EokEskiIkVXIpFIFhEpuhKJ\nRLKISNGVSCSSRUSKrkQikSwiUnQlEolkEZGiK5FIJIuIFF2JRCJZRKToSiQSySIiRVcikUgWESm6\nEolEsohI0ZVIJJJFRIquRCKRLCJSdCUSiWQRkaIrkUgki4gUXYlEIllEpOhKJBLJIrJkotvW1sbD\nDz8MQHt7O3feeSef+cxn+MxnPsPf/d3fAfDUU09x33338alPfYrnn39+qYYqkUgk84a+FCf97ne/\ny49//GOSySQAb775Jr/1W7/FZz/72XCbwcFBnnjiCZ5++mmKxSIPPvgg+/btwzCMpRiyRCKRzAtL\nYulu3ryZ73znO+Hrt956i+eff55/+S//JV/96lfJ5XIcP36cm2++GV3XSaVSbNmyhVOnTi3FcCUS\niWTeWBLR/dCHPoSmaeHrG2+8kS996Uv81V/9Fa2trXz7298mm82STqfDbRKJBJlMZimGK5FIJPPG\nkrgXJnPPPfeEAnvPPffw9a9/nb1795LNZsNtcrkcNTU1czpeQ0N69o0WCHlueW557pV77vngqhDd\nRx55hN///d/nhhtu4OWXX+Zd73oXN9xwA9/61rcwTZNSqcTZs2fZsWPHnI43MLA0FnFDQ1qeW55b\nnnsFn3s+uCpE92tf+xr/8T/+RwzDoKGhgf/wH/4DyWSShx9+mIceegghBI899hiRSGSphyqRSCRX\nhCKEEEs9iPlmtT6F5bnlueW5F/bc88FVYelKJJLFxxWCQ8d76BrI0dKQZN/uDaiKstTDWvFI0ZVI\nVimHjvfw3GvdAJzuGgXgjhs3LuWQVgUyDVgiWaV0DeRmfC1ZGKToSiSrlJaG5IyvJQuDdC9IJMsQ\n1xUcbLt4Rf7Yfbs3AFQcQ7LwSNGVSJYhzx7tuGJ/rKoo0oe7BEj3gkSyDDnfO17xWvpjlw9SdCWS\nZciW9ZUp8dIfu3yQ7gWJZBnywVs3kckUpT92GSJFV7IiWemB/6q6PPyxK/1zuByk6EpWJDLw/+pA\nfg5TkT5dyYpEBv5fHcjPYSpSdCUrEhn4f3UgP4epSPeCZEUiA/+vDuTnMBUpupIViQz8vzqQn8NU\npOhKVgRylVyyXJCiK1kRyFXySuRD6OpFiq5kRSBXySuZ7SEkRXnpkKIrWRG0NCRDcQler2ZmewjJ\nmcHSIUVXsiKQq+SVzPYQkjODpUOKrmRFIFfJK5ntIXSpM4Nyd8SubWvZvbVeuiMuEym6EskKZLaH\n0KXODMrdEed6x8lkivIhd5lI0ZWsWlbzYtKlzgykO2L+kKIrWbWspsWkK33AyIXK+UOKrmTVMl/W\n23KwmK/0AVPujgh8upLLQ4quZNUyX9bbcrCYr/QBU+6OaGhIMzCQmbexrTak6EpWLfMVZraY/s7A\nqh7KmaxNRuZsVUv3wNWDFF3JssUVgn84fIH2s0OXNa2/0jAz1xW80HaRN88NMZIpkYzppBKRBRW0\nwKo2dBXLdoHprepyt0dzQ5IP7NlI92BexjEvMVJ0JcuWQ8d7OPhGD5btTpnWL4af9dmjHfzk0HnG\ncyVcAbmizc5N9fMiaNON/1Ks6kCghRC8enqApjVx9u5qQgjBk8++fdX6n1c6Sya6bW1tfPOb3+SJ\nJ56go6ODr3zlK6iqyo4dO3j88ccBeOqpp3jyyScxDINHH32Uu+66a6mGK7kKmUmALsfPeqlCfb53\nHNN2UBQFTQFDV0nEjIp9ZjrmdH9zheB7P2vn+NkhIrpWMf5LcRME9yNXsMnkTUzboW+4AEAqYVy1\n/ueVzpKI7ne/+11+/OMfk0x6X5hvfOMbPPbYY9xyyy08/vjjHDhwgD179vDEE0/w9NNPUywWefDB\nB9m3bx+GYSzFkCWzsBQr+C0NSc71jle8DrgcP+ulCvWW9TVEdJV80UYAhitoXpcAJu7H4fY++oYL\nVUVuuvMdOt7D8bNDlEyHkulUjD+wost9utMRCLRpe8eI6Fr4bzDmfF+qsRwiNq5WlkR0N2/ezHe+\n8x2+9KUvAfDWW29xyy23AHDnnXdy6NAhVFXl5ptvRtd1UqkUW7Zs4dSpU7z73e9eiiFLZmEpVvD3\n7d5AOh2r8OkGXM7C0aUK9Qdv3cSxE70cOz0AQNTQwBee4H4MjxdD4UwljIpjTne+roEcEV0L9zNt\nJxx/4IeeSwRBcD/KhT+br9ym2n2Zi6Auh4iNq5UlEd0PfehDdHd3h6+FEOG/k8kk2WyWXC5HOp0O\n308kEmQyMkzlamUpMpZUReFDt21mz7Y1U/52OZEJlyrUqqqQiBk0rUmE73WXCScQiqdnYRoVx5zu\nfC0NSU51jgCe4O7etnbG8U8nkoFA79u9oWJBDSFmXFCbi6BeyuddbXzBeVajpXxVLKSp6kR/zFwu\nR01NDalUimw2O+X9udDQkJ59owVitZ5717a1FVP9XdvWsnZtimePdnC+d5wt62v44K2bUNX5/2FN\nd92/fs/cvi8BH7v7WtLp2KzjdV3hXddL53EBXVNQfMHYtW0tDQ1pdm1by9meMVQVNE2hLhXlo3ds\n5569E8ec7nxzHUdw3f9w+AIH3+gBvLoI6XSMD966qeLef+zua+d87wezJoWSTclyiBoag1kzPJfr\nehEjA+MFCiWbdDKCUnbd1QjGJ4C2dwZ57Z1B1tbEOd87juKPOZWKoSjM6buylN/z+eCqEN3rr7+e\no0ePcuutt/LCCy9w++23c8MNN/Ctb30L0zQplUqcPXuWHTt2zOl4SxW4vZRB40t97t1b68lkiqHl\nsntrPX/73OnQYmo7PbAgRVLKr3s+/Ix7tq0JLeehoWzVbQ62XawI22pZlyQRM8LrHhjIsHtrPcdO\nJOgfKZCIGqiqQjZbnHLM6c432zjKr7v97FAYPha8zmSKl33vh0fyjGZKABSKNsMj+fBcB9sucvCN\nHkzLwXEEUV3ltl1N4XVXIxhfNm+RyXuC/g5jRHSNVMLzLT979AK5oj3reJf6ez4fXBWi++Uvf5nf\n//3fx7Istm/fzv79+1EUhYcffpiHHnoIIQSPPfYYkUhkqYcqmYZqMa+L7XK4FD/jlQj05OtIxAwe\nvKfSIFAVz/WwpiY27X7zReCmEEKQK9h0D2bpHswihAgt8Es5dzyqk05EMG2HiK4Rj07IRHAcRVFI\nJQya16W448aNuEJwsO1i1ftZbUEPCF0u1VjJBXWWTHSbm5v54Q9/CMCWLVt44oknpmxz//33c//9\n9y/20CTzxEJkQU0Wy4/dfW34t+lEvprATifQcxHjatdVbb/prv9SBN92Xb7/85N09mdpbUzxG/9s\nJ3qZOw4qF8xyBZtc0SabtwBCS/JSFsxaG1Oc6R4jm/eEsVCyeaHtIt0DOfJFCwEVAn+w7SJCCH75\n+kWyeYtXTvRyqmMERVHo7M/S0pDkzj0bOHC0C8t2EUKQShjsbKgLZwjB/pPv1UrkqrB0JSuThejm\nMFks0+lYOA2fTuSqCex0Aj2X3mJCCJIxnYihs+eatbz3hvVV42qnu/5Lsci///OTHD3ZD0DvsBd6\n8Mi91+O6Uy3LroFcOEVPJQySMZ3mdakp9362cLZ9uzdwunM0vJ72CyOc7BgllTAQQrCjtZ6ewWwo\n8M+91k0ipjM4WqBgOijAkZP9CAGaqtA7nKd/tICuKWiqQqZgsaYmWvEAcX2rfDV0/pCiK1kwFqKb\nw2SxPN87HorudCJXTWCnE+jZrOVXTvTS0ZdFVRXiUZ0929fw8hu9VeNqp7v+S3G7dPZnq75+9mjH\nFOGefE237Wqqev7Zwtkmu0aGx4v+ngaKopD03QqBwAOMZkoUTAfhCgSAQujaAOgbLmDoKkXTQQhB\n10COl473cOee5vCc5TONlZwxJ0VXsmDMlHE11+n15G2bJwlLLm/xgwNnwuNU++FO3idftLxp77ok\n8ZhOa0MqFOhq/tHy6XP3QBbbEagKmJbLkZP9NK9LTRtXW+068kUL13XJF71QsnzRwhWi6j1obUyF\nFm7wGryHDUxM85891sXdN7dUra8w+R4Gwj2XcDYhBK4rcFxBJm+CgI6eDMmoVjHOulSU8Zy3SKYo\nCooCZZGgNK2J0zWQw3W9Ny3bu3eB6AashvhfKbqSBWOmjKu5/rAmb/uBPRu5+6bmULzO94xhO2LG\n40/ep6M/EwpeEAMbCF41/+hzr3WTjOlk8xaO44mGK8BxvYiB6eJqJwtt1+CERRs1NEazJhFdo2sw\nx6HjPVPugSsE21tqOdszTsl0uK61lu0ttfzgwBlcwHVdhsdLFEo2RVPnqefOEIvq7NpUz3tvWB9e\n0+T70bLO86EKIVBVhVTc4K49GxEQPsDee8P6svtgAQq5ouW7B0yGxx1aG1IVPtl8yfMlm7bDDVvX\nVPh0t7fW8v89f9azgPHcDtVYDR0qpOhKFoyZMq5m2i7AFYLD7X0MjxeJ6BrJuE73YD6MFPjBgTP+\nFNYTws5+zyp99lgXuaIdLiJN3idfdDyrDTh2eoCRbInbdjWF4nvHjRsr/KMBpu2gKuAIL/HM0DX2\n+vsF11Fuub/wejc/eekCpu1g2S6JqE466UXgmJY7a2TDoeM9/Or1i0QMjYihoaoqv/IXm3RdJWpo\nWLaLoigUSjaugKLlhj7gR+69vuqx4zGdlnVJXj0zGI7l7a6x8KEQWLiKovgLcgrJuB5GHyh4roPy\nqI1qPtlA9IMQu1hEp2g66JqCrqnh36pFOgSsxAU1KbqSy2Iu3WFnyriayw/r0PEe+obz5Is2WWEx\nllNoWZcMp+KTay8USp5VmivaoagCoYvgvTesJ1+0GM+buK5n6SmuZ40FAhtYm5PHuHdXE3XJCMdO\nD6C4nnuhNmXw7LEuznSO8vCHr0O80cvh9j4Ot/exd1cTh9v7w3E4riCPN/U2bYem+jiWMzH/zhc9\nN0nzugQoCt0DOboHp/pzFd9CVPDEMp2IMJr1qpwF7zuu4Pg7Q7zwejcCePPcEMPjJT+JA1rWJTnb\nM45puyhAJm/S3jFCOjERknnkZD+5os3QeJGC6VA0bWIRzT9D5ec2m7soEP3gISgQKChkCxbPHDrP\n4fa+8KG3EIuvVxtSdCWXxVy6w073A5rthxX8iJ891kXRdDyBBBxHcLJzNJyKT6690DngiVQwbfaE\nQidbsHjutW5Od47SNZgjoqsUTAdV9VJlg7jRcotwujGe6hzDtL0Fs8HRIkJA92COM12jOC6hyPYN\nF4gYE6FdmqoQNbQw9rVkOWxqTJOIGeQKJicuDFMoOdiOS0RXWVcXJ1eYiEQAz59b7qJobUxxsmOk\n4t45vvqWLIcfv3iOXNHGdlyEANOGRFTnZMcIYzlzYtELv25EOQIGRwvkSzZCQNG02X3NOq5rqWU4\nb1UU25nNXdTckOTV0wPhtTetSXgPUj9ZwrSdiofeSvPhTkaKruSymIuLYLrV+9miGoIfca5oUzAd\nzw8ovFoHpuVwuL2vIk43iF442HaR104Pki1YZUfzFppM23MppBMR1tXFyeat0OJKxr2fwUxTWVcI\njrT3hcJRNB1P4HzVGhwrEikTrlzRwhUaiuL5f1XF+5+uTWRhxWM6LQ1JfvxiP2NZE+EfznEdcgWb\nZFwnGTdC2/Ka5hp2tNbR7c8u3r2ljv/yVJsnWJZDmeGMEIJMwcJ2BIEDRsG/h7aLrqkI4XrV0XSV\nX7u1NazX29KQ5HTnKO0dI+FimOPC+Z5xUjFjysxm1u9C+YoaUJ+MkC/aU5IlVqL/thpSdCWXxaX6\n3uYSsVBu4eaKnuhkCyYly1+8cgVF0+FCb4ZcweLV0wO89s4gN21fF05ND5cJoxCCXNEOV8xdV6Bg\nkYzrFE2bqKFRXxNh8/o0LQ0pTnWM8OMXzxGNaGxdn6ajP8NIxsS0XX728nks2yVftMkJT9RF2ZQ+\n8FGCZ206vgvCdgSKAqYjsB0TUBBCkE5GKBQn3CFuuS4JP1ur4K3ym5ZLKmHwfFsPd9/UzIP37AjT\nYW/b1URnf9abDbgiHJSiKN71Tri8QXjH01SFWERF8wW4ed1EuFzw2XT2Z1FRcJgY2MBIgVetAY6c\n7OPdW9bwmx/ZFbp5ZkoCee5VzwoOfNiJmMHdNzXzSnsfHb0ZTNshm/cs4tWAFF3JZXGp3WHnknQQ\nJBi4rsC0PD9iIFpiwqjEsl0Gx4pYtsu57nE6e7OhX3DvzsZwqiqEIBrRyBYsIrpGIqaRTkQYyZQo\nmt7iVrZg0VSf4O2uMQ6394eujP7hAgIRiuHAaBFdC4qPe++pgRWrKugaJOMGjisomQ4RQ8W0PEvO\ncXz3iABFEZQsh39+4waOnBxgeLyIEBN1EwAM3YsoMC3Xm35b3t8nl4YMPocgkcFwBZbtYugqjiuI\nR72FNlcQuhhsx0VTNdbWxOkfLVCTiDCaNfnpSxcqkiRaG1MYhopjOhMnUzz3iaIoHD87VOHmCb4L\n1ZJAsgWLsZxJrmiR9KMd7rhxI0II+ocLEzV+J1nEKxUpupLLYnJ32L7+8Wlz76H6FHRySFWQYAB4\nq/WK4gmI43pTecVb9XZ9cQEvbCtb8BbH+oYL3Pu+zWF4WEtDEgH80hd713UxLc8X6wrhLYipKp39\nWQSEggvgTBIARZnwlwYI323guoKSBUWrRDyilS04uRXH9HbyrN9Db/bSNZDDst2K4yqKZxG+a8sa\nznSPQR5My6waSxt8Dr/5kV0cOt5D50DWK1CTKdI/UiQR08gXHZrWxBnJlBga85IcLNtlJFuqmvwA\nXpjYxrVJWhqSE7UWEFiOd98FYNmez1gIwfun8cN29mfJ5i3fLywmrtOP1T5ysj+clQSRKasBKbqS\nOTGbe6CaJVtexzVftCqO19yQrLRsbZfy0E1VVdi9bS0nO0Z964owttTQVXRNpeRbw67r/W0kW+J/\nH+1k6/o0nf1ZznaPki85jOdN0gmDZNygoy+L7fpWn4CI6ll1fSN5ZrOzElGNbHHC8vNqEHj/DgQl\nX/Ks7GhEI2KoWLaC40783dvH5UJvJszYCiz54HYq/pT9TPdY6G9urI+xJh2jc8CLxCivOaEqSsW9\nHsmWSMZ1vyiNSvM6L6EiEF2oXDjzfKqCgZG8l1Um4OSFEaKGyk3XNpKM6pzvHedcTya8R66A4UyJ\nZw6dR5mUTRZ8R/JFb6EsEGpD8az17oGcH5lSqMjim85FNZdImeWEFF1JyEzCOllUgzjOroEcO7eu\n5ZUTvWE8bTANfrHtYhinamgquzbXVwTTHz87RD7wufppo/GIhqoq3LB1DdtbahnOFLEdF9OyCWah\nJcsBBImozmjWCcVPEYK+4Tz9I14fsEDoFKBoOmTyVsXiF3gVta5priEa0Th7cRx3GuXVVMVLkCgT\n3WoI4QlvLKJT8ssfTj6maQsiuupN+ZkQ7uD/B0YLnOkaq8guK7fYz3SNVdScmPz5VCt207wuQZ8/\nlY/oGr92SwuqqoaFzQ+2dTM8XqoQ1YLpcr5nnP17N3G4va+i2UBArmhNW7ciEdO98peO6/mTHUE2\nb4UWdDA+0/as8enCw+YSKbOckKIrCZnJ79o5kA2zjSK6xuGT/eSD+qfvDJIrWFi2W2G1HG7vC0Oo\nisLmbM8479q6htOdo3T0ZaZMveMRjTU1MaKGSvuFEdreGSIW0SiUbFRVwXW9qbjno3Qomm6FG1D4\n//EeCJXvO/4iXPk5vcgCwU9f7sDQ/ULk0/gVFUUhW3QoX5easg0TUQIly/ZdItUP6QpRIbjl2I73\nQLq2ta4iqaOcoOZE+eJjtmCFccCpuMGO5lpaG1O894b1vPRGL01r4gDs3dnI+2/cWGEtPnesa0rq\nLkDJdOgcyE51k/g4rpi2bkUQLaGrCq7woiQCcT10vIfTXaO+8BrctqtpWut1pWWpSdFdpVSzasu/\nzNm8xbPHugDPTVAo2oxmPUsoj03EUMMQqZLl+BlaXkETb8rvIvxVfOELzEimxGunBxjJlEKBNHQV\nWwjiUZ2IrjI4WqDoC7eAyiIqk5jsYwV/qk55CH/ZNVfxyWbyFmBV2boS03aJRXVP/IWYIk4KkIzr\nYSRCvuglQghBVaG2y+K7ggW5yWMtL7aTL1r+TMK7xy8dv8jQcJ5rmmt4vq2HXNFmLGdOHM8VXBzK\n0dqY4qXjPRVlEwO3xsG2i3T0ZbjQl2FovDhlDIoCOzfVUSjamJaDqigVvm5FgWTMCF0ek2tc7N3V\nRH0qGlYrS8b1UFwvJQlipWWpSdFdpVSzaoMvdxC0DoTbDGd8f6AQuMBYrkQ8YpBKGEQNjZzthpZk\n0XT4m1+d9WNBPWelcAWOIxjNmhU/bkNTaW1McXEwy2jWniJOk4VyNurSUaK6RtG0yRUtLHuyZXtJ\nh6ugULRYWxMlX/IeLIHoq4pCQ12MQskJxdgVoKt+6JbAC+eahiACAl/MVdXPHCsrTdk5kCWia4zl\nSr7VbvPyiT7ePDdMXTrqC76F64qwo0WQaZeMVf7Mu3yf6nOvdXsJEEU7bI2jKp4rJRU32LW5ns9+\nZBdPPfs26USEfMlGcby046iheqUbNYUzXWOhO6R8EXPf7g28v8zXXC6ul1KB7lIjZa52pOiuUqpN\n2R744DUAoYUbLOJ0DeRQFK8Wqu14wlAyHeIRr2brR+/YzoEj53m7awzXt7I8S8+z8FT81X+/YlU5\nuZJNZ38G056fcKGS6fDRfVvpHsjx6ukBL0U2iNO9wlM4rlfUJWroWLaDrqkYukpE19i2oYajpwYq\ntnddqEt7qbUjZf7SyShAOm5QkzTCAjam5XKqY4T33rDeE08/wUP4obiOP1XIFix0TfXr5074SGEi\n6QC8mUuu6BXsMS2vehrCe0AGkRsKoGkqG9YmwmI2L7/RS0tDktfODOK6Xvp1NKKF1c7KE1GCimuT\n1wNmEte5xG9PjpRZqnY984UU3VXK5Clb87pE+OVvbUzROZANp6HNDUmy+RK244bC5bietdtYFwOE\nX85PqVhsCabWE/7L6rIzX4ILXv2Fn750jryfzTafoZ+ugFzBJl+0w2gDTVXYuamOaESb4tLQVMIS\njjMOQ/EW+qKGGhauMW2Xl0/0ca434y8Ylqq6NRACQ/dmEjs31XFNSy1HT/bT55eDHB4voqtxiqZN\nyRfYvuEC4znLv6aJqAqAiK4yNFakdyhPOhHhdOcILQ0pP3LEs2zjUZ2N65K0NqTCmVAmZzI4WuDs\nRa8WxqmOEX7r3utnjTKYskDr3Y4V3SVYiu4qZbJPTTDhSgCmlO073TV1Zd92BGd7xnnywGmiupfd\nNBlVVar6XhcKV8DgWGnBjj852iBXtDnc3kddKlLRHBK81frxnDWrha36D6uh8VLFtkJ46cXaNH5k\n8EO3xktEIxq5os2O1joee2BPRSeLvpE8jiu8YjlBRIcyEWccpK0pikJEVxnLeQkQmbxJtmBycSiP\nAv5DV2Esa1Io2hXfoTfPDTGWM8NZxatnBrmuSrnKyUyecR1p7wuTW5a6nq5lu3QPZunoy9LZl+UL\nn755Xo4rRXeVUl6p/8W2izzz0nnyRZtkzCAZ1znbM46hqeSLFtGISq5YfbHJsl1Gs6WwzuxkLtUn\nuxyxHU8wgxoRoQ+Z6V0aXiqu5kVjlC04TibiZ5dNNpWVsn/Yjgum51o50t7HnTdurGyKmQfbsfx0\nDcBf6JsYC6STUXZtqqezP4tpu2HShuuKcPEQxXuI1qYixKN6xbT/mz/M0ueH6nn3xK1YiJ3OWp08\n45rMYkUqFEo2HX0ZOvqy/uJilp6hXIXB8Nl7d1BXW3PF55Kiu8o5dLyHZw6dZyRb8ny1lkMmr/jp\ntwpdA1mSMT1MaZ2MKwjTVKux8iXXx3c3KJMse8X/T9AxI2Z4SROqolCfjpA3HfpHilUX2hS8h5br\nulXDucojKWzfJzs8XuCbP3yNkUwJ03J9v7ygPh0FAbmSRa5gV8Qwa5rK+jWJMLPt2Ve7yBVsxvOm\n78ef+Hws2mmGAAAgAElEQVRrEhFqkhFaG1MV/tj6VJR4RPMKFPnZZ/2jBZ45dB4B3FnFWi3vNwde\ntANi4RtUjmZLobB29GXo7MvSP1qYdb9cvkRd7ZWfX4ruKqdrIBeW7wP8/Hzvhxz4+zIF2wvmd6cu\nhEl8lCDqQAlThlV/lX/D2iSW7dLamGL7xjQ/faWD8ZzJaNb0/b5TDhWaskXLIR7RcIUXgqepihdh\noPlZeaaNaQsUP6pkeLzE8LgXeWLoql9e0kunVhTFr58wMWvxFtFc6lNRfnDgDBd6x8nkTWIRnVuv\na+BkxwijfgU0AMtx2H/7FvZsX8Oh4z0cONbJ8HiJkuWwJh1lU1OKjr6s55v2W/wE1vdkDk0OZQP2\n3bhx3hpUukIwMFoos149S3Y8Z86675p0hI3rErSsS7CpKcWO1jU0b2i47LGUI0V3hVCtO+xcFiCq\nWRKCqRZq4K+dKTlgNRM8rLyMWq/CmKGrxKM6d9y4MRSdb/7wNcayZmhpVnGDhzHMMBHm5i3cefV/\ndV3BcQR5y6tpENEVIn4XCdsRCC9DBIE3c0nFjXBRtGg6Uz5Ax4WTHSMUTScMH8sVbLZtqGH92iTj\neQvXLxGZLdi0nx9m97Z6fnGkg97hfLhgOjheYkdLHYqi8HbX2Kz3rFoEzeU2M7Udl+6BHB39Ey6C\nzv5sGPM9HaripVhvXJugtSHB1g21bGupJxWPzLjflSBFd4VQrTtskPkTFEKJR3VaG1Pc9u4mnvi7\nU3T0ZYhGNOpTEfpGCuEi0Uwr/lJwZ6Y86cExHYqmw1//71P84vAFFKB/tHhJC4tBeF6wh+kKLLvy\nc7Bsgar6s5PyRBLHzxD0QwLG/YWuqQEQXgnMoEZC4Mt9/e1Bor7fGYKHgeDwiV7eOjtI/0hhYoaE\nFxrY2Z/l7vc0V6Qc793ZWPXayuPCJzfonCmUrFCy6ezPVvhguwdzs97XiK6yfk2c5nUJNjUm2bax\nls0b6irqIC8GUnQXkUvpgnupnC9rWwOVAfBBskM6EeF01yg/e/k8g2PF0KKKRzXiUa/G7DTrYZIr\nwHIEvcOz+wynY4pIVnkdLHjFoxoly6tuFjE01tZEGct5n/90muQKpliEAkIhLj+hEF7MbzY/dWFV\nUxVaG1O8v4qLoPy7H7Qk6hrIYmgKJcuLfe4cyIblIoPvruO4tL0zyPF3hlBUhe7BHD2Dsy+uJWM6\nG9bEaWnw3APXtNSzcV06TAJZSqToLiIL2V56y/oa2k5PBOcHRUU8K8bCdlw/Y8n1fHRl++ZLM0/B\nJEvHXNw5ikLY38y0HeqiBpm8STJmkCs6lKyp8cqaqkxbT6Ecx89wK4/RroamKaytjXFNs7e6P/l7\nHTSnzOYtDh73ugqvq4szPF4kaugk4zq2Izh2eoCBsQKvvNXHSKYUWq/9I9M/tOpTETaui9PakGRz\nU5prWtewpiYWulSuNqToLiILWbjjg7duIpMphtbFe29Yz1/8rJ2Lg7lwylv0p7uShaNaHYUrYS6H\n8qqoeb7Ypvo4m9fXUCx5dXV7hwtY1tTkjFTc8B/EdrVDhjj+4l06EaFQmnBBTCZYNPubX53lyMl+\nbr9+fTiTC7o69w7lMO2JIkXWYBbHhaywGM54VvTFwRzH3xmqfp0KbFibYH1djNZGz/+6vaWe5AL6\nXxcCKbqLyHwW7pjsqvjY3ddW+HAP/vWrdPZnp42flSwMqqrgLtI9D6zgoMBOIuaVkwwEt6MvW9Vl\nlIhq1CQNIrrGOxfHqx26gppUhHRMo6NghdXbJlvOws/WE8DpzjEvDE4Ibrt+PT956RwXejOUJoUW\nzpSJqKpQmzCI6CotjQl+7dbNbNlQy8YNdcs+DVgR0+VmLmOW6kOZLS98Pn265dM103bYe/16WhuS\n/PK1bjI5M4y7lSwu01WHnM+oD11T0DWFklVZ2rI84WJyR4oAVfHaCkV0jbU1Bme6MjOOS8GLwnBd\nF8dfZK1NRkhENXrLFtKqnUfX1Gkt40oEhqZgaAoRQyUejfCurWt56EPXTtlyKWsvNDSk5+U40tJd\nRC43HKYaXQO5impgL71xkZjh9QOzpHW7dExz6+fzE3FdQX1dnJ6hyvY2jitmdRe4wovTVbDIFc1Z\nxyVgSnr3eM7k7pu30n5+hDNdY1V9w0H9iCnHEy4KAkV4KcWqpoOiUJ+OEfcL9gBhQZ3KYwr+4fAF\n2s8OLeu6DFeV6P76r/86qZR3s1taWnj00Uf5yle+gqqq7Nixg8cff3yJR3j10NKQ5JUTvYD3Y7Md\nl4JcEFtyFuNx5wroH5naT+xSrGkBFM2poqhrfhnOWcIGn37h3KxnSMcNrmlO09qYJF9ysGwvuy0e\n0+keyHGhd5yi6aAoCrmixY6WurDex+SIh6A+yItv9GDZ7pLXZbgSrhrRNU3PYvvLv/zL8L1//a//\nNY899hi33HILjz/+OAcOHOCee+5ZqiFeVbz3hvUcPH6RC31ZhJjaEkaysnGmS6q4AlQ/Xdm8DL+U\ngouiCCKaCqrKlvV1/O6DN01riXqJPFl0zTtXLGKQiBlhp4zy7tARXeN01yjJmI4QInSpHW7vW5bW\nrrrUAwg4efIk+XyeRx55hM9+9rO0tbVx4sQJbrnlFgDuvPNOXn755SUe5dVB0bb54p8e5O3u8bDF\ntkRyOZTL1XQugXKEELiug4pDU53B+rVxNjWlSMQiRIwIyUSchroEa2tiPPns2xxsuximk5ezb/cG\ndm9bSzSikU5ESCUqOx0fOt4TdofO5M0wLjjju9RKpkPfcIFDx3vm5T4sJleNpRuLxXjkkUe4//77\nOX/+PJ/73Ocqqi4lk0kymeW9ajlffO27R8jN0iBRsjKotjAXCGVY/YuJbTSFS0pwmXlT4XW/wGtV\nj+ItjCXjMdako3zgPS0oeC3bcwUvJdm0HeqNCF2DXoz4q6cHONzex227miqs0vLW8dXqLHQN5Ijo\nWthzz7Qd9u5s5PV3himU7IoGqMuNq0Z0t2zZwubNm8N/19XVceLEifDvuVyOmpq5lVWbr1XGy2Gh\nzu26gmePdnC+d5yBslbakpXNFMH1C+sYmkrU0MIWSZbvb7i8WY/XWfmG7WvZujFNvuhSclxsy+V8\n7zhCCAZGCkQNlYb6BBFdYzhT5LUzg9x5UwvbW+sqLGRXeAkV4zmTbMEi32PTPZijYyDH//XATRVZ\nYb9+T/Xf9K5taznbM4amKZQsh1t2NvHxD15HTU0H2YJZsd1S/t4vh6tGdH/0ox9x+vRpHn/8cfr6\n+shms+zbt48jR46wd+9eXnjhBW6//fY5HetqDRmbKxUpkw1JHNflxwfPhdWhpDdh9aL6/eZMd8IC\nnNIReQaEX1xDCE8kVU0jamhsXp9GV1UGR0u0NqTYt3sDTz77Nrbthn3UcqrCSMb0Q8gEfUN5uvqz\n7NxUV1HAvWVdkq7BnJ9M4SVmmJbCkRO9/O1zp+e0+LV7az2ZzMYKK3hoKDslCWj31vpF+72vuJCx\nT3ziE/y7f/fveOihh1BVlT/6oz+irq6Or371q1iWxfbt29m/f/9SD3PBcYXgf/70BK+eGfQK0Lhu\nRXNFyeqmPPZ2tuxCT2BdhF/1TNN1XFfxGoWiomte3ztdU+kfKfJO9zjpRIQzXWNe12e/A3FwniA6\nwrQddNVbDjJth3hMr2hIGRRUGhydiON1haBkOnMqbA7Th1eq6vyFXS4VMjliHrkcS7daWMwPnz1T\nUVlKIpkNEVivfvM2Va2snKUoXnRCEPWgKFCXihKLaBi6ykimFJZ1jOieoHo1gb3ykK7rFWhHeIV0\ngrrK6USEj+7bUiGEQeLO0FjBixv2q5whYE1NjFTC4O6bmquK52wJRKsmOWJsbIza2sqS6d3d3TQ3\nN8/LIFYz5UVwTnWOMJopyfoIkmnxbCSB8CufK6qKongFygOhrWZGTXQQnnhtWo4XIuZXJQv+N7nX\nW4gvtFFDIRGLUp+OstdfIAvEsrM/y1vnh8kWLIQAQ1O9BT+/cE4q4SVAHD7RxyvtfYxmStSloty2\nq5H3l1UXg6XvkbZQzCi6PT09CCH4/Oc/z5//+Z+H0QSO4/C5z32OX/ziF4syyJVMZ382jDsslmyZ\nTSYJCf2vuIASiisoKFr1aM+g4M5cvkWW7bKuziAZ0xnPeg/7mWrSBn8Zy9noms5tu5pCQXzh9W6e\neek84zkzLLCkqxCN6Gxen6Y+FaVzIAt4pSEHRwvhImD/SIH+kUJYDrKc5RidMBsziu6f/umfcvjw\nYfr7+/n0pz89sZOuc9dddy302FYFhZIdpvJKwV29lPtfUVRUNWivo6DMMZxewS9CP9M2wUxdeDUV\nAOpTUUYyparxtNNh2k6FIB452V/REQNA1zVqkhE2rk3S0pBkR2sd3QM5v8PuRM0HUXa8+SwKdbUy\no+h+4xvfAOB//I//wec///lFGdBqIZiOdfRn0TXFa+gnWRWU+1/L3QMoGleSWyXC/0xFAbZuSNG0\nNknbmUFsR5BOGOSLFkfa+/C9FnPGdQUb18bDFlEjmdKU3W3HZSRT4rUzg5zqHGFTY5pEzKA+FaV3\nKE+h5ITu3oiuVcTqzrVH2kI2Blgo5uTTfeCBB/jrv/5rRkdHKxIWfvu3f3vBBrbSCXxX+aIti4iv\nYIRwff+r4gts0Lxy/lrEqAphq6Xp2N5cw+8+dBPf/H9fC+sdBN0sZqvZ4HUM9oTMdgSGrhA1NN7u\nHqfL7+JQshwMTcHCG4iqKigoYZ3doqkymjXDtvC7NtcznC1V+HQDwbwUH+5y9AHPSXS/8IUvkE6n\n2bFjx1VbjX05ERR1Hh4vUjJnrgolWR5M9r8Gi1qKok7rf71SgrKL25trGMmUvD53VdRTU73GlYff\n7KOjL1vR6RlmFlxVVUhENTY1pcnmLXJFm7p0BNsRdPZnvYgGvKLoTfWJsnq7wusKbDqeqNsuNYmJ\nYuOJmMEj915/xfdgOfqA5yS6g4ODfO9731vosawaXmy7yNnuUYqW9OEuRyb8r16HXuUy/K/zguKJ\nZyZvkvOjBcr+BBBmf/UO5Tnc3oeuKVzSc16IsFlpwbQZyRQZz5vUJAyubanl+Nlh33KGW69rYMem\nev7xZD8XB/Pe+3iiHjNUTNshm2dKnYUrYTn6gOckurt27eLkyZPs3LlzocezKjhysl8K7jIh9L9C\n6HudD//r/IzNm+73DhcquhCDt2AWj+phQXOAvuEC8ajXi2y2wjbguS3SiQi6ptLRlwlr9bq2S7Zg\n0T9apFCyw9TjY6cHOHZ6MOynpgDRiOb1YxOCmKFj2g4t6+pm9dXOlUv1AV8NzEl0z5w5w8c//nHW\nrl1LNBpFCIGiKDz77LMLPb4VhSsEL7Zd5HzP7C1SJIuPCKIHoMw9ML/+10tF1zzfqFWllqOmKggh\nqrZkalqT4J/u3cRzx7rIFS1AwbS9uNyIoWE77oyRDpqqUJuMkIobKKoSRtjARBGe/pGCZ93798yy\nJ63GKRMREhFd92N0vRKO87XYNZ+NARaLOYnut7/97YUex4omENu/P9pJ71BeZpotMRP+V4HiL3BB\nYMku8eAm4Vmw1b8x03XzNTSFrRtqOHyil9FciWzeqhBYXVNmbeUkhGA0a6JrCvGojqGroXUshHeM\npjVxzvVMhH6V909TmGjJXh6jC8vDBbCQzEl0m5ub+clPfsLbb7/No48+yt///d/zsY99bKHHtmI4\ndLyHZ146z/B4aamHsuqoENgK94Cy5O6BK6WabqoKbGpK035+iJGsVXU/xxVhtTJNVaY0jAQvwUJR\nBMPjRbZsqGFTU5p3usewHRdNU7lxxzp2NNeSK9qM50xqkgZbmtKc7hoL3RA3X9vAb35kF8C0JRxX\nI3MS3W9+85v09vby1ltv8bnPfY4f/ehHnDx5kq985SsLPb5ljysEr5zok4K7CAT+VwVQytwDK0Fg\n54qiKHQP5jCt6cMQQ6vXFUQNbUohfAVPkFXFO55puWzdUEu+5ImpoauMZU1+1dZDxNBYWxujtSFF\nPKqzc1M98ZgeVioL3AjLzQWwkMxpqfXFF1/kT/7kT4hGo6RSKb73ve/xwgsvLPTYljWuEBxsu8g3\nf/AaJztGlno4Kw4v/tXBdSestKD+gKIunQ92qVFVBceZW/umoHV7xNB8gfXejxpe6UbbX4SLGCrN\n6xIV+wZtc4bHiwyNFWl7Z5Az3WN0DeZobUhxx40bq/ptg9/FDw6c4WDbRWzXrXh9KVlxy5U5Wbpq\n6PPybqJpmuF7kqkE/Z1ef3tAdni4QirdA179AfD8rygsawt2pnbthq7iOO6sXSAU/x4EIutUa542\nwzkFXpRDS0OSC31ZFPx0dH+7iO51i0BRwvKNu7at5diJXt7p9haEbVcQjwTFdrwY9OkyxCYnM5zu\nHA0TLJZLcsOVMifR3b9/P1/4whcYGxvjL/7iL3jmmWe49957F3psy5agv1NeCu4lsZL9r9WIR1Q2\nrE1xrmc8FE1V8SIP6tNReofzs7ql4hENV3gptxC4BRQsy5ki2EHm2mRiEY2tG2qxHEHfcL4inMxx\nBdmCTVd/loc+dC3glTg8cXaQdCKCaTsYrgjjgbMFi6GxIp39WSK6hgDuLBPRyckLHf1Z8kUb03aI\n6FrFgttKZU6i+/nPf56DBw+yceNGenp6+J3f+R0+8IEPLPTYli2d/dlpV5YlHqH/tcJ6XbkCW418\nyWU4UyRqaBRMB01RqEl5tQmEEBRLs2cxlCwXV4gwCUG4AisoaDAJb3Gs8r0gNOxczxhDYwWsSYtq\njp/GWyjZHGy7SOdAFoHCiXND5IoWyZhOKm7Q0pAiETN489xQ2ESyZDocae+rEN3JyQxRQ6Vv2Ay3\nLxRXfobmnDtHNDY2cvfdd4evjx49yq233rogg1rOuEJwvnc8XMGV4Nce8KpnBS6qpY5/vVoYyUzE\nvzp+mJZpuUR0dcaaHIF4Bg/34AEf7iGq11RQFW9n189eiEU0BkYLZPIWhq4y2VQIEiRGsiWee62b\nbN5iLFfCdT13T65os3NTPb/5kV2oisI3f5hlaIYefpOTGTr6M4xkzNDSjUevmmY2C8acrvCLX/wi\nJ06coLGxMXxPURT+8i//csEGtlx50a+6tBqZcA8w4RqAMA5WMjtCQK5oM9s3SFW9Xmlihmo1ut/L\nLAwRU5QwxlZTFc8VoSqYtpcU4vhuAncGR3J5TzZNVTB0lUTMK0x+sO1iWDJSUSBq6Ozd2Vix/+Rk\nhoNtF3m7exzwjtHamJrlypc/cxLdkydP8vOf/xxNk5bJbBw+0bcqOj9MhGeVJxesLvfAUlItC20y\ntu1O1KwVniXtGboKqXgEgde3zPEjFVzhTokeUBSIGCquEAyOFnB8yzr4jIOSjOULZLGITtOaeNh2\nfSaWYxrvlTIn0b3xxhu5cOEC27ZtW+jxLHu6h1aelTvhf5XugfkmEdUwbXdK7YT5oNoRvYLhLq5w\nsSyHQpkLQ1FAU5SKsQgBI5kSY9kSJdPBMDSSMZ1kTGdNTYy9OxvDzsEBqYRB87rUnKIQlmMa75Uy\nJ9G9/fbbuffee2lsbETTNFl7YRpcVzCeq54FtFyYaG4oBXYxKFxmLeW51MCd7u9CQCZvTRF6V3jR\nEEI4oc84qIlbKgtFa6yPs21DLQ/esyN870qqfS3HQuRXwpxE97/+1//K97//fTZuXF1PpEvlwJGO\npR7CnAn8r1Dpcw3iXyWLw+XYt5rqPQhnso5nO24194SqeFXB9lyzjnhU5+JQjr7hAqPZUujHtWyX\nTN6aIqrvvWE9pztH6ezP0tqY4r03rJ/z9SzHQuRXwpxEt76+nltuuUUWMJ8BVwh+9NzppR5GVYL6\nr1QJz5IsH1S/XGMipjM8XprRmlWAREwnnTCoTUYYHi8xNF4MSy5O3k/Bm+orKFzbWscdN24MW6nn\nihau5dXG1TWFjetSU3yvL7/RS9dgDkVV6BrM8fIbvXMWzuVYiPxKmJPo7ty5k09+8pO8733vwzCM\n8H3ZrmeCX7V1c3Eov9TDKBPYSvcA0j2wLAla5aTjBvmSjWW7DI+XPD/7TPsp8MkPXBMK38G2i/yv\n598mX7QrEjGCZArHdXFcQaFkcb5vnNM/HeVCX8Zv0z6xvRCwtjY2Zfp/JcK5HAuRXwlzEt2NGzdK\n18IsPP2rdxb9nBPNDScEVQrsykLglXfMFT3Bnas7IhrRKqb7nQNZyjxKgGc1r62NMTJeJFPwYqnz\nJYd/bO+nZHki7LoCXfOEWddVkjGDRHyqbFyJcK62CIY5ia60aGcnW1jYMDHhOlAWngXS/7paUBTC\nUK3JRHQvHlpTFS/8y9/ItBxeequXiKGRihlc11obxuUKITB0lTU1XnWwnrIZmqIQNq4MCrq7YiIm\nN5Uw2Lqhdso4rkQ4V1sEw4yi+/GPf5ynn36anTt3Vvhzg+iF9vb2BR/gaqNaeiywqitnrXaCguDO\npNJhwXRfADHdK7HYM5THdgRBsEHJdCiZDq+/bbF+TYKLtiewyZjBB29uoXsgR0RXsf16CwpQn44y\nmjVD4Y1FNGIRncb6GGvSMc5dHOPYiV6GxguM5Szq01H27mri/Ss86mC+mFF0n376acBLjpiMaZpT\n3lutFOaQI1+N6fyvMjxLEqAooGsqyZiOnTNDkQ0qi7mOQFUE2YJFoWRPEeaAfMmhayCLK7zjASAE\nLQ1J1tXGGKRI0XSI6Cr1NVFqkhGKpkMsorFpfZpSyWE4U+Jkxyha9xhDY0UvkcJv29M3XEBhZUcd\nzBdzci888MADPPnkk+Fr13W57777+MlPfrJgA7taGcuWuNCXpaMv4/8vS/9oYdb9PP8rfgSBXOBa\njZS3s5n272VtdbxUWg1DV6lJRcNaB5btVBxHALqqoKgKEUX103qnHt+0BaoClnDJFkz+/mgn129Z\n4xer0enoy2JaDucuZkgnInx035aKKIbh8SIl00HTVK/eQ1kkhGk7Kz7qYL6YUXQ/85nPcOTIEYCK\nTsC6rlcUv1lIhBB87Wtf49SpU0QiEf7Tf/pPtLa2Lvh5XSEYGC3Q4QvsBV9gx3OzW/jCL6wt/a+S\ngCC2VkHxmkJOs135+0KAZTvEoxqZvMl43sR1pxYoF35px62NNdiuV1w8WzCnFV/XFZhC0D9SwLQG\nSSUMkjEdVfUMgqCy2OH2Pvbt3hCKaVDYXOB35/AVV8FLB25uSHLQrz2yGpIcLpcZRTcoaPP1r3+d\nr371q4syoMkcOHAA0zT54Q9/SFtbG9/4xjf4sz/7s3k9h+24XBzMcaEvQ2dgxfZnZ62hoCrQWO/5\n0Sr9r7LAi6QSx4VkTMN2BOmYQbFkYU7ySqkKU2rgCgH1qSgd/dlZkyEa6+Ps3FRP10CO5oYkh9/q\n5VxvZqLWruItpJVMJ7RQc0XL79ILhqaSdbwmlo4reKd7nO/9rJ0dLbW8enoAszxWtzGFoSsVPl2E\n4LnXLwKrI8nhcpmTe+HLX/4yzz//PKOjoxXvL0ZzymPHjnHHHXcAXg2IN99884qOVzRtOvuzdPRl\nfes1Q/dAblpfWIChKzSvTbBlQ63fqC9F87okhq7xW3/03BWNSbIymC01t2g6nrXouFRr8FCTjJAr\nWJh25VH6Rgoztt/R/AaT3QM5PvfP31UxnrzpkM1bmLZDPOL1Q/POL7zau7ZLNm/xgZuaOdM5yisn\n+nB889i0HY6dHmBbSw0RQwVFIWaoNK2Js21jZRowwA8OnKl4Ld0N1ZmT6P7u7/4uFy9eZPv27RVR\nDIshutlslnQ6Hb7WdR3XdefULmg8Z4ZWq+ciyNI/PHsL9ERUo7UxydYNtWxe7wlsU30irI4vkVRj\nuvY7AcGDfboHvOuCoWvYzkQCg6IolCwnDN8KiEW0UDy9KAaBrsHv/fkrlCyHnZvq2dacZjRTYjxv\ngoBs3iIWUXGcoPymV4axsT6OEIL2jhEvUcL/hXjuDZcDR7swLc+lYNku4zmTcz1jfO1/HqG1McVv\n/LOd6Kq66pIcLpc5ie6pU6f4xS9+sdBjqUoqlSKXm3hizia4f/V37bzTPcbZ7jGGx6cvphywrjbK\n1o217NhUz/aWOrZtrGVtbUymPEumEDVUTGt6f2w1LfWe08qcGi4KBAm/etdIpojtuOia6rm5yro+\nNK2Jc+v169ncVEP7+SHO9YwT0VXOXRwP26m//FYvr54eoGQ6FeMtmGWNPPFEfWNDir873MF4zqx4\nIKiqgq4pWI5LXU0UTfMeAI4Lnf3eb7JvpEAsZvCFB9/Dx+6+lnQ6xvnecbasr+GDt25aEEOloSE9\n+0ZXMXMS3e3bt9Pf319RxHyxeM973sMvf/lL9u/fz+uvv86111474/ZPHqhe/0BVoLEuxqamNNs2\n1tLalGZTU4pkzKjYTlg2g4Mrv0+T5NKJRbwKe5On/zMRdEIolOwZXQSqAoamkYgZXNtSR0tDkmcO\nnSeTN73iNH4kQzJm8O4ta/n8x3YzMJBh97Z6vv/zkxw7NRAKLnhW6mxrEkJAybR5/lgnKF7LdVVR\nvLq7iufjTcUNrm2upXsoTzyqE4/q5It2heV9pmOEgYEMAHu2rWHPtjUADA3N/++ooSEdnmuxmS+x\nn5PoFotF9u/fz7XXXkskEgnfX4zOER/60Ic4dOgQn/rUpwD4xje+Mes+hqawcV2CzevTbNlQy+am\nNM3rkkQMGZ4luXyyBauqL3YmSpZD1P/e6dr0lcEURSHpp9cGK/8vn+hlLFcKux4HGWHl3RW+//OT\nHD3Z73XsnQOBC8Q3wCceBELg+qKrqYrXhl14Bcy3t9YihOBU5xjRiJeEMZadaJi5Gro9zCdzEt1/\n9a/+1UKPY1oUReEP//AP57z9//PluzGEkP5XybwzV8FVy8TMdgSOY6OqUJOIcF1rLef7svQN53GF\nt60C1CQMrmutrwi1sm039L8KPEG8+6bmihTbzn7PmlQnZax5Bcm9MQfvGtpESBh4IWCV4/aaVHoH\nwBV7w90AACAASURBVA89s/j/nj/r+5VBzXvRPpuaUpiWG/p0JXNnTqK7d+9ejh07xunTp7nvvvto\na2u7aptStjQu/vRjz/ZaXn9nbFHPKbl6iRhahS9V4C2SBX3jxnNmKMqB8K6tjfHAB6+piGstWa5X\nKwFfmJMR7rhxI64Q/MPhC7SfHfKiCvB9x6pCKm7Q3JCkLhmhazBHruC1N9+9bS3XtNRy9GQ/CC/V\n9/g7g2QKE3FrW9anuPPG5jARAjxhLvhRF4p/ItN22bphavSCZG7MSXS///3vc+DAAfr7+9m/fz9/\n8Ad/wCc+8QkeeeSRhR7fsuD/vO8mPv/Hzy/1MCRXCVWrgfkRBm3vDE3pFB00h/zez9pJxIzQ2t3U\nmKJveKIYzSZ/Gn/oeA8H3+jxziPEFKtTVRRePN7DaM4kFTfYu7OR99+4EVVR+Cd7msPjPfdaFz88\ncAbHEWiawntv2BBa0Yfb++gbLmDaTkUihBCCqKHJyIQrYE6i+/TTT/PUU0/xyU9+kvr6ev7mb/6G\n+++/X4quj66q/Jcv7uML3zq01EORLBKaqqAoVPXRur57SxFeyIHjClQ8n3A1XAFDY0VGMiXW1MTC\nsKtg2t7Zn6WlIcn2llp+cOAM3YPZcCFLURS2bqjlgQ9ew6HjPfyv594hX7ToGpyI+AkicSZni716\nasBLmNC9v796aoC7fFHeuDZJfSrKSKZE30gBIVzyJYf6dJRP3H0tN26rn58buQqZk+iqqlqxgBaN\nRmVn4EnURKNs3VjDuYvjSz0UySyU+1wvl6ihsqY2Tlf/1BV64bsMDEMjEdUpWU7YktzQFfonJTsE\n8a/pxMRvrGsgh66qPHLv9QBh/QPwfK2apoSREZO78Q6PF4noWphp1jWQq9oSpxrl2wF84KZmFCpL\nNjY11iyIC2+19Eqbs0/3P//n/0yhUODAgQM8+eST3H777Qs9tmXHto21dPRmZs1ukywuqqoQj2jo\nmkquaM2pfflsWLZLybSnTYgQQqD5q2SxiE4yrqMoCnft2cjbXWO8/vYgRdMmGtHQVJWm+jimnx1m\n2g75ouVZzIoX43u4vS8UU6/8p0IyplftxhvRNUqWDXnCY3X0ZcJjR3SNzv4se3c20jdcIFf0LPC6\nZITOgcqHSPdALrSiA/H+6F0pXmi7yJH2PoAK98WVsFp6pc1JdL/0pS/x1FNPcd111/G3f/u33HXX\nXTzwwAMLPbZlx2/fv4dCweTE+WFGl3lX4BWF8NJdo4aKpk6EbYUSMUsmWTVsV5At2BXHKzscuqZi\nWg6xiPcTS8UNbtvVxL7dG/gne5orrLrmhiTCdfnf/9jJWK6Erqm0XxjhxbaL3LmnmUPHe+gbLlAy\nHfK+P3htbYxc0UZRvK4O5dlgybhOfTpC30iBiK7RNZjD0BTGcyVcATlhca5nnOaGJBFDJVeERFSn\nazBHa0Nl+NdkK/p01ygdAzlePz1AJu8Vf+obLqD4hcivxFpdLb3S5uxeuPvuu/nUpz7F0aNHOX36\nNKZpoutz2n3VoOsq/4ef+/4nP3iV051juNNU/JcsHkEBl7GcVZEZFkQFzFYzAbwqYa5LGDMr8Op4\nBIcLFpsM3UuHNW2XfNEJp/gwMc2f3FXhdMcIJztHGcuafoscL1rgyMl+7tzT7FufIgwLM3SVdMLA\ndkQoTJM7N3QOZLHKHgYlyyViaBT8rhDdgzl++tIFTNuPTPAjK2IRjZZ1yYquvv/rucpWVOd6xjDt\nicSL8rKOV2KtrpY04jmp5uOPP46qqnz605/m3/7bf8v73vc+XnnlFf7bf/tvCz2+Zctt16+nf6RI\nybLJFi6vyLlkfhBMX+8gqAs7EwpQl4qF1mOx5GA5k9Npvf+k4hFsFzY1psPFrGzeIpu36Bsu8MqJ\nXk53jrKjtY5f+uLUO5TDst2KMLJyCkWbTH5i5qSpQU1mEQrT5JY3B9sucqZrIoxxU2OK0WwJvSx+\nPXA1lEzHF1GDoulM6eo7WQy3bqhlLGNS8jPeIvpENMOVWKurpVfanET3jTfe4Ec/+hHf/va3ue++\n+/id3/kd7rvvvoUe27Lm/bs3hAsQR9r7GJtDHV4J1CYNckULe55bzgWVuILMrWotyCe/pyqgqSrr\n6mL801tbed/uDXz/5yc57PsyAwzdK+od8TPGwEv/vfumZroGcnQPZukdyofRC8fPDjFSltHlTKqR\nqyiQTkS8con+sdKJCKbtYOie/3fH5jWsTUamCFMwve/sz9KyLkk8ptPa4Fms3//5SY6fHSKiTyyC\nB+NtWhPntl1NU3y6Xb5PN/h3S0OSe++8hv/7r/+R9o4RoobGr93SEo7jSqzV1dIrbU6i6zgOruvy\n7LPP8od/+IcUCgUKhdm7Jaxmyr9AXYNZMnlzyg/rUv2IKx1DV2luSFGXjPDSW32zbh+ksla7j+X3\n12ttI2AGV4+uedlbwWekawrXtNRi2y4l0+FM1xjv272BRMwI6w8E52mqj7OpKc3JjpFwsaulMVXR\n/vxvyrpFl4teNRJRnY/u2xIKWWtjijPdY4AnkLdfv55fv+c6BgYyuEJUhIIJCC1ogLtvag7H8Zsf\n2VXhR0YIugZzFIr2RCTEumSFhdzSkJwihr96rYuuwVwYbaGqaui3XS3W6pUwJ9H92Mc+xvvf/37e\n8573cOONN/LhD384rIUgmZ3bdjbSX7ZK/J4d6/jsR3bxvZ+188qJPim+PpbtcqZzdO7RHwpVF7LA\nK9ai+patK4Tn153xsF6UQDD53tyUZk0qyj/+/+3de3BTZf4/8Pc5uTRt0hvQQiHcFSliu1Bl50uR\nYRFGGUEXqbIgVJG1FKWjDEWoIFSkgIK7O1p3V539IuL+BKm4+tsd+QmyghQF6S4UKC2gQNMLNPRC\nm6S5nuf3R5rTJE3btE1Ob5/XzM6a08N5npMmnz7nuXyeEj0AZzYtABg3PAoKWXOWuxCFDDMnD8OV\nsjviogeFnPf4S5CcEIfLujqxlakOdc464DgOuipD0xboNoA5Z1oMH+wczNr/7VVoY9T4n/uGAPAd\nyLz7UNUqz6+0++M97zXYVao34PzP1ag1WKGU87hcpsLMyVqxhd5a0Lx+03NapK8ySOv8CrrLli1D\namqqODf373//OwYMcGYSevfdd5GRkRG8GvYB0xKHguO4FiO6y+dOwD0jovH/TpdCX9fY5s4A/YWt\ng++BUs5BKedhsjT3R8g4YEBECJQKGarvmGE029ucm8tzriW6rOk1B6VC1qI/UldlwLJH43FZV4eC\ny3px4OlqeT3OX6sRcxlw4FB+2+R2fc6jlen+Gfj+XAX+c9m5SIGBQcYDFXojPv32CsJUcpTonK3b\n1gJZe32mvh7vXYH6dl2j+IfCuRuwGeV6Y7vLe0cNicC5y/o2yyCt83v6gftiCFfABYCjR49S0G2H\ndwvD1YJJTojD9MShKNcbcbmsDgaTDQ0mK3ieQ2x0qBgwiLOLIETpXGwAADaHAJtdgFzOIzpcBYPJ\nBqPZBoWch0LGY8zQSNQZLKi+Y4Zcxjkbnj4eKVzZuxxC846QAmMovdUAuYwXk4RzHIfhsRrwHIcw\nlQKDB4RBIedhswvQVRnEASnAOUDlHYhaawGW6Y3QhCmaNp20we4QYGi0iQnEOXBtBtZhg8KcW+k0\nDYrNSIwDz/NttlRd17N6ZSaz2gW/AuhDD4xAQ4OZuhA6qctzvrwz2pPWtTadxjX4oAlTgDEGi82B\n2gYLQhQyyGQc6iWa86uUN2ehCmar2zWA7t7ylMuc5Xp/nDgAXNPihkFRodAOUnvMCuB55/9bbHZo\nY9QYFReBRrMdZbeNaDBanblf3a7lfVcRaiWGDlLjVo0RdYbmfnez1QGeFyCXcVCrFIgfGS0uy/Ue\nLBoeqxHr5Eou428gcv/du6Zh2Zgzp4Jrl902A6HXHFjmx5xYV5lKOQ+7XRBnXowcrPGr3jxPXQhd\n0eWgSzss+K+16TSuD7quyoDrN+tR1rRnm1LOQ61SwGxxwCGwNvs6ZU2RzDVC35mQabMzjBkajhCl\nDFfL6tpM1u3P3Faf/44DZDIech4w2wQxf2u0RgmO51DXYIHFbXcGpYLH8FgNRg2JEOeN/nD+pnMw\naFAYdLdNOF10EyEKOWwOhuExGuj0BhhMNtQaLB515DjnDAObg4lB3JUw/MsT1zzqyRgDB2c3Q9I9\nsR6P3K7fV7XRioFqpUedOrogwH3gyWS24VJpLe4YrM7+ao5rN4CXN7WUXYNsZ4qrxKej1ubJuq5X\nWtWAGzcbYLEJGOGWLIcEF61ukFBr02lcj57fn6vAqUu33PKccogKD4HVJjQt57SLg0LueJ7DgAgV\nrDbnRPfQEJkzH6vAPOZ3tocBMFnsyFqahB2fFOBqueeAiesx29X/2JHWMMcBKgWPUXERGBCuQk2D\nGaW3DOB5rmlrb0CjkiN2gHPzREOjTRx08k7ooo1Ri2kQ/3HyOgZEqMRyyvTO0fg6g6Vlq7lpHq2r\nCwdoThh+6tItmCx253vonOgAu0MAb+c8luQ63wfn78t9F4POtvzcux0ExvC//yzCf67cBgCEqeS4\nWxvZZiD0/kx589U10d5gV3/JgdBdKOhKqL3pNGV6Y4u+wV83jXKX6Y24VnkHN5pyOwhNe2ZxAEKV\nMgAMchkPTbgChqZAO0CjgMlsb9FCFncP8DHdqqbegvzCSowcEo7KahNMFmeriYcz0DoEJm5c2JFp\nb4w5556u+d0k8ByHT49cEQe/aurNsNodMDTlCtCEKjAgPASGRjtqGywwmW04UVgpToVyb8F5D+qY\nzDYU3ahpsScZzzm323HtzuCal+oKKL+OHwxdlUHcrRdobsm7L8ntivaCGc9xUIcqMXhAmHjMfUDO\nF+/PFGMM/27aBh3o3CDXicJK/N/862I/MQMw3e0PgyuXLwXkzvEr6N65cweRkZEex8rLyzFs2DCM\nHTs2KBXri9prYWhj1CjR1QJo7hsEmr9QC34zBnu/LkFplQEhCh4jYjWw2ASEquS4XW9GzR1n4mlN\nmAJqlRzDBmkwPWEIDp3WoabeAqWcR5hKhtoG5+MrD8B7DYLd4UDesZ8xODoUSjkPs5UDYwy+Oix4\njgPHOxNdu+K6jG8630cwNlkc4hfU1UJjjEEQnP3YpqZZBgoZD7VK3pycRW/wWEzgek8Az0Edk9kG\nnd6ARotzI0eeQ9OiBRnuvycGdw2LQEV1o89g4T6tK1yuhNFsc043ExgMjTZxSW5X+LNEtqOLC7w/\nUwJjHjNl/ue+IS1SOvoKku5/EC5cq0a90eLcidjqwOlLt8Sg657Lty8npQmmNoNuZWUlGGNIS0vD\nhx9+KA6aORwOPP/88zh06BB27dolSUX7gxatFqBF686V6s/b2V9q8OXx5gn4v44fLH4Z5DKZR9q/\nqPAQcOBgtTsgb0qg3Whx5hEQGGCxOnCrthGaUAXsAoNSLoPZahdbvUBTd4FSBrvDmSvAdSxSrRS3\nevEeHQ8LaZ4B454s29jobI1b7c78AOFqJaw2waPbwJvYNeM2qPN/Dl+GsbFp08SmObwR6hDMmzqy\n3YDpPa3rwrVq3K7zvQDIFaBcfbr+tvb8WSLb1cUFvpYD+5MLwf0PQm2DMzGOzMct9ZekNMHUZtB9\n5513cOrUKVRVVeHpp59u/kdyOWbMmBHsuvU73l+YT49c8fh5Wx/wtqbxuP87pZyH0WyHQs5DKZdh\n/IgolN02igNPHJzLUl3LZaPDQ8BxHMJUMgyRh0FXZYDNIUCl4DEwUgWbXUB1vdm5+wDvDOQOB2sR\ncGUccO+o5sTXrnst0xvFgR/Bra/VfUYAAHExgff9CULziqzrN+vFFhrPcRgUFYpHpozweC/aesR3\nf/+Pn1N7PGLfPz5WLMeVJNw1ZQzwr7XnTys20IsL/A2S7sfVKrnHZ2TK+OZdwLUxalxzWxxBc3Q7\nrs2g69p594MPPkBaWpokFSLNOvKo2dY0Hs/rcOJMBwC4SxuJccOjxJkTuioj7A4BDocAq03AmLgI\ncQsZBuDof8pgaLTBZLbDZhcwekg47A7BOeIOwGoTfO9My3Gw2lmLoDfMbcoU0NzX6u+MgG9/KvVM\n3q2Qgec5KOUy3Dt6QIv3xN8sWO65M1x9pd5JwqMjQgD439rrjiWy/n6G3M/ThCkxfkS0x9ZBLskJ\ncQgPV3n06ZKO8atP99lnn8Vf//pXXLt2Da+99ho++ugjpKWleewmQQIvUF9S9+uU3zbA0GgTp/pV\n3DaJ06EExvCH/WfFyf6aMAXCVArx558eueJcuQVO7OssrzbhvrGDcLPaGXhu3GxoEXRdsbLGYMGJ\ncxXiQM/lsjr85ldDWyw7dQVX775KX32T7ktShaYuiogwJdSh8ha5YV3vQVuvXdp66lDKZR6pDf1t\n7XXHEll/P0O+zvP1R47nOMz+9Uj8akzzAima7dAxfgXdLVu2YMCAAbh48SJkMhlKS0uxYcMG7Ny5\nM9j169cC9SV1v457Hx/gXNH0/bkK6KoMaGzqs3VN1QKaA4rAGExmG2rqzeKGiILAUH2nEYLAcM+I\nKAyP0aCm3iwmgxE1Ja6tqmnE6eIqjx+VuwX9trhaqA1GK74vtOPQ6VLMfmA4jCYraurNEJq6RJRy\nHla7A+NjonwGmM5mwXINchob7bDY7BgyIAzxowdikMbZp9tTA4+/n6GufNb6y44PgeJX0L148SK+\n+OILHD9+HKGhoXjzzTcxb968YNeNBIGvwbqj/y0XlyCHhykAcB47HQDOL1bZbeeUNptdAMc5F2E4\nBAaz1QiT2Y4rZXcwJi4CDSYbLDbnTAW5jIPAnJmz3BN6u7QW9LyDmGvBwx2jFYLAcKu2Efu/veJc\nbME7d2lQKmQYGKlq6oNW+Ax6nX16cJ/d4FqIMTouQmzx+Ttg1RfR4FrH+BV0OY6D1WoVH0lra2tp\nJVov1dpjs+tx2Wp3zhoYNkjjcZ7ri6QJU0ATpkCDyQqbXRBbva4k2GGhCjz1m7taTTU4JX5wi40O\nffFuPWkHqWG1O5qXnTMGq825m4LAC2Jfrutz2Vow9zXFyp8pVa6cC+4zKq7frBeDbn8OPP1lx4dA\n8SvopqamYtmyZdDr9cjJycGRI0fw4osvBrtuxE+uCetFP99Go8UuJq725xG3eR2+c1GGK9er9xfH\n+4sVPyLamd/AZEW90RmADSYbtDGalvNG0X5foTfvoBUaIkfCmIEouKwXAz3HNfcX87xzyayvwZ+2\ndOTR2Ps9GDUkotWf9afAQzl0O8avoPvoo4+ivr4e9fX1iIyMxLJly2h/tB7ENWG9tt7S1EWgFBNR\nt/eI6573wTtg+zrPfdL9D+dv4seim54DZy1WgnW8r9C9/9jVvzw81lmnu7WROF1chdoGC6w2B3ie\ng9nqXEiy7NF4MaD724LtSAvV+z146IERqK42+PxZe4Gnp/YBdwbl0O0YvyJnZmYmKioqMHbsWJSX\nNz8q/va3vw1axUj7XF/cbwvKYLY53LoInI/6/jzidmWgxTXP1uZontfb3rJVf7j3H7sPivEch+m/\nGobpbrvptrZAwd8WbIem5Xm9B7zb1LuOBh4afOq//Aq6JSUlOHToULDrQjrI9cU1mu0wNNogb1pC\n1FoXQTAEY7K8e/8xoPA5KOYr6Yyva7T22qW7Ho37cx9wf+dX0B07diyqqqoQGxvb/smdNH36dIwa\nNQoAMGnSJKxevRpnz57Ftm3bIJfLMXXqVKxatSpo5fdGri+qOlQOmYxDiJzHiMHhrXYRBEMwJssH\non/U32t016Nxf+4D7u/8CrpmsxmPPPIIxo0b57Eg4uOPPw5IJUpLS3HvvffiL3/5i8fx7Oxs5Obm\nQqvVIi0tDcXFxRg/fnxAyuwLXF9cjuMQoVbiwfviOh1AOtvH6GuyfFcFovXZ0wd3enr9SPD4FXRX\nrFgR1EpcuHABt27dQmpqKkJDQ5GVlYVBgwbBZrNBq9UCAKZNm4aTJ09S0HXj/sWNHzMQCaOj2/kX\nretJfYyBaH329MGdnl4/Ejx+Bd0pU6YErMC8vDzs2bPH49jmzZuxYsUKPPzwwygoKEBmZibee+89\naDTNyzjVajXKysoCVo++wP2L21rfpr+C0cfYl0boCQkUyed9paSkICUlxeOY2WwWN75MSkqCXq+H\nWq2GwWAQzzEajYiIiIA/YmLCA1fhDuqtZcePGegxIBY/ZmCHrufr3MOnbuD785UAgGs36xEersLs\nX4/sdB07UrZUqOz+VXYg9IjJtrm5uYiKisLvf/97FBcXIy4uDhqNBkqlEjqdDlqtFidOnPB7IK0r\nLb6u6GprszvLThgd7ZEaMmF0tN/Xa63sS79Ue8zhvfRLdUD7ftsqWwpUdv8rOxB6RNBNS0vD2rVr\ncezYMcjlcjGlZHZ2NjIzMyEIApKTk5GQkNDNNe27gtHHSCP0Tr2hm6U31LGv6BFBNyIiAu+//36L\n44mJidi/f3831IgEQl8foXdPoN5WoOpJg5St6Q117Ct6RNAlfVNfH6F3T6DeVqDqDQshekMd+wq+\nuytASLC58jB8euQKvj9X0WKn4M5yT6AOtB6ofCUP6ml6Qx37Cmrpkj4vWI/O3tu/txaoekM3S2+o\nY19BQZf0ecF6dG5rM1B3vaGbpTfUsa+goEv6vGDNomhrM1BCWkNBl/R59OhMehIKuqTPo0dn0pPQ\n7AVCCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1C\nCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJEQ\nBV1CCJEQBV1CCJEQBV1CCJEQBV1CCJFQtwXdw4cPY82aNeLrc+fO4amnnsLixYuRm5srHs/NzcWT\nTz6JRYsWobCwsDuqSgghASPvjkJzcnKQn5+P+Ph48djmzZuRm5sLrVaLtLQ0FBcXQxAEnDlzBgcO\nHEBlZSUyMjKQl5fXHVUmhJCA6JaW7uTJk5GdnS2+NhgMsNls0Gq1AIBp06YhPz8fBQUFSE5OBgDE\nxcVBEATU1tZ2R5UJISQggtrSzcvLw549ezyObd++HXPmzMHp06fFY0ajERqNRnytVquh0+mgUqkQ\nFRUlHg8LC4PBYEB0dHQwq00IIUET1KCbkpKClJSUds9Tq9UwGAzia6PRiMjISCgUChiNRo/j4eHh\n7V4vJqb9c4KFyqayqey+W3YgdEufrjeNRgOlUgmdTgetVosTJ05g1apVkMlk2LVrF5577jlUVlaC\nMebR8m2NXt8gQa1biokJp7KpbCq7D5cdCD0i6ALA66+/jszMTAiCgOTkZCQkJAAAkpKSsHDhQjDG\nsGnTpm6uJSGEdA3HGGPdXYlA669/halsKpvKDm7ZgUCLIwghREIUdAkhREIUdAkhREIUdAkhREIU\ndAkhREIUdAkhREI9Zp4uIaT7CIwhv7ASZXojtDFqJCfEgee47q5Wn0RBlxCC/MJKHP1vOQDgclkd\nAODBxKHdWaU+i7oXCCEo0xvbfE0Ch4IuIQTaGHWbr0ngUPcCIQTJCXEA4NGnS4KDgi4hBDzHUR+u\nRKh7gRBCJERBlxBCJERBlxBCJERBlxBCJEQDaYR0Ea3mIh1BQZeQLqLVXKQjqHuBkC6i1VykIyjo\nEtJFtJqLdAR1LxDSRbSai3QEBV1CuohWc5GOoO4FQgiREAVdQgiREAVdQgiREAVdQgiREAVdQgiR\nEAVdQgiREAVdQgiRULcF3cOHD2PNmjXi6yNHjmD27NlITU1Famoqzpw5AwDIzc3Fk08+iUWLFqGw\nsLC7qksIIQHRLYsjcnJykJ+fj/j4ePHYhQsX8Morr2D27NnisaKiIpw5cwYHDhxAZWUlMjIykJeX\n1x1VJoSQgOiWlu7kyZORnZ3tcezixYv4/PPP8fTTT+PNN9+Ew+FAQUEBkpOTAQBxcXEQBAG1tbXd\nUGNCCAmMoLZ08/LysGfPHo9j27dvx5w5c3D69GmP48nJyZg1axa0Wi02b96Mffv2wWAwIDo6Wjwn\nLCysxTFCCOlNghp0U1JSkJKS4te5CxYsQHh4OABg5syZ+OabbxAfHw+DwSCeYzQaxXPaEhPT/jnB\nQmVT2VR23y07EHpMwpvHHnsM+/btw+DBg/Hjjz9i4sSJSEhIwK5du7B8+XJUVlaCMYaoqKh2r6XX\nN0hQ45ZiYsKpbCqbyu7DZQdCjwm6OTk5WLVqFVQqFe666y489dRTkMlkSEpKwsKFC8EYw6ZNm7q7\nmoQQ0iUcY4x1dyUCrb/+FaayqWwqO7hlBwItjiCEEAlR0CWEEAlR0CWEEAlR0CWEEAlR0CWEEAlR\n0CWEEAlR0CWEEAlR0CWEEAlR0CWEEAn1mGXAxD8CY8gvrESZ3ghtjBrJCXHgOa67q0UI8RMF3V4m\nv7ASR/9bDgC4XFYHAHgwcWh3VokQ0gHUvdDLlOmNbb4mhPRsFHR7GW2Mus3XhJCejboXepnkhDgA\n8OjTJYT0HhR0exme46gPl5BejLoXCCFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0\nCSFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0CSFEQhR0CSFEQpLn0zUY\nDMjMzITRaITNZkNWVhYSExNx9uxZbNu2DXK5HFOnTsWqVasAALm5uTh27BjkcjmysrKQkJAgdZUJ\nISRgJA+6u3fvxtSpU5Gamopr165hzZo1OHjwILKzs5GbmwutVou0tDQUFxdDEAScOXMGBw4cQGVl\nJTIyMpCXlyd1lQkhJGAkD7rLli2DUqkEANjtdoSEhMBgMMBms0Gr1QIApk2bhvz8fCiVSiQnJwMA\n4uLiIAgCamtrER0dLXW1CSEkIIIadPPy8rBnzx6PY9u3b8fEiROh1+vxyiuvYMOGDTAajdBoV4EV\nqwAAC01JREFUNOI5arUaOp0OKpUKUVFR4vGwsDAYDAYKuoSQXiuoQTclJQUpKSktjpeUlCAzMxPr\n1q3D/fffD4PBAIPBIP7caDQiMjISCoUCRqPR43h4eHi75cbEtH9OsFDZVDaV3XfLDgTJZy9cvXoV\nL7/8Mnbt2oVp06YBADQaDZRKJXQ6HRhjOHHiBJKSkjBp0iScOHECjDFUVFSAMebR8iWEkN6GY4wx\nKQt84YUXUFJSgmHDhoExhoiICLz33ns4d+4ctm3bBkEQkJycjJdffhmAc/bC8ePHwRhDVlYWJk+e\nLGV1CSEkoCQPuoQQ0p/R4ghCCJEQBV1CCJEQBV1CCJGQ5IsjAmn69OkYNWoUAGDSpElYvXp1tywn\nZowhOzsbJSUlUCqVyMnJwfDhwwN2fZcnnnhCnM+s1WqRnp6O9evXg+d53H333di8eTMA4LPPPsP+\n/fuhUCiQnp6OGTNmdLrMc+fOYdeuXdi7dy9KS0v9Ls9isWDt2rWorq6GRqPBjh07Ojy/2r3sS5cu\nYcWKFeLve9GiRZgzZ07Ay7bb7Xj11VdRXl4Om82G9PR03HXXXZLct6+y4+LiJLlvQRCwceNGXLt2\nDTzP4/XXX4dSqZTkvn2VbbPZJLlvl+rqaixYsAC7d++GTCYL7n2zXurGjRssPT29xfHHH3+c6XQ6\nxhhjzz//PLt06RK7ePEie+aZZxhjjFVUVLAFCxYEtC7ffPMNW79+PWOMsbNnz7KVK1cG9PqMMWax\nWNj8+fM9jqWnp7OffvqJMcbYpk2b2OHDh5ler2dz585lNpuNNTQ0sLlz5zKr1dqpMj/88EM2d+5c\ntnDhwg6Xt3v3bvbuu+8yxhj717/+xbZu3dqlsj/77DO2e/duj3OCUfbnn3/Otm3bxhhj7M6dO2zG\njBmS3bd72XV1dWzGjBnswIEDktz34cOH2auvvsoYY+zUqVNs5cqVkt23r7Kl+n0zxpjNZmMvvvgi\ne/jhh9kvv/wS9Pvutd0LFy5cwK1bt5CamooVK1bg+vXrrS4nLigo8LmcOFAKCgrw4IMPAgASExNx\n4cKFgF3bpbi4GCaTCcuXL8ezzz6Lc+fOoaioCPfffz8AZ6v/5MmTKCwsRFJSEuRyOTQaDUaNGoWS\nkpJOlTly5Ei899574uuLFy/6VV5xcTEKCgowffp08dwffvihy2V/9913WLJkCTZu3Aij0RiUsufM\nmYOXXnoJAOBwOCCTyfx+nwNZtiAIkMvluHjxIv79738H/b5nzZqFN954AwBQUVGByMhIye7bvezy\n8nJERkZKdt8A8Oabb2LRokWIjY0FYyzo990rgm5eXh7mzZvn8b/Y2FisWLECH3/8MdLS0sTMZd7L\niRsaGlqsZHMtJw4Ug8HgcX25XA5BEAJ2fQBQqVRYvnw5/va3vyE7OxuZmZlgbrP91Go1DAaDz3tt\naGjoVJmzZ8+GTCYTX/tbnuu463fhOrcrZScmJuKVV17BJ598guHDhyM3N7fF+x6IskNDQ8XrvPTS\nS1i9erVk9+1d9ssvv4yEhASsW7cu6PcNADzPY/369di6dSvmzp0r6e/bVXZOTg7mzZuHxMRESe77\n4MGDGDhwIJKTk8X7df/uBuO+e0Wfrq/lxGazWfxSJiUlQa/Xt7jpri4n9pdGo/G4viAI4PnA/j0b\nNWoURo4cKf53VFQUioqKxJ8bjUZERERAo9G0eA8iIiICUgf3e2qvPPf3JBDv96xZs8RrzJo1C1u3\nbsWUKVOCUnZlZSVWrVqFJUuW4NFHH8XOnTt9liFF2Q0NDZLdNwDs2LED1dXVSElJgcVikey+3ct+\n8sknsW/fPsTGxgb9vg8ePAiO45Cfn4+SkhKsW7fO4yk4GPfdK1q6vuTm5orJdIqLixEXF9dty4kn\nT56MY8eOAQDOnj2LcePGBezaLp9//jl27NgBALh16xYMBgOSk5Nx+vRpAMDx48eRlJSE++67DwUF\nBbBarWhoaMAvv/yCu+++OyB1mDBhAn766Se/yps0aZL4nhw7dkx8XOus5cuX4/z58wCAH374Affe\ne29Qyr59+zaWL1+OtWvXYv78+QCA+Ph4Se7bV9lS3feXX36JDz74AAAQEhICnucxceJEvz9fgSyb\n4zhkZGSgsLAw6Pf9ySefYO/evdi7dy/Gjx+Pt956Cw8++GBQf9+9dkVafX091q5dC5PJBLlcjk2b\nNmH06NHdspyYuc1eAJyZ1EaPHh2w6wMQE75XVFSA53msXbsWUVFR2LhxI2w2G8aOHYutW7eC4zgc\nOHAA+/fvB2MMK1euxKxZszpdbnl5OdasWYN9+/bh+vXreO211/wqz2w2Y926ddDr9VAqlXj77bcx\ncODATpddVFSEN954AwqFAjExMdiyZQvUanXAy87JycHXX3+NMWPGgDEGjuOwYcMGbN26Nej37avs\n1atX46233gr6fTc2NiIrKwu3b9+G3W7HihUrMGbMGL8/X4EsOy0tDXFxcdiyZUvQ79tdamoqXn/9\ndXAcF9TPea8NuoQQ0hv12u4FQgjpjSjoEkKIhCjoEkKIhCjoEkKIhCjoEkKIhCjoEkKIhCjokh7H\nYDDgxRdf7PC/KywsxK5du9o85/Tp01i6dGlnq+ZTWVkZNmzYELTrk76Fgi7pcerq6lBcXNzhf/fz\nzz+jurq63fM4jutMtVpVXl4OnU4XtOuTvoWCLulxcnJyUFVVhYyMDPzjH//AE088gfnz52Pjxo2w\nWq0oKipCcnIyamtrUVdXh5kzZ6K4uBjvvPMOjh49ivfff9+vckpLS/Hcc8/hiSeewNNPPy0G+qys\nLOTk5GDx4sV46KGHcPDgQQDOFvgLL7yAefPmIT09HfPnz0d5eTlycnJw4cIFMVNWTU0N0tLS8Mgj\nj+CFF16AzWYLzhtFeqcOJZ4kRAJlZWVs5syZ7OrVq2zx4sXMYrEwxhh7++232Z///GfGGGPvvvsu\nW79+PVuzZg378MMPGWOMHTx4UMxr3JpTp06xpUuXMsYY+93vfscuXbrEGGPs6tWr7OGHH2aMMbZ+\n/XqWkZHBGGOspKSETZkyhTHG2Pbt29nOnTsZY4ydP3+eTZgwgZWXl3tc89SpU2zy5MmsvLycMcZY\nSkoK++677wLzxpA+oVdkGSP9D2MMP/74I27cuIGFCxeCMQa73Y4JEyYAANLT07FgwQKEhoa224/r\ni8lkwvnz55GVlSWm9DObzbhz5w4AiPmXx40bh/r6egDAyZMn8fbbbwMAJk6ciHvuucfntcePH4+h\nQ4cCAMaOHRvQ3M2k96OgS3osQRAwZ84ccZCqsbERDocDgDPhkdFohMlkQl1dXYezxgmCAJVKhS++\n+EI8duvWLURGRgJwZrvyJpPJPHKtslbSlrjnAab+XeKN+nRJj+NKAv/AAw/g8OHDqKmpAWMMmzdv\nxkcffQQA2LJlC5YsWYLFixcjOzsbgDPY2e12v8rQaDQYOXIkvvrqKwBAfn4+lixZ4vNcV3CdOnUq\n/vnPfwIASkpKcOXKFXAcB5lMJv4xIKQ9FHRJjzNw4EDExcVh+/btWLVqFZ555hnMmzcPAJCWloav\nv/4aOp0OzzzzDFJTU3Hjxg0cOnQICQkJKCwsxB/+8Ae/ytm5cycOHDiAxx57DH/84x/xpz/9yed5\nrtbqypUrcePGDTz++OPIzc1FTEwMQkJCMHbsWNTX12PdunWBeQNIn0apHQnx01dffYXhw4dj0qRJ\nqKysxNKlS3HkyJHurhbpZahPl/Q5Z86cERNPu7CmpOAffPABYmJiOnXdMWPGYPPmzRAEATKZTJwi\nRkhHUEuXEEIkRH26hBAiIQq6hBAiIQq6hBAiIQq6hBAiIQq6hBAiIQq6hBAiof8PiTg8rAGWtusA\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17e930690>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(data=df, x='text_length', y='net_sentiment')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1652b3610>"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFhCAYAAACh/xvXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1652b3a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Weak positive relation between stars and net_sentiment, IMO\n",
    "sns.lmplot(data=df, x='stars', y='net_sentiment')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df['sent_per_token'] = df.net_sentiment/df.text_length"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>stars</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th>text_length</th>\n",
       "      <th>sent_per_token</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>229875.000000</td>\n",
       "      <td>229875.000000</td>\n",
       "      <td>229875.000000</td>\n",
       "      <td>229875.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3.766808</td>\n",
       "      <td>12.055896</td>\n",
       "      <td>426.355998</td>\n",
       "      <td>0.039002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.216994</td>\n",
       "      <td>11.453223</td>\n",
       "      <td>372.945316</td>\n",
       "      <td>0.042616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-170.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>-1.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>175.000000</td>\n",
       "      <td>0.015974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>324.000000</td>\n",
       "      <td>0.032407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>558.000000</td>\n",
       "      <td>0.053571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>141.000000</td>\n",
       "      <td>3603.000000</td>\n",
       "      <td>1.333333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               stars  net_sentiment    text_length  sent_per_token\n",
       "count  229875.000000  229875.000000  229875.000000   229875.000000\n",
       "mean        3.766808      12.055896     426.355998        0.039002\n",
       "std         1.216994      11.453223     372.945316        0.042616\n",
       "min         1.000000    -170.000000       3.000000       -1.333333\n",
       "25%         3.000000       5.000000     175.000000        0.015974\n",
       "50%         4.000000      10.000000     324.000000        0.032407\n",
       "75%         5.000000      17.000000     558.000000        0.053571\n",
       "max         5.000000     141.000000    3603.000000        1.333333"
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x19ab9fa90>"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFhCAYAAACh/xvXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niASkEBYRCUghLCIS0P8H9aKGNBOdZvgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x19ad74f50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(data=df, x='stars', y='sent_per_token')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# a fast way to save a big df --to msgpack\n",
    "df.to_msgpack(\"data/yelp_df_tokens_sentiment.msg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_msgpack(\"data/yelp_df_tokens_sentiment.msg\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prep for Search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lynn/miniconda/envs/esnlp/lib/python2.7/site-packages/numpy/lib/function_base.py:3823: RuntimeWarning: Invalid value encountered in percentile\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>stars</th>\n",
       "      <th>text_length</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th>sent_per_token</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>229907.000000</td>\n",
       "      <td>229907.000000</td>\n",
       "      <td>229907.000000</td>\n",
       "      <td>229875.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3.766723</td>\n",
       "      <td>426.296655</td>\n",
       "      <td>12.054218</td>\n",
       "      <td>0.039002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.217010</td>\n",
       "      <td>372.953278</td>\n",
       "      <td>11.453309</td>\n",
       "      <td>0.042616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-170.000000</td>\n",
       "      <td>-1.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>175.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>324.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>558.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>3603.000000</td>\n",
       "      <td>141.000000</td>\n",
       "      <td>1.333333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               stars    text_length  net_sentiment  sent_per_token\n",
       "count  229907.000000  229907.000000  229907.000000   229875.000000\n",
       "mean        3.766723     426.296655      12.054218        0.039002\n",
       "std         1.217010     372.953278      11.453309        0.042616\n",
       "min         1.000000       0.000000    -170.000000       -1.333333\n",
       "25%         3.000000     175.000000       5.000000             NaN\n",
       "50%         4.000000     324.000000      10.000000             NaN\n",
       "75%         5.000000     558.000000      17.000000             NaN\n",
       "max         5.000000    3603.000000     141.000000        1.333333"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "      <th>tokens</th>\n",
       "      <th>text_length</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th>sent_per_token</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4851</th>\n",
       "      <td>kYkuqU9wJS01L8q4d2nTTg</td>\n",
       "      <td>2010-09-08</td>\n",
       "      <td>RNow4E11pxrE3Davw9Gynw</td>\n",
       "      <td>1</td>\n",
       "      <td>Just don't.</td>\n",
       "      <td>f3LA83yEEBMj9q92H28O7w</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6450</th>\n",
       "      <td>tbRu-3NR5za1SEZkszcp5Q</td>\n",
       "      <td>2012-01-03</td>\n",
       "      <td>IVoMGfn9QrZ4dATwc9HBcg</td>\n",
       "      <td>3</td>\n",
       "      <td>X</td>\n",
       "      <td>SHcX9hp5RG3-OqsbvissIQ</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6708</th>\n",
       "      <td>KGX7O-_WqOIy9o7u9NOa9A</td>\n",
       "      <td>2011-04-19</td>\n",
       "      <td>EdFx6mNJDfJWP6yDO42cUQ</td>\n",
       "      <td>4</td>\n",
       "      <td>Go</td>\n",
       "      <td>vnSz2VT6s-gdRBjlFDACYg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22740</th>\n",
       "      <td>B8ujMtvvpHyEQ2r_QlAT2w</td>\n",
       "      <td>2008-05-06</td>\n",
       "      <td>PY-Eir6BUMMfw1u0_OipiA</td>\n",
       "      <td>4</td>\n",
       "      <td></td>\n",
       "      <td>il6RQX17f_Qxs-pssa5zMg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41989</th>\n",
       "      <td>FeI75xIG8PF_XZ6P80gLBQ</td>\n",
       "      <td>2009-04-13</td>\n",
       "      <td>v-eOjkSTppvRxlB00LJAqg</td>\n",
       "      <td>3</td>\n",
       "      <td>eh...</td>\n",
       "      <td>oC0kwGvgaAeT3ZIxdxTcJA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43522</th>\n",
       "      <td>nZBAnK1rOmV7bLVVDPJs0A</td>\n",
       "      <td>2011-02-26</td>\n",
       "      <td>X7aXNuBFUKvx2O3kLM4hfA</td>\n",
       "      <td>3</td>\n",
       "      <td>eh it's chili's..</td>\n",
       "      <td>3Zi1udoY2iT9s8Q3X6tgZQ</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54130</th>\n",
       "      <td>I6yAKGBDclGh7UlvOfqV1A</td>\n",
       "      <td>2011-12-11</td>\n",
       "      <td>TZjjbt_bskQLDnSSF7uKFg</td>\n",
       "      <td>5</td>\n",
       "      <td>A+</td>\n",
       "      <td>9VmTOyq01oIUk5zuxOj1GA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69503</th>\n",
       "      <td>LfKPiwth-lGmCKTxlDoV9g</td>\n",
       "      <td>2008-05-06</td>\n",
       "      <td>NOhzie7YJMprviyeaEh_AQ</td>\n",
       "      <td>3</td>\n",
       "      <td></td>\n",
       "      <td>il6RQX17f_Qxs-pssa5zMg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72493</th>\n",
       "      <td>tbrFsCEUuq2S5lraOln4nQ</td>\n",
       "      <td>2012-08-13</td>\n",
       "      <td>rpm84EXAPoSbHEcD_JZYNA</td>\n",
       "      <td>3</td>\n",
       "      <td>Its ok.</td>\n",
       "      <td>PXUZAP_Y7YgXrDNKRpOLwA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74314</th>\n",
       "      <td>ZPkomfeRH2pflacWf7U3-A</td>\n",
       "      <td>2012-11-23</td>\n",
       "      <td>lbCGla9eVmJ0JEeLiUkrHg</td>\n",
       "      <td>5</td>\n",
       "      <td>A</td>\n",
       "      <td>qDBaslzhTlpzOPg7hARuzA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76155</th>\n",
       "      <td>s8cHXeSlX_aPI2x49UkCCg</td>\n",
       "      <td>2011-01-31</td>\n",
       "      <td>FYS4T68CvB__Z7L7jD65nw</td>\n",
       "      <td>2</td>\n",
       "      <td>Just ok!</td>\n",
       "      <td>zqfNAOCE1s_z-21Qs1wdeQ</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84811</th>\n",
       "      <td>Q5ZD7ZAs96NAVblFDV476g</td>\n",
       "      <td>2012-11-25</td>\n",
       "      <td>9YvOKZiNV2jip4FVCIwHRg</td>\n",
       "      <td>3</td>\n",
       "      <td>Was ok.</td>\n",
       "      <td>e-89w7FUWyUKgAELA75E3w</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85530</th>\n",
       "      <td>FRiT2RI6Y1nds1jjZqvnXQ</td>\n",
       "      <td>2012-07-17</td>\n",
       "      <td>4D-sxk_ivBYD6bHKtWM80g</td>\n",
       "      <td>3</td>\n",
       "      <td>just ok</td>\n",
       "      <td>1Ge4YW-QzFDd4xqFNrPEbQ</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87378</th>\n",
       "      <td>-4A5xmN21zi_TXnUESauUQ</td>\n",
       "      <td>2012-08-12</td>\n",
       "      <td>9tFeEejWpGSxAZEuYFagUA</td>\n",
       "      <td>3</td>\n",
       "      <td>So so</td>\n",
       "      <td>bKMFX12ewC0oXRX_n-6p0w</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90478</th>\n",
       "      <td>xcOncADGPr9eki8OU5Ln7g</td>\n",
       "      <td>2008-03-17</td>\n",
       "      <td>K7fQAOToFP_pXiE7f5OLUg</td>\n",
       "      <td>3</td>\n",
       "      <td></td>\n",
       "      <td>OLYVUMDfGUlBbXfl5Vcy9w</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122118</th>\n",
       "      <td>Oj79-vTnjQk_jLr6oNbT0Q</td>\n",
       "      <td>2012-08-12</td>\n",
       "      <td>Y3GCdXKyordRyCNfLaYLrw</td>\n",
       "      <td>3</td>\n",
       "      <td>So so</td>\n",
       "      <td>bKMFX12ewC0oXRX_n-6p0w</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128357</th>\n",
       "      <td>fF6m3qsD5blnwuZRuYhzWg</td>\n",
       "      <td>2008-05-06</td>\n",
       "      <td>w8LVmdgVqdSYsuPR0RoZZA</td>\n",
       "      <td>4</td>\n",
       "      <td></td>\n",
       "      <td>il6RQX17f_Qxs-pssa5zMg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157031</th>\n",
       "      <td>SMaed6DT2GCRlM-kFIl-eQ</td>\n",
       "      <td>2008-01-19</td>\n",
       "      <td>0g3vPgL4oW7sBxZA6R2ZXQ</td>\n",
       "      <td>5</td>\n",
       "      <td></td>\n",
       "      <td>_PrMbW8gLcU8R3aUzJ9Skg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160087</th>\n",
       "      <td>Ez2MFsVT5lJZ05yvK_0AXQ</td>\n",
       "      <td>2012-12-16</td>\n",
       "      <td>0RHjhvYSUB3q1Brl8_WR0Q</td>\n",
       "      <td>3</td>\n",
       "      <td>Just OK</td>\n",
       "      <td>bXpfj8725k6deWK6PZD-qw</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>163915</th>\n",
       "      <td>0j2UisBk89SHqihr0IdlJw</td>\n",
       "      <td>2011-12-18</td>\n",
       "      <td>5gGBQLW56ZQVhV_l3uJR_Q</td>\n",
       "      <td>3</td>\n",
       "      <td>just ok</td>\n",
       "      <td>i5UBvz_VPvAEFRMYjtqGSA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180691</th>\n",
       "      <td>Scy5uuSnzk4JcR8h6lctxQ</td>\n",
       "      <td>2011-07-17</td>\n",
       "      <td>KGTpZuHNazjrzU4HLamaLQ</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>I3xKqgX4XDiz72iUO_lN1Q</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185612</th>\n",
       "      <td>Bc4DoKgrKCtCuN-0O5He3A</td>\n",
       "      <td>2009-05-04</td>\n",
       "      <td>6ox90kpJwj_wYNUVIOQ4wg</td>\n",
       "      <td>5</td>\n",
       "      <td>mmmmmmmmmmmmmmmmmmmmmmmmmmwowmmmmmmmmmmmmmmmmm...</td>\n",
       "      <td>P2kVk4cIWyK4e4h14RhK-Q</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187880</th>\n",
       "      <td>lT3yCGUc-LQuphjSz1uoIw</td>\n",
       "      <td>2009-08-23</td>\n",
       "      <td>swBAQcYgeHdVOVLL21KNaQ</td>\n",
       "      <td>1</td>\n",
       "      <td>it is what it is</td>\n",
       "      <td>hIWlVkf-N1Yzj7I-dfF7qA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>199901</th>\n",
       "      <td>CFuzIBaREsa0nunnI_FFYg</td>\n",
       "      <td>2012-04-11</td>\n",
       "      <td>ZPgfMcrIkfh_6kYXUi3WVg</td>\n",
       "      <td>2</td>\n",
       "      <td>Just ok</td>\n",
       "      <td>BaHlhZ4dUO1AUY87ZXY6Tg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>208880</th>\n",
       "      <td>YnWb8fRudnNkXc-VjrmKGg</td>\n",
       "      <td>2013-01-01</td>\n",
       "      <td>58M4NbRF-b0H_mpJpSv-jg</td>\n",
       "      <td>3</td>\n",
       "      <td>Its ok.</td>\n",
       "      <td>vlaGl8X9yT9p8jltCk6ipg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>210488</th>\n",
       "      <td>RhrrkjyoAUZaZcV21bEteg</td>\n",
       "      <td>2012-02-07</td>\n",
       "      <td>5IiVhzGu5wH4Pk2FYs3fBw</td>\n",
       "      <td>1</td>\n",
       "      <td>Don't</td>\n",
       "      <td>-YtyX6MBmqRjz0C1V7cJyQ</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>211935</th>\n",
       "      <td>uyuiIvQ3aWZxc1-2i_dQtQ</td>\n",
       "      <td>2012-12-21</td>\n",
       "      <td>QCh9o0tVPIfPLxuAWm6sMQ</td>\n",
       "      <td>3</td>\n",
       "      <td>Was ok</td>\n",
       "      <td>9VmTOyq01oIUk5zuxOj1GA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>217771</th>\n",
       "      <td>8rEzGHaBgQaHT1v9nb4rNg</td>\n",
       "      <td>2012-02-07</td>\n",
       "      <td>fe_FmNKL8Xmu5Q7EoGVqMA</td>\n",
       "      <td>1</td>\n",
       "      <td>Don't</td>\n",
       "      <td>-YtyX6MBmqRjz0C1V7cJyQ</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>221155</th>\n",
       "      <td>oXKPSI-RUqOvmuSCh_DEQQ</td>\n",
       "      <td>2010-09-04</td>\n",
       "      <td>oSeLqyidjAPPniogkSenKw</td>\n",
       "      <td>5</td>\n",
       "      <td>Oh my!!!!</td>\n",
       "      <td>HquCK6xTmqpybGrysHo7Bw</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223001</th>\n",
       "      <td>b5cEoKR8iQliq-yT2_O0LQ</td>\n",
       "      <td>2011-10-27</td>\n",
       "      <td>GyJRDZibDV9-duNRg8I_HQ</td>\n",
       "      <td>4</td>\n",
       "      <td>Go!</td>\n",
       "      <td>Qmz4VWrpMb-Bg6qe4_Y3Sg</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>225979</th>\n",
       "      <td>i41M_To0DAxuzZIGvU1wyg</td>\n",
       "      <td>2012-08-11</td>\n",
       "      <td>Y8BwS2Ov83bQPjtiyxPnlg</td>\n",
       "      <td>2</td>\n",
       "      <td>Just so so!</td>\n",
       "      <td>acDuLeG4dALgdXMqIF4IkQ</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>229313</th>\n",
       "      <td>SMpL3z4FLF07bRA6-y22JQ</td>\n",
       "      <td>2007-07-25</td>\n",
       "      <td>bY6XggJUzpO6SNUOaS3zaw</td>\n",
       "      <td>4</td>\n",
       "      <td></td>\n",
       "      <td>b5ksiCn6IOi8Dmfp4EH5tA</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   business_id        date               review_id  stars  \\\n",
       "4851    kYkuqU9wJS01L8q4d2nTTg  2010-09-08  RNow4E11pxrE3Davw9Gynw      1   \n",
       "6450    tbRu-3NR5za1SEZkszcp5Q  2012-01-03  IVoMGfn9QrZ4dATwc9HBcg      3   \n",
       "6708    KGX7O-_WqOIy9o7u9NOa9A  2011-04-19  EdFx6mNJDfJWP6yDO42cUQ      4   \n",
       "22740   B8ujMtvvpHyEQ2r_QlAT2w  2008-05-06  PY-Eir6BUMMfw1u0_OipiA      4   \n",
       "41989   FeI75xIG8PF_XZ6P80gLBQ  2009-04-13  v-eOjkSTppvRxlB00LJAqg      3   \n",
       "43522   nZBAnK1rOmV7bLVVDPJs0A  2011-02-26  X7aXNuBFUKvx2O3kLM4hfA      3   \n",
       "54130   I6yAKGBDclGh7UlvOfqV1A  2011-12-11  TZjjbt_bskQLDnSSF7uKFg      5   \n",
       "69503   LfKPiwth-lGmCKTxlDoV9g  2008-05-06  NOhzie7YJMprviyeaEh_AQ      3   \n",
       "72493   tbrFsCEUuq2S5lraOln4nQ  2012-08-13  rpm84EXAPoSbHEcD_JZYNA      3   \n",
       "74314   ZPkomfeRH2pflacWf7U3-A  2012-11-23  lbCGla9eVmJ0JEeLiUkrHg      5   \n",
       "76155   s8cHXeSlX_aPI2x49UkCCg  2011-01-31  FYS4T68CvB__Z7L7jD65nw      2   \n",
       "84811   Q5ZD7ZAs96NAVblFDV476g  2012-11-25  9YvOKZiNV2jip4FVCIwHRg      3   \n",
       "85530   FRiT2RI6Y1nds1jjZqvnXQ  2012-07-17  4D-sxk_ivBYD6bHKtWM80g      3   \n",
       "87378   -4A5xmN21zi_TXnUESauUQ  2012-08-12  9tFeEejWpGSxAZEuYFagUA      3   \n",
       "90478   xcOncADGPr9eki8OU5Ln7g  2008-03-17  K7fQAOToFP_pXiE7f5OLUg      3   \n",
       "122118  Oj79-vTnjQk_jLr6oNbT0Q  2012-08-12  Y3GCdXKyordRyCNfLaYLrw      3   \n",
       "128357  fF6m3qsD5blnwuZRuYhzWg  2008-05-06  w8LVmdgVqdSYsuPR0RoZZA      4   \n",
       "157031  SMaed6DT2GCRlM-kFIl-eQ  2008-01-19  0g3vPgL4oW7sBxZA6R2ZXQ      5   \n",
       "160087  Ez2MFsVT5lJZ05yvK_0AXQ  2012-12-16  0RHjhvYSUB3q1Brl8_WR0Q      3   \n",
       "163915  0j2UisBk89SHqihr0IdlJw  2011-12-18  5gGBQLW56ZQVhV_l3uJR_Q      3   \n",
       "180691  Scy5uuSnzk4JcR8h6lctxQ  2011-07-17  KGTpZuHNazjrzU4HLamaLQ      4   \n",
       "185612  Bc4DoKgrKCtCuN-0O5He3A  2009-05-04  6ox90kpJwj_wYNUVIOQ4wg      5   \n",
       "187880  lT3yCGUc-LQuphjSz1uoIw  2009-08-23  swBAQcYgeHdVOVLL21KNaQ      1   \n",
       "199901  CFuzIBaREsa0nunnI_FFYg  2012-04-11  ZPgfMcrIkfh_6kYXUi3WVg      2   \n",
       "208880  YnWb8fRudnNkXc-VjrmKGg  2013-01-01  58M4NbRF-b0H_mpJpSv-jg      3   \n",
       "210488  RhrrkjyoAUZaZcV21bEteg  2012-02-07  5IiVhzGu5wH4Pk2FYs3fBw      1   \n",
       "211935  uyuiIvQ3aWZxc1-2i_dQtQ  2012-12-21  QCh9o0tVPIfPLxuAWm6sMQ      3   \n",
       "217771  8rEzGHaBgQaHT1v9nb4rNg  2012-02-07  fe_FmNKL8Xmu5Q7EoGVqMA      1   \n",
       "221155  oXKPSI-RUqOvmuSCh_DEQQ  2010-09-04  oSeLqyidjAPPniogkSenKw      5   \n",
       "223001  b5cEoKR8iQliq-yT2_O0LQ  2011-10-27  GyJRDZibDV9-duNRg8I_HQ      4   \n",
       "225979  i41M_To0DAxuzZIGvU1wyg  2012-08-11  Y8BwS2Ov83bQPjtiyxPnlg      2   \n",
       "229313  SMpL3z4FLF07bRA6-y22JQ  2007-07-25  bY6XggJUzpO6SNUOaS3zaw      4   \n",
       "\n",
       "                                                     text  \\\n",
       "4851                                          Just don't.   \n",
       "6450                                                    X   \n",
       "6708                                                   Go   \n",
       "22740                                                       \n",
       "41989                                               eh...   \n",
       "43522                                   eh it's chili's..   \n",
       "54130                                                  A+   \n",
       "69503                                                       \n",
       "72493                                             Its ok.   \n",
       "74314                                                   A   \n",
       "76155                                            Just ok!   \n",
       "84811                                             Was ok.   \n",
       "85530                                             just ok   \n",
       "87378                                               So so   \n",
       "90478                                                       \n",
       "122118                                              So so   \n",
       "128357                                                      \n",
       "157031                                                      \n",
       "160087                                            Just OK   \n",
       "163915                                            just ok   \n",
       "180691                                                ...   \n",
       "185612  mmmmmmmmmmmmmmmmmmmmmmmmmmwowmmmmmmmmmmmmmmmmm...   \n",
       "187880                                   it is what it is   \n",
       "199901                                            Just ok   \n",
       "208880                                            Its ok.   \n",
       "210488                                              Don't   \n",
       "211935                                             Was ok   \n",
       "217771                                              Don't   \n",
       "221155                                          Oh my!!!!   \n",
       "223001                                                Go!   \n",
       "225979                                        Just so so!   \n",
       "229313                                                      \n",
       "\n",
       "                       user_id tokens  text_length  net_sentiment  \\\n",
       "4851    f3LA83yEEBMj9q92H28O7w                   0            0.0   \n",
       "6450    SHcX9hp5RG3-OqsbvissIQ                   0            0.0   \n",
       "6708    vnSz2VT6s-gdRBjlFDACYg                   0            0.0   \n",
       "22740   il6RQX17f_Qxs-pssa5zMg                   0            0.0   \n",
       "41989   oC0kwGvgaAeT3ZIxdxTcJA                   0            0.0   \n",
       "43522   3Zi1udoY2iT9s8Q3X6tgZQ                   0            0.0   \n",
       "54130   9VmTOyq01oIUk5zuxOj1GA                   0            0.0   \n",
       "69503   il6RQX17f_Qxs-pssa5zMg                   0            0.0   \n",
       "72493   PXUZAP_Y7YgXrDNKRpOLwA                   0            0.0   \n",
       "74314   qDBaslzhTlpzOPg7hARuzA                   0            0.0   \n",
       "76155   zqfNAOCE1s_z-21Qs1wdeQ                   0            0.0   \n",
       "84811   e-89w7FUWyUKgAELA75E3w                   0            0.0   \n",
       "85530   1Ge4YW-QzFDd4xqFNrPEbQ                   0            0.0   \n",
       "87378   bKMFX12ewC0oXRX_n-6p0w                   0            0.0   \n",
       "90478   OLYVUMDfGUlBbXfl5Vcy9w                   0            0.0   \n",
       "122118  bKMFX12ewC0oXRX_n-6p0w                   0            0.0   \n",
       "128357  il6RQX17f_Qxs-pssa5zMg                   0            0.0   \n",
       "157031  _PrMbW8gLcU8R3aUzJ9Skg                   0            0.0   \n",
       "160087  bXpfj8725k6deWK6PZD-qw                   0            0.0   \n",
       "163915  i5UBvz_VPvAEFRMYjtqGSA                   0            0.0   \n",
       "180691  I3xKqgX4XDiz72iUO_lN1Q                   0            0.0   \n",
       "185612  P2kVk4cIWyK4e4h14RhK-Q                   0            0.0   \n",
       "187880  hIWlVkf-N1Yzj7I-dfF7qA                   0            0.0   \n",
       "199901  BaHlhZ4dUO1AUY87ZXY6Tg                   0            0.0   \n",
       "208880  vlaGl8X9yT9p8jltCk6ipg                   0            0.0   \n",
       "210488  -YtyX6MBmqRjz0C1V7cJyQ                   0            0.0   \n",
       "211935  9VmTOyq01oIUk5zuxOj1GA                   0            0.0   \n",
       "217771  -YtyX6MBmqRjz0C1V7cJyQ                   0            0.0   \n",
       "221155  HquCK6xTmqpybGrysHo7Bw                   0            0.0   \n",
       "223001  Qmz4VWrpMb-Bg6qe4_Y3Sg                   0            0.0   \n",
       "225979  acDuLeG4dALgdXMqIF4IkQ                   0            0.0   \n",
       "229313  b5ksiCn6IOi8Dmfp4EH5tA                   0            0.0   \n",
       "\n",
       "        sent_per_token  \n",
       "4851               NaN  \n",
       "6450               NaN  \n",
       "6708               NaN  \n",
       "22740              NaN  \n",
       "41989              NaN  \n",
       "43522              NaN  \n",
       "54130              NaN  \n",
       "69503              NaN  \n",
       "72493              NaN  \n",
       "74314              NaN  \n",
       "76155              NaN  \n",
       "84811              NaN  \n",
       "85530              NaN  \n",
       "87378              NaN  \n",
       "90478              NaN  \n",
       "122118             NaN  \n",
       "128357             NaN  \n",
       "157031             NaN  \n",
       "160087             NaN  \n",
       "163915             NaN  \n",
       "180691             NaN  \n",
       "185612             NaN  \n",
       "187880             NaN  \n",
       "199901             NaN  \n",
       "208880             NaN  \n",
       "210488             NaN  \n",
       "211935             NaN  \n",
       "217771             NaN  \n",
       "221155             NaN  \n",
       "223001             NaN  \n",
       "225979             NaN  \n",
       "229313             NaN  "
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# You could filter these out.  But be aware of what you are losing (the A+ and the \"go\")\n",
    "df.query('text_length == 0')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "229875"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Delete the rows with 0 text length, yes, this removes the ones with some text in it that doesn't tokenize :(\n",
    "dfes = df[df.text_length != 0]\n",
    "\n",
    "len(dfes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# cut text off at 558 (75%ile) for the index\n",
    "dfes = df[df.text_length <= 558]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "172474"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(dfes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dfes.to_msgpack(\"data/yelp_filtered_length.msg\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Work on Business Index Now..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>sent_per_token</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th colspan=\"3\" halign=\"left\">stars</th>\n",
       "      <th>text_length</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>len</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>median</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>hW0Ne_HTHEAgGF1rAdmR-g</th>\n",
       "      <td>693</td>\n",
       "      <td>0.015385</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.279942</td>\n",
       "      <td>1.018394</td>\n",
       "      <td>231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JokKtdXU7zXHcr20Lrk29A</th>\n",
       "      <td>571</td>\n",
       "      <td>0.052419</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.352014</td>\n",
       "      <td>0.756545</td>\n",
       "      <td>231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VVeogjZya58oiTxK7qUjAQ</th>\n",
       "      <td>486</td>\n",
       "      <td>0.034909</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.952675</td>\n",
       "      <td>1.259939</td>\n",
       "      <td>279.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ntN85eu27C04nwyPa8IHtw</th>\n",
       "      <td>471</td>\n",
       "      <td>0.038462</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.023355</td>\n",
       "      <td>1.121116</td>\n",
       "      <td>253.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWMwV5V9BxNs_U6nNVMeqw</th>\n",
       "      <td>458</td>\n",
       "      <td>0.046729</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.157205</td>\n",
       "      <td>0.952575</td>\n",
       "      <td>259.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SDwYQ6eSu1htn8vHWv128g</th>\n",
       "      <td>439</td>\n",
       "      <td>0.050269</td>\n",
       "      <td>12.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.555809</td>\n",
       "      <td>0.641742</td>\n",
       "      <td>253.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V1nEpIRmEa1768oj_tuxeQ</th>\n",
       "      <td>438</td>\n",
       "      <td>0.049184</td>\n",
       "      <td>11.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.399543</td>\n",
       "      <td>0.807686</td>\n",
       "      <td>263.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WNy1uzcmm_UHmTyR--o5IA</th>\n",
       "      <td>433</td>\n",
       "      <td>0.040254</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.408776</td>\n",
       "      <td>0.797343</td>\n",
       "      <td>263.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>QnAzW6KMSciUcuJ20oI3Bw</th>\n",
       "      <td>394</td>\n",
       "      <td>0.047080</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.258883</td>\n",
       "      <td>0.887565</td>\n",
       "      <td>236.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OFBJtqWGSx6n6CchtVmoRw</th>\n",
       "      <td>384</td>\n",
       "      <td>0.046154</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.921875</td>\n",
       "      <td>1.003462</td>\n",
       "      <td>257.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-sC66z4SO3tR7nFCjfQwuQ</th>\n",
       "      <td>357</td>\n",
       "      <td>0.045455</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.226891</td>\n",
       "      <td>0.881492</td>\n",
       "      <td>245.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c1yGkETheht_1vjda7G5sA</th>\n",
       "      <td>350</td>\n",
       "      <td>0.039239</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.380000</td>\n",
       "      <td>0.816134</td>\n",
       "      <td>270.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>aRkYtXfmEKYG-eTDf_qUsw</th>\n",
       "      <td>349</td>\n",
       "      <td>0.040000</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.257880</td>\n",
       "      <td>0.957113</td>\n",
       "      <td>248.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pwpl-rxwNRQdgqFz_-qMPg</th>\n",
       "      <td>339</td>\n",
       "      <td>0.048387</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.280236</td>\n",
       "      <td>0.821994</td>\n",
       "      <td>266.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3oZcTGb_oDHGwZFiP-7kxQ</th>\n",
       "      <td>327</td>\n",
       "      <td>0.049281</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.012232</td>\n",
       "      <td>0.943091</td>\n",
       "      <td>266.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K8pM6qQdYu5h6buRE1-_sw</th>\n",
       "      <td>326</td>\n",
       "      <td>0.039106</td>\n",
       "      <td>9.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.319018</td>\n",
       "      <td>0.881994</td>\n",
       "      <td>268.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Zx8_4zKdDBSO3qGrkukBIA</th>\n",
       "      <td>316</td>\n",
       "      <td>0.030632</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.034810</td>\n",
       "      <td>1.104992</td>\n",
       "      <td>254.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R8VwdLyvsp9iybNqRvm94g</th>\n",
       "      <td>307</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.094463</td>\n",
       "      <td>0.943261</td>\n",
       "      <td>278.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>_FXql6eVhbM923RdCi94SA</th>\n",
       "      <td>304</td>\n",
       "      <td>0.053964</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.115132</td>\n",
       "      <td>0.876858</td>\n",
       "      <td>265.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3n9mSKySEv3G03YjcU-YOQ</th>\n",
       "      <td>297</td>\n",
       "      <td>0.050109</td>\n",
       "      <td>11.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.380471</td>\n",
       "      <td>0.757857</td>\n",
       "      <td>248.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qjmCVYkwP-HDa35jwYucbQ</th>\n",
       "      <td>291</td>\n",
       "      <td>0.047244</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.982818</td>\n",
       "      <td>1.015253</td>\n",
       "      <td>250.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YKOvlBNkF4KpUP9q7x862w</th>\n",
       "      <td>290</td>\n",
       "      <td>0.050898</td>\n",
       "      <td>12.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.158621</td>\n",
       "      <td>1.016649</td>\n",
       "      <td>248.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5VMajxduxfLPSyQkVFKESg</th>\n",
       "      <td>279</td>\n",
       "      <td>0.044728</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.057348</td>\n",
       "      <td>1.061224</td>\n",
       "      <td>244.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>uKSX1n1RoAzGq4bV8GPHVg</th>\n",
       "      <td>278</td>\n",
       "      <td>0.041877</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.093525</td>\n",
       "      <td>0.979148</td>\n",
       "      <td>288.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>L9UYbtAUOcfTgZFimehlXw</th>\n",
       "      <td>275</td>\n",
       "      <td>0.046154</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.385455</td>\n",
       "      <td>0.894308</td>\n",
       "      <td>238.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AqbgC7Gul5Es1rRzGNLDFA</th>\n",
       "      <td>274</td>\n",
       "      <td>0.034822</td>\n",
       "      <td>8.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.543796</td>\n",
       "      <td>1.192121</td>\n",
       "      <td>248.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>xY1sPHTA2RGVFlh5tZhs9g</th>\n",
       "      <td>262</td>\n",
       "      <td>0.045014</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.778626</td>\n",
       "      <td>0.973126</td>\n",
       "      <td>268.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>uEJQSIjWui-TDWXaGlcqyQ</th>\n",
       "      <td>258</td>\n",
       "      <td>0.038529</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.019380</td>\n",
       "      <td>1.038000</td>\n",
       "      <td>282.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rZbHg4ACfN3iShdsT47WKQ</th>\n",
       "      <td>255</td>\n",
       "      <td>0.048387</td>\n",
       "      <td>12.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.419608</td>\n",
       "      <td>0.837370</td>\n",
       "      <td>231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53YGfwmbW73JhFiemNeyzQ</th>\n",
       "      <td>251</td>\n",
       "      <td>0.049618</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.924303</td>\n",
       "      <td>0.978901</td>\n",
       "      <td>271.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bNGR0iQTx8hXRmQv4hV5gw</th>\n",
       "      <td>1</td>\n",
       "      <td>0.037594</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>266.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GE0_TCcCbviNBQhtbNOcfg</th>\n",
       "      <td>1</td>\n",
       "      <td>0.058140</td>\n",
       "      <td>20.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>344.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ATk3ShTVfXD35kfMELPH7Q</th>\n",
       "      <td>1</td>\n",
       "      <td>0.049383</td>\n",
       "      <td>24.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>486.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OnRGW_PdhrO95s8-I7QL4g</th>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>146.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pvJuii1d2CTQd9BF3gT_Sg</th>\n",
       "      <td>1</td>\n",
       "      <td>0.131148</td>\n",
       "      <td>8.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>61.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pvukd8Pn1lNbzezXmGyx-Q</th>\n",
       "      <td>1</td>\n",
       "      <td>0.026829</td>\n",
       "      <td>11.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>410.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6ZLsBqO2cHECURCK8LeFsQ</th>\n",
       "      <td>1</td>\n",
       "      <td>0.051402</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>214.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mvGW6YW6LNULvdASs8IUCQ</th>\n",
       "      <td>1</td>\n",
       "      <td>0.011261</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>444.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37TCTrXk2kFZ07hSSd5PHQ</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.032520</td>\n",
       "      <td>-8.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>246.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>y09589YKY953yYzK4JxcRg</th>\n",
       "      <td>1</td>\n",
       "      <td>0.003846</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>260.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KqYY2rwD8ymjU9mVJwiuRQ</th>\n",
       "      <td>1</td>\n",
       "      <td>0.263158</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ERgRiC6UDxwSHSynlvmJtQ</th>\n",
       "      <td>1</td>\n",
       "      <td>0.012945</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>309.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vJOzlJdQjNT4YIh3KriqsA</th>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KtsxJfAcDoNjhmTZeZtFmw</th>\n",
       "      <td>1</td>\n",
       "      <td>0.022951</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>305.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OhPg5mP2v6CseLpIw7FlGg</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.001980</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>505.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bLbZRs9Kcy1aXsdRWYJIgg</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.012422</td>\n",
       "      <td>-4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>322.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XLm6u1RZ7540tiFmOAoH4g</th>\n",
       "      <td>1</td>\n",
       "      <td>0.068182</td>\n",
       "      <td>15.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>220.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KwWKimlUB64oJBtYJF7Bow</th>\n",
       "      <td>1</td>\n",
       "      <td>0.017857</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>392.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>xtlgvMFsv2Xpor-Cc-txoQ</th>\n",
       "      <td>1</td>\n",
       "      <td>0.034409</td>\n",
       "      <td>16.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>465.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q7w9tokdouwt0vEEzsRqxg</th>\n",
       "      <td>1</td>\n",
       "      <td>0.031746</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>126.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q8P4EZkDgzHQodbQSbxSiw</th>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bLk3tgJHryMAIQVcT-tXcQ</th>\n",
       "      <td>1</td>\n",
       "      <td>0.066116</td>\n",
       "      <td>8.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-njvGjJacPJMJ9YWSr6G3Q</th>\n",
       "      <td>1</td>\n",
       "      <td>0.039286</td>\n",
       "      <td>11.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>280.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-nmr-3l0EmQaZAGEoXgk4w</th>\n",
       "      <td>1</td>\n",
       "      <td>0.072816</td>\n",
       "      <td>30.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>412.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>l_HptXEKW5QdBNzgXtahRQ</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>194.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XQM004tccGb-zLz4huUKQg</th>\n",
       "      <td>1</td>\n",
       "      <td>0.090909</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>33.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>L3qC7bMuRBxqDtUnZ6G0tg</th>\n",
       "      <td>1</td>\n",
       "      <td>0.097561</td>\n",
       "      <td>12.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XNUP4pbOL8oj34wWDPVu5A</th>\n",
       "      <td>1</td>\n",
       "      <td>0.033730</td>\n",
       "      <td>17.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>504.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>T-FM14EE1tt-94eONQEaVA</th>\n",
       "      <td>1</td>\n",
       "      <td>0.015873</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>189.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>iQyh-df2rS2cnEpAjLYLog</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.028571</td>\n",
       "      <td>-6.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>210.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11479 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date sent_per_token net_sentiment  stars            \\\n",
       "                        len         median        median median      mean   \n",
       "business_id                                                                 \n",
       "hW0Ne_HTHEAgGF1rAdmR-g  693       0.015385           3.0    3.0  3.279942   \n",
       "JokKtdXU7zXHcr20Lrk29A  571       0.052419          11.0    4.0  4.352014   \n",
       "VVeogjZya58oiTxK7qUjAQ  486       0.034909           9.0    4.0  3.952675   \n",
       "ntN85eu27C04nwyPa8IHtw  471       0.038462           9.0    4.0  4.023355   \n",
       "EWMwV5V9BxNs_U6nNVMeqw  458       0.046729          11.0    4.0  4.157205   \n",
       "SDwYQ6eSu1htn8vHWv128g  439       0.050269          12.0    5.0  4.555809   \n",
       "V1nEpIRmEa1768oj_tuxeQ  438       0.049184          11.0    5.0  4.399543   \n",
       "WNy1uzcmm_UHmTyR--o5IA  433       0.040254          10.0    5.0  4.408776   \n",
       "QnAzW6KMSciUcuJ20oI3Bw  394       0.047080          11.0    4.0  4.258883   \n",
       "OFBJtqWGSx6n6CchtVmoRw  384       0.046154          10.0    4.0  3.921875   \n",
       "-sC66z4SO3tR7nFCjfQwuQ  357       0.045455          10.0    4.0  4.226891   \n",
       "c1yGkETheht_1vjda7G5sA  350       0.039239          10.0    5.0  4.380000   \n",
       "aRkYtXfmEKYG-eTDf_qUsw  349       0.040000          10.0    5.0  4.257880   \n",
       "pwpl-rxwNRQdgqFz_-qMPg  339       0.048387          11.0    4.0  4.280236   \n",
       "3oZcTGb_oDHGwZFiP-7kxQ  327       0.049281          11.0    4.0  4.012232   \n",
       "K8pM6qQdYu5h6buRE1-_sw  326       0.039106           9.0    5.0  4.319018   \n",
       "Zx8_4zKdDBSO3qGrkukBIA  316       0.030632           7.0    4.0  4.034810   \n",
       "R8VwdLyvsp9iybNqRvm94g  307       0.041667           9.0    4.0  4.094463   \n",
       "_FXql6eVhbM923RdCi94SA  304       0.053964          13.0    4.0  4.115132   \n",
       "3n9mSKySEv3G03YjcU-YOQ  297       0.050109          11.0    5.0  4.380471   \n",
       "qjmCVYkwP-HDa35jwYucbQ  291       0.047244          11.0    4.0  3.982818   \n",
       "YKOvlBNkF4KpUP9q7x862w  290       0.050898          12.0    4.0  4.158621   \n",
       "5VMajxduxfLPSyQkVFKESg  279       0.044728          10.0    4.0  4.057348   \n",
       "uKSX1n1RoAzGq4bV8GPHVg  278       0.041877          11.0    4.0  4.093525   \n",
       "L9UYbtAUOcfTgZFimehlXw  275       0.046154          10.0    5.0  4.385455   \n",
       "AqbgC7Gul5Es1rRzGNLDFA  274       0.034822           8.0    4.0  3.543796   \n",
       "xY1sPHTA2RGVFlh5tZhs9g  262       0.045014          10.0    4.0  3.778626   \n",
       "uEJQSIjWui-TDWXaGlcqyQ  258       0.038529          10.0    4.0  4.019380   \n",
       "rZbHg4ACfN3iShdsT47WKQ  255       0.048387          12.0    5.0  4.419608   \n",
       "53YGfwmbW73JhFiemNeyzQ  251       0.049618          13.0    4.0  3.924303   \n",
       "...                     ...            ...           ...    ...       ...   \n",
       "bNGR0iQTx8hXRmQv4hV5gw    1       0.037594          10.0    4.0  4.000000   \n",
       "GE0_TCcCbviNBQhtbNOcfg    1       0.058140          20.0    5.0  5.000000   \n",
       "ATk3ShTVfXD35kfMELPH7Q    1       0.049383          24.0    5.0  5.000000   \n",
       "OnRGW_PdhrO95s8-I7QL4g    1       0.000000           0.0    5.0  5.000000   \n",
       "pvJuii1d2CTQd9BF3gT_Sg    1       0.131148           8.0    5.0  5.000000   \n",
       "pvukd8Pn1lNbzezXmGyx-Q    1       0.026829          11.0    5.0  5.000000   \n",
       "6ZLsBqO2cHECURCK8LeFsQ    1       0.051402          11.0    4.0  4.000000   \n",
       "mvGW6YW6LNULvdASs8IUCQ    1       0.011261           5.0    4.0  4.000000   \n",
       "37TCTrXk2kFZ07hSSd5PHQ    1      -0.032520          -8.0    1.0  1.000000   \n",
       "y09589YKY953yYzK4JxcRg    1       0.003846           1.0    1.0  1.000000   \n",
       "KqYY2rwD8ymjU9mVJwiuRQ    1       0.263158           5.0    5.0  5.000000   \n",
       "ERgRiC6UDxwSHSynlvmJtQ    1       0.012945           4.0    3.0  3.000000   \n",
       "vJOzlJdQjNT4YIh3KriqsA    1       0.000000           0.0    1.0  1.000000   \n",
       "KtsxJfAcDoNjhmTZeZtFmw    1       0.022951           7.0    3.0  3.000000   \n",
       "OhPg5mP2v6CseLpIw7FlGg    1      -0.001980          -1.0    1.0  1.000000   \n",
       "bLbZRs9Kcy1aXsdRWYJIgg    1      -0.012422          -4.0    1.0  1.000000   \n",
       "XLm6u1RZ7540tiFmOAoH4g    1       0.068182          15.0    5.0  5.000000   \n",
       "KwWKimlUB64oJBtYJF7Bow    1       0.017857           7.0    3.0  3.000000   \n",
       "xtlgvMFsv2Xpor-Cc-txoQ    1       0.034409          16.0    5.0  5.000000   \n",
       "q7w9tokdouwt0vEEzsRqxg    1       0.031746           4.0    1.0  1.000000   \n",
       "q8P4EZkDgzHQodbQSbxSiw    1       0.000000           0.0    1.0  1.000000   \n",
       "bLk3tgJHryMAIQVcT-tXcQ    1       0.066116           8.0    4.0  4.000000   \n",
       "-njvGjJacPJMJ9YWSr6G3Q    1       0.039286          11.0    2.0  2.000000   \n",
       "-nmr-3l0EmQaZAGEoXgk4w    1       0.072816          30.0    4.0  4.000000   \n",
       "l_HptXEKW5QdBNzgXtahRQ    1      -0.010309          -2.0    1.0  1.000000   \n",
       "XQM004tccGb-zLz4huUKQg    1       0.090909           3.0    1.0  1.000000   \n",
       "L3qC7bMuRBxqDtUnZ6G0tg    1       0.097561          12.0    4.0  4.000000   \n",
       "XNUP4pbOL8oj34wWDPVu5A    1       0.033730          17.0    4.0  4.000000   \n",
       "T-FM14EE1tt-94eONQEaVA    1       0.015873           3.0    3.0  3.000000   \n",
       "iQyh-df2rS2cnEpAjLYLog    1      -0.028571          -6.0    2.0  2.000000   \n",
       "\n",
       "                                 text_length  \n",
       "                             std      median  \n",
       "business_id                                   \n",
       "hW0Ne_HTHEAgGF1rAdmR-g  1.018394       231.0  \n",
       "JokKtdXU7zXHcr20Lrk29A  0.756545       231.0  \n",
       "VVeogjZya58oiTxK7qUjAQ  1.259939       279.5  \n",
       "ntN85eu27C04nwyPa8IHtw  1.121116       253.0  \n",
       "EWMwV5V9BxNs_U6nNVMeqw  0.952575       259.5  \n",
       "SDwYQ6eSu1htn8vHWv128g  0.641742       253.0  \n",
       "V1nEpIRmEa1768oj_tuxeQ  0.807686       263.0  \n",
       "WNy1uzcmm_UHmTyR--o5IA  0.797343       263.0  \n",
       "QnAzW6KMSciUcuJ20oI3Bw  0.887565       236.5  \n",
       "OFBJtqWGSx6n6CchtVmoRw  1.003462       257.5  \n",
       "-sC66z4SO3tR7nFCjfQwuQ  0.881492       245.0  \n",
       "c1yGkETheht_1vjda7G5sA  0.816134       270.0  \n",
       "aRkYtXfmEKYG-eTDf_qUsw  0.957113       248.0  \n",
       "pwpl-rxwNRQdgqFz_-qMPg  0.821994       266.0  \n",
       "3oZcTGb_oDHGwZFiP-7kxQ  0.943091       266.0  \n",
       "K8pM6qQdYu5h6buRE1-_sw  0.881994       268.5  \n",
       "Zx8_4zKdDBSO3qGrkukBIA  1.104992       254.5  \n",
       "R8VwdLyvsp9iybNqRvm94g  0.943261       278.0  \n",
       "_FXql6eVhbM923RdCi94SA  0.876858       265.0  \n",
       "3n9mSKySEv3G03YjcU-YOQ  0.757857       248.0  \n",
       "qjmCVYkwP-HDa35jwYucbQ  1.015253       250.0  \n",
       "YKOvlBNkF4KpUP9q7x862w  1.016649       248.5  \n",
       "5VMajxduxfLPSyQkVFKESg  1.061224       244.0  \n",
       "uKSX1n1RoAzGq4bV8GPHVg  0.979148       288.5  \n",
       "L9UYbtAUOcfTgZFimehlXw  0.894308       238.0  \n",
       "AqbgC7Gul5Es1rRzGNLDFA  1.192121       248.5  \n",
       "xY1sPHTA2RGVFlh5tZhs9g  0.973126       268.5  \n",
       "uEJQSIjWui-TDWXaGlcqyQ  1.038000       282.5  \n",
       "rZbHg4ACfN3iShdsT47WKQ  0.837370       231.0  \n",
       "53YGfwmbW73JhFiemNeyzQ  0.978901       271.0  \n",
       "...                          ...         ...  \n",
       "bNGR0iQTx8hXRmQv4hV5gw       NaN       266.0  \n",
       "GE0_TCcCbviNBQhtbNOcfg       NaN       344.0  \n",
       "ATk3ShTVfXD35kfMELPH7Q       NaN       486.0  \n",
       "OnRGW_PdhrO95s8-I7QL4g       NaN       146.0  \n",
       "pvJuii1d2CTQd9BF3gT_Sg       NaN        61.0  \n",
       "pvukd8Pn1lNbzezXmGyx-Q       NaN       410.0  \n",
       "6ZLsBqO2cHECURCK8LeFsQ       NaN       214.0  \n",
       "mvGW6YW6LNULvdASs8IUCQ       NaN       444.0  \n",
       "37TCTrXk2kFZ07hSSd5PHQ       NaN       246.0  \n",
       "y09589YKY953yYzK4JxcRg       NaN       260.0  \n",
       "KqYY2rwD8ymjU9mVJwiuRQ       NaN        19.0  \n",
       "ERgRiC6UDxwSHSynlvmJtQ       NaN       309.0  \n",
       "vJOzlJdQjNT4YIh3KriqsA       NaN        12.0  \n",
       "KtsxJfAcDoNjhmTZeZtFmw       NaN       305.0  \n",
       "OhPg5mP2v6CseLpIw7FlGg       NaN       505.0  \n",
       "bLbZRs9Kcy1aXsdRWYJIgg       NaN       322.0  \n",
       "XLm6u1RZ7540tiFmOAoH4g       NaN       220.0  \n",
       "KwWKimlUB64oJBtYJF7Bow       NaN       392.0  \n",
       "xtlgvMFsv2Xpor-Cc-txoQ       NaN       465.0  \n",
       "q7w9tokdouwt0vEEzsRqxg       NaN       126.0  \n",
       "q8P4EZkDgzHQodbQSbxSiw       NaN        17.0  \n",
       "bLk3tgJHryMAIQVcT-tXcQ       NaN       121.0  \n",
       "-njvGjJacPJMJ9YWSr6G3Q       NaN       280.0  \n",
       "-nmr-3l0EmQaZAGEoXgk4w       NaN       412.0  \n",
       "l_HptXEKW5QdBNzgXtahRQ       NaN       194.0  \n",
       "XQM004tccGb-zLz4huUKQg       NaN        33.0  \n",
       "L3qC7bMuRBxqDtUnZ6G0tg       NaN       123.0  \n",
       "XNUP4pbOL8oj34wWDPVu5A       NaN       504.0  \n",
       "T-FM14EE1tt-94eONQEaVA       NaN       189.0  \n",
       "iQyh-df2rS2cnEpAjLYLog       NaN       210.0  \n",
       "\n",
       "[11479 rows x 7 columns]"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz = dfes.groupby(\"business_id\")\n",
    "# count of the dates\n",
    "biz_summary = biz.agg({'date': len, 'stars': [np.median, np.mean, np.std], 'text_length': np.median, 'net_sentiment': np.median, 'sent_per_token': np.median})\n",
    "biz_summary.sort_values([('date','len')], ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>sent_per_token</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th colspan=\"3\" halign=\"left\">stars</th>\n",
       "      <th>text_length</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>len</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>median</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>-8Z6AiHwXInpnl6Q1Ey_5w</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.028986</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>69.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-CZ78c-H3tTxpP-uQ09CWw</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.005970</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>335.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-njvGjJacPJMJ9YWSr6G3Q</th>\n",
       "      <td>1</td>\n",
       "      <td>0.039286</td>\n",
       "      <td>11.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>280.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-nmr-3l0EmQaZAGEoXgk4w</th>\n",
       "      <td>1</td>\n",
       "      <td>0.072816</td>\n",
       "      <td>30.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>412.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-uSHlr8Wk6f8ewVEgsvcWQ</th>\n",
       "      <td>1</td>\n",
       "      <td>0.062112</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>161.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date sent_per_token net_sentiment  stars           \\\n",
       "                        len         median        median median mean std   \n",
       "business_id                                                                \n",
       "-8Z6AiHwXInpnl6Q1Ey_5w    1      -0.028986          -2.0    1.0  1.0 NaN   \n",
       "-CZ78c-H3tTxpP-uQ09CWw    1      -0.005970          -2.0    5.0  5.0 NaN   \n",
       "-njvGjJacPJMJ9YWSr6G3Q    1       0.039286          11.0    2.0  2.0 NaN   \n",
       "-nmr-3l0EmQaZAGEoXgk4w    1       0.072816          30.0    4.0  4.0 NaN   \n",
       "-uSHlr8Wk6f8ewVEgsvcWQ    1       0.062112          10.0    4.0  4.0 NaN   \n",
       "\n",
       "                       text_length  \n",
       "                            median  \n",
       "business_id                         \n",
       "-8Z6AiHwXInpnl6Q1Ey_5w        69.0  \n",
       "-CZ78c-H3tTxpP-uQ09CWw       335.0  \n",
       "-njvGjJacPJMJ9YWSr6G3Q       280.0  \n",
       "-nmr-3l0EmQaZAGEoXgk4w       412.0  \n",
       "-uSHlr8Wk6f8ewVEgsvcWQ       161.0  "
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We could filter out the businesses with only 1 review, if we wanted:\n",
    "biz_summary[biz_summary.loc[:, ('date', 'len')]==1].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# But let's save it with the single reviews for now\n",
    "biz_summary.to_msgpack('data/biz_data.msg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "biz_summary = pd.read_msgpack('data/biz_data.msg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>sent_per_token</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th colspan=\"3\" halign=\"left\">stars</th>\n",
       "      <th>text_length</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>len</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>median</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>--5jkZ3-nUPZxUvtcbr8Uw</th>\n",
       "      <td>11</td>\n",
       "      <td>0.046753</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.545455</td>\n",
       "      <td>0.687552</td>\n",
       "      <td>327.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>--BlvDO_RG2yElKu9XA1_g</th>\n",
       "      <td>29</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.103448</td>\n",
       "      <td>1.291312</td>\n",
       "      <td>242.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-0D_CYhlD2ILkmLR0pBmnA</th>\n",
       "      <td>3</td>\n",
       "      <td>0.023339</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>298.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-0QBrNvhrPQCaeo7mTo0zQ</th>\n",
       "      <td>2</td>\n",
       "      <td>0.034889</td>\n",
       "      <td>14.5</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>0.707107</td>\n",
       "      <td>414.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-0bUDim5OGuv8R0Qqq6J4A</th>\n",
       "      <td>6</td>\n",
       "      <td>0.061905</td>\n",
       "      <td>4.5</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>1.211060</td>\n",
       "      <td>104.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date sent_per_token net_sentiment  stars            \\\n",
       "                        len         median        median median      mean   \n",
       "business_id                                                                 \n",
       "--5jkZ3-nUPZxUvtcbr8Uw   11       0.046753          10.0    5.0  4.545455   \n",
       "--BlvDO_RG2yElKu9XA1_g   29       0.041667          10.0    5.0  4.103448   \n",
       "-0D_CYhlD2ILkmLR0pBmnA    3       0.023339          13.0    4.0  3.333333   \n",
       "-0QBrNvhrPQCaeo7mTo0zQ    2       0.034889          14.5    4.5  4.500000   \n",
       "-0bUDim5OGuv8R0Qqq6J4A    6       0.061905           4.5    2.5  2.333333   \n",
       "\n",
       "                                 text_length  \n",
       "                             std      median  \n",
       "business_id                                   \n",
       "--5jkZ3-nUPZxUvtcbr8Uw  0.687552       327.0  \n",
       "--BlvDO_RG2yElKu9XA1_g  1.291312       242.0  \n",
       "-0D_CYhlD2ILkmLR0pBmnA  1.154701       298.0  \n",
       "-0QBrNvhrPQCaeo7mTo0zQ  0.707107       414.5  \n",
       "-0bUDim5OGuv8R0Qqq6J4A  1.211060       104.0  "
      ]
     },
     "execution_count": 246,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_summary.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create a Fake Name for Each Business from First Few Cap'd Words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_cap_words(text):\n",
    "    # get first few cap words for a joint fake title.\n",
    "    words = [word for sent in nltk.sent_tokenize(text) for word in nltk.wordpunct_tokenize(sent)\\\n",
    "              if word[0].isupper() and len(word) > 2]\n",
    "    if len(words) >= 1:\n",
    "        return ' '.join(words[0:3])\n",
    "    else:\n",
    "        return 'None'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "pairs = defaultdict(dict)\n",
    "for name, grouped in biz:\n",
    "    #print name, len(grouped), grouped.iloc[0]\n",
    "    if grouped.iloc[0].text:\n",
    "        pairs[name] = get_cap_words(grouped.iloc[0].text)\n",
    "    else:\n",
    "        pairs[name] = \"None\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11479"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(pairs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lynn/miniconda/envs/esnlp/lib/python2.7/site-packages/pandas/core/indexing.py:288: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.obj[key] = _infer_fill_value(value)\n",
      "/Users/lynn/miniconda/envs/esnlp/lib/python2.7/site-packages/pandas/core/indexing.py:465: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.obj[item] = s\n"
     ]
    }
   ],
   "source": [
    "for index, row in dfes.iterrows():\n",
    "    dfes.set_value(index, 'fake_name', pairs[row.business_id])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "      <th>tokens</th>\n",
       "      <th>text_length</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th>sent_per_token</th>\n",
       "      <th>fake_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9yKzy9PApeiPPOUJEtnvkg</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>fWKvX83p0-ka4JS3dc6E5A</td>\n",
       "      <td>5</td>\n",
       "      <td>My wife took me here on my birthday for breakf...</td>\n",
       "      <td>rLtl8ZkDX5vH5nAx9C3q5Q</td>\n",
       "      <td>wife took birthday breakfast excellent weather...</td>\n",
       "      <td>542</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0.064576</td>\n",
       "      <td>The Our Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6oRAC4uyJCsJl1X0WZpVSA</td>\n",
       "      <td>2012-06-14</td>\n",
       "      <td>IESLBzqUCLdSzSqm0eCSxQ</td>\n",
       "      <td>4</td>\n",
       "      <td>love the gyro plate. Rice is so good and I als...</td>\n",
       "      <td>0hT2KtfLiobPvh6cDC8JQg</td>\n",
       "      <td>love gyro plate rice good also dig candy selec...</td>\n",
       "      <td>50</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.120000</td>\n",
       "      <td>Rice</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>_1QQZuf4zZOyFCvXc0o6Vg</td>\n",
       "      <td>2010-05-27</td>\n",
       "      <td>G-WvGaISbqqaMHlNnByodA</td>\n",
       "      <td>5</td>\n",
       "      <td>Rosie, Dakota, and I LOVE Chaparral Dog Park!!...</td>\n",
       "      <td>uZetl9T0NcROGOyFfughhg</td>\n",
       "      <td>rosie dakota love chaparral dog park convenien...</td>\n",
       "      <td>280</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.035714</td>\n",
       "      <td>Rosie Dakota LOVE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6ozycU1RpktNG2-1BroVtw</td>\n",
       "      <td>2012-01-05</td>\n",
       "      <td>1uJFq2r5QfJG_6ExMRCaGw</td>\n",
       "      <td>5</td>\n",
       "      <td>General Manager Scott Petello is a good egg!!!...</td>\n",
       "      <td>vYmM4KTsC8ZfQBg-j5MWkw</td>\n",
       "      <td>general manager scott petello good egg detail ...</td>\n",
       "      <td>248</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.052419</td>\n",
       "      <td>General Manager Scott</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>hW0Ne_HTHEAgGF1rAdmR-g</td>\n",
       "      <td>2012-07-12</td>\n",
       "      <td>JL7GXJ9u4YMx7Rzs05NfiQ</td>\n",
       "      <td>4</td>\n",
       "      <td>Luckily, I didn't have to travel far to make m...</td>\n",
       "      <td>1ieuYcKS7zeAv_U15AB13A</td>\n",
       "      <td>luckily travel far make connecting flight than...</td>\n",
       "      <td>162</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.080247</td>\n",
       "      <td>Luckily And Phoenix</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id        date               review_id  stars  \\\n",
       "0  9yKzy9PApeiPPOUJEtnvkg  2011-01-26  fWKvX83p0-ka4JS3dc6E5A      5   \n",
       "2  6oRAC4uyJCsJl1X0WZpVSA  2012-06-14  IESLBzqUCLdSzSqm0eCSxQ      4   \n",
       "3  _1QQZuf4zZOyFCvXc0o6Vg  2010-05-27  G-WvGaISbqqaMHlNnByodA      5   \n",
       "4  6ozycU1RpktNG2-1BroVtw  2012-01-05  1uJFq2r5QfJG_6ExMRCaGw      5   \n",
       "7  hW0Ne_HTHEAgGF1rAdmR-g  2012-07-12  JL7GXJ9u4YMx7Rzs05NfiQ      4   \n",
       "\n",
       "                                                text                 user_id  \\\n",
       "0  My wife took me here on my birthday for breakf...  rLtl8ZkDX5vH5nAx9C3q5Q   \n",
       "2  love the gyro plate. Rice is so good and I als...  0hT2KtfLiobPvh6cDC8JQg   \n",
       "3  Rosie, Dakota, and I LOVE Chaparral Dog Park!!...  uZetl9T0NcROGOyFfughhg   \n",
       "4  General Manager Scott Petello is a good egg!!!...  vYmM4KTsC8ZfQBg-j5MWkw   \n",
       "7  Luckily, I didn't have to travel far to make m...  1ieuYcKS7zeAv_U15AB13A   \n",
       "\n",
       "                                              tokens  text_length  \\\n",
       "0  wife took birthday breakfast excellent weather...          542   \n",
       "2  love gyro plate rice good also dig candy selec...           50   \n",
       "3  rosie dakota love chaparral dog park convenien...          280   \n",
       "4  general manager scott petello good egg detail ...          248   \n",
       "7  luckily travel far make connecting flight than...          162   \n",
       "\n",
       "   net_sentiment  sent_per_token              fake_name  \n",
       "0           35.0        0.064576       The Our Saturday  \n",
       "2            6.0        0.120000                   Rice  \n",
       "3           10.0        0.035714      Rosie Dakota LOVE  \n",
       "4           13.0        0.052419  General Manager Scott  \n",
       "7           13.0        0.080247    Luckily And Phoenix  "
      ]
     },
     "execution_count": 247,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0               The Our Saturday\n",
       "2                           Rice\n",
       "3              Rosie Dakota LOVE\n",
       "4          General Manager Scott\n",
       "7            Luckily And Phoenix\n",
       "8        Definitely Happy Prices\n",
       "9          Nobuo Carefully Start\n",
       "10              The Perhaps Here\n",
       "11    Wonderful Vietnamese Their\n",
       "12             They BBQ Probably\n",
       "13                Good Clean Jet\n",
       "15            Was Absolutely Bad\n",
       "17      EVER Oakville Scottsdale\n",
       "19         They Groupons Groupon\n",
       "20                  DVAP You The\n",
       "21                          This\n",
       "22                      The ALOT\n",
       "23               That They Might\n",
       "24              Elsa Chicken The\n",
       "25                This The Sorta\n",
       "Name: fake_name, dtype: object"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfes['fake_name'].head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "229880            This Hands Fresh\n",
       "229883                      Prices\n",
       "229884             The However The\n",
       "229886                How Well Yet\n",
       "229888        Definitely Their THE\n",
       "229890          Maybe Pricing This\n",
       "229891             One Italian Fun\n",
       "229892    This Phoenix Wonderfully\n",
       "229893                  Great Hugh\n",
       "229895              First The What\n",
       "229896         Who Scottsdale They\n",
       "229897           Nice Breads Great\n",
       "229899                    This But\n",
       "229900              Good Not Small\n",
       "229901          Radio Milano Tried\n",
       "229902               Too There The\n",
       "229903        Pros Reasonable Cons\n",
       "229904         The GREAT Seriously\n",
       "229905                     Awesome\n",
       "229906                         You\n",
       "Name: fake_name, dtype: object"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfes.fake_name.tail(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# save it\n",
    "dfes.to_msgpack('data/yelp_df_forES.msg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date            len         5.000000\n",
       "sent_per_token  median      0.038961\n",
       "net_sentiment   median     10.000000\n",
       "stars           median      5.000000\n",
       "                mean        4.600000\n",
       "                std         0.894427\n",
       "text_length     median    248.000000\n",
       "Name: 6ozycU1RpktNG2-1BroVtw, dtype: float64"
      ]
     },
     "execution_count": 248,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_summary.ix['6ozycU1RpktNG2-1BroVtw']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "biz_summary['ids'] = biz_summary.index.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>sent_per_token</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th colspan=\"3\" halign=\"left\">stars</th>\n",
       "      <th>text_length</th>\n",
       "      <th>ids</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>len</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>median</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>--5jkZ3-nUPZxUvtcbr8Uw</th>\n",
       "      <td>11</td>\n",
       "      <td>0.046753</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.545455</td>\n",
       "      <td>0.687552</td>\n",
       "      <td>327.0</td>\n",
       "      <td>--5jkZ3-nUPZxUvtcbr8Uw</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>--BlvDO_RG2yElKu9XA1_g</th>\n",
       "      <td>29</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.103448</td>\n",
       "      <td>1.291312</td>\n",
       "      <td>242.0</td>\n",
       "      <td>--BlvDO_RG2yElKu9XA1_g</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-0D_CYhlD2ILkmLR0pBmnA</th>\n",
       "      <td>3</td>\n",
       "      <td>0.023339</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>298.0</td>\n",
       "      <td>-0D_CYhlD2ILkmLR0pBmnA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-0QBrNvhrPQCaeo7mTo0zQ</th>\n",
       "      <td>2</td>\n",
       "      <td>0.034889</td>\n",
       "      <td>14.5</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>0.707107</td>\n",
       "      <td>414.5</td>\n",
       "      <td>-0QBrNvhrPQCaeo7mTo0zQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-0bUDim5OGuv8R0Qqq6J4A</th>\n",
       "      <td>6</td>\n",
       "      <td>0.061905</td>\n",
       "      <td>4.5</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>1.211060</td>\n",
       "      <td>104.0</td>\n",
       "      <td>-0bUDim5OGuv8R0Qqq6J4A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-0xwm_AwS0d_5rwUprHGuw</th>\n",
       "      <td>10</td>\n",
       "      <td>0.061647</td>\n",
       "      <td>10.5</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>0.966092</td>\n",
       "      <td>174.5</td>\n",
       "      <td>-0xwm_AwS0d_5rwUprHGuw</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-1-sUowC7fJ-cn7mCxvJ5w</th>\n",
       "      <td>3</td>\n",
       "      <td>0.013699</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>2.309401</td>\n",
       "      <td>219.0</td>\n",
       "      <td>-1-sUowC7fJ-cn7mCxvJ5w</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-1N0Z3uM8xbxKS8XiAnaog</th>\n",
       "      <td>3</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>1.527525</td>\n",
       "      <td>378.0</td>\n",
       "      <td>-1N0Z3uM8xbxKS8XiAnaog</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-1bOb2izeJBZjHC7NWxiPA</th>\n",
       "      <td>42</td>\n",
       "      <td>0.036686</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.476190</td>\n",
       "      <td>1.064690</td>\n",
       "      <td>232.0</td>\n",
       "      <td>-1bOb2izeJBZjHC7NWxiPA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-2xCV0XGD9NxfWaVwA1-DQ</th>\n",
       "      <td>3</td>\n",
       "      <td>0.096774</td>\n",
       "      <td>9.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.333333</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>56.0</td>\n",
       "      <td>-2xCV0XGD9NxfWaVwA1-DQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-34jE_5dujSWMIOBudQsiQ</th>\n",
       "      <td>3</td>\n",
       "      <td>0.033520</td>\n",
       "      <td>12.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.666667</td>\n",
       "      <td>0.577350</td>\n",
       "      <td>379.0</td>\n",
       "      <td>-34jE_5dujSWMIOBudQsiQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-3WVw1TNQbPBzaKCaQQ1AQ</th>\n",
       "      <td>105</td>\n",
       "      <td>0.034946</td>\n",
       "      <td>8.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.723810</td>\n",
       "      <td>1.023712</td>\n",
       "      <td>245.0</td>\n",
       "      <td>-3WVw1TNQbPBzaKCaQQ1AQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-3xbryp44xhpN4BohxXDdQ</th>\n",
       "      <td>45</td>\n",
       "      <td>0.050633</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.644444</td>\n",
       "      <td>1.227628</td>\n",
       "      <td>205.0</td>\n",
       "      <td>-3xbryp44xhpN4BohxXDdQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-4A5xmN21zi_TXnUESauUQ</th>\n",
       "      <td>218</td>\n",
       "      <td>0.047733</td>\n",
       "      <td>11.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.366972</td>\n",
       "      <td>0.865739</td>\n",
       "      <td>220.0</td>\n",
       "      <td>-4A5xmN21zi_TXnUESauUQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-4QayefhaPXm6wPp-Zpb0A</th>\n",
       "      <td>3</td>\n",
       "      <td>0.081395</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.333333</td>\n",
       "      <td>0.577350</td>\n",
       "      <td>171.0</td>\n",
       "      <td>-4QayefhaPXm6wPp-Zpb0A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-4zMr3jk0ykmsk5sxsQMlA</th>\n",
       "      <td>3</td>\n",
       "      <td>0.038760</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>129.0</td>\n",
       "      <td>-4zMr3jk0ykmsk5sxsQMlA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-5EWgRkudDlcCbQSIL_twA</th>\n",
       "      <td>5</td>\n",
       "      <td>0.025063</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>1.341641</td>\n",
       "      <td>409.0</td>\n",
       "      <td>-5EWgRkudDlcCbQSIL_twA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-5IgAihccTd8_iENVgpd9A</th>\n",
       "      <td>3</td>\n",
       "      <td>0.028926</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>305.0</td>\n",
       "      <td>-5IgAihccTd8_iENVgpd9A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-5XAZC1hC6G0NDkmPyhmUg</th>\n",
       "      <td>5</td>\n",
       "      <td>0.035417</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>0.894427</td>\n",
       "      <td>274.0</td>\n",
       "      <td>-5XAZC1hC6G0NDkmPyhmUg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-5rFC4EVrT-v8g1PSEf6Xg</th>\n",
       "      <td>30</td>\n",
       "      <td>0.052325</td>\n",
       "      <td>12.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.100000</td>\n",
       "      <td>1.124952</td>\n",
       "      <td>278.0</td>\n",
       "      <td>-5rFC4EVrT-v8g1PSEf6Xg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date sent_per_token net_sentiment  stars            \\\n",
       "                        len         median        median median      mean   \n",
       "business_id                                                                 \n",
       "--5jkZ3-nUPZxUvtcbr8Uw   11       0.046753          10.0    5.0  4.545455   \n",
       "--BlvDO_RG2yElKu9XA1_g   29       0.041667          10.0    5.0  4.103448   \n",
       "-0D_CYhlD2ILkmLR0pBmnA    3       0.023339          13.0    4.0  3.333333   \n",
       "-0QBrNvhrPQCaeo7mTo0zQ    2       0.034889          14.5    4.5  4.500000   \n",
       "-0bUDim5OGuv8R0Qqq6J4A    6       0.061905           4.5    2.5  2.333333   \n",
       "-0xwm_AwS0d_5rwUprHGuw   10       0.061647          10.5    4.0  3.600000   \n",
       "-1-sUowC7fJ-cn7mCxvJ5w    3       0.013699           3.0    1.0  2.333333   \n",
       "-1N0Z3uM8xbxKS8XiAnaog    3       0.000000           0.0    2.0  2.333333   \n",
       "-1bOb2izeJBZjHC7NWxiPA   42       0.036686           9.0    4.0  3.476190   \n",
       "-2xCV0XGD9NxfWaVwA1-DQ    3       0.096774           9.0    5.0  4.333333   \n",
       "-34jE_5dujSWMIOBudQsiQ    3       0.033520          12.0    5.0  4.666667   \n",
       "-3WVw1TNQbPBzaKCaQQ1AQ  105       0.034946           8.0    4.0  3.723810   \n",
       "-3xbryp44xhpN4BohxXDdQ   45       0.050633           9.0    4.0  3.644444   \n",
       "-4A5xmN21zi_TXnUESauUQ  218       0.047733          11.0    5.0  4.366972   \n",
       "-4QayefhaPXm6wPp-Zpb0A    3       0.081395          14.0    4.0  4.333333   \n",
       "-4zMr3jk0ykmsk5sxsQMlA    3       0.038760           5.0    4.0  4.000000   \n",
       "-5EWgRkudDlcCbQSIL_twA    5       0.025063          10.0    1.0  1.600000   \n",
       "-5IgAihccTd8_iENVgpd9A    3       0.028926          13.0    4.0  3.333333   \n",
       "-5XAZC1hC6G0NDkmPyhmUg    5       0.035417           9.0    4.0  3.600000   \n",
       "-5rFC4EVrT-v8g1PSEf6Xg   30       0.052325          12.0    4.0  4.100000   \n",
       "\n",
       "                                 text_length                     ids  \n",
       "                             std      median                          \n",
       "business_id                                                           \n",
       "--5jkZ3-nUPZxUvtcbr8Uw  0.687552       327.0  --5jkZ3-nUPZxUvtcbr8Uw  \n",
       "--BlvDO_RG2yElKu9XA1_g  1.291312       242.0  --BlvDO_RG2yElKu9XA1_g  \n",
       "-0D_CYhlD2ILkmLR0pBmnA  1.154701       298.0  -0D_CYhlD2ILkmLR0pBmnA  \n",
       "-0QBrNvhrPQCaeo7mTo0zQ  0.707107       414.5  -0QBrNvhrPQCaeo7mTo0zQ  \n",
       "-0bUDim5OGuv8R0Qqq6J4A  1.211060       104.0  -0bUDim5OGuv8R0Qqq6J4A  \n",
       "-0xwm_AwS0d_5rwUprHGuw  0.966092       174.5  -0xwm_AwS0d_5rwUprHGuw  \n",
       "-1-sUowC7fJ-cn7mCxvJ5w  2.309401       219.0  -1-sUowC7fJ-cn7mCxvJ5w  \n",
       "-1N0Z3uM8xbxKS8XiAnaog  1.527525       378.0  -1N0Z3uM8xbxKS8XiAnaog  \n",
       "-1bOb2izeJBZjHC7NWxiPA  1.064690       232.0  -1bOb2izeJBZjHC7NWxiPA  \n",
       "-2xCV0XGD9NxfWaVwA1-DQ  1.154701        56.0  -2xCV0XGD9NxfWaVwA1-DQ  \n",
       "-34jE_5dujSWMIOBudQsiQ  0.577350       379.0  -34jE_5dujSWMIOBudQsiQ  \n",
       "-3WVw1TNQbPBzaKCaQQ1AQ  1.023712       245.0  -3WVw1TNQbPBzaKCaQQ1AQ  \n",
       "-3xbryp44xhpN4BohxXDdQ  1.227628       205.0  -3xbryp44xhpN4BohxXDdQ  \n",
       "-4A5xmN21zi_TXnUESauUQ  0.865739       220.0  -4A5xmN21zi_TXnUESauUQ  \n",
       "-4QayefhaPXm6wPp-Zpb0A  0.577350       171.0  -4QayefhaPXm6wPp-Zpb0A  \n",
       "-4zMr3jk0ykmsk5sxsQMlA  1.000000       129.0  -4zMr3jk0ykmsk5sxsQMlA  \n",
       "-5EWgRkudDlcCbQSIL_twA  1.341641       409.0  -5EWgRkudDlcCbQSIL_twA  \n",
       "-5IgAihccTd8_iENVgpd9A  1.154701       305.0  -5IgAihccTd8_iENVgpd9A  \n",
       "-5XAZC1hC6G0NDkmPyhmUg  0.894427       274.0  -5XAZC1hC6G0NDkmPyhmUg  \n",
       "-5rFC4EVrT-v8g1PSEf6Xg  1.124952       278.0  -5rFC4EVrT-v8g1PSEf6Xg  "
      ]
     },
     "execution_count": 251,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_summary.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ids = biz_summary.index.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11479"
      ]
     },
     "execution_count": 252,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ids)\n",
    "len(biz_summary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "biz_flat = pd.DataFrame(biz_summary.to_records())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11479"
      ]
     },
     "execution_count": 256,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(biz_flat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([u'business_id', u'('date', 'len')', u'('sent_per_token', 'median')',\n",
       "       u'('net_sentiment', 'median')', u'('stars', 'median')',\n",
       "       u'('stars', 'mean')', u'('stars', 'std')', u'('text_length', 'median')',\n",
       "       u'('ids', '')'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 268,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_flat.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "cols = ['business_id', 'reviews',\n",
    " 'sent_per_token_median',\n",
    " 'net_sentiment_median',\n",
    " 'stars_median',\n",
    " 'stars_mean',\n",
    " 'stars_std',\n",
    " 'text_length_median',\n",
    " 'ids']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "biz_flat.columns = cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>reviews</th>\n",
       "      <th>sent_per_token_median</th>\n",
       "      <th>net_sentiment_median</th>\n",
       "      <th>stars_median</th>\n",
       "      <th>stars_mean</th>\n",
       "      <th>stars_std</th>\n",
       "      <th>text_length_median</th>\n",
       "      <th>ids</th>\n",
       "      <th>fake_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>--5jkZ3-nUPZxUvtcbr8Uw</td>\n",
       "      <td>11</td>\n",
       "      <td>0.046753</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.545455</td>\n",
       "      <td>0.687552</td>\n",
       "      <td>327.0</td>\n",
       "      <td>--5jkZ3-nUPZxUvtcbr8Uw</td>\n",
       "      <td>George The Tzatziki</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>--BlvDO_RG2yElKu9XA1_g</td>\n",
       "      <td>29</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.103448</td>\n",
       "      <td>1.291312</td>\n",
       "      <td>242.0</td>\n",
       "      <td>--BlvDO_RG2yElKu9XA1_g</td>\n",
       "      <td>Hawaii Hawaii That</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0D_CYhlD2ILkmLR0pBmnA</td>\n",
       "      <td>3</td>\n",
       "      <td>0.023339</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>298.0</td>\n",
       "      <td>-0D_CYhlD2ILkmLR0pBmnA</td>\n",
       "      <td>Saw Olive Gourmet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0QBrNvhrPQCaeo7mTo0zQ</td>\n",
       "      <td>2</td>\n",
       "      <td>0.034889</td>\n",
       "      <td>14.5</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>0.707107</td>\n",
       "      <td>414.5</td>\n",
       "      <td>-0QBrNvhrPQCaeo7mTo0zQ</td>\n",
       "      <td>Best The Mexican</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0bUDim5OGuv8R0Qqq6J4A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.061905</td>\n",
       "      <td>4.5</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>1.211060</td>\n",
       "      <td>104.0</td>\n",
       "      <td>-0bUDim5OGuv8R0Qqq6J4A</td>\n",
       "      <td>This</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id  reviews  sent_per_token_median  \\\n",
       "0  --5jkZ3-nUPZxUvtcbr8Uw       11               0.046753   \n",
       "1  --BlvDO_RG2yElKu9XA1_g       29               0.041667   \n",
       "2  -0D_CYhlD2ILkmLR0pBmnA        3               0.023339   \n",
       "3  -0QBrNvhrPQCaeo7mTo0zQ        2               0.034889   \n",
       "4  -0bUDim5OGuv8R0Qqq6J4A        6               0.061905   \n",
       "\n",
       "   net_sentiment_median  stars_median  stars_mean  stars_std  \\\n",
       "0                  10.0           5.0    4.545455   0.687552   \n",
       "1                  10.0           5.0    4.103448   1.291312   \n",
       "2                  13.0           4.0    3.333333   1.154701   \n",
       "3                  14.5           4.5    4.500000   0.707107   \n",
       "4                   4.5           2.5    2.333333   1.211060   \n",
       "\n",
       "   text_length_median                     ids            fake_name  \n",
       "0               327.0  --5jkZ3-nUPZxUvtcbr8Uw  George The Tzatziki  \n",
       "1               242.0  --BlvDO_RG2yElKu9XA1_g   Hawaii Hawaii That  \n",
       "2               298.0  -0D_CYhlD2ILkmLR0pBmnA    Saw Olive Gourmet  \n",
       "3               414.5  -0QBrNvhrPQCaeo7mTo0zQ     Best The Mexican  \n",
       "4               104.0  -0bUDim5OGuv8R0Qqq6J4A                 This  "
      ]
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_flat.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11479"
      ]
     },
     "execution_count": 280,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ids)\n",
    "len(biz_flat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for index, id in enumerate(ids):\n",
    "    biz_flat.set_value(index, 'fake_name', pairs[id])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>reviews</th>\n",
       "      <th>sent_per_token_median</th>\n",
       "      <th>net_sentiment_median</th>\n",
       "      <th>stars_median</th>\n",
       "      <th>stars_mean</th>\n",
       "      <th>stars_std</th>\n",
       "      <th>text_length_median</th>\n",
       "      <th>ids</th>\n",
       "      <th>fake_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>--5jkZ3-nUPZxUvtcbr8Uw</td>\n",
       "      <td>11</td>\n",
       "      <td>0.046753</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.545455</td>\n",
       "      <td>0.687552</td>\n",
       "      <td>327.0</td>\n",
       "      <td>--5jkZ3-nUPZxUvtcbr8Uw</td>\n",
       "      <td>George The Tzatziki</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>--BlvDO_RG2yElKu9XA1_g</td>\n",
       "      <td>29</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.103448</td>\n",
       "      <td>1.291312</td>\n",
       "      <td>242.0</td>\n",
       "      <td>--BlvDO_RG2yElKu9XA1_g</td>\n",
       "      <td>Hawaii Hawaii That</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0D_CYhlD2ILkmLR0pBmnA</td>\n",
       "      <td>3</td>\n",
       "      <td>0.023339</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>298.0</td>\n",
       "      <td>-0D_CYhlD2ILkmLR0pBmnA</td>\n",
       "      <td>Saw Olive Gourmet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0QBrNvhrPQCaeo7mTo0zQ</td>\n",
       "      <td>2</td>\n",
       "      <td>0.034889</td>\n",
       "      <td>14.5</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>0.707107</td>\n",
       "      <td>414.5</td>\n",
       "      <td>-0QBrNvhrPQCaeo7mTo0zQ</td>\n",
       "      <td>Best The Mexican</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0bUDim5OGuv8R0Qqq6J4A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.061905</td>\n",
       "      <td>4.5</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>1.211060</td>\n",
       "      <td>104.0</td>\n",
       "      <td>-0bUDim5OGuv8R0Qqq6J4A</td>\n",
       "      <td>This</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id  reviews  sent_per_token_median  \\\n",
       "0  --5jkZ3-nUPZxUvtcbr8Uw       11               0.046753   \n",
       "1  --BlvDO_RG2yElKu9XA1_g       29               0.041667   \n",
       "2  -0D_CYhlD2ILkmLR0pBmnA        3               0.023339   \n",
       "3  -0QBrNvhrPQCaeo7mTo0zQ        2               0.034889   \n",
       "4  -0bUDim5OGuv8R0Qqq6J4A        6               0.061905   \n",
       "\n",
       "   net_sentiment_median  stars_median  stars_mean  stars_std  \\\n",
       "0                  10.0           5.0    4.545455   0.687552   \n",
       "1                  10.0           5.0    4.103448   1.291312   \n",
       "2                  13.0           4.0    3.333333   1.154701   \n",
       "3                  14.5           4.5    4.500000   0.707107   \n",
       "4                   4.5           2.5    2.333333   1.211060   \n",
       "\n",
       "   text_length_median                     ids            fake_name  \n",
       "0               327.0  --5jkZ3-nUPZxUvtcbr8Uw  George The Tzatziki  \n",
       "1               242.0  --BlvDO_RG2yElKu9XA1_g   Hawaii Hawaii That  \n",
       "2               298.0  -0D_CYhlD2ILkmLR0pBmnA    Saw Olive Gourmet  \n",
       "3               414.5  -0QBrNvhrPQCaeo7mTo0zQ     Best The Mexican  \n",
       "4               104.0  -0bUDim5OGuv8R0Qqq6J4A                 This  "
      ]
     },
     "execution_count": 290,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biz_flat.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# save it\n",
    "biz_flat.to_msgpack('data/biz_statsdf_for_ES.msg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### We can use both tables in ES for different things now."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Topic Modeling Aside\n",
    "\n",
    "This section is an exploration of the contents using gensim's LDA.  The topics could be used for filtering contents before you index in ES.  Also, they might help you in determining aspects of your UI and data cleaning issues. \n",
    "\n",
    "This is an \"aside\" because I'm not using the results of this in the current JS app in the web directory."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\"\"\" An alternate from gensim tutorials - just use all words in the model in a rewiew. \"\"\"\n",
    "\n",
    "import re\n",
    "\n",
    "class YelpReviews(object):\n",
    "    \"\"\"Iterate over sentences of all plaintext files in a directory \"\"\"\n",
    "    SPLIT_SENTENCES = re.compile(u\"[.!?:]\\s+\")  # split sentences on these characters\n",
    "\n",
    "    def __init__(self, objs, field):\n",
    "        self.field = field\n",
    "        self.objs = objs\n",
    "\n",
    "    def __iter__(self):\n",
    "        for obj in self.objs:\n",
    "            text = obj[self.field]\n",
    "            for sentence in self.SPLIT_SENTENCES.split(text):\n",
    "                yield gensim.utils.simple_preprocess(sentence, deacc=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from gensim.utils import simple_preprocess\n",
    "from gensim.parsing.preprocessing import STOPWORDS\n",
    "import string\n",
    "\n",
    "def simple_tokenize(text):\n",
    "    return [token for token in simple_preprocess(text) if token not in STOPWORDS]\n",
    "\n",
    "#strip any cap words as proxy for proper names from a text...unfortunately right now this is yanking the first word from a sentence too.\n",
    "def tokenize_strip_proppers(text):\n",
    "    # first tokenize by sentence, then by word to ensure that punctuation is caught as it's own token\n",
    "    tokens = [word for sent in nltk.sent_tokenize(text) for word in nltk.wordpunct_tokenize(sent)\\\n",
    "              if word.islower() and len(word) > 2]\n",
    "    return \"\".join([\" \"+i if not i.startswith(\"'\") and i not in string.punctuation else i for i in tokens]).strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load nltk's SnowballStemmer as variabled 'stemmer'\n",
    "from nltk.stem.snowball import SnowballStemmer\n",
    "stemmer = SnowballStemmer(\"english\")\n",
    "import re\n",
    "\n",
    "def tokenize_and_stem(text):\n",
    "    # first tokenize by sentence, then by word to ensure that punctuation is caught as it's own token\n",
    "    # also remove upper case words and stop words\n",
    "    tokens = [word for sent in nltk.sent_tokenize(text) for word in nltk.wordpunct_tokenize(sent)\\\n",
    "              if word.islower() and len(word) > 2]\n",
    "    filtered_tokens = []\n",
    "    # filter out any tokens not containing letters (e.g., numeric tokens, raw punctuation)\n",
    "    for token in tokens:\n",
    "        if re.search('[a-zA-Z]', token):\n",
    "            filtered_tokens.append(token)\n",
    "    stems = [stemmer.stem(t) for t in filtered_tokens]\n",
    "    return \"\".join([\" \"+i if not i.startswith(\"'\") else i for i in stems]).strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'wife took here birthday for breakfast and was excellent weather was perfect which made sitting outside overlooking their grounds absolute pleasure waitress was excellent and our food arrived quickly the semi busy morning looked like the place fills pretty quickly the earlier you get here the better yourself favor and get their was phenomenal and simply the best ever had pretty sure they only use ingredients from their garden and blend them fresh when you order was amazing the menu looks excellent had the white truffle scrambled eggs vegetable skillet and was tasty and delicious came with pieces their griddled bread with was amazing and absolutely made the meal complete was the best toast ever had can wait back'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenize_strip_proppers(data[0]['text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'wife took here birthday for breakfast and was excel weather was perfect which made sit outsid overlook their ground absolut pleasur waitress was excel and our food arriv quick the semi busi morn look like the place fill pretti quick the earlier you get here the better yourself favor and get their was phenomen and simpli the best ever had pretti sure they onli use ingredi from their garden and blend them fresh when you order was amaz the menu look excel had the white truffl scrambl egg veget skillet and was tasti and delici came with piec their griddl bread with was amaz and absolut made the meal complet was the best toast ever had can wait back'"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenize_and_stem(data[0]['text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{u'business_id': u'9yKzy9PApeiPPOUJEtnvkg',\n",
       " u'date': u'2011-01-26',\n",
       " u'review_id': u'fWKvX83p0-ka4JS3dc6E5A',\n",
       " u'stars': 5,\n",
       " u'text': u'My wife took me here on my birthday for breakfast and it was excellent.  The weather was perfect which made sitting outside overlooking their grounds an absolute pleasure.  Our waitress was excellent and our food arrived quickly on the semi-busy Saturday morning.  It looked like the place fills up pretty quickly so the earlier you get here the better.\\n\\nDo yourself a favor and get their Bloody Mary.  It was phenomenal and simply the best I\\'ve ever had.  I\\'m pretty sure they only use ingredients from their garden and blend them fresh when you order it.  It was amazing.\\n\\nWhile EVERYTHING on the menu looks excellent, I had the white truffle scrambled eggs vegetable skillet and it was tasty and delicious.  It came with 2 pieces of their griddled bread with was amazing and it absolutely made the meal complete.  It was the best \"toast\" I\\'ve ever had.\\n\\nAnyway, I can\\'t wait to go back!',\n",
       " u'type': u'review',\n",
       " u'user_id': u'rLtl8ZkDX5vH5nAx9C3q5Q',\n",
       " u'votes': {u'cool': 2, u'funny': 0, u'useful': 5}}"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_review_tokens(dataset):\n",
    "    for doc in dataset:\n",
    "        text = tokenize_and_stem(doc[u'text'])\n",
    "        idn = doc[u'review_id']\n",
    "        tokens = simple_tokenize(text)\n",
    "        if len(tokens) < 10:\n",
    "            continue  # ignore short articles and various meta-articles\n",
    "        yield idn, tokens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\n",
    "class MyCorpus(object):\n",
    "    def __init__(self, data_file, dictionary):\n",
    "        \"\"\"\n",
    "        Parse the docs in data file.\n",
    "        Yield each document in turn, as a list of tokens (unicode strings).\n",
    "        \n",
    "        \"\"\"\n",
    "        self.data_file = data_file\n",
    "        self.dictionary = dictionary\n",
    "    \n",
    "    def __iter__(self):\n",
    "        self.ids = []\n",
    "        for idn, tokens in get_review_tokens(self.data_file):\n",
    "            self.ids.append(idn)\n",
    "            yield self.dictionary.doc2bow(tokens)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Beware - will do stems or not depending on current def of get_review_tokens\n",
    "doc_stream = (tokens for _, tokens in get_review_tokens(data)) # dep on def of the review_tokens - stems or words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 5min 54s, sys: 8.01 s, total: 6min 2s\n",
      "Wall time: 6min 53s\n",
      "Dictionary(84992 unique tokens: [u'fawn', u'remastered', u'belligerence', u'vang', u'saimin']...)\n"
     ]
    }
   ],
   "source": [
    "%time id2word_yelp = gensim.corpora.Dictionary(doc_stream)\n",
    "print(id2word_yelp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 17min 48s, sys: 15.4 s, total: 18min 4s\n",
      "Wall time: 18min 4s\n",
      "Dictionary(55969 unique tokens: [u'fawn', u'circuitri', u'gai', u'mustachio', u'woodi']...)\n"
     ]
    }
   ],
   "source": [
    "%time id2wordstems_yelp = gensim.corpora.Dictionary(doc_stream)\n",
    "print(id2wordstems_yelp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "id2wordstems_yelp.save('data/id2wordstems_yelp_dictionary.dict')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# create a stream of bag-of-words vectors\n",
    "review_corpus = MyCorpus(data, id2word_yelp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# create a stream of bag-of-words vectors\n",
    "reviewstems_corpus = MyCorpus(data, id2wordstems_yelp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(0, 1), (1, 1), (2, 1), (3, 1), (4, 1), (5, 1), (6, 1), (7, 1), (8, 1), (9, 1), (10, 1), (11, 2), (12, 1), (13, 1), (14, 1), (15, 1), (16, 1), (17, 1), (18, 2), (19, 2), (20, 1), (21, 1), (22, 1), (23, 2), (24, 1), (25, 1), (26, 1), (27, 1), (28, 1), (29, 1), (30, 1), (31, 1), (32, 1), (33, 1), (34, 1), (35, 1), (36, 1), (37, 1), (38, 1), (39, 1), (40, 3), (41, 1), (42, 1), (43, 1), (44, 1), (45, 1), (46, 1), (47, 1), (48, 1), (49, 1), (50, 1), (51, 1), (52, 1), (53, 1), (54, 1), (55, 1), (56, 1), (57, 1)]\n"
     ]
    }
   ],
   "source": [
    "import itertools\n",
    "vector = next(iter(review_corpus))\n",
    "print(vector)  # print the first vector in the stream"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(u'excellent', 3)\n"
     ]
    }
   ],
   "source": [
    "# what is the most common word in that first article?\n",
    "most_index, most_count = max(vector, key=lambda (word_index, count): count)\n",
    "print(id2word_yelp[most_index], most_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 7min 49s, sys: 11.8 s, total: 8min\n",
      "Wall time: 9min\n"
     ]
    }
   ],
   "source": [
    "%time gensim.corpora.MmCorpus.serialize('data/yelp_bow.mm', review_corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 19min 40s, sys: 14.9 s, total: 19min 55s\n",
      "Wall time: 20min 24s\n"
     ]
    }
   ],
   "source": [
    "%time gensim.corpora.MmCorpus.serialize('data/yelpstems_bow.mm', reviewstems_corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MmCorpus(208951 documents, 84992 features, 9612630 non-zero entries)\n"
     ]
    }
   ],
   "source": [
    "mm_corpus = gensim.corpora.MmCorpus('data/yelp_bow.mm')\n",
    "print(mm_corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MmCorpus(210280 documents, 55969 features, 9693003 non-zero entries)\n"
     ]
    }
   ],
   "source": [
    "mmstems_corpus = gensim.corpora.MmCorpus('data/yelpstems_bow.mm')\n",
    "print(mmstems_corpus)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Topic Models for the Reviews -- Using Stemmed and Unstemmed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 8min 21s, sys: 7.67 s, total: 8min 29s\n",
      "Wall time: 8min 37s\n"
     ]
    }
   ],
   "source": [
    "clipped_corpus = gensim.utils.ClippedCorpus(mm_corpus, 6000)  # use fewer documents during training, LDA is slow\n",
    "\n",
    "%time lda_model = gensim.models.LdaModel(clipped_corpus, num_topics=15, update_every=5, id2word=id2word_yelp, passes=20)\n",
    "#update_every=5, chunksize=10000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 8min 38s, sys: 6.32 s, total: 8min 44s\n",
      "Wall time: 8min 53s\n"
     ]
    }
   ],
   "source": [
    "clipped_corpus2 = gensim.utils.ClippedCorpus(mmstems_corpus, 6000)  # use fewer documents during training, LDA is slow\n",
    "\n",
    "%time ldastems_model = gensim.models.LdaModel(clipped_corpus2, num_topics=15, update_every=5, id2word=id2wordstems_yelp, passes=20)\n",
    "#update_every=5, chunksize=10000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0,\n",
       "  u'0.006*good + 0.005*great + 0.005*place + 0.004*ordered + 0.004*food + 0.004*like + 0.003*time + 0.003*wine + 0.003*salad + 0.003*cheese'),\n",
       " (1,\n",
       "  u'0.015*food + 0.013*good + 0.013*place + 0.010*time + 0.009*like + 0.009*service + 0.006*order + 0.006*got + 0.005*came + 0.005*ordered'),\n",
       " (2,\n",
       "  u'0.019*good + 0.018*food + 0.010*chicken + 0.010*place + 0.010*great + 0.008*like + 0.006*cheese + 0.006*tacos + 0.006*salsa + 0.005*ordered'),\n",
       " (3,\n",
       "  u'0.037*pizza + 0.009*good + 0.008*service + 0.007*crust + 0.006*great + 0.006*food + 0.005*place + 0.004*restaurant + 0.004*like + 0.004*location'),\n",
       " (4,\n",
       "  u'0.003*chocolate + 0.003*cupcakes + 0.002*cupcake + 0.002*yogurt + 0.001*kids + 0.001*venue + 0.001*fabric + 0.001*cake + 0.001*backyard + 0.001*wedding'),\n",
       " (5,\n",
       "  u'0.006*books + 0.004*library + 0.003*donuts + 0.002*donut + 0.001*book + 0.001*museum + 0.001*river + 0.001*gross + 0.001*read + 0.001*pets'),\n",
       " (6,\n",
       "  u'0.004*place + 0.003*service + 0.003*pool + 0.003*food + 0.003*great + 0.003*love + 0.003*hotel + 0.003*new + 0.002*nice + 0.002*hour'),\n",
       " (7,\n",
       "  u'0.002*car + 0.002*time + 0.002*office + 0.002*insurance + 0.002*water + 0.001*years + 0.001*campus + 0.001*called + 0.001*home + 0.001*animals'),\n",
       " (8,\n",
       "  u'0.009*ice + 0.007*cream + 0.005*cupcakes + 0.005*cupcake + 0.002*frosting + 0.002*store + 0.002*truth + 0.002*got + 0.002*daughter + 0.001*helpful'),\n",
       " (9,\n",
       "  u'0.004*http + 0.004*www + 0.004*com + 0.004*select + 0.003*biz_photos + 0.003*yelp + 0.002*course + 0.001*courses + 0.001*classes + 0.001*pump'),\n",
       " (10,\n",
       "  u'0.005*like + 0.005*store + 0.004*time + 0.003*place + 0.003*little + 0.003*shoes + 0.003*new + 0.003*fries + 0.002*day + 0.002*service'),\n",
       " (11,\n",
       "  u'0.010*good + 0.007*salad + 0.006*cheese + 0.006*sandwich + 0.006*pork + 0.005*sauce + 0.005*ordered + 0.004*menu + 0.004*bread + 0.004*like'),\n",
       " (12,\n",
       "  u'0.017*place + 0.013*like + 0.011*good + 0.011*great + 0.009*food + 0.006*time + 0.006*love + 0.006*nice + 0.005*little + 0.005*best'),\n",
       " (13,\n",
       "  u'0.002*classes + 0.002*address + 0.001*bag + 0.001*money + 0.001*trail + 0.001*camera + 0.001*degree + 0.001*shipping + 0.001*surgery + 0.001*level'),\n",
       " (14,\n",
       "  u'0.003*nails + 0.003*manicure + 0.002*gel + 0.002*park + 0.002*dog + 0.002*soap + 0.001*people + 0.001*nail + 0.001*moisturizer + 0.001*dogs')]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lda_model.print_topics(-1)  # print a few most important words for each LDA topic\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0,\n",
       "  u'0.003*lake + 0.001*gas + 0.001*like + 0.001*want + 0.001*yeah + 0.001*roll + 0.001*boat + 0.001*panang + 0.001*desert + 0.001*station'),\n",
       " (1,\n",
       "  u'0.003*soap + 0.003*bus + 0.002*moistur + 0.001*lavend + 0.001*deejay + 0.001*brim + 0.001*mint + 0.001*arrang + 0.001*wax + 0.001*inclus'),\n",
       " (2,\n",
       "  u'0.012*donut + 0.003*order + 0.002*pretzel + 0.002*want + 0.002*place + 0.002*good + 0.002*best + 0.002*sum + 0.002*food + 0.002*dim'),\n",
       " (3,\n",
       "  u'0.085*pizza + 0.015*crust + 0.011*good + 0.009*slice + 0.009*salad + 0.007*order + 0.006*wing + 0.006*chees + 0.005*pie + 0.005*like'),\n",
       " (4,\n",
       "  u'0.004*trail + 0.004*hike + 0.003*tea + 0.002*water + 0.002*grass + 0.002*hiker + 0.002*like + 0.002*backyard + 0.001*tree + 0.001*center'),\n",
       " (5,\n",
       "  u'0.003*wash + 0.002*thier + 0.002*car + 0.001*map + 0.001*cow + 0.001*windshield + 0.001*model + 0.001*nephew + 0.001*blade + 0.001*dairi'),\n",
       " (6,\n",
       "  u'0.021*cupcak + 0.004*frost + 0.002*great + 0.002*fro + 0.002*good + 0.002*review + 0.002*custard + 0.001*bra + 0.001*hubster + 0.001*order'),\n",
       " (7,\n",
       "  u'0.010*place + 0.010*time + 0.009*veri + 0.008*like + 0.007*good + 0.006*servic + 0.006*great + 0.006*look + 0.005*onli + 0.005*friend'),\n",
       " (8,\n",
       "  u'0.007*cake + 0.006*time + 0.006*pancak + 0.006*order + 0.005*good + 0.005*breakfast + 0.005*love + 0.004*dessert + 0.004*servic + 0.004*toast'),\n",
       " (9,\n",
       "  u'0.020*place + 0.012*great + 0.012*like + 0.010*bar + 0.009*time + 0.008*good + 0.008*food + 0.008*beer + 0.008*realli + 0.007*drink'),\n",
       " (10,\n",
       "  u'0.021*good + 0.019*food + 0.015*place + 0.013*order + 0.011*like + 0.011*time + 0.009*veri + 0.009*tri + 0.008*restaur + 0.008*great'),\n",
       " (11,\n",
       "  u'0.013*like + 0.013*place + 0.010*food + 0.009*great + 0.009*sandwich + 0.008*good + 0.008*love + 0.007*flavor + 0.007*salad + 0.007*chees'),\n",
       " (12,\n",
       "  u'0.002*river + 0.002*tube + 0.001*apparel + 0.001*layov + 0.001*ostrich + 0.001*princess + 0.001*carpet + 0.001*poser + 0.001*fianc\\xe9 + 0.001*diaper'),\n",
       " (13,\n",
       "  u'0.006*librari + 0.006*book + 0.001*nibbl + 0.001*physician + 0.001*mama + 0.001*giggl + 0.001*knish + 0.001*lollipop + 0.001*poker + 0.001*greek'),\n",
       " (14,\n",
       "  u'0.014*nail + 0.013*hair + 0.008*pedicur + 0.007*manicur + 0.006*salon + 0.005*massag + 0.005*gel + 0.004*haircut + 0.003*cut + 0.003*pedi')]"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ldastems_model.print_topics(-1)  # print a few most important words for each LDA topic\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sample = (tokens for _, tokens in get_review_tokens(data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[u'wife', u'took', u'birthday', u'breakfast', u'excellent', u'weather', u'perfect', u'sitting', u'outside', u'overlooking', u'grounds', u'absolute', u'pleasure', u'waitress', u'excellent', u'food', u'arrived', u'quickly', u'semi', u'busy', u'morning', u'looked', u'like', u'place', u'fills', u'pretty', u'quickly', u'earlier', u'better', u'favor', u'phenomenal', u'simply', u'best', u'pretty', u'sure', u'use', u'ingredients', u'garden', u'blend', u'fresh', u'order', u'amazing', u'menu', u'looks', u'excellent', u'white', u'truffle', u'scrambled', u'eggs', u'vegetable', u'skillet', u'tasty', u'delicious', u'came', u'pieces', u'griddled', u'bread', u'amazing', u'absolutely', u'meal', u'complete', u'best', u'toast', u'wait']\n",
      "[(4, 0.1118140268140206), (6, 0.52080972789300939), (11, 0.35506853596995919)]\n",
      "[u'idea', u'people', u'bad', u'reviews', u'place', u'goes', u'probably', u'griping', u'fault', u'people', u'like', u'case', u'friend', u'arrived', u'past', u'pretty', u'crowded', u'thought', u'evening', u'thought', u'wait', u'forever', u'seat', u'said', u'seated', u'girl', u'comes', u'seating', u'seated', u'waiter', u'came', u'got', u'drink', u'orders', u'pleasant', u'host', u'seated', u'waiter', u'server', u'prices', u'good', u'placed', u'orders', u'decided', u'wanted', u'shared', u'baked', u'spaghetti', u'calzone', u'small', u'pizza', u'try', u'calzone', u'huge', u'got', u'smallest', u'personal', u'got', u'small', u'pizza', u'awesome', u'friend', u'liked', u'pizza', u'better', u'liked', u'calzone', u'better', u'calzone', u'sweetish', u'sauce', u'like', u'sauce', u'box', u'pizza', u'home', u'door', u'great', u'like', u'bad', u'reviewers', u'goes', u'try', u'things', u'bad', u'reviewers', u'issues']\n",
      "[(7, 0.049765628699221527), (11, 0.94038587423601294)]\n",
      "[u'convenient', u'surrounded', u'lot', u'paths', u'desert', u'xeriscape', u'baseball', u'fields', u'ballparks', u'lake', u'ducks', u'wonderful', u'job', u'keeping', u'park', u'clean', u'shaded', u'trash', u'cans', u'poopy', u'pick', u'mitts', u'located', u'park', u'paths', u'fenced', u'area', u'huge', u'let', u'dogs', u'run', u'play', u'sniff']\n",
      "[(2, 0.70157999557823014), (8, 0.27292977258354051)]\n",
      "[u'good', u'egg', u'let', u'assure', u'issues', u'albeit', u'rare', u'speak', u'treat', u'guy', u'respect', u'state', u'case', u'surprised', u'walk', u'totally', u'satisfied', u'inevitable', u'recover', u'important', u'awesome', u'staff', u'got', u'customer', u'life']\n",
      "[(11, 0.96410252655779416)]\n",
      "[u'simply', u'beautiful', u'windows', u'earthy', u'wooden', u'walls', u'feeling', u'warmth', u'inside', u'restaurant', u'perched', u'middle', u'farm', u'restaurant', u'fairly', u'evening', u'secured', u'reservations', u'couple', u'days', u'friend', u'sampled', u'sandwiches', u'earlier', u'week', u'impressed', u'want', u'eat', u'restaurant', u'crisp', u'fresh', u'veggies', u'disappoint', u'ordered', u'salad', u'orange', u'grapefruit', u'slices', u'crudites', u'start', u'good', u'know', u'liked', u'raw', u'radishes', u'turnips', u'tried', u'pesto', u'aioli', u'sauces', u'entrees', u'ordered', u'lamb', u'friend', u'ordered', u'pork', u'shoulder', u'started', u'good', u'trailed', u'quickly', u'food', u'took', u'long', u'time', u'couple', u'seated', u'received', u'finished', u'entrees', u'received', u'bothered', u'explain', u'situation', u'maitre', u'apologized', u'minutes', u'later', u'chef', u'unhappy', u'sauce', u'entree', u'started', u'anew', u'isn', u'problem', u'communicated', u'earlier', u'troubles', u'comped', u'glass', u'wine', u'ordered', u'forgot', u'bring', u'entree', u'requested', u'offer', u'bread', u'echo', u'lady', u'whispered', u'way', u'ask', u'bread', u'received', u'warm', u'foccacia', u'apple', u'walnut', u'pomegranate', u'slices', u'wonder', u'honey', u'butter', u'entrees', u'solid', u'live', u'innovation', u'freshness', u'vegetables', u'lamb', u'sauce', u'delicious', u'meat', u'tough', u'vegetarian', u'entrees', u'way', u'dessert', u'gingerbread', u'pear', u'cake', u'winner', u'entrees', u'tad', u'inspired', u'service', u'weren', u'spotty', u'place', u'definitely', u'warranted', u'stars', u'return', u'like', u'try', u'tasting', u'menu', u'came', u'people', u'including', u'tip', u'drinks']\n",
      "[(11, 0.80235991185143773), (13, 0.19204868282282439)]\n",
      "[u'drive', u'ate', u'day', u'food', u'good', u'cute', u'little', u'green', u'building', u'gone', u'competely', u'unoticed', u'hadn', u'driving', u'avoid', u'construction', u'waiting', u'turn', u'th', u'sign', u'caught', u'eye', u'little', u'yelping', u'soul', u'leaped', u'joy', u'new', u'place', u'try', u'looked', u'desolate', u'outside', u'opened', u'door', u'easy', u'decor', u'smell', u'cleanliness', u'inside', u'ordered', u'dinner', u'menu', u'awesome', u'loved', u'seeing', u'variety', u'poblano', u'peppers', u'mole', u'mahi', u'mahi', u'mushrooms', u'wrapped', u'banana', u'leaves', u'difficult', u'choose', u'far', u'delicious', u'meals', u'shrimp', u'burro', u'stole', u'flavor', u'snagged', u'bites', u'hubbys', u'mole', u'mahi', u'mahi', u'burros', u'mmmm', u'delight', u'salsa', u'bar', u'endless', u'stocked', u'excited', u'try', u'strawberry', u'salsa', u'hot', u'fact', u'big', u'wimp', u'comes', u'hot', u'peppers', u'horchata', u'handmade', u'delicious', u'throw', u'pecans', u'fruit', u'yummy', u'bonus', u'good', u'food', u'wasn', u'win', u'art', u'restaurant', u'sho', u'sucker', u'folk', u'art', u'painting', u'hanging', u'salsa', u'bar', u'amazing', u'paintings', u'great', u'love', u'artist']\n",
      "[(11, 0.99202278073168604)]\n",
      "[u'travel', u'far', u'connecting', u'flight', u'thank', u'brief', u'layover', u'pleasant', u'employees', u'kind', u'flight', u'time', u'time', u'grace', u'presence', u'little', u'longer']\n",
      "[(2, 0.36047897905578408), (5, 0.11323330634107021), (9, 0.48184318950964283)]\n",
      "[u'come', u'hour', u'amazing', u'sake', u'bombers', u'atmosphere', u'wait', u'staff', u'incredibly', u'nice', u'right', u'needs', u'ask', u'thing', u'spot', u'gets', u'crowded', u'evening', u'especially', u'plan', u'sitting', u'outside', u'wish']\n",
      "[(11, 0.96111107525553308)]\n",
      "[u'shows', u'unique', u'talents', u'menu', u'crafted', u'features', u'drink', u'pork', u'belly', u'buns', u'stout', u'longer']\n",
      "[(6, 0.53571047753174728), (7, 0.39762275446052803)]\n"
     ]
    }
   ],
   "source": [
    "for doc in sample:\n",
    "    print doc\n",
    "    print lda_model.get_document_topics(id2word_yelp.doc2bow(doc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "wife took birthday breakfast excel weather perfect sit outsid overlook ground absolut pleasur waitress excel food arriv quick semi busi morn look like place pretti quick earlier better favor phenomen simpli best pretti sure onli use ingredi garden blend fresh order amaz menu look excel white truffl scrambl egg veget skillet tasti delici came piec griddl bread amaz absolut meal complet best toast wait\n",
      "[(8, 0.12423999650546777), (10, 0.58247846428765693), (11, 0.28097383789597324)]\n",
      "idea whi peopl bad review place goe pleas everyon probabl gripe someth fault mani peopl like ani case friend arriv past pretti crowd thought thought wait forev seat said seat girl come seat someon els seat waiter came got drink order veri pleasant host seat waiter server price veri good place order onc decid want share bake spaghetti calzon small pizza tri calzon huge got smallest person got small pizza awesom friend like pizza better like calzon better calzon doe sweetish sauc like sauc box pizza home door everyth great like bad review goe tri thing becaus bad review issu\n",
      "[(3, 0.12778427049932581), (9, 0.27922210937204894), (10, 0.58507282332688282)]\n",
      "veri conveni surround lot path desert xeriscap basebal field ballpark lake duck doe wonder job park clean shade trash poopi pick mitt locat park path fenc area huge let dog run play sniff\n",
      "[(7, 0.1497386913792251), (9, 0.434863913687003), (14, 0.39186796181549494)]\n",
      "good egg let assur ani issu albeit rare speak treat guy respect state case surpris walk total satisfi alway inevit recov import awesom staff got custom life\n",
      "[(7, 0.96666662766010136)]\n",
      "simpli beauti window earthi wooden wall feel warmth insid restaur perch middl farm restaur fair secur reserv coupl day befor friend sampl sandwich earlier week impress want eat restaur crisp fresh veggi disappoint order salad orang grapefruit slice crudit start veri good know like raw radish turnip tri pesto aioli sauc entre order lamb friend order pork shoulder start veri good trail quick food took veri long time coupl seat receiv finish entre befor receiv bother explain situat maitr apolog minut later chef unhappi sauc entre start anew isn realli problem communic earlier troubl comp glass wine order forgot bring entre request offer bread echo ladi whisper way ask bread receiv warm foccacia appl walnut pomegran slice wonder honey butter entre solid quit live innov fresh veget lamb sauc delici meat tough vegetarian entre way dessert gingerbread pear cake anoth winner entre tad inspir servic weren spotti place definit warrant star return like tri tast menu came peopl includ tip drink\n",
      "[(8, 0.30765371587035056), (10, 0.6869964831379699)]\n",
      "drive ate day food good cute littl green build gone compet unot hadn drive avoid construct wait turn th sign caught eye littl yelp soul leap joy new place tri look desol outsid open door easi decor smell cleanli insid order dinner menu awesom love varieti poblano pepper mole mahi mahi mushroom someth wrap banana leav difficult choos someth far veri delici meal shrimp burro stole flavor snag bite hubbi mole mahi mahi burro mmmm delight salsa bar endless realli stock excit tri strawberri salsa hot fact big wimp come hot pepper horchata handmad delici throw pecan fruit yummi bonus good food wasn win art restaur sho sucker folk art paint hang salsa bar amaz paint great love artist\n",
      "[(2, 0.093811508738370852), (9, 0.14880250886113106), (10, 0.32445316756730397), (11, 0.42682169890036714)]\n",
      "travel far connect flight thank brief layov pleasant employe kind flight time time grace presenc littl longer\n",
      "[(7, 0.83645192275407554), (12, 0.11539987567494486)]\n",
      "come hour amaz sake bomber atmospher wait staff incred nice right need ask thing alway spot crowd especi plan sit outsid onli wish\n",
      "[(7, 0.54826320494742631), (9, 0.41562565934996537)]\n",
      "uniqu talent everyth menu craft featur drink pork belli bun stout longer\n",
      "[(2, 0.36774951091466024), (10, 0.56558376553819578)]\n"
     ]
    }
   ],
   "source": [
    "for doc in sample:\n",
    "    print \" \".join(doc)\n",
    "    print ldastems_model.get_document_topics(id2wordstems_yelp.doc2bow(doc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "ldastems_model.save('data/ldastems_model.mod')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "lda_model.save('data/lda_model.mod')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ldastems_model = gensim.models.LdaModel.load('data/ldastems_model.mod')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9yKzy9PApeiPPOUJEtnvkg</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>fWKvX83p0-ka4JS3dc6E5A</td>\n",
       "      <td>5</td>\n",
       "      <td>My wife took me here on my birthday for breakf...</td>\n",
       "      <td>rLtl8ZkDX5vH5nAx9C3q5Q</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ZRJwVLyzEJq1VAihDhYiow</td>\n",
       "      <td>2011-07-27</td>\n",
       "      <td>IjZ33sJrzXqU-0X6U8NwyA</td>\n",
       "      <td>5</td>\n",
       "      <td>I have no idea why some people give bad review...</td>\n",
       "      <td>0a2KyEL0d3Yb1V6aivbIuQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6oRAC4uyJCsJl1X0WZpVSA</td>\n",
       "      <td>2012-06-14</td>\n",
       "      <td>IESLBzqUCLdSzSqm0eCSxQ</td>\n",
       "      <td>4</td>\n",
       "      <td>love the gyro plate. Rice is so good and I als...</td>\n",
       "      <td>0hT2KtfLiobPvh6cDC8JQg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>_1QQZuf4zZOyFCvXc0o6Vg</td>\n",
       "      <td>2010-05-27</td>\n",
       "      <td>G-WvGaISbqqaMHlNnByodA</td>\n",
       "      <td>5</td>\n",
       "      <td>Rosie, Dakota, and I LOVE Chaparral Dog Park!!...</td>\n",
       "      <td>uZetl9T0NcROGOyFfughhg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6ozycU1RpktNG2-1BroVtw</td>\n",
       "      <td>2012-01-05</td>\n",
       "      <td>1uJFq2r5QfJG_6ExMRCaGw</td>\n",
       "      <td>5</td>\n",
       "      <td>General Manager Scott Petello is a good egg!!!...</td>\n",
       "      <td>vYmM4KTsC8ZfQBg-j5MWkw</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id        date               review_id  stars  \\\n",
       "0  9yKzy9PApeiPPOUJEtnvkg  2011-01-26  fWKvX83p0-ka4JS3dc6E5A      5   \n",
       "1  ZRJwVLyzEJq1VAihDhYiow  2011-07-27  IjZ33sJrzXqU-0X6U8NwyA      5   \n",
       "2  6oRAC4uyJCsJl1X0WZpVSA  2012-06-14  IESLBzqUCLdSzSqm0eCSxQ      4   \n",
       "3  _1QQZuf4zZOyFCvXc0o6Vg  2010-05-27  G-WvGaISbqqaMHlNnByodA      5   \n",
       "4  6ozycU1RpktNG2-1BroVtw  2012-01-05  1uJFq2r5QfJG_6ExMRCaGw      5   \n",
       "\n",
       "                                                text                 user_id  \n",
       "0  My wife took me here on my birthday for breakf...  rLtl8ZkDX5vH5nAx9C3q5Q  \n",
       "1  I have no idea why some people give bad review...  0a2KyEL0d3Yb1V6aivbIuQ  \n",
       "2  love the gyro plate. Rice is so good and I als...  0hT2KtfLiobPvh6cDC8JQg  \n",
       "3  Rosie, Dakota, and I LOVE Chaparral Dog Park!!...  uZetl9T0NcROGOyFfughhg  \n",
       "4  General Manager Scott Petello is a good egg!!!...  vYmM4KTsC8ZfQBg-j5MWkw  "
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df['topics_stemmed'] = \"\"\n",
    "for row,vals in df.iterrows():\n",
    "    tokens = tokenize_and_stem(vals['text'])\n",
    "    df.set_value(row, 'tokens_stemmed', tokens)\n",
    "    topics = ldastems_model.get_document_topics(id2wordstems_yelp.doc2bow(tokens.split(' ')))\n",
    "    topstring = ';'.join([str(w[0]) + ':' + str(w[1]) for w in topics])\n",
    "    df.set_value(row, 'topics_stemmed', str(topstring))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "topics = [(8, 0.12424965619294445), (10, 0.58240970356859401), (11, 0.28103293892798464)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>business_id</th>\n",
       "      <th>date</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>user_id</th>\n",
       "      <th>tokens</th>\n",
       "      <th>net_sentiment</th>\n",
       "      <th>text_length</th>\n",
       "      <th>tokens_stemmed</th>\n",
       "      <th>topics_stemmed</th>\n",
       "      <th>sent_per_token</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9yKzy9PApeiPPOUJEtnvkg</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>fWKvX83p0-ka4JS3dc6E5A</td>\n",
       "      <td>5</td>\n",
       "      <td>My wife took me here on my birthday for breakf...</td>\n",
       "      <td>rLtl8ZkDX5vH5nAx9C3q5Q</td>\n",
       "      <td>wife took birthday breakfast excellent weather...</td>\n",
       "      <td>35.0</td>\n",
       "      <td>542</td>\n",
       "      <td>wife took here birthday for breakfast and was ...</td>\n",
       "      <td>8:0.124248119419;10:0.582431169676;11:0.281013...</td>\n",
       "      <td>0.064576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ZRJwVLyzEJq1VAihDhYiow</td>\n",
       "      <td>2011-07-27</td>\n",
       "      <td>IjZ33sJrzXqU-0X6U8NwyA</td>\n",
       "      <td>5</td>\n",
       "      <td>I have no idea why some people give bad review...</td>\n",
       "      <td>0a2KyEL0d3Yb1V6aivbIuQ</td>\n",
       "      <td>idea people give bad reviews place goes show p...</td>\n",
       "      <td>22.0</td>\n",
       "      <td>703</td>\n",
       "      <td>have idea whi some peopl give bad review about...</td>\n",
       "      <td>3:0.127784743918;9:0.279232475457;10:0.5850619...</td>\n",
       "      <td>0.031294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6oRAC4uyJCsJl1X0WZpVSA</td>\n",
       "      <td>2012-06-14</td>\n",
       "      <td>IESLBzqUCLdSzSqm0eCSxQ</td>\n",
       "      <td>4</td>\n",
       "      <td>love the gyro plate. Rice is so good and I als...</td>\n",
       "      <td>0hT2KtfLiobPvh6cDC8JQg</td>\n",
       "      <td>love gyro plate rice good also dig candy selec...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>50</td>\n",
       "      <td>love the gyro plate good and also dig their ca...</td>\n",
       "      <td>11:0.883333204766</td>\n",
       "      <td>0.120000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>_1QQZuf4zZOyFCvXc0o6Vg</td>\n",
       "      <td>2010-05-27</td>\n",
       "      <td>G-WvGaISbqqaMHlNnByodA</td>\n",
       "      <td>5</td>\n",
       "      <td>Rosie, Dakota, and I LOVE Chaparral Dog Park!!...</td>\n",
       "      <td>uZetl9T0NcROGOyFfughhg</td>\n",
       "      <td>rosie dakota love chaparral dog park convenien...</td>\n",
       "      <td>10.0</td>\n",
       "      <td>280</td>\n",
       "      <td>and veri conveni and surround lot path desert ...</td>\n",
       "      <td>7:0.14959518433;9:0.434998933203;14:0.39187644...</td>\n",
       "      <td>0.035714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6ozycU1RpktNG2-1BroVtw</td>\n",
       "      <td>2012-01-05</td>\n",
       "      <td>1uJFq2r5QfJG_6ExMRCaGw</td>\n",
       "      <td>5</td>\n",
       "      <td>General Manager Scott Petello is a good egg!!!...</td>\n",
       "      <td>vYmM4KTsC8ZfQBg-j5MWkw</td>\n",
       "      <td>general manager scott petello good egg detail ...</td>\n",
       "      <td>13.0</td>\n",
       "      <td>248</td>\n",
       "      <td>good egg into detail but let assur you you hav...</td>\n",
       "      <td>7:0.966666626696</td>\n",
       "      <td>0.052419</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id        date               review_id  stars  \\\n",
       "0  9yKzy9PApeiPPOUJEtnvkg  2011-01-26  fWKvX83p0-ka4JS3dc6E5A      5   \n",
       "1  ZRJwVLyzEJq1VAihDhYiow  2011-07-27  IjZ33sJrzXqU-0X6U8NwyA      5   \n",
       "2  6oRAC4uyJCsJl1X0WZpVSA  2012-06-14  IESLBzqUCLdSzSqm0eCSxQ      4   \n",
       "3  _1QQZuf4zZOyFCvXc0o6Vg  2010-05-27  G-WvGaISbqqaMHlNnByodA      5   \n",
       "4  6ozycU1RpktNG2-1BroVtw  2012-01-05  1uJFq2r5QfJG_6ExMRCaGw      5   \n",
       "\n",
       "                                                text                 user_id  \\\n",
       "0  My wife took me here on my birthday for breakf...  rLtl8ZkDX5vH5nAx9C3q5Q   \n",
       "1  I have no idea why some people give bad review...  0a2KyEL0d3Yb1V6aivbIuQ   \n",
       "2  love the gyro plate. Rice is so good and I als...  0hT2KtfLiobPvh6cDC8JQg   \n",
       "3  Rosie, Dakota, and I LOVE Chaparral Dog Park!!...  uZetl9T0NcROGOyFfughhg   \n",
       "4  General Manager Scott Petello is a good egg!!!...  vYmM4KTsC8ZfQBg-j5MWkw   \n",
       "\n",
       "                                              tokens  net_sentiment  \\\n",
       "0  wife took birthday breakfast excellent weather...           35.0   \n",
       "1  idea people give bad reviews place goes show p...           22.0   \n",
       "2  love gyro plate rice good also dig candy selec...            6.0   \n",
       "3  rosie dakota love chaparral dog park convenien...           10.0   \n",
       "4  general manager scott petello good egg detail ...           13.0   \n",
       "\n",
       "   text_length                                     tokens_stemmed  \\\n",
       "0          542  wife took here birthday for breakfast and was ...   \n",
       "1          703  have idea whi some peopl give bad review about...   \n",
       "2           50  love the gyro plate good and also dig their ca...   \n",
       "3          280  and veri conveni and surround lot path desert ...   \n",
       "4          248  good egg into detail but let assur you you hav...   \n",
       "\n",
       "                                      topics_stemmed  sent_per_token  \n",
       "0  8:0.124248119419;10:0.582431169676;11:0.281013...        0.064576  \n",
       "1  3:0.127784743918;9:0.279232475457;10:0.5850619...        0.031294  \n",
       "2                                  11:0.883333204766        0.120000  \n",
       "3  7:0.14959518433;9:0.434998933203;14:0.39187644...        0.035714  \n",
       "4                                   7:0.966666626696        0.052419  "
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What You Could Do Next\n",
    "\n",
    "You could use the topics to filter out topics of lesser interest, like the dog parks and salons, if you're interested in restaurants.\n",
    "Beware, though, because \"hot dogs\" and restaurants that are pet-friendly may get folded in with dog parks.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}