{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Occupancy plots with Python and ggplot2 via rmagic\n",
      "===================================================\n",
      "\n",
      "## Background ##\n",
      "\n",
      "In two previous posts ([here](http://hselab.org/occupancy-analysis-with-python-pandas-part-1-create-by-date-data-frame.html) and [here](http://hselab.org/occupancy-analysis-with-python-pandas-part-2-compute-occupancy-summary-stats.html)), I showed how to do occupancy analysis with the beginnings of a Python based version of [Hillmaker](http://hillmaker.sourceforge.net/). The example is based on data from a hospital short stay unit.\n",
      "\n",
      "In this short tutorial, I'll show how to use `rmagic` from within an IPython notebook so that we can make occupancy plots using `ggplot2`. In particular, we want to create a grid of occupancy histograms with one grid axis being patient type and the other axis being day of week. We want to make sure that the plots are ordered Sunday, Monday, ..., Saturday and NOT in alphabetical order (Friday, Monday, ..., Wednesday).\n",
      "\n",
      "The first part of such an analysis leads to a table we call the \"by datetime\" table. At the end of Part 1 of this tutorial series, we ended up with a csv file called bydate_shortstay_csv.csv. Let's read it in and take a look at it.\n",
      "\n",
      "You can find the data and the `.ipynb` file in my [hselab-tutorials](https://github.com/misken/hselab-tutorials) github repo. Clone or download a zip.\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import pandas as pd\n",
      "import numpy as np\n",
      "import matplotlib.pyplot as plt\n",
      "\n",
      "\n",
      "# Make the graphs a bit prettier, and bigger\n",
      "pd.set_option('display.mpl_style', 'default') \n",
      "pd.set_option('display.line_width', 5000) \n",
      "pd.set_option('display.max_columns', 60) "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "line_width has been deprecated, use display.width instead (currently both are\n",
        "identical)\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "## Read sample data set and convert string dates to datetimes\n",
      "bydatetime_df = pd.read_csv('data/bydate_shortstay_csv.csv',parse_dates=['datetime'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "bydatetime_df.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>category</th>\n",
        "      <th>datetime</th>\n",
        "      <th>arrivals</th>\n",
        "      <th>binofday</th>\n",
        "      <th>binofweek</th>\n",
        "      <th>dayofweek</th>\n",
        "      <th>departures</th>\n",
        "      <th>occupancy</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> IVT</td>\n",
        "      <td>1996-01-07 00:00:00</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 288</td>\n",
        "      <td> 6</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> IVT</td>\n",
        "      <td>1996-01-07 00:30:00</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "      <td> 289</td>\n",
        "      <td> 6</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0.5</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> IVT</td>\n",
        "      <td>1996-01-07 01:00:00</td>\n",
        "      <td> 0</td>\n",
        "      <td> 2</td>\n",
        "      <td> 290</td>\n",
        "      <td> 6</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> IVT</td>\n",
        "      <td>1996-01-07 01:30:00</td>\n",
        "      <td> 0</td>\n",
        "      <td> 3</td>\n",
        "      <td> 291</td>\n",
        "      <td> 6</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> IVT</td>\n",
        "      <td>1996-01-07 02:00:00</td>\n",
        "      <td> 0</td>\n",
        "      <td> 4</td>\n",
        "      <td> 292</td>\n",
        "      <td> 6</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0.0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 8 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 3,
       "text": [
        "  category            datetime  arrivals  binofday  binofweek  dayofweek  departures  occupancy\n",
        "0      IVT 1996-01-07 00:00:00         0         0        288          6           0        0.0\n",
        "1      IVT 1996-01-07 00:30:00         0         1        289          6           1        0.5\n",
        "2      IVT 1996-01-07 01:00:00         0         2        290          6           0        0.0\n",
        "3      IVT 1996-01-07 01:30:00         0         3        291          6           0        0.0\n",
        "4      IVT 1996-01-07 02:00:00         0         4        292          6           0        0.0\n",
        "\n",
        "[5 rows x 8 columns]"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's say that we are interested in average occupancy by day of week (ignoring time of day) by patient type (category). Since each day consists of the same number of equally sized time bins, we can use the Pandas `groupby` function to easily compute mean occupancy by day of week."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanocc_df = bydatetime_df.groupby(['category','dayofweek'])['occupancy'].mean()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanocc_df"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 5,
       "text": [
        "category  dayofweek\n",
        "ART       0             1.841435\n",
        "          1             1.700810\n",
        "          2             1.770544\n",
        "          3             1.868866\n",
        "          4             2.144907\n",
        "          5             0.000000\n",
        "          6             0.000000\n",
        "CAT       0             2.562963\n",
        "          1             2.055556\n",
        "          2             2.181134\n",
        "          3             2.136053\n",
        "          4             2.540683\n",
        "          5             0.620833\n",
        "          6             0.278312\n",
        "IVT       0             8.670891\n",
        "          1             7.403183\n",
        "          2             7.350810\n",
        "          3             7.557870\n",
        "          4             9.250231\n",
        "          5             0.969792\n",
        "          6             0.283494\n",
        "MYE       0             2.056250\n",
        "          1             1.873669\n",
        "          2             1.841030\n",
        "          3             1.826505\n",
        "          4             2.022743\n",
        "          5             0.147280\n",
        "          6             0.091293\n",
        "OTH       0             1.455093\n",
        "          1             1.174363\n",
        "          2             1.124248\n",
        "          3             1.147454\n",
        "          4             1.270544\n",
        "          5             0.000000\n",
        "          6             0.000000\n",
        "Total     0            16.586632\n",
        "          1            14.207581\n",
        "          2            14.267766\n",
        "          3            14.536748\n",
        "          4            17.229109\n",
        "          5             1.737905\n",
        "          6             0.653098\n",
        "Name: occupancy, dtype: float64"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Even better than just the means would be occupancy histograms by patient type and day of week. That will require a little more work. On the way we'll see how to access individual levels of a hierarchical index, how to push data into an R workspace and use ggplot2 to make nice plots.\n",
      "\n",
      "Let's start by computing average occupancy by category by date (i.e. roll up over the time bins)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate = bydatetime_df.groupby(['category',bydatetime_df['datetime'].map(lambda x: x.date())])['occupancy'].mean()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "category  datetime  \n",
        "ART       1996-01-07    0.000000\n",
        "          1996-01-08    1.732639\n",
        "          1996-01-09    1.532639\n",
        "          1996-01-10    1.541667\n",
        "          1996-01-11    1.872222\n",
        "Name: occupancy, dtype: float64"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now we can use this to create a grid of histograms of occupancy by category and day of week. To make our life a little easier, let's add a string day of week column to this new dataframe (yes, the output of the `groupby` is a `DataFrame` or `Series`). This will require us to access the datetime level of the hierarchical index of our `Series` object, meanoccdate. What does the index look like for one specific row in meanoccdate?"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate.index[0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 8,
       "text": [
        "('ART', datetime.date(1996, 1, 7))"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "It's a tuple with the index values for each level of the hierarchy. We want the 2nd element in the tuple."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate.index[0][1]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 9,
       "text": [
        "datetime.date(1996, 1, 7)"
       ]
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Ok, we've got that part. Now, how to convert a `Timestamp` (equivalently, a `datetime`), to its day of week label? Reading the [datetime docs](https://docs.python.org/2/library/datetime.html) leads to the `strftime` function for creating formatted string representations of datetimes."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "ts = pd.Timestamp('6/26/2014')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "ts.strftime('%a')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 11,
       "text": [
        "'Thu'"
       ]
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Great. Now just put it all together."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate_df = pd.DataFrame(meanoccdate)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate_df['dayofweek'] = meanoccdate_df.index.map(lambda x: pd.Timestamp(x[1]).strftime('%a'))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate_df.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th>occupancy</th>\n",
        "      <th>dayofweek</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>category</th>\n",
        "      <th>datetime</th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th rowspan=\"5\" valign=\"top\">ART</th>\n",
        "      <th>1996-01-07</th>\n",
        "      <td> 0.000000</td>\n",
        "      <td> Sun</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-01-08</th>\n",
        "      <td> 1.732639</td>\n",
        "      <td> Mon</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-01-09</th>\n",
        "      <td> 1.532639</td>\n",
        "      <td> Tue</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-01-10</th>\n",
        "      <td> 1.541667</td>\n",
        "      <td> Wed</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-01-11</th>\n",
        "      <td> 1.872222</td>\n",
        "      <td> Thu</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 2 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 14,
       "text": [
        "                     occupancy dayofweek\n",
        "category datetime                       \n",
        "ART      1996-01-07   0.000000       Sun\n",
        "         1996-01-08   1.732639       Mon\n",
        "         1996-01-09   1.532639       Tue\n",
        "         1996-01-10   1.541667       Wed\n",
        "         1996-01-11   1.872222       Thu\n",
        "\n",
        "[5 rows x 2 columns]"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate_df.tail()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th>occupancy</th>\n",
        "      <th>dayofweek</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>category</th>\n",
        "      <th>datetime</th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th rowspan=\"5\" valign=\"top\">Total</th>\n",
        "      <th>1996-03-27</th>\n",
        "      <td> 15.801389</td>\n",
        "      <td> Wed</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-03-28</th>\n",
        "      <td> 13.538194</td>\n",
        "      <td> Thu</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-03-29</th>\n",
        "      <td> 18.146528</td>\n",
        "      <td> Fri</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-03-30</th>\n",
        "      <td>  1.436111</td>\n",
        "      <td> Sat</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1996-03-31</th>\n",
        "      <td>  0.227083</td>\n",
        "      <td> Sun</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 2 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 15,
       "text": [
        "                     occupancy dayofweek\n",
        "category datetime                       \n",
        "Total    1996-03-27  15.801389       Wed\n",
        "         1996-03-28  13.538194       Thu\n",
        "         1996-03-29  18.146528       Fri\n",
        "         1996-03-30   1.436111       Sat\n",
        "         1996-03-31   0.227083       Sun\n",
        "\n",
        "[5 rows x 2 columns]"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate_df['occupancy']['Total']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 16,
       "text": [
        "datetime\n",
        "1996-01-07     0.871528\n",
        "1996-01-08    17.759028\n",
        "1996-01-09    15.146528\n",
        "1996-01-10    14.095139\n",
        "1996-01-11    16.148611\n",
        "1996-01-12    16.236111\n",
        "1996-01-13     1.586806\n",
        "1996-01-14     0.725694\n",
        "1996-01-15    18.181250\n",
        "1996-01-16    16.000000\n",
        "1996-01-17    14.304861\n",
        "1996-01-18    16.816667\n",
        "1996-01-19    16.011111\n",
        "1996-01-20     2.631944\n",
        "1996-01-21     0.682639\n",
        "...\n",
        "1996-03-17     0.753472\n",
        "1996-03-18    17.765278\n",
        "1996-03-19    11.013194\n",
        "1996-03-20    12.238194\n",
        "1996-03-21    13.309722\n",
        "1996-03-22    15.611806\n",
        "1996-03-23     0.915972\n",
        "1996-03-24     0.570139\n",
        "1996-03-25    15.807639\n",
        "1996-03-26    14.565972\n",
        "1996-03-27    15.801389\n",
        "1996-03-28    13.538194\n",
        "1996-03-29    18.146528\n",
        "1996-03-30     1.436111\n",
        "1996-03-31     0.227083\n",
        "Name: occupancy, Length: 85, dtype: float64"
       ]
      }
     ],
     "prompt_number": 16
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Exploring rmagic - a prelude to graphing occupancy\n",
      "--------------------------------------------------\n",
      "\n",
      "The *rmagic* extension allows us to magically use R from within IPython. Well, [it's not really magic](http://ipython.org/ipython-doc/dev/config/extensions/rmagic.html). It relies on the [RPy2](http://rpy.sourceforge.net/) Python module.\n",
      "\n",
      "http://nbviewer.ipython.org/github/ipython/ipython/blob/3607712653c66d63e0d7f13f073bde8c0f209ba8/docs/examples/notebooks/rmagic_extension.ipynb\n",
      "\n",
      "After loading the `rmagic` extension, you can use `%R` for a single R *line magic* command or `%%R` for *cell magic* to handle a sequence of R commands."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Load the extension\n",
      "%load_ext rmagic"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 17
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 18
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Create some data in Python and scatter it\n",
      "\n",
      "import pylab\n",
      "X = np.array([0,1,2,3,4])\n",
      "Y = np.array([3,5,4,6,7])\n",
      "pylab.scatter(X, Y)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 19,
       "text": [
        "<matplotlib.collections.PathCollection at 0xba87cac>"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "/home/pcba/anaconda/lib/python2.7/site-packages/matplotlib/font_manager.py:1236: UserWarning: findfont: Font family ['monospace'] not found. Falling back to Bitstream Vera Sans\n",
        "  (prop.get_family(), self.defaultFamily[fontext]))\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAECCAYAAADXf53UAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYBP5AAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0xbab926c>"
       ]
      }
     ],
     "prompt_number": 19
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# \"Push\" these two numpy arrays into the R \"space\"\n",
      "%Rpush X Y"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 20
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%%R\n",
      "linmodel <- lm(Y~X)\n",
      "print(summary(linmodel))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "\n",
        "Call:\n",
        "lm(formula = Y ~ X)\n",
        "\n",
        "Residuals:\n",
        "   1    2    3    4    5 \n",
        "-0.2  0.9 -1.0  0.1  0.2 \n",
        "\n",
        "Coefficients:\n",
        "            Estimate Std. Error t value Pr(>|t|)  \n",
        "(Intercept)   3.2000     0.6164   5.191   0.0139 *\n",
        "X             0.9000     0.2517   3.576   0.0374 *\n",
        "---\n",
        "Signif. codes:  0 \u2018***\u2019 0.001 \u2018**\u2019 0.01 \u2018*\u2019 0.05 \u2018.\u2019 0.1 \u2018 \u2019 1\n",
        "\n",
        "Residual standard error: 0.7958 on 3 degrees of freedom\n",
        "Multiple R-squared:   0.81,\tAdjusted R-squared:  0.7467 \n",
        "F-statistic: 12.79 on 1 and 3 DF,  p-value: 0.03739\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 21
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "So, pretty easy to run R commands from within IPython notebooks. Now let's exploit rmagic to do some ggplotting."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Occupancy Histograms ###\n",
      "\n",
      "A popular plotting module for Python is [matplotlib](http://matplotlib.org/). The project homepage has many links to resources for learning it, with a very good place to start\n",
      "being the official [documentation](http://matplotlib.org/contents.html) and the [gallery of graphs](http://matplotlib.org/gallery.html). The are a few modes of using matplotlib. There is a **pyplot** mode which is particularly well suited for interactive plotting\n",
      "in a Python shell like [IPython](http://ipython.org/) (much like one would work in Mathematica or MATLAB). You can also use matplotlib within Python scripts either with\n",
      "the pyplot commands or via an objected oriented API (similar to [ggplot2](http://ggplot2.org/) for plotting in [R](http://www.r-project.org/))."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here is a basic histogram for overall occupancy (there's a 'Total' cateogry in our data frame). For a more API based approach, see [this example from the matplotlib page](http://matplotlib.org/examples/api/histogram_demo.html) as\n",
      "well as the next version of our histogram below."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# normed=1 plots probs instead of counts, alpha in [0,1] is transparency level (RGBA colors)\n",
      "plt.hist(meanoccdate_df['occupancy']['Total'], 20, normed=1, facecolor='green', alpha=0.75) \n",
      "plt.xlabel('Occupancy')\n",
      "plt.ylabel('Probability')\n",
      "plt.title(r'Histogram of Short Stay Occupancy')\n",
      "plt.grid(True)\n",
      "plt.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEfCAYAAAC5/EqkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0xbab242c>"
       ]
      }
     ],
     "prompt_number": 22
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Clearly there's a day of the week effect - the lower mode is due to the weekends.\n",
      "\n",
      "\n",
      "Now, let's create the same plot, but using R's `ggplot2` package. Our basic strategy is:\n",
      "\n",
      "- use the `%Rpush` command to \"push\" the meanoccdate_df `DataFrame` into the R workspace.\n",
      "- use `%R` line magic to load the `ggplot2` library\n",
      "- use `%%R` cell magic to create and show the plot\n",
      "\n",
      "A big question is, what happens to the hiearchical index in the pandas `DataFrame` when pushed to R?"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%Rpush meanoccdate_df\n",
      "%R str(meanoccdate_df)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "'data.frame':\t510 obs. of  2 variables:\n",
        " $ occupancy: num [1:510(1d)] 0 1.73 1.53 1.54 1.87 ...\n",
        " $ dayofweek: Factor w/ 7 levels \"Fri\",\"Mon\",\"Sat\",..: 4 2 6 7 5 1 3 4 2 6 ...\n"
       ]
      }
     ],
     "prompt_number": 23
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Bummer, the category (patient type) and date fields that made up the pandas `MultiIndex` did not come through. No problem, we'll just create columns from the index before pushing the data frame to R."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanoccdate_df['patient_type'] = meanoccdate_df.index.map(lambda x: x[0])\n",
      "meanoccdate_df['date'] = meanoccdate_df.index.map(lambda x: x[1])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 24
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%Rpush meanoccdate_df\n",
      "%R str(meanoccdate_df)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "'data.frame':\t510 obs. of  4 variables:\n",
        " $ occupancy   : num [1:510(1d)] 0 1.73 1.53 1.54 1.87 ...\n",
        " $ dayofweek   : Factor w/ 7 levels \"Fri\",\"Mon\",\"Sat\",..: 4 2 6 7 5 1 3 4 2 6 ...\n",
        " $ patient_type: Factor w/ 6 levels \"ART\",\"CAT\",\"IVT\",..: 1 1 1 1 1 1 1 1 1 1 ...\n",
        " $ date        : Factor w/ 85 levels \"1996-01-07\",\"1996-01-08\",..: 1 2 3 4 5 6 7 8 9 10 ...\n"
       ]
      }
     ],
     "prompt_number": 25
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%R library(ggplot2)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 26,
       "text": [
        "array(['ggplot2', 'tools', 'stats', 'graphics', 'grDevices', 'utils',\n",
        "       'datasets', 'methods', 'base'], \n",
        "      dtype='|S9')"
       ]
      }
     ],
     "prompt_number": 26
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%%R\n",
      "g <- ggplot(data=meanoccdate_df[meanoccdate_df$patient_type == 'Total',]) + geom_histogram(aes(x=occupancy, y=..density..), fill=\"#FF9999\", colour=\"black\")\n",
      "print(g)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHgCAIAAADytinCAAAgAElEQVR4nO3de3RU9d3v8d/eeyaT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\nAAAAAElFTkSuQmCC\n"
      }
     ],
     "prompt_number": 27
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now let's do a facet plot by patient_type and day of week."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%%R\n",
      "g2 <- ggplot(data=meanoccdate_df) + geom_histogram(aes(x=occupancy, y=..density..), binwidth=2, fill=\"#FF9999\", colour=\"black\")\n",
      "print(g2 + facet_grid(patient_type ~ dayofweek))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHgCAIAAADytinCAAAgAElEQVR4nOydeXxU1fn/z519ySQz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\nQmCC\n"
      }
     ],
     "prompt_number": 28
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Notice that the days of the week are not sorted correctly. Why? Recall the structure of our R `data.frame`."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%R str(meanoccdate_df)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "'data.frame':\t510 obs. of  4 variables:\n",
        " $ occupancy   : num [1:510(1d)] 0 1.73 1.53 1.54 1.87 ...\n",
        " $ dayofweek   : Factor w/ 7 levels \"Fri\",\"Mon\",\"Sat\",..: 4 2 6 7 5 1 3 4 2 6 ...\n",
        " $ patient_type: Factor w/ 6 levels \"ART\",\"CAT\",\"IVT\",..: 1 1 1 1 1 1 1 1 1 1 ...\n",
        " $ date        : Factor w/ 85 levels \"1996-01-07\",\"1996-01-08\",..: 1 2 3 4 5 6 7 8 9 10 ...\n"
       ]
      }
     ],
     "prompt_number": 29
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The dayofweek column is a Factor. We want it to be an Ordered Factor and we want to tell R the order to use."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%%R\n",
      "# Create vector with DOWs ordered as you wish\n",
      "DOW_order <- c(\"Sun\", \"Mon\", \"Tue\", \"Wed\", \"Thu\", \"Fri\", \"Sat\")\n",
      "\n",
      "# Change DOW from factor to ordered factor using vector you just made\n",
      "meanoccdate_df$dayofweek <- factor(meanoccdate_df$dayofweek,levels=DOW_order,ordered=TRUE)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 30
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%R str(meanoccdate_df)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "'data.frame':\t510 obs. of  4 variables:\n",
        " $ occupancy   : num [1:510(1d)] 0 1.73 1.53 1.54 1.87 ...\n",
        " $ dayofweek   : Ord.factor w/ 7 levels \"Sun\"<\"Mon\"<\"Tue\"<..: 1 2 3 4 5 6 7 1 2 3 ...\n",
        " $ patient_type: Factor w/ 6 levels \"ART\",\"CAT\",\"IVT\",..: 1 1 1 1 1 1 1 1 1 1 ...\n",
        " $ date        : Factor w/ 85 levels \"1996-01-07\",\"1996-01-08\",..: 1 2 3 4 5 6 7 8 9 10 ...\n"
       ]
      }
     ],
     "prompt_number": 31
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%%R\n",
      "g2 <- ggplot(data=meanoccdate_df) + geom_histogram(aes(x=occupancy, y=..density..), binwidth=2, fill=\"#FF9999\", colour=\"black\")\n",
      "print(g2 + facet_grid(patient_type ~ dayofweek))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHgCAIAAADytinCAAAgAElEQVR4nOydeXxU1fn/z519yWRm\nspJgCAgVARFZFARxQarWoiBo0YqlVkFbcPnZfr/6rS0qWr/9VqvV2oq0Fv1pVLSUFlncEIogSCmL\nLAIa9iSQfZYks9/fH8+P831y7tybSZhhbpLn/UdeM3d57nPOfc6Zm3Oe87mSLMuMIAiC0B+GbDtA\nEARBJIc6aIIgCJ1CHTRBEIROoQ6aIAhCp1AHTRAEoVOogyYIgtAp1EETBEHoFOqgCYIgdAp10ARB\nEDqFOmiCIAidQh00QRCETqEOmiAIQqdQB00QBKFTTNl2gDU1NTU1NWXbi44599xz4UNNTU1bW1t2\nnekQ7i1j7MiRI4lEIovOpAJ2+NChQ1n0JEW6r8P19fV+vz+7zqQCd7iqqiocDmfXmQ7B8ZBGst9B\nt7W1dYsOmhMMBrtFfHOam5vj8Xi2vegE3SseWHdzuNu1uEAg0NLSkm0vsgMNcRAEQegU6qAJgiB0\nSvaHOM4EWZYXLVq0adMms9nscrkWLFhQUFCQbadU+fzzz++///4VK1aUlpYyxk6dOvXd7373+eef\nnzhxYrZdU2XhwoVHjx49efKk2WzOz88fOXLk/Pnzs+1Ucn7+859Pnjx50qRJPp9v8uTJ77zzzsCB\nA/fv3//rX//6tddeS3pKa2vrgw8+uHjx4rPr6f+irN4hQ4YcP378hz/8YbZcSpHjx4/PmjVr0KBB\n8HXy5Mm33XYbfF6zZk1zczP/mnU67CUaGhr27t17+eWXZ8tDDbp3B7179+7t27e/8cYbkiR9/PHH\nf/7znx955JFsO6VFnz59Pv7449mzZzPG1q5d26dPn2x71AELFixgjC1evLioqGjatGnZdkeLMWPG\n7Nq1a9KkSVu3bi0rK9u8efPAgQO//PLLMWPGZNs1VZTVu3bt2mw7lSqDBw9O+tv2ne985+w7o0GH\nvURDQ8M///lP6qDTj8fjOXXq1L/+9a/Ro0dfffXVl1566fvvv9/W1va9732vsrLyjTfemDhx4qZN\nm2KxWFVV1ezZs7N+D8aPH79p0ybooDdu3DhhwgTGWEtLyxNPPBEMBhOJxMMPP3zo0CFd+Swg1PDP\nf/7zp5566sSJE2az+Re/+EVZWVm2HBszZsyKFSsYY1u2bLnvvvvee++9WbNm7dy588Ybb4xEIthJ\nh8OxYMECg8FQVFSULW812LVr1y9/+cvq6uqf/OQn1dXVuLYff/zxbHunykcfffTFF1/s2LFj5syZ\n8Xj8+9//frY9+v8oe4nW1tZHHnkkFosxxubNm/e3v/1t+/btW7ZsGTduXLadFeneY9D9+vV76qmn\nVq5cedttt/3iF79obGxUHlNdXf3EE0889NBDq1evPvseClit1uLi4mPHjp06dcrj8djtdsbYu+++\nO2jQoD/+8Y933XXXiy++yHTmszYrV64sKCj4y1/+8sADD/zP//xPFj0pKytrbGwMh8NfffXVFVdc\nEQgE2tra9u3bd+GFFwpOLlmy5Morr/z9738/atSoLDqsRiQSefLJJ++///5//OMf2fZFi4MHD849\nDeQ1VVdX//Wvf3W5XNl2rR3KXqKuru7mm2/+4x//OGXKlPXr18+cOXPUqFE67J1Zd3+CrqysLCgo\nWLhwoSzL69atW7BgwYwZM2AXTywbNWqUJEler1cnucDf/va3P/roI4fDMXny5C+//JIxduzYMfiv\ncMSIEdDH6c3npEANf/PNN8eOHYMnO7PZnEV/JEkaMWLEqlWr+vXrZzAYRo8evWbNmry8PIfDITh5\n7Nixm2++mTE2cuRIHXaC8LORl5cHj3iADhMlzzvvPGGIY/To0QaD7p75lL3Eb37zm02bNm3btq22\ntrZv377ZdlAL3dVmp6ipqXnppZcSiYQkSd/61rdisZjBYIAcz3//+99wjNFozKqPIuPGjdu8efOm\nTZvGjx8PW/r167dz507G2JdffnnOOecw/fmMEWq4X79+Y8eOffzxx++9996sj8aMGTPm1VdfhUeh\n8ePHv/rqqzAALThZXl6+detWxtj27duz63BS8N1XxrOeMZn0+MCn7CXefffdESNGPPTQQ2PHjpVl\nmTEGf3WIHis0dS699NK9e/fOmTMnGo1ardZf/vKXHo9nxYoV//Ef/5HFwVBtLBZLWVlZJBKx2Wyw\n5Xvf+97jjz8+b968eDwOY9DZ9VAbGOrlNTxt2rSFCxfOmzfPZrPdeeed2fXt4osvfvLJJ8eOHcsY\nu+iii3w+H3TQgpOXX375L3/5y/Xr1/fv3z+7T/0dItQ20QWUvURLS8uSJUvWrVtXVlb2+eefT5ky\nZe/evRs3brzsssuy7ayIlPWfjurq6pqamuz6kAqjR4+GD19//bX+VxJybxljO3fu1OE/yALY4W7x\ntNh9HT5+/HhtbW12nUkF7vD+/fv1v5IQx0Ma6d5DHARBED0Y6qAJgiB0Sno6aFmWV61a9frrr1dU\nVOB/RmKx2K9//etFixbBSh61000mE55eyM3N5eOzjDH8WZKkgoICSZKS7rXZbLm5ufyr2WzGZt1u\nt9VqTXqiwWAQFhfhvXa7XcgcslqteLY6Ly8PXwifazabvV4vvpDFYuFfc3JyHA4HNouLlp+fj6+C\nzVosFo/Hw78ajUY8lqrMc+pUYXNycrD/2AePx4MvhE80Go15eXn8qyRJuLYdDofT6VQrKT4xaWHd\nbjf/mnq0MMbOTrQYDAZ8sNPp1C5shqIFzy5qRAtT1GG6mobZbMY+aEdLfn4+/6qMFhyEUFj8Vbtp\n4GgxGo24pC6Xq2vRYrVacbQoK5BlhvRMEh45cqS1tXX27Nk7duzYvHnz5MmTYXtTU9PgwYNvuukm\njXNdLpfRaJQkiScV5eXlhUKh1tZW+Gq1WrnYoCRJJSUlFouFD53jvQ6Hw2az8Wxok8kkyzIffs3P\nz29tbeVKofhEg8FQXFyMgwnvdTqdeFdeXp4sy9FolPtQXFzc0NDA/cfnQrjwO20wGIxGYzQaha9u\ntzsejweDQX4wNltSUmK327n/2KzNZsvJyYE0ajBrMBi4A7iRM8b69OljNpv5uUajsaioCJfIYrFE\nIhH4nJOTYzKZmpubeR0mEgme7VdYWOj3+7kp7JLJZCooKOANSZIks9nMzbpcLkmS+Ni9xWKJxWLc\nrLBgpKSkRCis0+nkfZMQLV6vNxwOn3m0CGa1o0VYAgo1zAsLLZkX1mw2x2IxHC2NjY08BoRo8Xg8\nGtGSSCQCgYDyRPBBLVqsVivuRnNzc61Wq9A0WlpaQqFQ0sJqNA1ltGCzHUYL/vnBQehyuXD/C78Q\nnWoa9fX18FUZLZFIhD9ECnVYWlqqFi12u91ut6tFS15e3tGjR/lFOyT1Aev0PEEfO3YM8sPOOeec\n48eP8+2NjY319fXvvPPOe++95/P50nItgiCIXkJ6sjjWrFnTv3//IUOGRCKRP/3pT/PmzYPtx48f\n9/v9w4YN27Nnz+7du7l+yl/+8pdjx47B58svv3zSpEnY2i9+8Ytf/epX8Nnlcu3YsWPgwIFn7uSZ\nEI/H+f9u0WhU57lZAolEQofLB4iziSzL/J/3SCSCH131SSwW42MI+DMwduxYyGRnjF100UU7duw4\n2/4pyNATdHqGOGw2G/yb4/P5+P/djDGevDl48GCsAnP77bfzf29DoVBdXR1Da6X27t374q23zho7\nljF2+TPP7Nu3j4/oSZJUXFx86tQp/ruC/zOy2+1Wq5X/wwVdKjebl5fX2trK/4/DJxoMhsLCwlOn\nTnEP8V6n02kymXi5AoGALMv4n9bCwsKmpib+/w4+12Kx5Obm8jsnjEXk5ubG43H+D5fyf+H6+nru\nPzZrtVqdTif/h8tgMEiSxI/E/yYzxurr600mk0ZhzWYz/z8aCsv/3TEajbIs85uVn58fCARw6fhn\nk8mUl5fH87ckSTKZTNxsTk6OJEn833Oz2RyPx/HICR67PHnyJLZss9kcDgf+75Kh2+rxeCKRCB/i\nwCcyxvr06aMRLTabjUvXC2a9Xm9bW5tatAhjMqdOncKFhfEEXliTyRSPx9MSLYlEAg+I4ZIWFRU1\nNDRoRAvvlEOhENzfdDUNHC3YrEa0wBg0zvbDQQjRwjtlsI+bxldffbX7scf6ejx1gcDwp55qbGxU\naxpnLVpYZkhPB11eXg6Lsqqrq/v168e3b9y40Wq1XnzxxSdOnMAxjYf8Y7FYJBKRJAkva3ZYLF6H\ngzFmNBgYY3wXPAUkEgleibj7gI38q8FgEPYKX4WF1Pgr3osvh+3gjXigVnkutoxLmtSlVMwKJYXC\nCnuxEXwur8OkhRVcArMaB+OSsvZ3SsNh2JWiw0r/hTpUq6UOo0XDrEZJlQgeJi2sEEKp1CFLIVoE\nNzTM4qsnNdWFpqGMFuFcNYcN7du18kg8WcfrE9eh2273OhyRWEx5Lmsfh52tw6TR0mEQssyQnv98\nBwwY4HQ6Kyoq9u3bN378+Nra2ldeeYUxNnr06MrKyiVLlmzcuPHaa69Ny7UIgiB6Cel5gpYkCYvA\nOhyOe+65hzFmt9tvvfXWtFyCIAiit0FzRwRBEDpFF2JJZrNZkiQ+TYSHn1j7VQCwC7I4YYvRaOR7\nTSYTPlgY5zIYDCaTie/FJ8IkGx4ZV5rF/kiSBMNt3CuLxcL9x+fCQg/sP5zLj8TXNRqN2CxjzGKx\ncP87NMsdgIFjfhVYPiBUi1BYXucmk0k4GIbt+IVwBgs+Eq6uYVajpDCcB47Jsmy1WlMvLCzSSXpb\ngRSjRVmHatEixGc8Hoc1I2pVAenMQh3iGMAuYTsdRgsuKQShWrRgn8EBWZb5LJzBYIAtynOTRgv2\nQRkt3GyXo0VQc4RdQrULBwiF7cJtBdSipcMWl6FhaF100NFoFGd9K6eM8NIDxlg4HE6aTA6rhnAy\nPE6bTyQSsVgsado81C9OWReS6nFwwHQBTpuXZTkSiSRdqCLLst1uxxfCSw9sNls8HsfrFLBZxlgk\nEkmajQ8hiM3i6X6Hw4F/UeByuJbwV9Z+thpaFC67MH0UjUbVlh7gE4WFKhaLRZIkXFK8UEWWZfzz\nHA6HhcJaLBbsP44Wh8Oh5lKnokUwqx0tDGE0GsPhMC4sNGN+vJCcA3WYdKEK/BZqRAtuC8IiCwjC\npNHC2i+K49GLm4ZaHUJh1ZoG9IZqLS71aGGKbBZcyWBQaBoY3IiEpqGMFrXbCqhFC/TsamadTqfG\n78eZQEMcBEEQOiWzWhxAKBR6/vnn03IhgiCI3kN6OmiuxTF06NDNmzcLe9etW8czw4msEwwGy8rK\n+p7m3nvvzbZHBEEkJz1j0FiLA97exKmqqgqHw5lTeyI6SzAYbKytbXz+ecbY55WV/4FWeBIEoSvS\n00G3trbCQkG3240flhOJxNq1a2fMmLFkyRJ8vLYWhzCHm5ubW1paireUlJRoOCMcjLFarRqLMjVO\nxG8kyc3NVWpxCGt/U7fMToufJaW4uLjLZjlFRUXCvJYkSbBQ02Wzmc1mQU+SMYZlLQWwSmRnXRI0\nJDtlRMOy3W4X5DQxGYqWDi+hUdjCwsKuucSSaclyNKIFT6/Z7XblA5OQzNAplzIRLfiFuR6PR+Nt\nh5IkKa+iHS0az4tdjpYMkVktjq1btw4bNgxr4wLaWhzC+5n8fv/JkyfhM2lxwOcz0eIQJF1isVhD\nQwNpcZAWRzfS4sDIskxaHB2gpsVRU1MTDAa/+uorn89XUVFx++23w/YOtTgESItDu6SsM1ocwl6l\nD4JmAmlxsGTRouFw0sIKIURaHNpmO9TiEEpEWhxaqGlx3HTTTXfcccesWbPcbjfvnQmiU+zZs8fj\n8fBZzZdffjnbHnXAW2+95XK5wNvy8nKNdwkRhDaZ1eLgzJ8/Py0XInohx44dG9237/vz5zPGXlq3\n7sA332Tbow44ePDgA1df/diUKYyxO1977dChQ2PHjs22U0S3RBcrCQlCG7PRCLOa9m7yqgSbyQQO\nW9RntwiiQ3QRPaTFwb+eBS0O4W0aUC161uIQsnpAOELPWhzKteDYB9LiYKTFkTK66KBJiwO+nh0t\nDjxtzatFz1ocQlYPCEfoWYsjkUjgI+AW42ghLQ7S4kgR0uJIA9ddd11xcTFMCk2YMAGncBJE92LW\nrFkFBQUQzOPHj+c/D0RWSM8TtCzLq1evhnzbadOm8cTnSCSybNmyUCiUSCSmT5+euWzB7LJ///5N\n//mfQ0pKwrFYyX/8h/DERxDdiK+//nrDz352Qd++0Xi8+Gc/ow66CzzyyCPXX3/9pZdeeuZvl86s\nFseuXbtKS0vvvPPOUaNGbdmyJS3X0ie5drvX4fCgRToE0U2BYPaqrw8ktLnxxhs/+eST7373u08/\n/fTOnTs1Mug7JD0dNNbiOH78ON9eXl4+atQodnpgKC3XIgiC0DPjx49fuHDhoEGDJkyY8Pbbb99y\nyy1/+MMfumYqs1ocsHHp0qVHjhyZO3cu397ztDgweXl5eE5G2zLLhhYHRv9aHMJZdrtdWGmtNy0O\ni8WCH5lcLpdwgJ61OITWV1xcLEkSaXF0aFaJ0WicOHGiw+FYvnz50qVL582bl/q5nMxqccDs9syZ\nMw8fPrxy5co77rgDtvcwLQ6BpqYmPqVLWhzw9Uy0OASFcQgYPWtxRKNR3MkFg0Hsg861OITWV1dX\n53A4SIujU1ociURi165dhw4duuGGGwYPHnzdddctWLBA6XkqZFaLY8OGDQUFBSNHjgS9GL6952lx\nYHDiMGlxJHVYKf5wJg7rU4tDsCOEkJ61OASbTL1pKKNF8ErN4Z6txTFlypR+/frNmTNn0qRJGv8f\np0J6OugBAwYcOHCgoqLCYDBMnTq1trZ2+fLl99xzz7hx45YvX759+/ZEIjFlypS0XIsgCELPPPPM\nMzCKFQqF+D9h2orEamRWi8Plcv3gBz9IyyUIgiC6Bc888wx8OHTo0LnnngsfNmzY0AVTulhJSBAE\n0WN47bXX4MPcuXMXL17MGLvrrru6ZkoXHXR31+IQAGUDXjTS4mCkxdF9tDiE0oER0uI4Qy0O7b0a\n6KKD7u5aHALYYZm0OE4XjbQ4cB3qVotDiGrYSFocXdDiCAaD+/fvb21ttdlsPKuks+iigyYIgugx\nPProozU1NTU1NXfffTekFHdZEDyzWhyxWGzlypU+ny8UCk2ZMqVv375puRxBEIRuueqqqy688MKC\nggKDwXDDDTfU1dUNGjSoa6Yyq8VRWVlpsVhmz559ww03rFmzJi3XIgiC0DOTJ0/my3e9Xu955533\n1FNPdc1UZrU4cnNzL7nkEsaYw+HIkF4qQRCEzjl69GjXTsysFgcsbK+qqlq5ciUW3CAtDgxpcWgb\nIS0ODGlxCOhQiwNS6zBcrKKzZFaLQ5blTz/99NixY1OnTu3Tpw/fTloc8JW0OOAraXHwz6TFwbq/\nFody0WCXhaEzq8Wxb9++pqam2bNnC09wpMWh4RI2S1ocqThMWhz8dNLiECItK1oc06ZNEzx8//33\nk3reIekZgx4wYIDT6ayoqNi3b9/48eNra2tfeeUVxlhlZeWJEycWL168aNGit956Ky3XIgiC6F6M\nHDmyaydmVovjxhtvTIv97kIskWCM8WzCyy+/fNmyZVn1iCA65sEHH1y0aBF81piuILrM/Pnzu3Yi\nLVRJJ7F4nDFW88wzVpPpq5qam994I9seEUTHHDhw4PU777zhwgsZY0U//Wm23SH+F1100D1Mi8Nj\nt9vM5ly7HeQRSIuDkRaHvrU4JElyWixJX0JIWhzc7JlocXQZXXTQPUyLgyPLciQSIS0ORloc+tbi\n0Ihk0uKAz13Q4kgLNN5EEAShU9LTQcuyvGrVqtdff72iokLIWmWMbdiwYd++fWm5ENHL+aqm5q23\n3iouLu7bt+8555zzwQcfZNujDthbXf3Tn/60tLS0b9++55133pEjR7LtEdGdyKwWRyKRWLJkyfr1\n69NyFYKoDQRmjR3b+Pzzjc8//8NLL/3mm2+y7VEH+EOh333ve+DwQK+3uro62x4R3YnManFIkjR7\n9uwJEyak5SoEwRizmkxeh8PrcFjVl//qCofFAg4bSY6G6CSZ1eKQTiMc37O1ODhGoxFkLkiLA0iL\nFgcmNzfXbrfrXIsDk5+fbzQa9abFoREepMWRotkMkVktDjV6thYHJx6PNzY2khYHS6sWByYQCIRC\nIZ1rcWAaGxvj8bjetDg0VrGTFgd87pQWRxrJrBaHGj1bi0OwKVxIV1oc1T7f/v37+RP0/fff/+ST\nTwqaCXrT4sDwCtSzFofgsIblbGlxaEBaHHh7ilocaSSzWhyEzqkLBIb06QNTWL+/9Vb9z7kRRK8i\ns1ocgDDETOgKo8EAS8gc7VcYEgSRdWihCkEQhE7RRaJSD9Pi4HQLLQ4MVJHOtTgw4K3OtTgwEOo6\n1OJQc5i0OLhZ0uIgLY4saHFgoIqYvrU4MLFYLB6P61yLAwP3l7Q4SIsjRWiIgyAIQqek5wlaluXV\nq1dDvu20adOcTqf2doIgCKJDMqvFobadIAiC6JDManGobScIgiA6Rk4Hq1ev3rdvH8xvvPTSSx1u\nf/XVVx87zdq1a4X1PHi4HevL4BESvNYWL5/Hq/L5PAOecMCL+rGGAJZHwAYffvhh+fTCbiASiciy\njP8hwPYLkVQCdgbrJPCC4JJiIzjRAusG9OnTJ+lnrs8gSdJf//pX4e7E43FZll944QV+fC6aMjKj\nGXOsaMF9O5Pawyof/FzB4KFDh5IG1axZs/hhho5mYFKpPewYrj2u9aF2O3AqwjnnnAP1qeRb3/qW\ntpM4mFOpPRw/OGFAuVEwiKt92rRp8ukVqgDMg/HF0GLBsZOoJtNSe6kE89NPPy2fnsYE4DOWlsUF\nx1I2Xe4KsEAFrj21YOaFMhgMH3/8sSzLR44c2ZYyHfaonMxqcahtT6rFsX//ftjidDqj0ShM6cqy\nDLkNjLFYLGY2mwsLC+vq6iKRCNSR2WwOhUJww2w2m9Fo5NINiUQCJyHk5eW1tbW1tLTAjTGbza2t\nrWBEkiSPx8MX2sdiMbvdDhe12+2BQECpxVFWVsYdzs3NbWlpgbnmeDxus9ngXHDe5XKBAgPM/BoM\nhnA4DD7k5OTwae54PA5pUvB7YDKZCgsLGxoaIpGI0mGLxWKxWLgsgyzLRqMRiul0OkOhkFKLY9q0\naddeey0U1u12c8WSRCIBORVgB9QV/H5/NBqF1CLojyCyc3JywuEwvx28lmKxGEgL1dfXg/OQhBQO\nh+HWOJ1OPnueSCRgoyzLUMNQOdxh0OJ44oknfvGLXzDGrFar2WzmheWrbGVZhrJEIpFgMAjpj2az\nua2tjbe9vLy8pqYmuFA0GoXXr0BtO51Oq9Xq8/ng6pDJE41G+a2JRCIQhLikRqOxvr4+qRbHypUr\n4UJCYc1mM+8fQdmnubmZxwB22Gw22+12v9/Pb6vJZOI5DHl5eYlEAvRM4L1fPPIZY263OxAIQAxg\nhy0WS1NTk1KLQ5blpC0ukUhYrVYewAaDoaCgoK6uDuJBaHHwqwDSFlB2iMN4PG4wGPLy8oLBIC4d\n/2wwGNxuN25xNpsNgtDhcASDQaUWx7Bhw7jDHo/H7/fzZejc4Vgs5nA4HA5Hc3Mzv62MMUizYYy5\nXK5IJAK3BtcSlK6oqAhKqnTYZrOZTKakLQ4aBcsMmdXiUNueVIuDPwbm5eWFQiEuaCIkTnm93lAo\nJKNUGN4ZORwOm83GfySFpB+PxwP9mvJEEEvCAlr4okm1OAwGA3cYdGr46fhci8Xidrt5ppSQOOV2\nu+PxOJa/wblEXq83Go3ixCnusM1mA0EZbhan2cmKPEVo6uCw0WgsKirCuWs4wyknJwfqjdchTrMr\nLCz0+/1qiVNer1ctzc7lckmSxLsenGan1FiALTabDcprs9lwGpOQ4eR2u3lnJ9SSJElwW+VkiVMQ\nLThxGJvNz89vbW1ta2tTnqiUsACH+bMkPNPxwgoaWF6vF34JlA5bLBaPx8OPVEYLTm/HJzLGioqK\nsGCWhsP86YwHcH5+fktLC1eGEtLLoMXx0/F1IVr4j4TQ4rxer9Fo5AfjE00mU0FBAW5xOAjlZFoc\nRqMRtziG8i+xw9A0+C7htnq93kgkwp/hhDQ7jb4FRBPVglDOmBaH8fHHHz9zKx6P59ixY1u3bq2v\nr//2t7/d1NT05ptvjhkzRtieVKWT/6DhQuIkWWGXJElCJi/eyzN5k56obVb4GeR74XmNP0G3tbVh\nJTbuktCJ48+8jbHTzYOfGI/HcV/ZNbOgTITN4v9XgsEg3gto1GE8Hteow2g0qlaH2CyEslCHGmbx\nbzY8JGrUoWBZqMPUo0WoQ6HytaMFD6QoHcYuKS0LKb0at0YjWuCmCyVNahaO5H0N/A/Utduq3Itv\na9KSphgt+FwoKY/h1tbWTrW4RCKh0TQgdz7piRrRAnY0zIbDYay0rE3qsqWSUGUEQRCETqCFKgRB\nEDqFOmiCIAidktmVhLFY7Nlnn4XkreHDhyd9OSHM8+L5DZfLFY1G+bwEftUCTII3NDTwkRm812az\nWSwWPjkjvA3E7XaHQiEsI4Dn7rxeb0NDA/cK77Xb7UajkRcKJoghtwG25OXlQTKA8lyz2ex0OnnK\nBOQ28DE7mM3gM1EwZc/N5ufnNzU1cf+xWYvFYrfb+ZssBFkol8uFk67q6urwXJOysCaTibsEheXz\nljBVzV3yeDwtLS24dPyz0Wh0u904eQubdTgckiTxyRnBrNfrxUO6p06d0igsZMKkEi2MsYKCArVo\nsVqtVqtVLVpyc3PD4XDSaIEgxGkntbW1+LZCqPDCCrfV6/UGAgF+sBAtOTk5PLdBGS2JRIIPdCqj\nBSJTadZisTgcDj7KHwgEINMON422tjY+/NrZpoGjBZs9w2jhk66NjY2Q2sQPPpOmkZZoEczm5uY2\nNjbyG9chF1xwQYpHpqeD5isGd+zYsXnz5oecetAAACAASURBVMmTJ8P2pqamwYMH33TTTRrnwh3F\nHTQkPPGvPNOInQ5ZHJd4L0z+CrMoal+FE8Es90pplu+Cq2MfQD0uqWVI9MFFwx5yg/xgbJZnLCV1\nCZsVvkoKuTW8l19I7atQhziLI5Vbk7S2lSUVzAqHadxWoVo0bqt2tAglFb5qlBSSKwSHhRrGheWp\nikn973K04B6NXyWpWWhf/Eio+bQ0DaaIFmwn9WhhyYIQO8zaZ/uk3jSUHgp1qPQhlb5FaRZrMKWR\n9HTQeMXgzp07+fbGxsb6+vp33nnHaDRec801PE39wIED/Fc3Ly8vPz8fl/bEiRMffPABBJ/L5Zo5\nc6bQHhwOB/6V43stFovRaOQPj8LvudFoBE015YkwiYyfOgWzwjoI6bRgIz8dknkZY+vXrz9+/Dh8\nHj58+NixYw0GA7fMc3XhK6Qo8b0mkwmbZYzZ7Xb8mMBdggxlwSxu57i67HY7zoWCsuDCmkwmnqYG\nV1GrQ1BcxG7wzzgxlp1ugdgsrmEhe0/ooB0Oh1BY7JJQWJwUKLjEraUSLco61IgW1h6Hw4ELC2ng\nuA5x5yJJks1m4zGALQsnKqNFlmW+VygpBKFGtPAjjUaj3W4XmobVahX0ablZjabR5WiBomkEIb6n\n0ADxxo0bN+7YsQMudN55511xxRWpNw2LxZI0HoAuRwvLDJl9q7fD4Rg/fvywYcP27NmzevXq2267\nDbZXV1fzN2MajcaSkhKcZPPzn//88NatAwoKGGMff/XVwIEDx44diy+H/x3muej8s5qWMdxj4WD4\nLJ2WmU5qVnijsNFohNaImxysMQmHwzOmT5920UVGg8HX1vZ8MHjw4EFJoU6LG4Ysy9hhpVn81KBW\nUqVZobqEwgppbfiRUJAkTqoHjTtlNbPwFZvFNaw0ix2GVp1iYeG2SihBVejuU4wWZaaXdrQIDneq\nsIJwM17nlnq0JC2pWrRgn/mqnLQ0DY0Wpx0tglkchHwNCAC+4aZx4403Tigrc1gsoWj00W++aWpq\nSr1pKKtCqMOkezsMQpYZMruSsKysDD4MHjx47dq1fPtVV13FPwcCAZ/PJ6Gs72g0eu8VV9w1YQJj\n7KInnwwEAnhUzm63Nzc381ulXHrADxbS5jWWHhgMhuLiYjyEhPc6nU6z2cwfZNra2uT2etDFxcU+\nnw/+x0nIcsVdd9nM5j3V1dcsXhwIBPCKqU4tVLHZbHhoW5mNj83ihSperxffBZ/PZzabsZSt1WrF\nhVUuVOGD5sqFKsFgUG2hSkFBAb5TKS5UYYwVFRXhX8GmpiahsE6nE79QWWq/9ACnoArLmlKPFmPK\nC1UMBgNemgwO48JqL1QpLi6GhZpKy7BQRSNa+EpCplhk0adPH7VosVqtLpeLd3+w8kBoGhoLVTSa\nhjJasNnUo4W1D0KXy4V7QxjNx00jHo+/fPvtZV7vKb+/36OPwqg67BKaRqcWqmhECyxUUTObl5fH\nMkN6Ov7y8vKqqiqmWDG4cePGf/3rX4yxEydOCKtjCYIgCG0y+1bv0aNHV1ZWLlmyZOPGjSAEQRAE\nQaRIZt/qbbfbb7311rRcgiCIs8CuXbv2798P4yr9+vW78sors+1Rr0YX7ySEYXvl3ItwgNqRGnth\nLD/pwZ0yK5wlJN7xY5RFECwLhyX9qjSb1HkNs6x9aqDSB2U9pG5ZeW4XSpr0ooLNTlVaKrWk/bXD\nahGO5PCqTjF+OrSsUVKlhymaVVa12u348Y9/fJ7D4XE4ErL8Xzt2HDp8WKNo2v6nXlI1J9UOVjsm\nRctdiJYOg1DDtzNBFx00TF+olVZCwoOwC8/nwvQu/4wPhqlVbBbnQmmcqNwr+KNMt4CDhZRVuKLg\nP7YsfBXMsvaTzvjIpGax89gNmIAWiqNWWOFg+IpzG9TqEG6iRh1ih+FgPKfPuzlZliE3QK2wYApX\nsrIOcfHVokVplrWPFjWzQvUqHRZqOOltVXMJn5hKtOB6MKG3GyvNYpQtjt/WeDz+zM03jzjnnGg8\nbvnJT7SjBWanu9DiUokWXDQJJcko0e4EsCmYd01LtGj08mlEFx10LBaT1LX7BBEp1n4+12Aw4KVK\n+GCT4h3D8XgcL5HCn+X2eml4r5AHrXyrN3yNxWJKxTWYvleTG4Vke1w6QWYMK29hl6CFY7M4i0NY\nmADVi89lp1/1ja8Ln+PxOP4Kaa349dtqdSif1lzmNjXMSu214oTn/Wg0KhQW65MZFa/fVnOpU9Fi\nVAhIakQLQ8AuvOoMyoVrGGdxwFdth/kuIVpwPeAT2ekgTBotEN7YZ77Yip+rjF6AyyUL5YXPVqsV\nu6RscWolFaKFtQ9CrKrK61OoQ4wQHhq3VSNaeHmTRgusUdAwm9SxM4e0OAiCIHRKejpoWZZXrVr1\n+uuvV1RU8BxDTigUev7559NyIYIgiN5DZt/qDaxbty51KWuCIAgCyKwWB2OsqqoqHA7jl0WyjrQ4\nhBF30OLCW3SrxaGcToRXcJEWB2lxdBctDqF08J4nvWlxYKA+SYtDCzUtjkQisXbt2hkzZixZsgQf\nr63FIYQIRA/eolstDiGAoHVJpMVBWhzdR4tD+XjE9KfFIaC8U6TF0Q41LY6tW7cOGzaMKylztLU4\nhCnRtra2bqTFgT0H5QTS4mCkxdF9tDiE1ufz+eDN6HyLHrQ4MLIskxZHB6hpcdTU1Ozbt+/NN9/0\n+XwVFRVpuRZBEEQvIT1P0AMGDDhw4EBFRYXBYJg6dWptbe3y5cvvueceLtX/0ksv3X777Wm5FkEQ\nRC8hs1ocnPnz56flQgRBEL0HXawklDpaa4+XmSmP1NgLY/lJD+6UWeEs0uJIemTqJU16UcFmpyot\nlVrS/tphtQhHcnqYFofyMI2iafufekk1HFDzIekxKVruQrR0GIQavp0JuuigSYsDX5S0OEiLI6kP\n2iWV0qTFIQAbSYuDtDhIi4NJpMVx2hmDihZHW1vbP/7xD3BDkqRp06bBm9F1q8Vx4sSJzz77DGLG\n6XROmzaNkRYHaXGkjC46aIJIkTfffPP1l14aU17OGNt+7FggEPiv//qvbDulxdNPP31i585zT79g\ns7S09MYbb8y2U0S3IT0dtCzLq1evrq+vN5lM06ZN44nPkUhk2bJloVAokUhMnz7d6/Wm5XJEryUe\nj199/vl/uuMOxtiPKyr4I4xuicfj90ycOGfiRMbYqKeeEgbBiJ6Kz+fbsGFDdXV13759r7jiCpfL\n1TU7mdXi2LVrV2lp6Z133jlq1KgtW7ak5VoEQRB6Zs+ePXPmzDl69GhhYeGRI0fmzJmzb9++rpnK\nrBZHeXk5LGmT2q8tJi0O+EpaHNgljoYWB686wGKx6FyLQ5iDgkXhpMVh6NFaHL/73e+eeeaZ8vJy\n+Dp16tSnn3765ZdfVjrfIZnV4oCNS5cuPXLkyNy5c/l20uLgprKrxSHL8nvvvdfS0gJXufrqqwcO\nHCjpWItDeS8M+tbiEJwBV0mLQ9niDD1IiyMWi/HemTFWVlbWZTnPzGpxhMNhs9k8c+bMw4cPr1y5\n8o477oDtpMUBX7OuxbFt27YHfvzja4YOZYwdrq//4IMPKioq9KzFwW8fvxegLKFbLY5IJMLQc2Iw\nGEwkEqTF0bO1OCKRiN/vhzBgjPn9fn7XOktmtTg2bNiwa9cuxpjJZKLpER0Si8XK8/PfnTv33blz\nf3zFFZnLFiKI3sOsWbN+9KMfffjhh4yx2traO+6446677uqaqcxqcYwbN2758uXbt29PJBJTpkxJ\ny7UIgiD0zPXXX3/++eefOnWKMeZ0Ol9++eXS0tKumcqsFofL5frBD36QlksQBEF0Cx555JHrr7/+\n0ksvZYw5nU6l3nLq6GKhitTRWnu8zEx5pMZeGMtPenCnzApn9RgtDrU617DMUqtD7ZImLaxgU9sH\nYVcqtaT9tcPgEY7kaGtxaLjdoQ/KOlR6mKJZpRsp3g6NonXof+ol1XBAzYekx6RouQvRoh3Pwlk3\n3njjBx988NRTT1155ZXXX3/9hRdeKMwipo4uOmjS4sAXPZtaHEK+BzgsXMigJy0OwWEwq6xDfEx2\ntTiEYAb/SYujZ2txjB8/fvz48T/5yU8mTJjw9ttvP/nkk5MmTZo3b56a/xroooMmLQ5eurOsxSGs\nxOMaF7rV4hB+AuPxuIa6QqeiJUNaHMK8K/hPWhy9QYvDaDROnDjR4XAsX7586dKlXeugM/UqLYIg\niN5JIpHYsWPHoUOHbrjhhrfffvvKK6/86KOPumYqs1ocsVhs5cqVPp8vFApNmTKlb9++abkcQRCE\nbpkyZUq/fv3mzJkzadIkng3dNdLTQXMtjh07dmzevHny5MmwvbKy0mKxzJ49u7q6evXq1XfffXda\nLkecIatWrQJxgMOHD2fbF4Loabz++uuFhYV4y8KFCxcsWNAFU5nV4sjNzb3kkksYYw6HA4+KkhYH\nfM2KFkd9ff1NU6feNHIkY+zr2lp8pNFoJC0OXhzS4mCkxdElLQ6hd2aMHT16VOl5KmRWi6OkpIQx\nVlVVtXLlykmTJvHtpMXBTZ19LQ6j0Wgxmd6dO5cx9vt16/68cSM/EhyQSIvjdEk1okVwWK2wgjOk\nxYHrsKdqcSxevFjw8OTJk0k975DManHIsvzpp58eO3Zs6tSpWL6AtDjga1a0OLgghpJoNNrW1kZa\nHIy0OE5DWhx8V+paHPC0ihH+80ud9HTQ5eXl27dvZwotjn379jU1Nc2ePVv4hSEIguipwIvNMO+/\n/37XTGVWi6OysvLEiRPwwJ+bm/v9738/LZcjehXxeLyiogLGxLZv316SbX86pLm5efny5fD0ffz4\ncebxZNsjIsuMHDmyaydmVouDXr9GnDmffPLJIw8+eOV55zHGtn/zzXeHD8+2Rx3w+9//fsX//b/D\n+/ZljB386iume4eJTDN//vyunaiLlYRSR2vt8TIz5ZEae2EsP+nBnTIrnNUDtDi00bDMUqtD7ZIm\nLaxgk++Nx+PDSkpgVvOGl14SjkxqWc1bja8dBo9aBSq1OOLx+NQRI349fTpjrOyRR5IWsGt1qPQw\nRbNJIzOV26G8qMbXztZh1+JB45gULXchWrTjWdu3M0EXHTRpceCLZlqLQ0PTQNKlFofGBIZBl1oc\nGs2VtDiSllRpVtnDdiMtjjSiiw6atDh46c6CFofGm7D1qcWh8RoBfWpxqGWDMdLi6E1aHGmBkisI\ngiB0Sma1OIANGzYUFBQMHTo0LdciCILoJaTnCZprcQwdOnTz5s18eyKRWLJkyfr169NyFYIgiF5F\nZrU4JEmaPXv2unXrhONJiwO+ZkWLw2azMRX0qcWRdPIK0KcWh8asJmlx4KL1VC2ONJJZLQ4hGYVD\nWhzc1NnX4sC1J2DQpRaHRgetTy0OjQ6atDhwHfZULY40klktDjVIiwO+khYHfNXW4gCNiKToU4sj\nFouptSvS4uBH9mAtjjSSno6/vLy8qqqKKbQ4CIIgiC6Tng56wIABTqezoqJi375948ePr62tfeWV\nV9JimSAIoteSWS0OACtBEwRBECmii5WEEmlxqHub1OGkHuK9PUmLQ9vPpJbVvNX42mHwqFWgUosj\nRbc79EFZh0oPUzSbNDJTuR1Ji5auOkxjPHTKcheiRTueU7/1nUUXHTRpceCLkhYHaXEk9UG7pNIZ\naHF88803a9euhdlI4ZUIpMXBHdYw28M7aNLi4KUjLQ6JtDjOuhbHgw8+2HLo0DleL2OssbERH0la\nHLxoWdHi0EUHTRBEFonH4z/99rdnjBrFGDP/+MfZdof4XzKrxaGt0UEQBEFokFktDrXtBEEQRIdk\nVotDbXtSLY5XX30Vthw9enSD1RpPJBhjtYHAJ598UllZyRgLBoM5OTkOh6OtrS0YDMLzuNFo9Pv9\n8NlkMrW2tvLFmn6/3+PxwPhUNBrNy8uLRqN+vx9W7huNxubmZpfLxRiTJAkrabS0tOTm5sJQ2sUX\nXzxq1Cg8IgYSDc3NzStXroQt0WgUBuNgTOrVTZvMRmN1c3MkEnn77bfNZnMoFALnwU5zczO8+Nlo\nNAaDQVj119bWBusVYWUUlDQcDgcCAWVJjUZjKBTiixuDwWBubi4MPV933XUFBQX47oAWx8GDB7/4\n4gvGWCAQiCUSiz/7jDH2eWVlQzAInzd9882paPSNN96QJCkSiTQ3N3s8HoPBEA6Ho9EoqBDEYjE+\nbxkIBDweD9RSS0uLy+WyWq1tbW0tLS1OpxNmtHw+X05ODmPMbDb7fD4wAkOlDocjkUiYTKZbbrlF\nGNUFLY7PP/8c7vvevXurfT5w8lhjY3NbG3zeW13t3r37tddei8fjgUAAZBbgtubk5ICT8XicT/vg\n29rS0uJ2u41GYzgc9vv9ubm5MBcaCAQgJGAsFSaFYDkojDPm5eVNnTpViH/Q4lixYgUsldy9e7dU\nVQVOtoTDn33zDQxqnvL7169f39DQABeFmhFuazgc5jNR/LYyxnw+X2FhoSzLra2tcDuMRqPP57PZ\nbHA8D0KhpIMHD7788stxDYMWRygUevvtt2HLyZMnP47HG1paGGMJWV62ffsXhw9DA1y2bBnoN8Bt\nFRw2m82BQADUXRKJRGtrKzgcCoUkSXK73dFolN8OcBhKDcO7fN4vGAzCqlrG2IQJE4YPH67U4qir\nq/vwww95i3tr61avw+Fva5Nl+a233uK3NTc3V9nimpqa3G43nBuJRKCL4CViZ9C3GAyG7373u91V\ni0Ntu1KLgzH22WefwRabzVbtdH7Y2soYu2j8+MOHD584cYIx1tjY6PF4zGZzPB73+XxQ45Ik+f1+\n3s/6fD7o+xhjdXV1hYWFjLFwOBwKhbxeryzLzc3NVqvVZrNJktTQ0AB9EGOMt0z4nJubCyFis9ku\nueQSXFjo3EOhEHcYwg5mJ4YNG7YuEpFlmVksY8eO/ec//ylJEqzuBcfgurA2NJFINDc3w2dYfgrR\n0NTUlJOTY7PZ4vF4Y2MjHIBLJ0kSDyM4Pi8vDxweOnTooEGDsMPQbdXW1oLDiUTi/KFDoXojxcUl\nsvxRW5ssy6y0tKClZd26deBwbW1tUVGRJEmhUCgUCkFtBwIBi8UCa3AbGxu9Xi/0g9CbG41G6Cv5\n7WhuboYTDQZDbW0t/HJEIpHW1lbP6VepXn/99cJr6kGL4+uvv960aRPcPtc554DDheefHwqFwGHv\n0KEtLS2ffvqpLMvBYFCWZegIGhsbc3Jy4NcLHAOzfr/f7XZDLcEthpLyGobP4GRbW1sikYAahi4G\n+sGcnJybb75ZiH/Q4vj3v/998uRJxlg0Go16veDw4AsvrHW74fOI8eOPHj1aW1ubSCR47OFaEm6r\nz+fzeDzgcF1dXXFxsSzLoVCora3N4/FAPFgsFugffT4frJAGhyHaITauvvpqPGkMmTmRSIQHcE5O\nzlGHo661lTE2ctSoL02mva2tjLHrrrtuy5YtcFtxAOMW19TUBDUcj8ebm5vz8/MhmGVZhkbU1NQE\nYglwa8BzdrpPVLa4vLy8iy66SKnFEQgEuMPf+ta3tiQShtZWxtjkyZP/+c9/wrlwi+G24i6itrYW\nuoJoNNra2grb/X6/1Wrlwcz7Fh4zQkn5Z7gdRUVFcNExY8b079+fZQZJY8Y5ddatW2ez2S699NK6\nurr333//Rz/6kfZ2TCAQaG1tldovbA+FQmrqCiUlJTU1NRrqCnwaurNaHDU1NdwrpRYHF7JqaGhQ\nanE0NDRw/wV1BbfbXVdXxy+UuhZHSUlJbW2thhYH/5HT1uI4efIk1uIwGo1FRUW4sFgGQamuIGhx\n+P1+DXUF6KHgTnVZi6O6uloorNPpbGho4P5L7dUVNLQ4Uo+WM9HiqKmpwYWFjowX1mw24wyE4uLi\nxsZGDS2O2tpafqFOaXHU1dVpaHHwpz+fz6ehxaEsrEbT6FCLI8VoYcm0OPgPFcR515pGp7Q4SktL\n1aIFtDjUoiUvL+/UqVP8oh0yevToFI/MrBYHaXQQBEF0mfQ8Qcuy/MEHHzQ2NhoMhqlTpwaDweXL\nl99zzz3CdkHWWbCgluytsStDJ2Zrb+9xSfvEzFnOUEkzZ5miJaN7z/CmnwXS00ETBEEQaYdeGksQ\nBKFTMrtQJRaLPfvsszAlOnz48AkTJqidztTXswvrPlPfmyGzOnRJ26w+Hdb+51FvdahtVvsAvVV+\n5ixTi0s76emg+YKUHTt2bN68efLkybC9qalp8ODBN910k8a5lMVBWRyUxcGrhbI4GGVxINIzxIEX\npBw/fpxvb2xsrK+vf+edd9577z2fz5eWaxEEQfQS0jNJuGbNmv79+w8ZMiQSifzpT3+aN28ebD9+\n/Ljf7x82bNiePXt279592223wfYVK1bwH88RI0aMGjVKkv7XkzfeeOMPf/gDfHa5XBUVFXghA9aU\nYqeVKuGzof2rdvEupngYFPZqm00kEnzZXigUEkRBTSYTyIMpz4UFVFhADu81tn8LstIl/OTVKbP4\nX7NIJIL1zCRJMplMXa5DjZJqmIWlVhpm8X+L0WhUcAn/f9DZOkyxpMLeDqOFIQSHOyzs2Y8WqEP4\nmjSAu9w0MhEtBoMhHo/zR/5wOIwFWhlj8+fP37ZtG3wePnz4n//857MQLRpmTSbTsWPHMvEEndmX\nxpaVlcGHwYMHr127lm8fM2YM/n8qGAzi27xy5cqr8vOvGzaMMXbPm2/u2rVr7NixsEuSpLy8vEAg\nwKsG16nVarVYLPx/QOFm5ObmhkIh/m8UPtFgMHi9Xvw2VbwXlobzNhkOh8PhMI48r9cbDAa5//hc\ns9nsdDq5ZaEFOp3ORCLB/48WAhpKyv3HZi0Wi91ux2YhpuFrTk4Ofl9yMBjEjcFgMHg8HlxYk8nE\nXbLb7bAAndchfls5rAjFpcMCjLm5uWoldTgckiTxfy2FksIqRO6P3+8XCmuz2bhloVNwuVzRaJSH\nk9DG8vPzU4wWwaxGtEAQMgQsNMeFZYzxURehsBrRYjKZcnJyUowWoaQa0QIDdLyDhupKvWkI0SI0\nDZPJhKMFm009WpgiCBl6wXZbWxvcGl6Hb7/99iszZxbm5DS1ts567bUXX3wxxWjJycmJxWJpjxa+\nwjDtpKeDLi8v3759O1MsSNm4caPVar344otPnDiBn4JLS0v5Z+UYtCzLQ0pKJg8Zwhhz2WzxeByP\nKjLGYASNW8CjqyaTCY9z4REx0MrAQ054oE2WZeGd3NgOfr7jEs/cB1mWI5EI/nXl58qybLfb8YXw\nqKKtfekEs4yxSCSCH/FwPQgDhfjnXcg3j0QiuHTQFeLCgv/wGX6HcNkFPehoNJq0DmEQGXuIh2VB\nwASXVNCDxg7DYdiUxWLB/uNocTgcai51KlqUCr8a0cLaEw6HcWHh15EfD6ZwtIA4jNKyLMv4Timj\nJZFIJHWJnb6JSaOFtRcsTiQSEBK4aajVIRRWrWnAAm61Fpd6tLD2QSgosMNNEZrGZYMGlXm9p/x+\n2KXWNJTRonZb+dek0WIwGDSiJXM6nZl9aezo0aMrKyuXLFmycePGa6+9Ni3XIgiC6CVk9qWxdrv9\n1ltvTcslCIIgehu0UIUgCEKn6OKVVzCSxaeJhPRvg8HAp7xgl9VqxdOyfC+8u5N/hWEsPvoGo0h8\nLz4RhvzxxJrSLPYHJuXwlDRIeirPBaVE7L+EXvAKI1nYJUP718dZLBbuf4dmuQOQdsKvYjabhcKy\n0+Ok2A1eWOFgGBjlF8IJDPhIuLqGWY2SSpLEHZZlGeRGUywszN8mva1AitGirEO1aBHiMx6Pg9yo\nWlXAULJQhzgGsEvYTofRgksKQagWLdhncECWZfwKVNiiPDdptGAflNGC33LbtWgR5JVhl6S+KkRZ\n2C7cVkAtWjpscWlJh1Oiiw4aspTUXhqLJ0aU0z7ChEDqk4TCJJswSSgk1ePgSPrSWDxJiM/t1CSh\nsFCFtZ8kFJZgaE8S4l8UuJzGJCFeI9CpSUKhlljKk4TCQhVZlvHPczgcFgqb+iShcCJLOVq0JwmF\n2mYIUNnXmCQUEuCESUIhWlKfJBQWWQiThMJe0IzGDmhMEioLq9Y0OjVJqBEtrH0QCpOEYFBoGhgh\njyD1SUKhlph6tHQ4Sajx+3Em0BAHQRCETklPBy3L8qpVq15//fWKigqe68oJhULPP/98Wi5EEATR\ne8jsS2OBdevW4fddEQRBEKmQWS0OxlhVVVU4HOZvbCQIgiBSJLMvjU0kEmvXrp0xY8aSJUvw8dpa\nHMIkjNPphBc+cvBbq5XKAPxgpTIALKVNuhefqDSLF2I5nU5BysBgMPB3tgrnwlSvmmWY3OeL45Uu\n5efnd80srq68vDwhZUK7sNp1yN+7qnTJYDBou4QnwbFZweHCwkKlDIKGZYvFwpdyCZZZytEi7NWO\nFoGCgoLUCwu6Al2OFj7dpzSrES24TVmtVqfTqWwafL1yp5pGJqIF6xYwxlwul6DFgZEkCV8llWjh\nS207FS0aZrEWY3rJrBbH1q1bhw0bplwHqa3FIVRZKBTC6gSkxcFIi4O0OHq3FgdGlmV8FdLiSIKa\nFkdNTU0wGPzqq698Pl9FRcXtt98O2zvU4sDGSYtDabbDNDtcENLiSCVaSIuje2lxYHqwFkd6OugB\nAwYcOHCgoqLCYDBMnTq1trYWXhrLpfpfeukl3jv3cnbs2PHwww9DgzGZTL/5zW+GDBmSbadENm3a\n9NRTT7HTz+bPPvvs+eefn22nCF1w4sSJuXPnwkOowWB46KGHrr766mw71WPJrBYHZ/78+Wm5UA9g\n8+bNJbHYQ5MnM8Ze+PTTjRs36rCD3rhx4/kWy5zLLmOMPbV69bZt26iDJoC9e/e2VVU9d8stjLFl\n27d//PHH1EFnDl2sJOxt9PV4QEx12fbt2fZFlf75+eDknzZuzLYvhL7wOBwQG7urqnZk25mejS46\n6F6lxSGMYOK9+tHiENIquNhFl9UVPV7FJQAAIABJREFUSIsDF7Zba3EIAYxdIi2OtKOLDrpXaXHg\n6RqhdPrR4ojH4/idTnwSJnV1Be1JQtLi6L5aHEIAY4dJiyPt6KKDJogz5C9/+ct7773HTmfFvvLK\nK/BqbYLo1qSng5ZlefXq1fX19SaTadq0aTzpJBKJLFu2LBQKJRKJ6dOne73etFyOIAQ++uijaf36\nTRw0iDE26y9/OX78+LBhw7LtFEGcKZnV4ti1a1dpaemdd945atSoLVu2pOVaBJGU4X37Th4yZPKQ\nIfb2r9wmiO5Lep6gsRbHzp07+fby8nJYFCRMwREEQfRUkmrDCcvHUiSzWhywcenSpUeOHJk7dy7f\n3pu1OPBSeMaY2WxWc1jKnhaHw+HAM0EulwuO160WB18WDHi9Xn48aXFIadXiEDoa7LCyxfVOLY7v\nf//7Sj///ve/J/Vfm8xqccDs9syZMw8fPrxy5co77rgDtvdmLQ7uABCPx/le/WhxhMNhPEzAb4Fu\ntTjwOnvGWEtLCxxPWhws3VocQtpDIpHge0mLA+haX5yUzGpxbNiwoaCgYOTIkRCjfHtv1uLA9cDa\np9npR4sjxTQ7pq6uoJ1ml3YtDiH3KxKJhMPhTkULaXGkqMUhBHDqaXZMPVpYj9bi+Prrr9esWXP/\n/fdrHKNGZrU4xo0bt3z58u3btycSiSlTpqTlWgRBEPrn1KlTH3744Zo1a+x2+7XXXts1I5nV4nC5\nXD/4wQ/ScgmCIIhuwYoVK1avXh0MBq+55prnnnuupKSky6ZooQpBEEQ6eeGFF4YNG/azn/3swgsv\nVA6IdQpddNCkxcFPJC2OrmlxCLVqsVj48TrX4ojFYnPmzKmqqoJdN9xwwwMPPIBdIi2ObqfF8eGH\nH37xxRd/+9vfFi5cOHbs2KuvvnrkyJFCoVJEFx00aXHAZ9LiYF3V4khxklCHWhx+v//jDz9cfd99\njLG91dWLly279957+YVIi4N1Qy0Ok8k0YcKECRMmhMPhTZs2vfvuu4899tiqVauSOq9NelYSEgTR\nZQySBGsgx/Tvn21fiHRitVonTZr0m9/85oILLuiahcxqccRisZUrV/p8vlAoNGXKlL59+6blcgRB\nEN2I+vr6rp2Yng6aa3Hs2LFj8+bNkydPhu2VlZUWi2X27NnV1dWrV6++++6703I5giAI3bJ48WJh\nC1843Vkyq8WRm5t7ySWXsNMLydJyLYIgCD0DEhcYc1cFvDKrxQEJgFVVVStXrpw0aRLfTlocHNLi\nUDrcq7Q4BLMaLpEWB0uGpD8tjmnTpglOvv/++0md75DManHIsvzpp58eO3Zs6tSpffr04dtJi4ND\nWhysd2txcG0NQJZl0uLo7locSkaOHKl9gBqZ1eLYt29fU1PT7NmzhYdi0uLgkBYH691aHMJPi3Ah\n0uJg3V+LgzE2f/587QPUSE+a3YABA5xOZ0VFxb59+8aPH19bW/vKK68wxiorK0+cOLF48eJFixa9\n9dZbabkWQRBELyGzWhw33nhjWuwTBEH0QmihCkEQhE7RxVJv0uLgJ5IWR2/T4lCOaAsukRZHt9Pi\nSCO66KBJiwM+kxYH631aHILnLOVJQtLiwAfoSosjjdAQB0EQhE7JrBYHAC++Gjp0aFquRRDaNLW2\nzps3D3JpBw8e/NxzzymHEbLCCy+88MEHH0BadENDg8fjUaZdEgQm1cC9//77N2zYoBZMXItj6NCh\nmzdv5tsTicSSJUvWr19/5o4SRIq0hMM/veSS306a9Jsrr3zvvff4qoSss3r16h+cd97/XH75s1dd\ndeLEiZ+PG/fbSZMeHDUq234R+iXVJ2iv13vfffedOnVq+vTpN9988+WXX45XN6ppcUiSNHv27HXr\n1qXXaYLQZuyAAReVlcUUw7tZZ1S/flcNHswYkxi7bNCgMq93T1VVtp0i9EuqHfQTTzzxxBNPHDp0\n6O9///vjjz9+8ODBG2+88eabb77qqqvMZrOaFod0GsEaaXFwSItD6fCZa3Fg8vPzk970s6/FISg2\nJEXDJdLiUKsxvWlxpJHO2c3LyysrKxs4cOCePXs+//zzPXv2/OhHP/rDH/6Qm5ubVItDDdLi4JAW\nB8uAFgcmGAwmrcOzr8WRynAzaXGwnqjF0WVS7aCfeeaZVatWbdu2bcKECVOmTFmwYMGAAQMYY+vX\nr7/ttts2bdqUVItDjd6mxVFZWTl//nzYWF9ff/255/Ij9a/FcaShYdNvf/vaa68xxjwez6JFi9xu\nd9JaYimn2WVai0Owhnv/LGpxpJgq2420OE40Na369FMY1ZQk6Ve/+tWYMWNIiyONpDpJuG3btvvv\nv7+mpubDDz+87777BgwYAE9DF1988R//+Ec1LQ4C2LNnT6Sm5reTJv120qR+7RPj9U+Nz3fL+eeD\n8zX79x84cCDbHhF6obKubkyfPhAb5Yxt2rQp2x71NDp+goZfiS1btlRUVPCvwWCwX79+fr/f6XTe\ndNNNjLGkWhwAVoLutXgdjslDhjDGXli7Ntu+dJohJSXgvAutdyAIxlip2w2x8XeUHUCki447aJiU\niMfjtvaN8+abb86UUwRBEETqT9DXXHPNRx99lCEnerwWh8YLb7qFFgdGrQ51pcWBsVqtSR0++1oc\nKa4G7l5aHBiwQ1ocaSTVScLM9c6sF2hx4JlizOGGhhWLFv31r39ljFkslldeeaWkpCRbWhwHDx58\n6KGHoB+pqqr6oco7INTq8OxrcXzwwQcvvPAC9H2HDh1i55+f1GE8SZhdLY70ThKeTS2OhoaGe++9\nt6GhgTEWCATOUckqAzukxZFGOu6gx4wZs3DhwgULFih3bdu2LQMu9SKONzZ+59xzb734YsbYw3/7\n265du+Atjllh+/btzkDgsSlTGGMPLl2aLTdSZ/369Re7XFB70/fvz7Y7PZmamppdW7cunTOHMfba\n5s3HGhuz7VFvoeMOetGiRQMGDFi0aJHGMWpaHNoaHQRjbFBREcyx5OmgcopcLnDG3T5RT7fw2jOr\nj8kQacFqMkFV//Prr6mDPmt0nGY3ZsyY/Pz8MWPGeDyeCy+8cNiwYVu2bPnyyy9HjBjBj1HT4lDb\nThAEQXRIqnnQCxcuvOCCC/x+/7PPPvvGG2+88MILDzzwAN+LtTiOHz/e4XaCIAiiQ7QUBjD5+flr\n164dMWJEaWnpF198EYvFLrnkkvr6eti7Zs2a/v37DxkyJBKJ/OlPf5o3b572dqUWB2Ns3LhxsOXI\nkSMuSSrIyWGM7amq6j9wIIyNhMNhmD2A+Y2kmQZ4u9/vz83NZafnEEB/IBwO87SKUCgE8++MMZz2\ngFfZ33777Q888EAikeDz0aFQyGQy7d69++6774YtMEcEp0ejUSFnQ5Ikn8936sSJ84qLGWOVdXUm\ng6E8P58xtq+mJs/h6ON2M8YOnDxZfM45sEgvFArBbDhMs3CDapkAMKsG13rqqaeuueYaPMcSiUQM\nBsPSpUufe+45DTuNjY3NtbUDCwvBmRyrta/XyxjbXVVV4nbD7dhbXV1+7rlwO3DtQbVDveHai8Vi\nMB0kSVIgEABpi3g8HgqFwIjJZHrnnXf69++P/YFJ48cee+yDDz5QFjYUCsGU14kTJ8yRCNTeruPH\n++XleZ1Oxtj2Y8fO79PHYbHIsrzz+PELR4xQJh6An+AqNyjLcjAYhGW7sVgsFovBdrgdEFeFhYX/\n+Mc/hKl/cHjGjBknTpzgdxxiLxaLRSIRWNt58ODBUqcTcsn/ffTo8L59LSZTKBo9cPLkiLIyxlgw\nFGqSZXimYe2DORAI5ObmQuTz2sPBDK0D3w44ceLEic8++yxMI/MKNJlMPp/vmmuu4bcJclGEwkYi\nEbAZCoW+PnBgeN++jLEan8/f1ja4Tx/GWGVdnVGS+hcUMMaONjRY3e7i4mKokHg8brfboXoFg8p7\nirnnnnvuuuuueDzOj4Rp3s2bNz/44IOwZffu3ecXFVlMpmg8frC+nksZ4whM2hUkEolQKAS3Q9kV\nKPsWHMy4iwgGgxDMBoPhueeeu+yyy44dO8b7ww4ZPXp0ikemmsURj8c9Hs+//vWv4uLifv36nTx5\nEmtK2Gy2pFocatuTanE8+uijsCUajcL8NWMMghtmSyFGXS5XIBDw+/3QkEwmU1NTE3wG2SbenQUC\nAbfbDQtb4/F4fn5+JBJpbm4GuQmTydTY2Ai3TZBDCgQCXq8XLtq/f/9QKKTU4vB6vdzhRCJhMpnA\ngt/vz8vL4w57vV6bzRYIBOrq6oqKiiCfrKWlhXdVZrMZToxEIjabDeb6oXS5ubktLS3Nzc3gJC4p\nTCjj1gslZYwNHDgwGo0qtTiGDx8ODoNqB+8QA4GAx+OJx+OJRKKxsbFv374Gg6G5uRl0eQwGg9/v\n52kh0NtC6YSSut1up9MZCAQCgYDL5YL8sKamJnDearUGAgFocrAyGwrLGHM4HIlEQqnF8Z3vfAfi\nWK2wLS0tsiwXFRXFYrGGhoacnBxJkvBtZYzZ7XbQnRBuayAQyMvLg4CB1gsxwBseKC1Aqf1+v9fr\nhVwCu93e0tIidNCgxXHvvfeCUIbVavX5fNC5QzOB5K36+vq8vDy32612WyHBkbcsfmvAh6KiIvgJ\nicVidrvdZDLxYGanM2EgBnBJi4uLW1tblVoc8XicBzBDv9m4sH6/3+PxwG2tra0FQTSj0chvq8Fg\niEQiEBWJRCKRSECqX1tbm9Fo9Hq94XC4sbER4kFocbFYjLuESzpo0CBIlhC0OEpKSrjDra2tOTk5\nUNhoNGq323GLs1gsra2tEIdwCV7bUL1QY/BjDLEBMZCbm6vRt7S1tfEMQtziysrK1NK0zpxUO+hb\nb731uuuui0ajjz766OHDh2+77bZvf/vbfG95eXlSLQ617Um1OPgTdF5eXigU4lozQsZVSUlJTU1N\n0lQYh8Nhs9kaT89gCGl2+fn5ra2tXGtGyKEpLi6uqanhXuG9sVhMqcVhtVq5w8XFxQ0NDUnT7CwW\ni9vtrqur4xfCiVNutzsejweDQX4wziUqKSmpra1Nmjhls9lycnL4L7aQZqd8/5MsywUFBaDUZTQa\ni4qKcGFxhlNOTg60fF6H0PDga2Fhod/vV0ucKigo4P8YCWl20ES5hA1Os1OK4IDNc88999xzz4XC\nOp1OyPFiigwn6AXOPFoEsxrRIshdsdP/QvFZGegLeGGFV14VFxc3NjZiFR4cLR6Pp7a2Fr4qoyWR\nSHC9HiFFrE+fPnV1dUmjRdAq4VocPIDz8/NbWlrwM1OKTUMZLbjFpR4trH0Q8p9GAG6K0+nkDqfe\nNJTREolEuGiXUIelpaVq0WK32+12u1q0aEjBnCGpdtAvvfTS8uXLGWPTp08/dOjQLbfcghdzDxgw\n4MCBAxUVFQaDYerUqbW1tcuXL7/nnnuE7WrGhfaJUymZovBC3iLeG4/H8U9Zp8wKrU4wKwwICJYh\n3JOeC2MUapZxLrDSrKDbIrQxwSw+Ep4p1C6qlIXC5yoV14S8bMENNbNC2q9Qh5BLjs0Kgo0ahRVq\nSahD7ZJ2Klq0zQoP0cJtxbuU0aJmGbpODbPKsOTGNYIQKhCPLQh3R6OwTLNpKEUE8V7taNE2K4xK\nCQ5rNw1sWRktXetblEGILWuIKZ4hqY5BEwRBEGeZVJ+g165d+8tf/rKxff7jflodQBAEkTFSfYIu\nLy+/7bbbZs2ahf8VPV9lcS1BEARx5qTaQZeWllZWVqq9LUVtxWAsFnv22Wc9Hg9jbPjw4RMmTFCe\n29bWJowo2Ww2SHL6/y62fy2N0+nkY/zCXnhzD35JKN6rYVaSJIfDoWYWUui4Cn4gEBDmtRwORygU\n4iNWkuIVO3yuCcZh8SwEHttSvlurtbUVO6lmVthrtVqxZj/M4WgUFl8XCiuIn/Bz7XY7FnyQ2r8T\nyGazCS8843thAJQPDgr31OFw4CHdpqYm4bbCHHrSc202Gx5N7nK0CCXtMFrwkK7gcIeFzUq0cDmO\n1tbWcDicxqaBowWXNPVoEUoKFci7Eb/fL6MXtaRSWI2mgetQGS1qZjuMlpaWFvyCGG3ORa/s0CbV\nIY6HHnroxRdf/NnPfpY0+5ivGNyxY8fmzZsnT54M25uamgYPHgyC0WqAdgmeErXb7fF4POkEuiRJ\nXq+3oaFBbV4et2RhThnS9dSyONxuN76pSikivgtmY3C6RW5ubjgcVsvisNvt3LIwL89zkvhXbNbj\n8YTDYbWpapvNhs3iLA5BGBYSUbHOi1BYPIEOOnO4DnEWR05Ojob8jcvlwp0LzuIAwSlcUjzlJbwx\nqK2tTSgsbnLCBDp0LmceLUlVddSiBZ45hBoW3kqD302Fszhyc3MjkYhaFofD4dCIFlmWk7rEGHO7\n3WrRYrVacUhEo1FBLAneSqOWxaHRNJTRgs2mHi2sfRAK+pGwvWtNQxktan0LY0wjWux2u0a0QPJl\nU1MTSzepdtB///vfd+7c+fTTT/fp04dPx/MxaLW3ejc2NtbX17/zzjtGo/Gaa67hr0qqr6/nN8Ng\nMMAPJjfb2tq6d+9eqBqHwzFs2DD8uMEYw9KIPAGZna41nrAMvyV4+QlOZ8YngoIoXmCiNMt38Z8o\n/EuLFR3xudAxcctCXw+fsUus/bwzjlSNkoL/3AFhBhzWvODC4ouy0+2KW8b6qMJb5jTqEKqF74K8\nV2wWOwzVItxWNYehEoTbigur5hK3lnq0cLPa0cLaw1cVJa1hKKxwW/FnXFJ8YofRgksKZUkaLYIc\nIwizsfZNA/5NUSusWtMADzVanOCGWrSw9kEI8cx38QbID6ipqdm/fz9Yg5UZKd7WLkdLh0HIMkOq\nQxxJ5wP5GLTaisHjx4/7/f5hw4bt2bNn9+7dt912G2x/7733qqur4fPFF188duxYbPbee+/9x9Kl\nLquVMVbt823asgXrfhiNRrWXb0InpZaTCMuE1MqrbRaWosDXcDgs5ITBAoGkljPnkrZZqf3CPKVC\n5lmuQ0N7bW7lifirYCRbdahtVsPhDgt79qOFoU4EnvfTZTlDDsMaLvjKl5Zwzj33XBYMmgyGuCyH\nzWZYwJlpl7TNZnklIfTF8Xi8trYWP0QDaisGy8rK4MPgwYPXolc93XLLLfxzIBCor6/H/y/U19f/\n90033TVhAmPsoiefrKqq6tOnD+ySJAly1NX+aT2ThSqnTp3iXuG9TqcTy5AHg0FhiCNzC1Xq6+u7\nsFDF6/Xiu9DQ0CAMcRQVFeHCpnGhCjcrpbxQhTFWVFSEf/NOnTolFPZMFqqkGC2pL1QxGAw8IIHa\n2lpc2GwtVFGLFqvV6nK58MppGNXFTUN7oYpa0zjDhSpqQQgrAHkHDTWJm0ZTU9PeRx8t83pP+f39\nHn20ubk5XQtV1KJFe6GK8Jb3NJLqk3lVVdVVV12Vm5s7dOjQf//73xMnTjx8+DDfW15eXlVVxRQr\nBjdu3Pivf/2LMXbixAlYJ0oQBEGkSKod9J133nnBBRc0NDS43e6LLrpo3Lhxc+bM4XvV3uo9evTo\nysrKJUuWbNy48dprr81ICQiCIHooqQ5xbNy48d1334W5YJPJ9PDDD5eXl/O9kiQlfau33W6/9dZb\n0+owQRBEbyHVJ+hvfetbGzdu5F+/+OKL1FP5CIIgiC6Q6hP0iy++OGPGjCuvvLKxsXHGjBmfffbZ\nm2++mS4nevxbvVN/T7NBH2/1xgfjWW+pq+9pzvRbvbvwnuaz/1ZvXIc4BrBLkuKN1BrRgksqna23\negs3XRkt9FbvNJJqB33FFVccOHDg/fffv+iii0pKSv7whz8IE9lnQo9/qze+kLH9e5rxC4mFLA52\ndt/qDZ+5+jsvO87ikFN+T7OU+bd688KqudSpaFEKSKb9rd64DpNmccBvoUa0JLL0Vm+1pgG9oVqL\nSz1aGL3VW4VUO2jGWH5+/g9/+MNMOEEQBEEo6biDFla1ApIkOZ1Onh8ua769OxQKvfzyy//n//yf\ntHhMEATRS+h4kvDIkSNHjhxZsGDB6NGj16xZs3///jVr1owZM+bJJ5/Ex2i8vXvdunVYQ4cgCIJI\nhVSfoJ9//vkvvvgCXlXVp0+f1157bdy4cXfeeScco6bFwRirqqoKh8NchQPQ1uIQRnPw2nmJtDiS\nlZS0OEiLQ+pNWhwCqd/WbqfFkeoYtCzLlZWV/F2ClZWVuAZbW1thoaDb7cYPy4lEYu3atTNmzFiy\nZAm2tm7dOg0tDqG0drtdWEnp9XqTOgmdlNqyS4hLYfgFo3aisAYf3tcpWHa73RrqChouybIsiM/h\nc5OOL6ViFn+FN6IKp2sUVvitEizDu6XV9mq7pPZGZ2HWXjDSYWHNZjOWVxXIULRoXEKZ+YAxGo0Z\nihbhMQjvwl+V1aVd2DOJFv5W1qR7NcxiTQyn0ylEiHAwvOcXu6QdLWqyyewMokXN4BmSagf9n//5\nn1OnTr3nnnsGDhxYWVn5yiuvLFy4kO9V0+LYunXrsGHDlDdeW4tD0CsJBoNY4YG0OBhpcZAWR+/W\n4sDIstyDtThS7aDvv//+kSNHLlu27JNPPiktLV25cuX48eP5XrW3d9fU1ASDwa+++srn81VUVNx+\n++3p9Z7oJezbt4835mHDhqk9nBJED6MTaXYTJ06cOHFi0l0DVN7qzaX6X3rpJeqdia6xadOmmTNm\nlLjdjDFfW9vM2bMfe+yxbDtFEGeDTnTQGkgqWhyc+fPnp+VCRC/E5/ONHTBg3U9/yhh79qOPdvp8\n2faIIM4SmRrbJgiCIM6Q9DxBnyGkxcG/khaHUotDmMGH6SPS4iAtjqSWe6kWR0YhLQ74SlocLJkW\nh5DVA8IRetbiMJlMX375JaQKmEymESNGkBYH04wWRlocKuiigyaInsTvf//73/73f+c5HIyxU4HA\ni3/8I56hIXo8n3/+uXIjTntLnfR00GpaHJFIZNmyZaFQKJFITJ8+XS0JnCB6Es3NzfdMnPjr6dMZ\nY99bvJinCRO9BCx3EQ6HN2/eXFNTs23bti6YSs8koZoWx65du0pLS++8885Ro0Zt2bIlLdciCILQ\nMz/96U/vu+++iy++2OfzHTlyZMaMGStWrOiaqfQ8QatpcZSXl8OSNmEKjrQ4+FdGWhwdaXEIxYHR\nQEnHWhxJA5i0OHqPFscTTzxx7NixcePG3XXXXfjVgF0gPR20mhYHbFy6dOmRI0fmzp3Lt/dgLY7j\nx48fOXIEtp9zzjn5+fnaLmVFi+Po0aN89Xn//v1dLpdgWVdaHIJyAqyex5b1psVhNptxo7fZbPgA\nnWtx1NXVHTx4ELYXFhaWlZVpuNRhtPROLY5wOAz6BGo3K3XS00GraXHA7PbMmTMPHz68cuXKO+64\nA7b3VC2OWCw2YsiQ/JwcibFIPJ5TXPzZZ5/pTYsjEAiMuvDCPrm5jLFQLFY+ZMiKFSv0rMXBlROA\nUCiElU90qMURiURwu2ppacE+6FyL45prrvFVV1uMRpmx+mBwz4EDZWVlpMXRKS2Op59+OhQKbdq0\n6Xe/+53P57vsssuuvvrq4uJipfMdkp4OWk2LY8OGDQUFBSNHjjSZTEK32yOJx+P+UOj/tffm8VFV\naf7/ubWvWUggEJYQBdEgKuICAfGlDeJX0YAbgi2MdiuIy7S27czY3eq09mvabhRt9TeDojRLbBRt\nmiYs47DYNBAEWcVgAgESskBlqVSSqrq13t8fz3DmcG7dm0rlFlzI8/4jr7s+5znnPufUzVk+98yC\nBTaz+XBDwx0ffnihPUoA1MPq3/6WELLt6NF569ZdaI8QHeHz+cqeeebaQYMisZhl/nzaBiHJ8847\n78BGXl5eRkbGp59++vbbb6c2SKhNA62kxTF27NjVq1fv27cvHo9PnTpVk7QQBEH0zJgxY9jdCRMm\npGwqvVocbrd79uzZmiSBIAhyUcCJygUCgf/6r/9SUppTB7U4EARB0sWOHTseeeQRblpB8uhiJeEl\npsVBgeN60+LgplJws9ZQi4MtltS0OORrwVkfBH1rcXC5AyOoxdEtLY7q6mpCSDQaXbZsmdfrffvt\ntwsLC5WcV0cXDfQlo8XBreuH43rT4qBj5WyxoBaHhloc8XicvQIe8cWixcHVPjiIWhzd0uKAQcJo\nNFpTU/PWW2+l3DoTnTTQCIIglwzvvfcebBw5cuTNN9+84oornnrqqdSELtKrxRGNRsvKynw+nyiK\nU6dOHThwoCbJIQiC6J+rrrrqk08++eKLL2bPnr127doULGjTQFMtjv3795eXl0+aNAmOV1dXWyyW\nOXPmNDQ0rF+//qc//akmyekNj8dz+PDhWCzG9R5cFITD4YMHD8L/0cOGDWMXZSG9hPb29sOHD8P/\n7zjxueecOnWK3R03btzw4cNTM5VeLY6MjIybbrqJEOJwONg+mktMi+NHP/qRt67OYjTGz+0jEwRB\nh1oc7IDysaam+traJx9+mBDSEQrdce+9ixYtQi0O9Zxeeloc8+bN27Fpk9NiIYScOfv1dJo1gloc\n3dTi+PGPfzxixAhCSFVV1RVXXEEIqays/Pvf/57QeXXSq8UxYMAAQkh9fX1ZWdntt99Oj19iWhxe\nr3fDc89dnZ/fFghkP/88m2JGRoa6S+dfiwNWEgL+UOiKvLyDv/41IeSTHTvKOjrsdruAWhy9TIvD\n5/N9MHPm/ddfTwgxP/UUeyozM1PFJXAYtTi4Gjds2LAPP/yQEPLkk0/Cxk9+8hOlFNVJrxaHJElb\ntmypra0tKSlh5QsuMS0OJWKxWGtrq960OGj5yAmFQp2dnajFQXqZFgf3FsnS3NzsdrtRi6NbWhwa\nos1ClYKCgvr6eiLT4qioqPB6vXPmzOECGkEQBOmS9GpxVFdX19XVwUt+RkbGrFmzNEkOQS4WYvF4\nXV3doUOHJElyOBzDhg270B4haScnJwc26HSJa6+9NjVT6dXiuPfeezWxjyAXKQdOnfrH4sVrli8n\nhDT4fGvKylKTnUQuIn7/+98LpChdAAAgAElEQVTDxkMPPQQbzz33XGqmcKEKgqSRcCz225KSJ265\nhRBy/Rtv0F54BEkGXTTQF7sWhxLCxaDFwQJFhFocRDVahG5qcbBAqOtNi0NQlrlALQ5qNnktDg3R\nRQN9sWtxKHFRaHGwQBER1OLQVIuDBZ6v3rQ4VCIZtThgu1taHBqiiwYaQZJEIsTr9R48eDAWiwmC\ncPXVV19ojxAkjaRXiwOAD18VFRVpkhbSm9lz8uQ/vv/+h927CSHNnZ0vv/ZaysMvCKJ/tJkHTbU4\nioqKysvL6fF4PL5kyZKvv/5ak1QQJBgOz7jhhurf/rb6t7+dddNNPp/vQnuEIGkkvVocgiDMmTNn\n69at3PWXmBaHEoLutTjkDsPwnZ61OFjAW51rcXAOE/1pcaiAWhx0O3ktDg1JrxaHcBbu+ktMi0MJ\n/WtxcFgsFv1rcXCnLBaLzrU4WDIyMnSoxaHSvqAWR5JmlQz2kPRqcSiBWhywi1ocsNstLQ6Wzs5O\nURR1rsXB4vV6ISpQiwO1OJIhvVocCIIgSMpo00AXFhY6nc7S0tKKiori4mKPx7No0SJNLCMIgvRa\n0qvFAbBK0AiCIEiSpKtvG0EQBOkhulhJiFocdBe1OLrU4mABb1GLg6AWB2pxpA/U4oBd1OIgSWhx\nsESj0VgshlocBLU4LlEtDuziQBAE0Snp1eJQ1+hAEARBVEivFofScQRBEKRL0qvFoXQ8oRbHoUOH\n4IjP56uJRPbW1BBCOkOhEydOwLq4cDhstVobGhra2trC4TBVCfD7/WDBarWCWDDYiUajVqs1FovF\nYjFJkrKyssLhcGdnJx2ooTcaDIaTJ0/SFX2RSARUhgkhQ4YMgeXa1HnoLQ0EAlVVVfT6ioaGUCTS\nGQoRQvafOmUxGqubmuLx+OHDhx0OR3t7ezgchp41QRACgQCMNkD3NHTFRqNRi8UCvZBwcUNDQ3t7\nuyiK4CTrMGSBliHNqSAII0aMSKjF4fF4YDGR1+uNSxIU76nW1mA4DNs1ra2dnZ2HDh0yGo1+vx86\ngo1GI/TzQqehzWaLRqPQ9RYOh51OJ5RSJBKBFVw+ny8SidBRX7/fL89pPB6Hs/F43GQyFRUVydci\nm0ymkydPer1eQkhNTU1HKARO+kTRZDTCtqejI+LxHDhwIBqN+v1+qgtBS4mcfUWAXkXOYVgF5/f7\nobRBdQR6kyGWIOPcjW63W/5dQXD4yJEj0Efp8XgM7e3gZCQWq21tpcF88uTJffv2wQJIGsCBQAC2\nYcyTdnRCqYIPoVAoOzs7Ho8HAoFYLGY2myGndADwxIkTtA+aDeC8vLwhQ4YIMi2OWCz23XffwZHO\nzs7jzc3gpETIkcbGaCwWjccJIT/88ENrayvUOBp71GGIOogH0GyxWCxwxGAwZGVlBYNBNmjpNoy/\n0fWfkUjEbreDncLCQlhfztY4QRA6Ojqqq6vhSDwe/66+3tPe3hoISJJ0+PBhyCw8KZvN1tHRQQOY\nEBIMBiFdm80WiUToY4Uuadi2WCyNjY0qbQtUTK7GGQwGeY3TEEGTwccNGzYMHTr0qquuCofDH330\n0dNPP61+fNWqVZwWhyRJI0aMgCNtbW02mw2GNeLxOIQjISQUCsFYEyFEEP7Pc7PZTBtl9rjf74dO\nFWhTwKAoiiaTiY5RJDRoMplo9/+jjz76yiuvQFMCR0KhkMlkOnjwIF2w3tTUlJOTAw+ppaWlT58+\nwllRJ+qY0WikgyfQhEHuRFEEYQR49hANEE8QW9FolA5ws46x27FYjA6yLViwYNq0aWx8RyIRg8Gw\nbNmyN954gxAiSVJrayt81zIajXZ0dFD9gZRLj4V1jDUYCAQgp7FYDMZO4fhXX311+eWXsxagoF58\n8cW//e1vUDJ+vx+cDAaDkUgkIyMjydKDKqReetFolMpUBQIByHU4HIb5JJBr+g2XrKysXbt2ceoT\n4PDkyZNramqg6CRJAoEIn89ntVp7Hsy09KLRKDRn4BiV/lF6HLfccssnn3xCmLFNyI7P57vxxhvh\nSGtrq8vlgtLzer0gGAKnJElKMvaSKT02mNkawRqcP3/+888/TwsKrBmNxu3btz/++ONwpKmpKTc3\nF+zAnCX10mODuVulxxpkEwoGgzSAP/zww9tvv722tpauL++SMWPGJHmlNg301q1bbTbbuHHjmpqa\n1q5dS8tR6ThLR0dHIBDgFraLoqiirtDY2JgOLY7GxkbqlVyLg+rLtLS0cLM48vLyWlpaqP+cukJP\ntDg8Hk/PtThOnz7NanEYjcZ+/fqxmWUH0OXqCt3S4jh9+jR9UilrcTQ0NHCZdTqdLS0t1H9OXUFd\niyPJaOmJFkdjYyObWfj9oJk1m83RaJSNljRpcTQ1NSWjxeHz+WCuAls1/H4/VdFKvmr0UIuDRgtJ\npMVBh6wgzlOrGt3V4lCKli61OM6cOZOOBjq9Whyo0YEgCJIy2rxBS5K0cePG1tZWg8FQUlLS2dm5\nevXquXPncseVNCHBB6WJhOyai26dTZPZHrokSZLS2TTltCeWob9Y8zJUN6t+rw6jRf0CvUULSVsZ\npilaenL2gkSLhmjTQCMIgiCagwtVEARBdEp6F6pEo9EFCxbAZ0FGjRo1fvx4TZJDEATpDWjTQNMF\nKfv37y8vL580aRIc93q9I0aMmD59usq9OIsDZ3HgLA5aLDiLg+AsDgZtujjYBSmnTp2ix1tbW5ub\nm1euXLlq1Sr8ADOCIEi3SO9HYx0OR3Fx8ciRIw8fPrx+/fqZM2fC8R07dtDfosLCQlinQH+4ysvL\nP/30U3i9crlcr776KvftVPbjmAaDgb4twkIserFw7peATSaTw+FgpRHpjTD0zKbCmWVTt9lssAqO\nHlm2bNmOHTsgocsvv/z555+n98JLEGeZ3gtLB6l99hTAfqGVdQmU2Fiz3GIH1ojb7TYajfReSZJe\nf/11WFJBCLntttseeOABdqo/WxSsWUjX5XKxuWO35WXImaVj4lxOOXnJrKws1nJnZ+cvfvEL+tbz\n2GOPjRkzhnoFa1Lo6yF7I5COaCHnkpmZyebo+PHj7777LuyaTKZf/vKXubm5rA9ut5tVwWXLEBbg\nJSxDiBZW4pLNqSAIKtHClrDZbIbiYjMLC/ASZla9arAOc2WoEi0JzbLRwn2mmatx//mf/7l3717Y\nHjly5Pz585WqBucSRAv73XQuWrKyshKWoXq0KH02t+ek96OxgwcPho0RI0Zs3ryZHme/9QsrJgkj\n3Pzee+8ZampuHDqUEPLul1/edddd48aNY5Oj/1kQ2bo19j99w7l60KCWm3A1FBS3klnDuR+Bp+uV\naRZeeeWVB6+8Mi8jIxqL/fKTT+bNm0fbXFhAzK2sYxdQsWchCtkGMRaLsR9+V8m4wHx6nAsXmPxE\nLz59+vQbv/nN6yUlhJATzc2/++ab6dOnU5fgF4XNO0z2gl1wPmEZglquUk65fwm5nNLlagAs7aMX\nb926dd2f//zkxImEkG1VVYtjseuuu46rzAldomWYcI1oD6NFngTN7PLly49u2/b/rr6aEFL6zTd/\nGzVqzpw5XBkmXERHl9slLEMuWhLmVCla2AYa1mR3K7PqVUOpDNWjhTPLLiyEWkNPyWvcv/7rvz4z\nYUKm3d4pii+Xls6bN0/FQ9alLqOF7YzqVrSQ9KBNA11QULBv3z4iW5Cyfft2q9V644031tXVwSs2\ncM0119Dtjo4Ov9/P6UH/v6uv/sn48YSQP+/eDQIacApeE2ApLRzhehUFQaAXcz1iVqtVFEWlXkWX\ny0VvJIn6oOkpWMbKqdM+Pn781fn5YiTyqzVraF8eIcRisVgsFmqZ61WEoFTqg3a73X6/X6mjzWQy\nsWbZ6AGhBuoDKDbQe4PBoNlo/JcpUwgh244e3bZuXTAY5Pqg2TJk+6DtdnswGFTqVXQ4HOyTYrtl\n4SeEzSnbB80JWHd2drKWI5HIkD59wGGjIByIRgOBAJtZlT7ojIwM+Aq7/CwXLdxPiHq0uN1uwgAC\nLzSzsVhsbGEhOLy3pgYCmPrgdDoDgYBSH7TValWJlng8Ts9y/acul0spWqxWK/tfIAgic1UjGAwq\n9UGrVA15tLBmk48Wcm4fNPvPFjkrG83VuKduvXVwdvaZ9vbfb9rEpsJVDe6xwmNS6oNWiRa73a4S\nLelroNP70dgxY8ZUV1cvWbJk+/btU6ZM0SQtBEGQXkJ6Pxprt9sffvhhTZJAEATpbejik1cIglxA\nvvrqqy+//BK6JrKzs//t3/7tQnuE/C+6aKChg4yOvXCDMALz6Ukq48mOtNKz0DHEDtHSv3AvN4DL\nbgvnfuBSbpY9JUkS6wMHdB3SrHH+s51rsM0mRM79QBz7vVqVnIJZ6iR4SHfBCL2Ym5TCFQt8zpXd\npWNK6mUIznMfCVVymMspjIbRkXH4QCf3HNlCY78SCaWd0CWgW9HCBqFKtLD24/E45zAXwHCKe6zc\nWfZ4ktHC5VQ491u0nFnOJSh/evHChQsHBgJXDxxICPndX/7y4IMPXnnllQkTlReFvGTYGgdjGPIb\n1aOFG5aHXRVRCvXHysqDpBwtXCWSm1XyrYfoooGGR6j0CUt22AFKhB2XZwd/YZyX+8YouwuCvPIb\nYaICeyV3lvMHLlaKGG72AusDPEjWMuuSwWDgzCo5zOVUbjahw/ICkacCAjFsgXMfjVVyiT4a9kbW\nrCAIrMOsWRa4TKX8wazSs2NvpPlVihZ5aSg9GhV/wGE2d3K4x8r6z81eIOeWIeeSSk65ySHynLIO\ny/M+7brr7r/+ekLIf/7975Lse7IqVYOrOOzPuYpLtBzoNht1CQOYm+DEXaBUNSRJ4iwrPVbqUjJt\nC7xPKDmsIXppoNlZYgkvgA0IWW6GFhsQ7MXwIsme5Xa5FNld9iwXGcmECz0Fl7GW2ZwmdIlroBM6\nzOWU85+btQai9dy98uwk3I2fJWHJcFeSc5+UisNwSslhrtDk5Sz3IaFLXUYLdzEXhOrRQoG7VBwG\n97gQSqYMUwhgJbPyAFbJUfJVg7PDOaySU26amvxKNh5o8KjUOKVIk79hdFmGCaOFe8Q0tNhE04E2\nb+aSJK1bt27p0qWlpaV0CgtFFMWFCxdqkhCCIEjvIb0fjQW2bt3KLi9EEARBkiG9WhyEkPr6+lAo\nxC63RRAEQZIhvVoc8Xh88+bN999//5IlS9jr1bU4uCFRm82mcy0ODofDwUpD6E2Lg13oSM4KJtB0\ndajFQR8ZNcVKBehQi0MeMJwPetPi4HJkt9vZYtGhFgeLIAhOpxO1ONRQ0uLYvXv3yJEjqWwgRV2L\nQz7GIl8v/38Z0IEWBwebig61OLjClM4datehFgdXM8EsV5l1pcUhSRJ7RTweZ32Q9KfFwcFFLNGf\nFgcH10SgFgePkhZHY2NjZ2fnkSNHfD5faWnpI488Ase71OJgjV8UWhws+tfiYL2Nx+P61+JgHY5e\nDFocbL3SvxYHF8miKLI+6FOLgyJJ0iWsxaFNA11YWFhZWVlaWmowGEpKSjweD3w0lkr1v//++7R1\nRhAEQZIhvVoclGeeeUaThBAEQXoP+NFYBEEQnaKLlYSoxUETIqjFgVoc512Lg0OeWdTiQC0O1OJA\nLY7/vRG1OKhlcr60OFjk44eoxSE3i1oc51wAG6jFkTCnnP8SanGc9YGoRgt3MWpxqFhWcp7dRS0O\nzdGmgZYkaf369c3NzSaTadq0aXTiczgc/vLLL0VRjMfj9913X3Z2tibJIQiC6Jnm5uaysrKGhoaB\nAwfefffd7FeDu0V6tTgOHjyYn5//2GOPXX/99bt27dIkLQRBED1TWVk5b948i8UyceJEk8k0b968\nqqqq1Exp8wbNanEcOHCAHi8oKICFhYIgcAt2EQRBLkn++Mc//uEPfygsLCSETJgwYdy4cQsXLnzv\nvfdSMJVeLQ44+Nlnn508efLJJ5+kx1GLA7ZRi4OaZV1CLQ72Yi5aUIuDQ4daHMFgEFpn4LLLLpOL\nMCdJerU4QqGQ2WyeMWPGiRMnysrKHn30UTiOWhw0FQm1OFCLA7U4Li0tjkgkAjORYFdlxleXpFeL\nY9u2bbm5uaNHjzaZTOx0FtTigF3U4oBd1OKgxYJaHOTi1+IoLi5euHDhc889B5l69913i4uL5Z4n\ngzaDhIWFhU6ns7S0tKKiori42OPxLFq0iBAyduzY77777uOPP964ceOUKVM0SQtBEETPzJ071+12\nP/DAAw8//PC0adPsdvtjjz2Wmqn0anG43e7Zs2drkgSCIMhFgcFgmDt37pNPPtnW1gbjE7/5zW9e\neeWVFEzpYqEKgiDIJYYgCHTlR01NTWpGdNFAoxYHTYigFgdqcaAWx0WuxfHhhx9yHp4+fVrJeXV0\n0UDDqILKSnz5mDJbiOxwPzcpgjPLjilzY7LsjXKz8kFw1gcOdriZs8wNXlNTbKLcLA52foXS5BAu\nFW50HpQi6FlOfIDzgbuYyBYQK5UhoDTqzc0kAR2DhEt74TLuXq54WR8SliHnUsJogbF1lWhRMsvV\nRnCYHfFPGMDcY1Wx3K1o4VJRihZyLlyN4+BSIedOt+AeDTcpght7TDJa5GbZQoZ5RCpzIeQ+sIly\nlrt0KcloUZnXANOLWbifgeTRRQMN9VPlF1Ie8ewRdgITdzF3pfys0o3cWfldcstK98pNJbmrktOE\nHrJn5VIGnEucneQtk6TLkMuaylm5Pwn9VLKsVEr0lTzJMuSCUD3MKPQulewk+Vi7jBaiUPhdmk3m\noXNnlRJV2U05p+r+0ItTczjJAO7SYZVnwRmZNm0a597atWuVPFcnvVoc0Wi0rKzM5/OJojh16tSB\nAwdqkhyCIMhFxOjRo1O7UZsGmmpx7N+/v7y8fNKkSXC8urraYrHMmTOnoaFh/fr1P/3pTzVJDkEQ\nRLdUVFRwR26//fbUTKVXiyMjI+Omm24iZxcFaJIWgiCInqGDhIcOHaLvzm+99VYKptKrxTFgwABC\nSH19fVlZGfsbglocsI1aHNQs6xJqcbAXoxbHRafF8c4778DGk08+mVq7TEmvFockSVu2bKmtrS0p\nKenfvz89npuba7PZYNvpdEYiEUH58wTxeJyuASWEOBwOdtGn2WxmlwILgkAv5qaIwfJiejF7oyAI\ndrudTYU7y/5mxM9+W0FpyCIWi1FTJpOJXQQMrRinOcBezJq12+2RSISNWtZ5o9GoZJZrIGCsmd7L\nrZyWJCkajbLzK1jLXBlardZIJMIunGUXInOrdTlND/bRcCvIbTYb20aEw2HWsnzaCTtdwWw2R6NR\napmLh+SjhSvDLqOFdSkSibCZjcfj7CtGLBZjfbDZbKDVILfcrWjh8tJltNArIYAlZcF+tsZ1WTUI\nISlEi8FgsNlsrFlOQIOtXFTqRKXGJV812OopjxYQckh4ViVaUp6k0SXp1eKoqKjwer1z5szhXopH\njBhBtzs6OgKBgLoWB30rh1/dQCBAr+FW0xsMBnqxXBkgHA4rqStkZGSw7/6cpEO3tDhYf0BdgVo2\nnKuuYDabY7EYPctpcWRmZgaDQSUtDrPZzJplpwRxr5zBYJDV4uDeoOPxuCiKbORJksSWIduSOp3O\nUCikpK4AKhNsobFVVxAENqesFofL5WIjJBAIsJblE6GCwSCbWTZCuAeXfLRwE6fUo4V9IwaH2cxy\nDXQkEmF9cLvdoigqaXHYbDaVaInH4wlzSgjJyMhQihar1coGcCwWU9fiYLVN1KuGPFpYs+rR4nK5\nWLPsrztMxqen4LiKFocoikpVQy6xohQthBCVaJEkSSVa6Oum5mjTQBcWFlZWVpaWlhoMhpKSEo/H\ns3r16rlz51ZXV9fV1UGPTEZGxqxZszRJDkEQRLc8++yzsHHs2DG6fSH1oJW0OO69915N7CMIglws\n/OxnP9PKlC4WqiAIglwywKwHTdBFA41aHDQhglocqMWBWhwXuRaHhuiigUYtDjZRbvwatThQiyOZ\nnEqoxcH4cGG1ODREFw20hFociXZVcprQQ/YsanEQ1OJQCGCVg126pJRuD3Oq7g+9ODWHkwzgLh1W\neRYqjvWQ9GpxAPDhq6KiIk3SQhAE6SVo03VCtTiKiorKy8vp8Xg8vmTJkq+//lqTVBAEQXoV2jTQ\nrBbHqVOn6HFBEObMmTN+/HhNUkEQBOlVpFeLQzgLd/3WrVubm5the/jw4VdeeaUgCOxoL3ux3W6n\nX44B2CX8RqPR4XDQbaPRSC+GoTPaPWQ2m51OJ13zw94ITrKpsGe5aQ+gDKDSI+ZyudjRXtYlSIhd\njytJEnsxa1YQhMzMTLrL5dRkMimZ5b4xnJmZyd7LLp0CU6w0BIz4K5UhqCuwbiiVoTynhBmtkptl\nXcrOzmYtc8u0LBZLZmYmm1mTyUQXT7I3AklGC+cwLIxUihZyLtnZ2dy97FkIYJpZWJtHL2Ytw4iW\nerRQ41xO1aOFLWFYr8hFGuuw0+nkfFCpGqzDKUcLOVeLg537Af5wllkEQXC73clXDZPJRCuIVtGi\n96XeSlocSuTn52dkZMB2ZmZmOBwWlLU4YGUq3YU1uGyzS9fCWiwWQRDoxdwUMbhSSV3BZrOxqXDL\n8NnfDJh+oDKmzEoimEwmk8nELo1lcwphx65/Zc2CWIGSuoLBYFAyy1XIcDhsMpnovawAAiFEkiRW\nMAFuVypDWIyuosXBLrNm5SkgoNmcspNDrFYr20aEQiHWMjfODkoRrLpCJBJRUahIMlrk6grq0cK6\nFAqFVLQ4otEoLK2mmWX957Q4ko8WLqddRgu9EqYxSMpaHOAwm1mVqpFatBgMBm6ZNTsNhhutjUQi\nBoNBvcYlXzXY9kSraOF+kjUkvVocSqAWB+yiFgfsohYHLRbU4iCoxcGgTR90YWGh0+ksLS2tqKgo\nLi72eDyLFi3SxDKCIEivJb1aHMDtqX5NAEEQpDeTrhWKCIIgSA/RxUpC1OKgCRHU4kAtDtTiQC2O\ns+iigUYtDjZR1ixqccizg1ocCXMqaaTF0SGKr7zyitvtJoQUFBT88pe/JKjFgVoc6r+Q8ohnj9Bt\nKdVl+PIbubPyu5RmZapbls6SzK5KThN6yJ4VUIsDtTgUAljlICGkPRi8xekclJkZi8fnv//+888/\nnzChhLsp51TFH/Zi9ZBI3nLKZajypJQc6yHp1eJQOo4giG65//rrrx00KBKLzf/00wvtS28nvVoc\nSscRBEGQLkmvFofScQRBEKRL0qvFoXRcrsVBCHnkkUfgyN69e+sMhv/+/ntCyPHm5rfeeisnJwes\n2e12WA8tiiKs3hEEQRRFWDsnCEIwGKSrerxeL6ydj0QioVAoIyNDkqRAIACLkgVB6OzstNvtMBwR\nDofpCv1QKARiBYSQ6dOnz5gxg80saHEcPXr01VdfpQn9fNWqTLs9Fo8LgvDMM89A7yfYgTX7rMM+\nnw9WukuS1NHRAduwuAuuCQQCsAQR1hmCIACX01AoRB0GI9CL+sILL+Tl5bEOgxbH9u3bly1bRggR\nRTEUjT704YeEkKaODm97+7x58yCzgUDA6XSCw62trX369IGEwuGwy+UCJ+lYtt/vdzgckFPIHV1H\nS52EawghBoOhtbUVhA7gcYBBq9X65ptvJtTiWLp06Y4dOwghjY2NxxobweHKM2dEh2P+/PmSJMVi\nsc7OTpC5EEUxHo/b7XZIFCQXqGPyxyqKIlwcj8c7OjpgTEySJJ/PR52MRqOgWxAMBi0WCziZl5e3\nYMECLv5Bi+OFF144ffo0IaSiooK0tR1vbiaEbD927Pinn0JGgsGg0Wi02+3xeJzGHhu0giCEw2E6\nVQAeBzjs9XpzcnIkSQqHw6Iout1uuNFgMEAYiKJosVggmCF3cOPYsWP/+Z//Wa7F0dnZOX/+fDhy\n5MiRt2pqPvv2W0JITJJ+8cUXWQ5HXJIIIS+88ILFYuEeK+twZ2cndF3G4/H29nYovVAoFIvFXC4X\n1Diz2QwzSfx+PxQ7lDDNKftoHnnkkXvuuUeuxXHw4ME333yTXv9UaanDYhEjEUmS5s6dC/eyNY6t\nOK2trdAUQNMBDsPjgNIDx2BlbzJtS1tbGwSewWB4+eWXuRqnIenV4lA6nlCL45ZbboEj11xzjcVi\ngetHB4NOpxOeFjQZDocjGAzSEjcajS0tLRAWJpPJ5/PRnu6Wlpa+ffvG43FoYvr27RuJRFpbW51O\np8lkMhqNHo8Hmn5YhUw99Hq9ubm5MBidn58Pkwqo8zD9wGazUYeHDx/ev39/2L7xoYecTifc29bW\nlpOTYzabRVGEXwuw09TUBOkaDIb29nZoqjo7O6FRliQJ2jJYKdvS0iLPqdFo7OjooEoubW1tffr0\ngVJyuVyxWEyuxdGvXz9w2GAwXH311QMGDICzU0SROtza2pqXlwfNRHNzc25ursFggHW00ISxTVVL\nS0u/fv3gRmg+rFZrMBiEnMIsjubmZnDebDY3NzfDCulQKBSJRKhgEAy7y7U4hg8fTov99MSJUMLF\nhAQCAZfLBcPxHR0deXl5sVisra0N2j6j0djU1JSVlQX3sm0cPA76aHJzc0FHAuIK1B6g4hFCYCU6\n1F6v19unTx8w4nK5WPUG6rDRaBwzZkx7ezsh5MYbbxRFEZ7UmFAI5nVJktTW1uZyudxudygUam5u\n7tOnDzxKWkpGoxEabvljbW5uzs/PlySps7MzHA673W6j0dja2upwOKCZ41pzGsAFBQWgZcEGMOSO\nBnBRUVFGRgbYuXzy5Pz8fLj3hgcftNvtdrs9EAi0tbXR2KNxaDabW1tbITbo72U8Hvf7/QaDITs7\nOxKJwHOH+W00pwaDAdpEeY2DoGKH3cB/t9tNHR42bBiUBiGkmAlgsAM1jj5WQojH48nNzSWEQGsA\n6fp8PrfbDRek3LYQQhwOh5KqSc9JrxaH0vGEWhwPPfQQHOnTp48oikrqCgMGDGhsbKTPjz3rcDhs\nNltra+v/5u3cST85OTBf4IoAACAASURBVDmBQEBJXSEvL6+xsZF6xZ5lX2rIWS0Op9NJHc7Ly2tp\naWHX+7PqCpmZmU1NTTQhVl0hMzMTwppezAoODBgwwOPxKGlxuFwu+l8IN5eImz4FWhwFBQUFBQWE\nEKPR2K9fPzazrAyCy+UymUzws0pkWhx9+/Ztb29XUlfIzc2FV0gi0+KANz5ov8i5WhySJHHTqkCL\nY8yYMWPGjIHMOp3OlpYWOMvNcMrOzma1I1KOFs6sSrQEg8GEWhx33nkn7MLLB82s2WxmJ2/l5eW1\ntrYqaXFkZWV5PB7YlUcLvO/LbySE9O/fv6mpKWG0iKKYUIuDBnBOTo7f76ciLclXDXm0sDUu+Wgh\n5wYh94sCx7OysqjDyVcNebSEw2G/35+wDPPz85WiBX6olKJF6eNhPSe9WhzccaXbuUkq3O+n/KzS\nvWxwyGEn3nbXLPcLmfy93FnuFDtjWn6Wm6GpklNufpj89zz5UmKbYznqj0bFYS6n8pna6g6rPDiV\nMpTnlPNB5bF2K6ddWmZRscw5LM8paxamlrNmlfyRB7Dc4dRqXMpl2N0gTLlqqDusVSOQvjdotdnH\nCIIgyAUEtTgQBEF0CjbQCIIgOiW9Kwmj0eiCBQtgGHTUqFFKHyeEHhwlwREQOlBKWuVsmswSQmKx\nGKfnkuS90JMl10booUuwCFVFsSVlyynnFO5Vyqm6WXXLOowW9ceapmhRKd4uHU5fGbISQt0y25Ps\npFw10hQtGqJNA01XDO7fv7+8vHzSpElw3Ov1jhgxYvr06Sr3yr+oordZHE6n02w20zltLS0t0rlf\nVNHbLI7s7Gx2UuPp06fZL6robRYHIaRfv35sTWhoaOAyq6tZHAaDgc6qZEuYZlZvszisVqvb7aaz\n5n0+H/dFFb3N4oDZb/Q9D+I8tapxfmZx9OnT58yZMzTRLoHpScmgTQPNrhg8cOAAPd7a2trc3Lxy\n5Uqj0XjHHXfQyUmhUIhWTjrLmP4cxeNxn88H4WIymWw2GytgSM6Kt9Ej8rN0l53eALvsWVabkZz7\ne8idZV8K5D6Qs02k0r1Kuwld4swmk1O5WcKQ8KxSZlVKKeHFSmUot6NiVt1h+S6XkFLhd1mGyceA\n/EoVh+VxyD3WJMswmWhJJqfyJyivGioBrF4s8t1kctplEHIOk0Q1LsmqQVTDMvkyVA9Ckh60mcWx\nYcOGoUOHXnXVVeFw+KOPPnr66afh+KlTp9rb20eOHHn48OHvvvtu5syZcPyTTz6pra2F7YkTJ95+\n7idXfv7zn7/99tuwbbPZDhw4wM6bviCw/3+xK6AuCrhZxkgvhF0KxC6a1S1s94J6R4ROqKmp0e8b\ntNKKwcGDB8PGiBEjNm/eTI8//vjjdLujo+PMmTPs/wvHjx9fPHv2T8aPJ4Rc9/rrVVVVsFSJ6KOL\no729/eLq4vB4PNjFIT+rYRdHY2PjRdTFEQwG29ra9N/FQUMC7Ou8i4OkB21erAoKCurr64lsxeD2\n7dv37NlDCKmrqwNRDgRBECRJ0ruScMyYMdXV1UuWLNm+ffuUKVM0SQtBEKSXkN6vetvt9ocffliT\nJBAEQXobOHaEIAiiU3QxNmo2mwVBoNMk5JOuQJKVngJNTjhiNBrpWfjyLt2FqQvsR4VNJhM9y94I\nc2jobkKzrD/cxClBEKgIJ3cvfGiZ9Z+drANDDaxL3Fwii8VC/e/SLHXAYDCwMzfMZjOXWShDNrO0\nzEGIlb0YpvrThLgvMbPOq5tVyakgCNRhSZKsVmvymQUxz4SPFUgyWuRlqBQtXHzGYjGr1crlji0K\nGOvjypCNAdYl1k6X0cLmFIJQKVqEc+eJgvwpHYUznPvd8S6jhfVBHi3UbMrRwq1YgVNcsatnNoXH\nCihFS5c1Lk2iRrpooCORiKD8iVyQcIVt4awQfsKRVqPRaDKZ2JFidkwZPv2bcEwZypcdz+VGnNng\nAOkvdkxZkqRwOJxwFockSXa7nU2IHZe32Wyg/Qi73CwOQkg4HE44VA0hyJplZ3E4HA72FwWSY0uJ\n3SXnDqBDjWLzzn3VG+T2E5YSeyM3iwO+kMDmlJ3FIUkS+/MM2vCsKYvFwvrPRovD4VByqVvRwplV\njxbCAKLSbGahGtPruVkcUIYJZ3HAb6FKtLB1gZuBAEGYMFrgXvZKCDO2aiiVIWRWqWpAa6hU45KP\nFnJuENIvDwBgMLWqIY8WpccKKEULtOxKZuEzFyQNYBcHgiCITtGmgZYkad26dUuXLi0tLaVzDCmi\nKC5cuFCThBAEQXoP6f2qN7B161b2g4QIgiBIMqRXi4MQUl9fHwqFuE8EqWtxyJEvtEctDtbhhFoQ\nrJFuySCoixUkWYZyOypm1R3uUgZBqfC7LMPkY0B+pYrD8jg0oBYHanGkRHq/6h2Pxzdv3nz//fcv\nWbKEvb60tFRFi4Mbw83IyOBW1qp/Q5e7mEVdgkDlRvZbOG63W67FAZ+kTMEyGFQ61bdv35TNUuDb\nr926l10pzqG+qlXdLNUnUyehERXLNpuNfoNYTsrRwo3vq5AwCZXM9iRa4BPDCVGJFrZpS1hcKVcN\nkp5oYT9qlZmZKdfiSLlqpCla0kR6tTh27949cuRIeaSqa3FwXwZrb29vaGiAbQG1OM4WOGpx0Myi\nFgdBLQ7U4lBBSYujsbGxoqJixYoVPp+vtLRUk7QQBOnlhEIh31m4eXKXGNq8QRcWFlZWVpaWlhoM\nhpKSEo/Hs3r16rlz51Kp/vfff/+RRx7RJC0EQXo5N91003fffQfbl19++TfffHNh/Ukf6dXioDzz\nzDOaJIQgCHLy5Mna3/1ucHb2mfb2Ib/85YV2J43gQhUEQRCdooul3qjFQXcNqMWBWhyoxZGEFgd3\nAWpxpBHU4oBd1OIgqMWBWhzJaXGwsJVIQC0OBEEQJHl+85vfpHajNm/QkiStX7++ubnZZDJNmzaN\nTnwOh8NffvmlKIrxePy+++7Lzs7WJDkEQZCLiJqamtRu1KaBploc+/fvLy8vnzRpEhw/ePBgfn7+\nrbfeun///l27drEzPRAEQS5JPvzwQ+4IuySnW6RXi6OgoACWtHFDcKjFoeISanGoF5o8IaXC77IM\nk48B+ZUqDsvj0IBaHFprcXAXqFQNohqWyZehehCyRuQfyJaLQyRJerU44OBnn3128uTJJ598kh5H\nLQ6WS0CLIxgMiqLo8/msVqvL5VIKR9TiAHqVFge7tDozM5O9uCdaHBRBEOQxeWG1OKZNm8YdWbt2\nbTI3ykmvFgeMbs+YMePEiRNlZWWPPvooHEctDti9NLQ4/H5/cXExdeb2229fvny5gFocqMXRt297\ne3tJScmOHTvgSL9+/Xbt2mW1WnuoxcEiSVJLS4vOtThGjx6tfoES2jTQBQUF+/btIzItjm3btuXm\n5o4ePdpkMnHNLnLJEAwG29vapEWLCCE7q6sf++tfL7RHiI6oq6v79uWXxxQUSJJkmDcvHA4n/6/J\nRYrf7//iiy9qamoKCgoefPBBh8OR8lJqbabZFRYWOp3O0tLSioqK4uJij8ezaNEiQsjYsWO/++67\njz/+eOPGjVOmTNEkLQRBED3z2muviaJ45513iqL42muv9cRUerU43G737NmzNUkCQRDkoiAQCEAD\neNNNNz399NM9MYULVfQOVVZsa2uj3ZFID+ns7Gxra2tra/P5fLQrHEE0IRaL7d27NxwO7927t4dd\nu7pY6o1aHHSXm0tkMpkmTpxYV1cHuzfccMPGjRuVzF4oLQ6uS5HmSLdaHIcOHZoyZQr9tXvppZde\neukleRmiFgfRQouDKy6LxQLPV25WuFS0OP7lX/5l4cKF9fX1gwYNeumll5TcTgZdNNCoxQG7ci0O\nURTr6uo6/vhHl9X6w+nTt37wAVws6EmLg1MzgJIUdKzFUVtbO3bo0G0vvkgIWbhp0zc1NaFQCLU4\n0qTFwVXncDgcCoUubS2OzMxMOipYWlo6Y8aMnJycnJychM6ro4sGGkEQ5JLhqaeeGjlyJDTZML3t\n1ltvnThxYgqm0qvFEY1Gy8rKfD6fKIpTp04dOHCgJskhCILolmeffXbChAmwvWvXrrFjx6ZsSptB\nQqrFUVRUVF5eTo9XV1dbLJY5c+bcc889GzZs0CQtBEmNQCAAo4J0nQKCpAOn0/nBBx80NTXV1NRc\nc801PTGVXi2OjIyMm266iRDicDjYPhrU4lBxiTObsCjkggMqUgbqShHyMlQXK5CfTeik3I6KWXWH\nu5RBUCp8er0kSR6P55ZbbqGr+24ZPpzLkXoMyB1QcVgeh4ZerMUhyIQ1aIVSMcvdQmQ1Tm6Tuzih\n/z0pQ/UgZI188sknd95550cffTRhwoS33nrrj3/8Y0K3kyG9WhwDBgwghNTX15eVlbGCG6jFwZK8\nFofh3EXGOtHi4IybTCYl4xdQi8Pr9ToFwbdoESHk13/729+rqugFdrs9oXHU4kjSJfVo4apzXl5e\nZmZml2Yvai0Om8129913V1ZWTpw4ceXKlcncokR6tTgkSdqyZUttbW1JSQmbN9TigN0utTjOnDnD\nFkU8Hoei0JUWB3UDiEajDQ0Ngs60OKjAhZxAINDQ0IBaHGnS4mCbWkLI6dOn/X7/pa3FMXjw4NWr\nVzc1Na1fv17ucLfQpg+6oKCgvr6eyLQ4KioqvF7vnDlzkvzlQRAEudjZuXPnqVOn8vPzjx8/Pn/+\n/J6Y0uYNurCwsLKysrS01GAwlJSUeDye1atXz507t7q6uq6uDuSrMzIyZs2apUlylzyRSCQQCNhs\nNvoWhiAAfYm2WCyxWEylxwO5UNx3331ZWVmw3dDQAP/13nHHHSmYSq8Wx7333quJ/d7GzJkz2ckw\nCEJ5880333//fdi2WCwbNmy48sorL6xLCEcoFOI6J1MGF6rokbq6uv2/+tV1gweL0ai9Z2IryCVG\nXV3d/zdr1lO33koIuek//sPj8WADrTeo8H3P0UUDjVocdBfMCsqyA3AxanF0V4tDZZYC3CIvQ31q\ncXBzZmjeBdTiYDJ7YbU4NEQXDTRqccAuncWh8rBRiyM1LQ7qiRx4BBeLFgenvUfzjlocsC3oQItD\nQ3TRQCMXI+FwOBgMhsNhOvEL0RYoWPbFBVrSC+qUHmlra4PGPcmJ9hcR6dXiAODDV0VFRZqkheiE\nu+6668iRI7BtUV5HgKTGzp07Z8yYkVCr+m6cDXWWzlAoHA4PGzYMdn/0ox99/vnnF9YlbUmvFkc8\nHl+yZMnXX3+tSSqIrqirq6v5j/+QFi36vmcf9UESUl9ff8+oUdKiRdKiRbcMH/7zyZNhu6Cr75P2\nKgLhsM1shpL5xy9+AasxLiXSq8UhCMKcOXO2bt3KXY9aHCouqY82GPSkxaHipNyOill1h7uUQVAq\nfHq9JEkqPrMWVIqFc0DFYXkcGrqvxZFknyZ3e+/U4kh4cUL/uwxvlTJUD8IuHUuN9Gpx0FDgrkct\nDha5FgdXAhSDnrQ4lIJSt1ocShfoU4uDClao06dPH+p879TiYDGZTNnZ2eqWz4MWh4akV4tDCdTi\ngF0lLQ6l75jFYjHo9rVarf3796caCNwsjvOjxaH0OoNaHD3R4jCZTPBoaMyr09LSAlfqSosDojon\nJ6ejo4Mb1ayqqnK5XEajcfDgweyvZspaHCzRaLS1tfWCa3FoiDYNdEFBAXw4gNPiQDSkpbOzubmZ\njrXOnDlzwYIFF9YlRFtOnDhx66230l+FkmuvvbD+pMzLL7+8YsUK2Kb/aUHDB/rDhJAbb7zxr3/9\n6wVx7yIivVocmhhHAJ8o5rndjX/4AyGk7NCh3zF9/cilQVNT02XZ2YffeYcQ8sTy5U0X7fzFurq6\nlU88MeOGGwghlnPVgtreeSfTbj/q8YzvgUpy7yG9WhwA18WMIAiCJIM20+wQBEEQzdHF+gLU4qC7\nhq60OFjOvxZHNBr1+/0w7KM0YiNJEvSiGo1Gh8MBXulZiyMejweDQXgudEaNSrQIadPiUJq9o4I+\ntTi6nIhJmDghZwcbhVS1OLisCajFoS2oxQG7yWhxsJx/LY6f/OQna9asge2EFabO6z127Bhd2fX4\n44+//vrrRMdaHN/V16/auXPVqlWw+9577913333kAmlxKM3eUUGfWhwJF0CytPr9LS0tNE5KSkoW\nL16cshYHlzXU4kB6KbW1tZuff/72K68khBgSDQJ7A4Gr+veHtYXLdu0qras7zx52l1a//7Hi4k/m\nzCGEPPPnP9fp3uFLg3ZRzHW5PAsWEEI2HD786507L7RHOiW9WhzqGh0IgiCICunV4lA6jiAIgnRJ\nerU4lI4n1OKgqpXRaDQQDnsDAUJINBbr7OykpwRBsNlsHR0dtJ9IFMVwOAwdQJFIRBRFejEdJIRh\nN6PRGAwGaS+SKIq0HxB6OemNMKYEZ10ulyGRFkc8HmdXALa3t0NPGXsvIQTGXsBhOkLIdjXGYjFY\nBSdJktlsjsVi0AfdLoreQECMRAghbYFAJBbrDIXikgTF0hkKxWIxtlhgnE0QhIyMDPkoK/SXwZgM\nDKSwmYVOUtiOxWImk4ktQ7ZPPB6Pd4ZC3rPL9nzBoDcQaBdFQggcDITDsXj8/7bPOgm+QYmxOTUa\njS6XK6HD9AFB/778sdLdcDgcCATACPtYCSE0Wvx+Py09MRKJxmKwHYnHQ9EoPW4OhSAhg8EgiqI8\nWqxWK11Tyjnc2dkJBQV/aWahH5+WocVi8fl8NAZYh/1+Py29cDQaPutkNBYTIxHYjksSrR2RWCwQ\nCIDDVqu1vb2dFgs1yw1I0GghTI1LvmqQc2ucPFrg0cRiMT8TJx2hkDcQAN/aAoG4JHWKosQEczwe\n9/l8EISw6JRzGNKiKwAlSfIFgy6rtYOx0yGKYCcSidAax0WLIAiRSIQOTnLR4vV6VdoWVlmXmk1Y\n47RE0oL169dXVFTAONv777/f5fGPP/741bNs3rxZkiSlUQVWx5JdvsyqGbDrbtk1l10OJbMDuDk5\nOQkNvvDCC5IkwSAPEA6HJUli/yFglymzy/zZdf2gVQJ02dXDPm8jk4tMpgRY4QX2y6ErV67kng40\nhe+++y69JpnS6zLmWMf6MXIidkaohDWoJKdw/PjxhEH14x//+P/sd6f02DkArM4JW3rm7syXYAMp\nLy+P/ohyDB8+nF7WZemxo3Zs7LGlZ+jKCJsLNpjZYr/77ruhflE/YRyMrlrmUCo9pWDu8qu1bJyw\n21nM7xxr8N///d+ls8OYAGxv3LiRXsM+kb6Mk6wdzZsC9jirxrNx40ZJkk6ePPlt0nTZolLOkZ5K\nma1bt9pstnHjxjU1Na1du5ZKbSgdZ+no6IDXH3ZhuyiKXaoryM+eHy2OlpYWKc1aHLA7YMAAj8eT\ncFzeZrOxggPqWhynT59OhxYHkc11yc3NPX36NH1SutLikJ/lokUnWhxZWVlUNkQeLfST3kQ2AyF5\nLQ6fzwdtdPq0OGA3+WghibQ46C8xxHlqVeO8aXGcOXOGJtolY8aMSfJKbfqgCwoKQImV0+JQOo4g\nCIJ0ifE1LdTWs7Kyamtrd+/e3dzcPHnyZK/Xu2LFihtuuIE7LlfpJISEw2HoNKA/XEeOHAmFQuw7\noMT0gf7jH//o168f+y8kPevxeE6ePJmVlcUep2e///77eDzOTQiFjUgksnPnTu61iJ5tbGysr6+H\nznRCSDAYZF+ICCG7d++GVwn5ve3t7YcPH+bkJenZ6urqjo4O9n92tqtn586d2dnZ7L+c9Eav11tZ\nWckqf7IFWFVVRZjOFugbpWfD4XB5eblSZhsaGurr61mtSzanBw4cMJlM7KIPejYYDH777bf0f0x4\nQPRsbW1tc3Mz/ZdZOrdTa8+ePbm5ufTfTHhJpPe2trYePXpUKbOVlZXq0ZKXl5dMtHA5raioiMVi\nCaMlGo3u2LGD7dCAjkt6QU1NDZtZzvKePXscDgdbF+jZjo6OQ4cOKUXL8ePHfT6fUg+PSrS0tbX9\n8MMPQ4cOhd1QKMQtNfr+++8lSWL/zWerhnq0NDQ0sN16bE4PHjzILu5gz4qiuHv3btptwkXLqVOn\nPB5Pfn4+7AYCAW5u9c6dO7OyslKoGpWVlaIoJowWSZJU2pampqYTJ06Alqk8p99//z10fJHkoFnr\nkvRqcXDHE+J2uzlB5O3btw8bNixhHqLR6OLFi++4446Ebf2ZM2dOnTpVXFycMKHNmzdnZ2cnNOv3\n+/fu3XvXXXclvLG2tpb9X4ztigJWrlx57bXXsv1flJqamh9++OG2225LaHnfvn1ZWVnXKoiWLV26\ndNy4cQl1gauqqo4dOzZx4sSEN5aXlxuNRvqLwtWujo6Offv2KT2XkydPNjc333zzzQnPrl27tqCg\nIGEZNjc3Hzp06I477kh4Y2VlZTgcVvrP7osvvigqKqL1nLPf3t5eXV19yy23JLx3x44dl112WUKX\nYrHY4sWLJ0+enHAZocfjqa2tVYqWLVu2ZGZmJjQbCAS+/fZbNlq4y44cOSJJklJmP//886uvvjqh\nsnBtbe0PP/ygpF1z4MABl8t13XXXJTy7fPnym2++mW1BKEePHj127Nitt94Ku5mZmVxQbdq0qU+f\nPgkz29nZuXfvXqVoqamp8Xg8VKCOY926dYMGDUpotqWl5dChQ1OmTEl4Y1VVFe1vIYnks1WqRmVl\nZXV1tVLV2Llz59ChQxO6JEnSRx99NGnSpIQ64M3NzTU1NePHj09oduvWrddcc82oUaMSnu0JqMWB\nIAiiU7CBRhAE0Sl6XOqdl5en9E0aQRAuu+wypTlMbrdb5Ys1/fv3V5oSZDQaCwsLlW6EuRYqDg8Z\nMkRJi8dms9GuBjm5ubkqs5SGDh2qNDXN4XCodGP169dP5aM+XWZWZYrYoEGDlL6YYzabCwoKlG7M\nzs5mP2LEMXjwYJXvSzmdTpXM5uXlyb8ZBkC0KE2xSlO0EEIS9jNQhgwZopRZ9WjJycmRT8GmFBQU\nJOz3I11FCyFkwIABSl3bXUaLSsfrwIEDlRzuMlqUtFOANFUNlWhxuVwq37tSiZYeos00OwRBEERz\nsIsDQRBEp2ADjSAIolP01QctqarfRaPRBQsWwMTVUaNGsVNetm3blpubW1RUlNACPctaGDlyZEtL\ni8/nE0Vx6tSp+fn57I1Wq7WsrIyezcvLS5g0Oux0OhN6O2rUqJtvvpl1SXOHVbzV3OGpU6du3bq1\nS29TdjiFeNDEYQzgdDvcQ7RZqKIVJ0+erKmpmTVrViwWq6qquuyyy9izLS0tgUDg0UcfveGGG+i6\nxHg8/qc//enQoUNFRUV9+/blLAwdOpQ9y1oQRbG9vf3BBx/Mz8/fsGFDnz592BtjsRh7tqCgQJ40\nOlxZWblt27aE3g4ZMuTo0aNpdTiht2lyePfu3Xa7vUtvU3A45XjQxOFeHsDnweEeoq8G+tChQ9nZ\n2YMHDzaZTN9+++3o0aPZs3V1dUeOHDl69OiRI0cGDRpEV3lde+21sVjM7Xb37dtXboE9y1oYMGDA\nsGHDHA5HLBY7cuSI0Whkbxw5cuSQIUPo2b59+yZMupc7vGfPnn/6p39K6O2gQYMMBgPrkuYOK3mb\nDof3798/efLkLr1NzeHU4kETh3t5AJ8Hh3uIvvqgA4EA/I+QmZlJ5W8oDoejuLj44Ycfvuqqq9av\nXw8HQVGQzgzjLHBnWQvffvttbm5ufX39Z599NnHiRO7GAQMGsGcTJo0OB4NBJW/Xr1/PuaS5w0q3\npMPhaDSajLcpOJxyPGjicC8P4PPgcA/RVx+0zWYDZSyfzyefbzt48GDYGDFixObNm3toYdOmTZs3\nb66trS0pKenfv39dXR17oyRJW7ZsoWepBS5pdFjlds4lzR1OxlutHLbZbGzhK3nbc4e79YB67jAG\ncLod7iH6eoNWV7/bvn37nj17CCF1dXUJhS+6ZcHhcHi93jlz5kDhcjdWVFSwZ5WSRodVbudc0tzh\nZLzVymGn05mMtz13uFsPqOcOYwCn2+Eeoq+FKpIkbdy4sbW11WAwlJSUcMuQgsHgmjVrgsGgyWS6\n++67WUHuLVu29O/fH4Zo5RboWdaCzWarr6+HrqKMjIyZM2eyN27atOn48eP07PTp0xMmjQ47HI6E\n3t59993bt29nXdLcYRVvNXfYarXW1tZ26W3KDqcQD5o4jAGcbod7iL4aaARBEISiry4OBEEQhIIN\nNIIgiE7BBhpBEESnYAONIAiiU7CBRvTFX/7ylxEjRmRmZt5///30a+iff/758OHDc3JynnrqKfjQ\nMndk165dY8eOhYvp9ooVK5544onZs2dnZWWNHz++srISLvjoo48KCwvtdvvYsWPh4A8//DBhwoQF\nCxYMHDiwsLBwy5YtCdN9/PHH//CHP8CpV1999bnnnjuPBYP0RrCBRnTEiRMnHn/88ffff//EiRMZ\nGRnPPvssIaSqqmr+/PnLli3bs2fPnj17VqxYIT+iZPBPf/rTuHHjjh49OmHChBkzZkiSdOrUqWee\neWbp0qWnTp266qqr3n77bbjywIED0Wj06NGjDz300K9+9auE6U6dOpUuEvvrX//6wAMPpL9IkF6N\nvlYSIr2cNWvWTJs2bfLkyYSQ3//+9wMHDozFYp9//vmsWbPGjRtHCPn44499Pp/8iJLBoqKip556\nihDyxhtvLF68+NixY4MHDz569OiQIUP8fn9ubu6pU6fgSqPR+OKLL5pMpkcffXTNmjWEEHkqo0eP\nnj17ts/na2lpOXPmjFaKZQiiBDbQiI44ffr00KFDYbtv374Wi6Wpqamurm748OFwEL6A/umnn3JH\ndu3aRY2wU/vp55rMZvPQoUPr6+sLCwsXL168YcOGzMxMq9VKP5fVv39/+IoS/ZaSPF1CyPjx4//n\nf/6ntrb2vvvuMxqNWhcAgpwDdnEgOqJ///41NTWw3dTUFAqFcnNz8/Ly6urq4GB5efny5cvlRwgh\n9JuH9BQh5MSJHj9pfAAAAcpJREFUE7ARjUZra2sHDBiwatWqdevW/fd///emTZtmzpxJr5R/iTFh\nKlOnTl23bh32byDnB2ygER1xzz33/OUvf9m8ebPX633xxRenT59uMpnuv//+5cuXf/PNN8ePH//Z\nz37W3NwsP5KZmXnw4MEDBw60tLR88MEH1OChQ4cWLVrU3Nz861//Oj8/f/jw4S0tLS6Xy263ezye\n9957LxgMKjkjT4UQMnXq1DVr1lRVVU2cOPF8lAjSy5EQRE988cUXV1xxhdvtnjZt2pkzZ+Dg0qVL\nCwsLMzIyHnvssVAoJD8Sj8efffZZl8s1atSoVatW3XzzzZIkLV++/K677nrggQfcbvfYsWMrKiok\nSWpra5s8eXKfPn2Ki4vXrl2bl5e3bNmyI0eOjBgxAtJit+XpSpI0cuTIJ5544jwXC9I7QS0O5JJl\nxYoVZWVlK1eu1NZscXHxa6+9dscdd2hrFkHk4CAhgiSL3+/fs2dPbW3tbbfddqF9QXoF2AeNIMny\n1VdfzZw584MPPjCbzRfaF6RXgF0cCIIgOgXfoBEEQXQKNtAIgiA6BRtoBEEQnYINNIIgiE7BBhpB\nEESnYAONIAiiU7CBRhAE0Sn/P/iKJNsbKS7VAAAAAElFTkSuQmCC\n"
      }
     ],
     "prompt_number": 32
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "And now we've got the DOW ordering we really wanted. This has been a quick tour of plotting occupancy data using ggplot2 from within an IPython notebook. Future installments of this series of occupancy analysis will focus on more occupancy visualizations using some combination of R and Python.\n"
     ]
    }
   ],
   "metadata": {}
  }
 ]
}