dask-dataframe-basics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Basics of Dask Dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook goes through the basics of using ``dask.dataframe``. We do the following\n",
    "\n",
    "1. Create a fake dataset\n",
    "2. Load datafrom csvs\n",
    "3. Perform some computations\n",
    "4. Output results to file or as plots"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Create a fake dataset\n",
    "\n",
    "We create a few fake dataframes, and save them as csvs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "for year in range(2000, 2005):\n",
    "    idx = pd.date_range(str(year), str(year + 1), freq='d', closed='left')\n",
    "    pd.DataFrame({'A': np.random.randn(len(idx)), \n",
    "                  'B': np.random.randint(len(idx))},\n",
    "                 index=idx).to_csv('{0}.csv'.format(year), index_label='date')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Load data from csvs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can load multiple `csv` files into `dask.dataframe` using the `read_csv` function. This supports [globbing](https://en.wikipedia.org/wiki/Glob_%28programming%29) on filenames, and will sort them alphabetically. This results in a single `dask.dataframe.DataFrame` of all data from 2000 to 2005.\n",
    "\n",
    "To tell `dask` to use the `'date'` column as the index, we'll also use the `set_index` method. These can be chained, allowing the dataframe to be created in a single line."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>A</th>\n",
       "      <th>B</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2000-01-01</th>\n",
       "      <td>-0.928750</td>\n",
       "      <td>280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-02</th>\n",
       "      <td>-0.711634</td>\n",
       "      <td>280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-03</th>\n",
       "      <td>-0.426995</td>\n",
       "      <td>280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-04</th>\n",
       "      <td>-1.698917</td>\n",
       "      <td>280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-05</th>\n",
       "      <td>1.901336</td>\n",
       "      <td>280</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   A    B\n",
       "date                     \n",
       "2000-01-01 -0.928750  280\n",
       "2000-01-02 -0.711634  280\n",
       "2000-01-03 -0.426995  280\n",
       "2000-01-04 -1.698917  280\n",
       "2000-01-05  1.901336  280"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import dask.dataframe as dd\n",
    "\n",
    "df = dd.read_csv('*.csv').set_index('date')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dask `DataFrame` objects are partitioned into multiple frames along the index. To see how the dataframe is partitioned, you can look at its `divisions` attribute:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('2000-01-01',\n",
       " '2000-12-31',\n",
       " '2001-12-31',\n",
       " '2003-01-01',\n",
       " '2004-01-01',\n",
       " '2004-12-31')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.divisions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Perform some computations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`dask.dataframe` mirrors the [pandas](http://pandas.pydata.org/) api. We can perform arithmetic, reductions, selections, etc... with the same syntax used by pandas.\n",
    "\n",
    "The main difference is that we'll need to add a `.compute()` to our final result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dd.Series<series-index-set-partition--collect-f2f64a639a376601d6c253dc1ca8e4dc.A[elemwise-ab0affb8c9739c38e73d8d01a529faef], divisions=('2000-01-01', '2000-12-31', '2001-12-31', '2003-01-01', '2004-01-01', '2004-12-31')>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.A[df.A > 3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Instead of a pandas `Series` object, this returns a dask `Series`. To perform the computation and get a pandas object, one needs to append a call to `compute()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date\n",
       "2001-03-20    3.116674\n",
       "2004-12-31    3.067005\n",
       "Name: A, dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.A[df.A > 3].compute()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`dask.dataframe` supports a common subset of the pandas api. For a full list of supported operations, see the [documentation](http://dask.pydata.org/en/latest/dataframe.html). We'll do a few more computations below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "106258"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.B[(df.A > 0) & (df.A < 3)].sum().compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "B\n",
       "0    1096\n",
       "1     731\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.B.groupby(df.B % 2).count().compute()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Output results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After doing our heavy computation or data reduction in dask.array we would like to either store our results on disk, or transfer our smaller results to numpy arrays so that we can interact with the rest of the PyData ecosystem.\n",
    "\n",
    "- Use `.compute()` to create a pandas object\n",
    "- Use `.to_csv()`, `.to_hdf()` or `.to_castra()` to store to disk\n",
    "- Depend upon the `__array__` protocol for automatic conversion with other projects"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use `.compute()` to create a pandas object\n",
    "\n",
    "As we've already seen, the `.compute()` method performs computation and turns our lazily evaluated dask objects into pandas objects:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date\n",
       "2001-03-20    3.116674\n",
       "2004-12-31    3.067005\n",
       "Name: A, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = df.A[df.A > 3]\n",
    "x.compute()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Store to file\n",
    "\n",
    "You can store your result to a file using one of the `to_*` methods:\n",
    "\n",
    "- `to_csv`: writes a csv\n",
    "- `to_hdf`: writes a hdf5 file using [PyTables](http://www.pytables.org/)\n",
    "- `to_castra`: writes your data as a [castra](https://github.com/blaze/castra)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "res = df[df.A > 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "res.to_csv('my_data.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot with Matplotlib\n",
    "\n",
    "Because `dask.dataframe` objects implement the `__array__` protocol they often just work with other libraries in the ecosystem. In the example below we pass a `dd.Series` object directly into a `matplotlib` function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x110c8ddd0>]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEACAYAAAC9Gb03AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110b5d450>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(df.A)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Of course, as performing `compute()` returns a `pandas.Series`, you can also just have the `Series` plot itself:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x110c3f710>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEPCAYAAACneLThAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJztnXe8HLXV93+6uIFNLzHFlNB5ML0ZsLnAA5jQMaEk9JeS\nhISEFmrgEtpLSWiB0AmBYCCYEDoYyKWXBGwwYMA4QMAvNqG5YoxtvX/MitVqVWc0Zfee7+dj390d\njaTRaM4cHR0dMc45CIIgiPako+wKEARBEPlBQp4gCKKNISFPEATRxpCQJwiCaGNIyBMEQbQxJOQJ\ngiDamMKFPGNsAcbYGMbY/UWXTRAE0dMoQ5P/JYC3AJCDPkEQRM4UKuQZYysA+AGAGwCwIssmCILo\niRStyV8K4CQA8wsulyAIokdSmJBnjO0K4FPO+RiQFk8QBFEIrKjYNYyx8wEcBGAugH4AFgEwinN+\nsJSG7PQEQRAp4JxrlefCNHnO+Wmc80Gc81UA7A/gSVnAS+noX0H/zjrrrNLr0JP+UXtTe+f1z0aZ\nfvKktRMEQeRMrzIK5Zw/BeCpMsomCILoSdCK1x5MZ2dn2VXoUVB7Fwu1d0JhE68+MMZ4lepDEATR\nCjDGwMueeCUIgiCKh4Q8QRBEG0NCniAIoo0hIU8QBNHGkJAnCIJoY0jIEwRBtDEk5AmCINoYEvIE\nQRBtDAl5giCINoaEPEEQRBtDQp4gCKKNISFPEATRxpCQJwiCaGNIyBMEQbQxJOQJgiDaGBLyBEEQ\nbQwJeYIgiDamUCHPGOvHGHuJMTaWMfYWY+yCIsvvCXz2GTBrVtm1IAiiKhQq5DnnswFsyznfAMB6\nALZljG1dZB3anaWXBg45pOxaEARRFQo313DOhZ7ZB8ACAL4oug7tzscfl10DgiCK4tVX7ccLF/KM\nsQ7G2FgAUwD8g3P+VtF1aHeYdjtfgiDakdmz7cfL0OTn18w1KwAYxhjrLLoOBEEQ7QLn9uO9iqlG\nM5zzqYyxBwFsAqBb/N7V1fVdms7OTnR2dhZdtZaHNHmCaG+6u7vR3d0NAPjPf+xpGXe9BiLCGFsK\nwFzO+VeMsQUBPArgbM75E7XjvMj6tCOMAVtuCTz3XNk1IQiiCJ55Bhg2jIFzrlXvitbklwVwC2Os\nA4mp6FYh4Il4kCZPEISgaBfKcZzzjTjnG3DO1+OcX1xk+a3IQgsBt90Wdk4WIf/558D666c/nyie\nKVOA+fPLrgVRFi7jB614rThffw289FJx5b37LvD668WVVybPPFN2DeIwcCBwzTVl14IoCxLyBGFg\n2DBgzpyyaxGH//637BoQZUFCngiip8170/wF0eqQkO+BkODyp6e91Ij2g4Q8QRBED4aEfBvio8n3\n6pV4ZRAEUeeWW4DBg8uuRRikyfdAfIT8vHnulXLtjHgw2sVcQya6ODzyCPDGG2XXIgwS8kQQZQs9\nxoAvCohLmuY6f/lLYObMsHNOPhmYPj28LILwhYR8D6TVtbpp04orK0TYX3FF+BqCiy4CXn658bc5\nc4DHHw/Lhyie228Hfv7zsmvhhoQ8QWgo0lyjlnHnncAOO+Rfbkxuuy15YfUkLr8cuOqqsmvhhoR8\nD6TVNfkQXnyx9SaQ580ruwbhnHxy8o+oHiTkibZmyBDgpz8NP69MTZ5oDdrlvpGQb0OyaPKt2LHT\naMZphXya9mnFNiVaB9LkCSNVNevEEopLLAFcf32cvAiiqpCQbwFOOQV44YWya9E6qEHFTJ38yy/N\nm6e0m588EYeqKj42SMi3ABdeCFx9dbz8qthRjzwyXqTEvn2BqVPr37MI6nYR8nnc8zfeAD78MH6+\nRFzaTsh//jnw7LNl1yI+VRA2++8PjBuXT9433BD3vsk71NvaTj42aVKzBk8Tr2YGDwa2377sWlSD\nV16prldU2wn5004Dhg4ttkzGgOuuS3/+888X+6Cn1eruvBO49964dfnss7qg8G0Dn3TyNfoK+RVW\nAJ54IqwuMQgp65//BAYNyq8uoVRVsBWBfN822QS4//58ytltt3w3sGkpIT9ypH3J+6RJwOTJ+ZT9\nyivpz91qK+CDD6JVJVdibyM3bhzw5JNx8wxBFbBffWU/XjbPPw98/HHZtagj2qeKJsCiyauvPPAA\ncM897nR//zuwzDLNv1dKk2eMDWKM/YMx9iZj7A3G2LEh5//oR8Ddd5uPr7EGsPHGWWupJ+sN1gnP\nvDT8LA9kbM2tbHOI6RhNvBKh9O2bjEzzeOH55PnUU/p5rUoJeQDfAjiOc/4/ALYAcAxjbO1Ymc+a\nVUxwq1hstVX1vGryHJ7nZa6ZN89/pFSGUA8pM60AyUvTppdgnT59kjlBwT33FBvJ1fRsVkrIc84n\nc87H1j7PADAewHIxy0jT2b/9Flg72qsmjLlzk7+2GxV6Tb7pdemqYoP96COz+6PKY48Bq6yiP0aa\nfLlMm9Y+UTj79q1/ZgwYMQI466ziyjeZUl19uFf8qvjBGFsZwIYAXiqrDoLp04G337anySoMXMIm\nZtlZtLrYNvm0HHww0N0dXwjn6V1zzz3JcP6oo4BDDgH6949fRtFkrfsGGwD9+gFvvWVP94c/JGli\nuhLHpm/f/EZMvntA6KikkGeMDQBwN4Bf1jT67+jq6vruc2dnJzo7OwPzDq9PVQRbFTB1pKeeSgTv\no48Ciy8OfO97xdYrLbffDrz/fjL/AQAnnQTsu28+ZR1zTDLxf9RRwJ//3Kj59VTef9/vmbzsMmDi\nxPKE/D/+kWwYovPaEn97986vfJ82EqN+AOju7kZ3dzcAYPx4+3mFC3nGWG8AowDcxjlvctiThXwa\nZs1KGiyW9jRpUpx8TITan33IQ5N/+unE/rj22sDmmyfRH/MidpvI8x7Chlpk7JpWJuR6Z89OtHaf\nPGbOBI4+OglhXAUuvzzxXjnggOZj4pnoKNkXUX42ZQV41CjgrrvONp5XtHcNA3AjgLc455fp0rjM\nJnlg68jXXutO00742OS//to/v7TtVpTLXtnePyqt6qr47bfAggs2/24SjO++C/zlL/XvZT1fn3+e\nrL2xjeZlhUC9P7Hqnae5puh301YADgSwLWNsTO3fcDlBGROgPjc4L6r28tB1JM6rt7w9qzA0tfvM\nmfEUjSzzMK2G6Rnq5WkrKKtNHnsMuOAC/XoAXZ1a0SZftHfNs5zzDs75BpzzDWv/HimyDvp65V/G\nMceUW4fJk/20dPGwjhxZ/+3xx4Ebb7SfN2GCO+8qabQmc80ZZ5TnaVVl0t6PBRaIW4/Y+JjtqvJS\nTutd01IrXvPCdqNjeWE89ljj93XXtec7axbw//5furJ0D+Syy/ptZSZeBG++Wf9txgx9WsE77yQL\n0VoRuf3nzk0mANNiGsrvuqu/O2jVyLq2wSTk1fRlC9I8vNxi0xKafJHss09zSNoHH9SnFW9I0Vg3\n3gjsskvjbx98kH37s2++qX8WQtR0g37yE2D55dOVY9K61NVyvn7yrgknOVhYDNJOvIY8hLqXtxry\nICa+sUmqthjKF5My5GuuicnAgckkagi+Jtsy25mEvMKoUY1xbDhPNCq1QRZdtG5vFsduvx146KHG\n3558Mmwj42eeae44IcG/Pv3UP60vIXY/OW2sjv3668Bdd2XLw9ahTS/xGHkTdqqkyU+ZEh7wSyfk\nqxZUMO1CxbYV8kCj5ixuotpQ06bV/Uxj3tRhwxLfWxehD4cPpg7j05HUUY3Peb4vgXPOAfbbzy+t\n4Mwz3QtpBHKMeRc6zTPr/bd5XlTlBXLeeUkIYV9CzTVpNfm8Fr3FSF/E6ui2mXgtGl3McXlBgUAn\n2NTz0pBlkVVZPrlpzDU2srpQnnNO3Y1Vzs+Vr+uh6alhDR55JNkMxMXYsWH5mtrTV1lx3QfOEw3d\nlylTwiK/Ft0PPvsMWGmlsHNIyGuQNXnRELqGEr+VMWTLQ9j4avJHH+1Xn7IXgXz7bf1znvcjS96v\nvJJMlsfKT+XSS4Gbb86ej++oS0y2Z9XkY3nXPPpoYmv35S9/Mccz0uGryZva77HHwsyaEyc2BjcL\nGWWb6meiZYR8mgfGV8iLYX5sTT4kL3Hsm28SM08egcLUjvSvf6U7LzZXXgn88Y/17+rDJE+gx3op\nxjbXbLKJfeL2m2/8Vk+b2vr444ETTkhXN5/8s2K6L7GeHzn6Yx6kHXWL65s4MVv5LmeJU06pzxOa\n6mCiZYT800+Hn+Mr5E85pTGNoIjVt2qZt90GbLedu9ONGWPu+LYHef78ZALUhZxH3pr8sccCP/uZ\nWVCoXlIxkctadtm4ect97bzzkp2p1DJDiCmg583zf8kD7lXOWYV82WYz3WIo9Zj6GUhiFM2end/L\nc5llEgeSlyxhHNtGyMtDdl90N8emIauNtfbaxZlrBDbTkcxGGwE//an+mM1cc++9wPrrh9XLV8jP\nnx/HDdEm5GNr8lnPu/vuRMt2oZsLCkW+D/fck+6ZEIwaBWy6qX/6adOSv7LiJOO6L++9p09v+u7i\n6acTu3ZWRLk+plpTHadPrz9z776bvU4qWV2UW0bIZ8VHyG+3XfOQuGghL/ARljNnmo/NmKHXgF0a\nme4F4RLy4pxrr00iVLo07yefBF5+ufl3k0ePTshnJdbL4ne/S+zlRSC3yYgRSSjmtHncd589ndw+\np59ef6GYXHtdmvzqq/uV53t8m23qI/AY+JhSbenEM7LmmsmktSpnZs60rwq3jQRscwG2On1XN/vh\n6hArVolNyL/0EvC3v+nPC0HY0bIIkH/+053GJEynTweWWAI46KDG3xnz9zqRUYW8yVVR7E0qh9jV\n5bf99vXFZj51yEPIp81P1fh8+6UueJcOW37qfeC8vhOaTz1mzEhCRgONwcFcnH9+/bOpz6keasKD\nJ8b92mUX4E9/av49pokkq5CX67Lhhs1rYk46yb4q/KuvzNfTY4S8iuxvfeWVbu3RR8jH4qab/Ori\n+s2F6Zpnz040r3feafw91mrKvIKVmbRBWbDaNPCQ60uryW+zTVh6QQwhrx6bPh1Yckn/OqQxJajt\n4zLXAMlCpBBffF05Mg89BDzxRPPvMeeKfPrBppuaR0BqXdRnU1WM1Htpm1gOHeU01c1+uBo8+2yy\n+YCMvHLy2GMbY60IQm3y6jm676GE3qCQ8kw2WfEgZhWEgjwmXm0vurIn4ULw3VNY14ahsdQZS9w0\nhf93yIT0/PnAgAF+aWfPNt8DU5nyvRs2rPn3PIipyfuGNXjEEE5Rvb/qd1ddbc9YjxDyQ4cmsVxC\nkS/etOI1JA9fTOFKDzss2RMyVsc3CXkxUaOWYzPXCFNLmhWvoelc2IansjBJY5PW5ZOW119vHi0B\nidLhg2pOk5k6tdnvvqMjmeQN8f8WrLqqfl2EjgUXBO64I/kcqskX6V1TlLlGxjSnpdYltG49XsgD\nbuHs6/pUhCZvEvKzZwPXXZc9f4H6wImgTELbCsn7N78xH/OdeA0hrclKTrPttuHlqvlwDhx+uNvm\n+dFHzb9/+aU+/ZVXpq+XYLnlgD33bPyto6NxQj7k/n7wQVgkzH//W/+7jyafhjTnxRxh2rxr1GdY\nh9p/YmvyWV5oLSPkdbiWOetsuaFCPgumbQhNEzkHHhg2YlE7phAKYoInRJO3dbKiV7z6ToKlTfPj\nHycuqIK//tWe/rHHgBVXrH8fNy6xh6uT9DGZNat5gQ1jjStIQ/tqjL4dS8j7uG/Ont046atSJU1e\nfUZCV/qK8z/5pPlYj9HkdbgEYhpN3pZHHsgPx913x8lzq60a8xaETOr5HnNhaj+XTV4u8/nnmyd7\nQ18E8n1/5pl6rH4fLUl1Z11vPeCGG5J9QWNji/7JWKMwCTWphfRlU9oizTVjxiTumyY6OpK9hj/7\nLHn5yKumbeVMn95crvjrsgj4avKu7+r1ivv629/ay9fR1kLedXE6Tf700+0BmvI014hjIYLPtxxd\nPmp+L72ULkKlbxTKmDZ5malTgR/9SH/Ml9VWS/4uvXSzi2eakUrotWbpRyLGiU2TjzHC8cHVd3Xz\nYDHLl+noAIYMAY47Dnj11WTVtIsHHgAWWaT5d9+RiEnIuyZeVVRTnLjXpvPazoXSd9XkG2/YF/fo\nHoL77gOuv96/Llk6JufN53/ySWMMCtklK/ZDoOZn20hBdK40ttVY9TZp8kDzPEOoJi88UrbeulHI\nA27trewNOYTbZkdH8dvp6ZSem24C1lmn8XfdhP28eXahbyvH9JuMuC/z5/u3i1jMpQZ7862byXbv\nssGr3+W9LoD6+aa+2DJCnjF2E2NsCmNsnC3d4ov75TdxInDxxebjOk1e/aySt6AFklWSAvmFYyr7\nk0/qXhwTJwK33mpPn0ZQp+1EjDXuB+uLLc9PP21+mIQnUQyN1UeTv/XWRNlwbX3oS5oXhVi2L15w\nqpDXvYiuvDIxb2XF1rcefTTZg+Hkk+tl6frc3LnNQj+0PBvyffNtX3HvDz9cX74rH5OQd028unAJ\neRuVEvIAbgYwPGaGqpuZTOhwFqi7EgrSeErYzDXqcRlTHe++G1hrreTz2WcDBx/sVw/dg5dm4tVU\nLxFWQaxRUOOThCLK2Wmn5odJ9f3P8jKWhfznn+vb5MMPk3jfwyP11jT1FfFiZAEg3yedwDn22MRN\n96uvspkeTe0sl3nRRfXnwzUPAsSJ3yMj7luIyU3ce3XNgGuOSKB7UcUQ8j71SEuhQp5z/gwAg+NZ\nfNJo8rHJsyzfEAV77JE9Lx3nnddYjvCt9sHVmdXjPgt/5Ifehizkbb7q06aZXQlDSbPWQPRfWcjL\nmrzJxMBYMhpWV2fG6IsPPND43WZGK1KT9zXXiHuv2uV9Qw2HmGsmTMi2cZCcf5bjlbTJy7zySvpz\nXYJdxPEI5aGH6h4sAlfgL1e5Ovu9jhAvigkTkompJ590n6/a5HX5ucrTkbaN1YfDx/ffV8uX22DW\nrHQCxhUb3mST3Xff9BPsHR1uTR4ARo9O/qp23xBMu6XdcIM+eJxJ65WX64dq8iH30Vdz7tdPn973\nBRSiya+xRt091/TcCTdZm4LSUjZ5P7oAdKGrqwvd3d1WP1mgfoG6h05+CHRavSlGs6vRHnmkbouc\nNQv4/e+BhRaqH3eZa0yEPvw+6V980f1imDu3uZP5jHzEkNdWD9f9k7GVGctco3tgbCY1UzmXXBJW\nrgha5/LJ1xGqyQuyaO6yYmDLd84c4Jpr6qtw5WOq6dMkIOW/O+wAnHtuYxqTAJcVE18hL9rv44+B\n/v3r5fq+gIQJTcZmrnHN6QivL1u/1v3W3d2Nrq4ubLVVF447rstahuc2u0XSlfyf/MEVV/idteuu\nzb+5BFVaTwWhDZx6ahJz3rZjT2whnzWNTuD37t08EeVD//7J3zyCvplswQ8/bD5n5MgkzrorwqKP\nkJePpTFl2VxlQxECQ514NYX9jYnrGj77rHGiV06vBuXSCVLRvuK8xx9PruuMM9x1S2OukZHn80Qf\n1sUT8hnJmkZuvu7FoowZM5r7jq78zs5OdHZ24uyzxS9nNyeqUUFNvpEspgJxbPZs/QRtWiEvIgre\nfnu+2pSLNHZe0znqIhef0UivmooQw+6olqMrc+rUZLWq6fgDDySbYYQI7ZjCOAa6+yOE2ZtvNtZV\nDkVRpJunHBzw1Vfr3l6A3bTho8kD9f609dbNx8aPr3+Wr9ln0+7Jk4H//V/9sSzzBb6mVhviWnQB\n79IqGoKiXShHAngewBqMsY8YY4flWZ7oLDvumGjcAnFDTEO8116z5yuE/Pz59qG+XFaRqGXKK/xM\naU1/AfN2gTaTxsCB+hjgpjqq6F4ciy1mP8eH+fP9XnqCtA9YrBcf4NefqvKicgn52bMbPdZ8vHJk\nXnih/lm0y8yZwM47m/MTaW17NNheQC58TCy+mnyaxYouCjXXcM4PCD/Hftx28X/7W3J83Dj9UNEk\n5F3hY/v0Sf7OnWsOPVpV1PYSwshmk3StJNTdoylTkvABJqZNS0ZX8lxGiAeU7bgtEJevJi/SFGES\ncSH3U13USxMxBH+WeSK1T82dmwhpOUqnLv/58xv3V5bTyCNO0S6ygLa9XG07qdn6f5q+mLbtffzk\nP/gAWHll/zwrb65xYWvMu+4C9t/ffE7WwFuTJzcOXXVk0QbSpgvpYOIBUTt5iKeP8OYI5YgjzGVm\nWRJvi3fiO/FaFLZRlkCur2lOIi9zTeiIxKXJqxORJnONPPKWkRUO3Ugyrdkly7yS6UUFhIf88Hlh\nrLJK2L6vlRfyMR7ANAuAYpVXlWG0QKddAek2hc4qWFTvixBN3kYvy/hUZ64p8x69804Sm4Yx/wiH\nJg49NLx8l/eTaatHH3R9TX2p2WzyLnRuv2mFfFZNXk0Taq5x5a8S4o7a8kI+7QM6f378FXjCI0im\nDAFi28V+330bv4tJs/vvb/zdp1P6Bi4zoT7Mcnx210Nja1fbhLqqyX/wQbzQBWkZOzb5q/qgC3yF\n/C23NH736Xu2UU8a5DJVxWHevGaTiU7IqyGWTQjPkixCXqwe1503zhp8pTFfU/8MFe669Do51RZC\nfsyY/KIacg6ccw7wi1/EyU9Qd2dyp1UZNcqd5uuv/baMu+oqvzKB5nDNuocuL1QhL2vg6rEQs0GI\nkDdRRlAy0zX6CHk16FpZyP1GXaU8d27cyJS69kprdsniXfP8881pTBu+yxPHujJ0z98WWzTP/YWM\nvCsr5MWmDnlp8rZwwzbGj9cH9pfxXV4fim7orEbTi41uA2VBVkFoE+Suh8bWtiHmmjIw1f1Xv0qC\nf6nYrkfgu9esjtGjgaWWSlZHn3VW+nwAu5BXo1LK6bM8K/L91AnrF190l5HFJv+Tn5j752GHNUZ/\n3XJLfR433JD8Na00Vpkzx08xBCos5AWui7W5RQl0E21pN+hYZ53GKJIu8tSI58xJt4jJhlpfk18x\nUKyQ95mgFMTQ5IvimGPqG5gA+gd38GB3PuoqUcDfLvzQQ0n4gRdfBB580F2WDXkVeYiQz4LLnj9k\niPmYIItNXpdGLkvVwm0v5IEDm8/X8e23SRRQHyq44jWMp58OP6dIO3nMskwPbUzB5ZuPulNTGmza\neRY/89hCPs+Xwssv15e2mxDrMrLSq5d+mH/ZZfXPWa91n33qn9UFdmrAMqD+3ScAnYkYE/ZpHA9M\ndQDsAtgWmnvQoOTvjBn2kOtz5vi3Wctr8lVj3XXrn6dOzVfIi+95hBW44gpzbB8gCX4WS5Pv3Tv5\nK19HFjOdrV6+5ho5TdYQyi6WWsp87Pzz/ba188HH7BPT4yxEk58zx38F+hZb6PNQP6uEPIvXXBN2\n3gHKCiDZz3/GDP9AbeIZcHk1ffutv5BveU3ehyKH5yK2OgDst1+YaSct48cDl14aJy8heGfP1nsL\nCXr39mvX7bc32/VFWX36JHmZAsqFYnso//MfP+8NeZn8Kaekr4sPIhaSjpjeL717u6OlikBqMXj3\n3cbvOk3+yy+BhRdOPvsqK6ryYRPy8i5zIS6mL7/sn9aF6ixhu06fxYlAMkoy7berQpp8Dsh1Vn3B\n8yhnvfXymYC1CfFevfyEvE0zlAW5T1AmGdtxm13Z1z3vuuv80sXAdC0uM45PvvfdV79PYsRk4/33\ns5Upo7ahTrittFL2cmz95swz0+Up740Qe9WzTciLY2JfXxMTJ5K5pvA888jfNbyeMCFOOTJy3W1C\n3Edg3HijvS1kT4KOjjBNPq37q01rLgvTtWbtRx9/3BgyOJZtPy1z5zZr90DigZIXaUeE8ohn/fXj\n1EUg7/GsIuq73Xb2PKZN89fke4S5pkyyrtyUh21FjGp8hbyvJu8r5L/+Wh9EzkRWL5AqYTLpZQ1y\nduGFjd/zmLsJYd48/XaatkB2Ptg0+SJMtaFldHebj/neo3nz/NP2CE0+iw9xGmItz1cnoooQ8vKm\nCDGEvE1Q2Y7Z9u7NQszokHkT+3671nfkTV4vGZvrbV6hS/IiyzaEJlqsCeLQLuaaPJBHDrrwCGJD\ncd+6+WryKj7xwdNQBSHv2z9acT7KRuwwIgKbUnX99fmUmRchmrwvlRfyrd7RL788/bmqIM1L8JkQ\nKwVlxIPa0WHfpUmQVsjnRV6CJg9ave+r5PWCtQn5kL2XqwAJ+QrnKXPiiXHyUc01Q4fGyTcLYtHI\n9OnuMMuA31LydhNmLnwf0CqMOmKS1322mWtajauv9kv36qv+eVZeyPdkyjDXuBALd2KYHEJiYrcT\nvtfd6gJLJa/refbZ/MuoGmrEURuVF/Kum2Zb+ps2z6pQpRgrgtC2s+0ONXlyEhCrVe5HLHxNCO3W\nLkVcjxpKm2gDF8pWmz0PYdKksmvQTGyb9vnnt5adPAY9VZMvwvxkC8WRF1W/T0Vv5D2cMfY2Y2wC\nY8wrhpqrAX3jXRBxEJt6h3TsX/7SfKxdBPzmm/un9RXyZJNvDap+XYUJecbYAgD+AGA4gHUAHMAY\nM+zkWMfVgGlMGlW/KVnIc/WgTEgbykHb2hU5+qKLnqrJt9v1CKp+XUVq8psBeI9z/gHn/FsAdwDY\nQ0106KFhJpg+fWJVrz0YMKCYckK0zCrOLZRJT7XJZwknXGWqfp+KFPLLA/hI+v5x7bcGGAvbCNsn\nhopK1W9KFtK0RxpC2rCd500AYPfdw15kPVWTP+KIsmuQD1W/T0VOvHo1xZgxXZKW2AnOO63pSZNv\npKg5iqBl1W0u5NdZJ0zI+2rytg3ZW5F2m2MQlHNd3bV/booU8pMADJK+D0KizTew0UZdePPN+qIb\nOZaKjqI011ahKN96MtfUueKKZGN4X3rq+oB2pRxNvrP2T3C2MWWROta/AKzOGFuZMdYHwH4A7tNW\nSqqVK6hSGs216sOrLBQh5NWwwD7p25nQYGqtttQ+C8s3GWTbj6rLk8IeP875XAA/B/AogLcA3Mk5\nH6+mU23yeUTOa9dhI5CPkFcnc9VdnFy0u5AHwkYr7eI26oMutHC7sddeZdfATqGPH+f8Yc75mpzz\n1TjnF5jShTwwaUwBrSLkbRv+mshDoKp5dnTQxGsWyo7rXiQLLVR2DfJn0CB3mjKp3OOnavJ50CpC\nfoklws//4UfGAAAgAElEQVQpQsiHavLtbpMPvb6eIuQHD87uCNAKfafq8qRyQh7Q31jTisJ2XgyV\nl319becStEZIk3cT0g97ipBX22S99cLzaIW+U/X7WbkmNGnyYkd3XfpQqv7mFaRxD/URvttvH5an\nTsiTTb4REvLNqM9ymueuFdoqr13MYlHJx4/MNQl5CflQSJOPyyuvlF2DYlD7SBWeu5VXjp/nTTel\nP3fppePVw0QlHz+dUIgp+EM722WXhZdhE2zHHeeXR5WFfNDONJXsZfEowm7cim34zTeN36sg5NdY\nw51mzz3zr4egkL6TfxFhmMw1psZIoxWFDgH33z+8jEMOCT9HJY2Q93mQQjuWOnnGWJgbYCtMni24\nYNk1CKNfvzj5XHppnHx0zJ7deO9bwfQCAMstV1xZRby8Kyfkv/oqTMi7uOOO5t9Ctd00Zce4eX37\nhp/jc22h16OuKp46Fdh7b//zW0ELXWml9Odynv+LTL2v8ktphRXS57vxxunPdaGu7K2CJu9DkW6f\nPVKTv+uuuBeu04ZDO1vRQn748ORvXuaarEI+lFbQ5LMIyjxMZAceaC9DHl1lCeWcp6fZYos1fich\n30yPFPIA8MUXzb+lbQydn24Rnc3mH+y6FqFdVcUmnzUIXCto8iNHAltvne7cPB7UffcF9tvPfFzu\nX1nMIHkKeXXrx7yeuxBTm8/1VkXI+8wf+NACj19C2gdJJ2BCXZ7KMtek8ZMXgd1sFK3Jt8K6hAED\n7PvR2sjLXKNTdgRy/8oi5GXBe9JJ6fPRsfTSjfc+r36w227+aX0WZ/Xv759fVvu9TU7Eaq+WEfJp\n0TViaAjXooW8uLlp8sgjLkrWRVlffhmnHnmSVUjnIeRHj/YrL0TIP/lk43dZyOc94spr4jWk3j73\nKWRk8IMf+KfVYav7hAnZ8v6ujDjZ5I/u5lxzjfs8XSN+/nn2sl3EiOueJg8fTT6UrOaaLbZwp8nT\ny6NPH2DLLe1pXPd4qaXi1ceXFVc0H5P7dYgZRM0zRFvcZRf/tAJ1MZSv+3AIIULeJ22IkK+icqDS\n0kL+6KPd58UQ8mkoS8j7aPKiLQ891C/PLOaazTbzGwmEDLlDWWMN98PkOn62OVx3Idg8rUIW+Kh9\nSn5BuNrAdI/GN8WSrSO/RObPz2e0ECIofcoPGblmvZ6Q82+5JWUZ6U4rnrRvPN15oXtNlmWTz8tc\nI67Hdz/YLELe13xUtAviaqs12qBt5ffrB3R2pi97k02ARx8NP0+u89VXAyNG1L9/LG23c/31wE9/\n6penLOQXXDDOZOjqq/ulmz8/230+4QT977GFvJrGtio163O+6qr+adO6u7a1kFc3BX/4YX26r76K\nX3bVNfmDDgIOOMAvzyxCvggf8jTsu2+j26StjvPmue+F7fxllwV23DGsfkCjkF9kEWDNNfXp+vYF\nFl3UnteFFyZ/xfMwfDgwfXpjGT6jnZ120v/uQ1ZN3uRtlLeQP+UUc9qsfVs3ajAJ/tTOJ+lOqw6m\ntzsA3Hxz4w3bZBN9OtcDUpYmn6dN/s9/dtupAeCSS4Bhw8LrEUpeL4Kf/zy5Bp3t2XdYPm9e8kJY\nf/10dUjbF2Qtm/Ns3hbCZ13UZYEFkn+hmvyttzb/Jt87W5+dNy/bfTadG3v/CfUabPfvL3/xL1uH\n7p7GfhZaRsirF64bIuqC95flox1Dk8/LXLPHHv75cW7XZHzO9+m0PmmWXTa8/GOOSbRP9WHivFHI\n28rnPHGrGzvWnEY9X1YoYvRBl4B3HRf1E/1SfA+xyZvKkc/78EPzuVnNNaZ2jD3xqgvjYWLqVP+y\ndRThXtyyQl6HbpgjnxfTru/C1JnUVYAq8gKINOX6bDQydKh/fr6a3muv+ed5113Nv/lca8z4+pw3\nmqFcQl7H3XebzznyyPrntEJe9TGPoQULASbqJO6vKqx0JjrG3G6Qtj1d0wj5229vLF9HbCGvC8iX\nFyGafNuZa3wnBV3IN6gIIf+97zWXKzNmjPncL74A3nnH/nZ3+eWeey5w5pn2NCG4hLzw0LFtCCG3\n3/jx6WMTZXnYhgxpdo2TNbY0fUOeCFXPl7/HEvK+aXWIOoi/on7ivEUWaayzyTmh6NW18j0ytWMM\nm/y11+rLDMk/lpmlZc01jLEfMsbeZIzNY4xt5EqvTlSpFy46jEtTl8OdFiHkBw9O/prMNTYzzuKL\nu/N3PSiLLqqfHMuLtdZyp0k7mlK1wixC/ve/B6ZMafwthknNRNFC3oUq5FVN3gfGsrVZVnON7UXq\ni2lXNNv98i0nTRiCdjPXjAOwF4CnbYnkiSGZtJ1jxoyw9Lvuaj/+4x/bj4t6mh7svATLcsvVhZjo\nODGWqbs6oRznQ2cG0p2/xx7AVVfVv99zj/7+/uEPjd99hKU63yCXr56fp5CXqYImbzLX2M7797+T\ndQ4yaeZFBGkmXn0UBN2LSufTf9ZZwBlnuMtJK+TTCGzdOTFGLA35pTstHM7525zzd33TxxLycpwa\nnzx0bmryeb430meS6PXX659NnU9FV/748cDzzwPLLJN8z+L7/PbbwHPP1c1OLmTf3e5ufRr1Qe3d\nG9hmm/pve+2lvzfqtfrMJaj5mCYWOc/X1hpbk8+KqI/aPrIJS/WcWWWV7CYtdTGULQ/TPICpfN3k\nsUDnSdevX/jE67Rp9nPy2NmpZc01vpjitqQVXF9/Xf+c1u7LmN/qWrkMH01emHYA4De/8ctf9+Cv\ntVZjPHTRVmmExEILJa6V4lxXHnLaLK5sPkL+ppsS/36VI46ou8H6Cnkg3SScjrXWsl97Fbxr1BGy\nSL/zzvUVq6o5KzZz59aVLtWk+Oijfi69OnSyIYtrojqPZztPDq0SS5M3UQlNnjE2mjE2TvPPe9E6\n510AuvDGG10AuoPKN5kG6vVz52HSJnzi5Kjn6DCZCNIMCVdZxZ1GIE8S2lC9c7J2QtUrJMRzQC77\n6KPND9y++zbbmnV5pPGa8LkvLm+hPM01vvFkRB1EeAT5xSzmVXT1jB0L6f33k79XXNH4u+ne+rS/\nTshn2TNCney13b88dhRzXXOyEVI3gC5svnkXgC5r+qhCnnO+A+d8sObf/f65dAHowoYbdgHoBJAM\nI33Cf+oaZ8klm3976CFzHlmDcblQO4ywafuahORjv/61Pk1aTd63nWMQ4pM9bJh90ZstX5u5JpZN\nvk8f+/Wk0cDUkAOme/nAA/bjah1UrxpdGhnZyyaGGcEUAVYnSB94INwmv846yd8su7+FaPILL9z8\nmzCburjzzsZ6iheGq52TMjsBdGHLLbtQqJAPwNld5AdwtdWaj4fexB12qDeeLQa0a8s9xoCLL7Yf\ntxGymk6Hz+KVtLbcNNsN+uCjoenSiHmBp56yx0eRRwvir1ila2qvRRe1+93fdFPyN622n0WTP/hg\nYPvt699vucU8b+GLyTtNRlfP0DhPLkyL9XTCdIMN/PKU52qE44ROk09rrrHdP52QD3G5lO+DLeqo\nnK/s7OBlgvarTnYYY3sxxj4CsAWABxljhkgyCb4Phm+D7ruvefJJxjX5AwAnnmg+X9w003DRNKHs\nug4RgW7IEHs6uWz1QXYFktKNYrJO/h12WDYhr1upakP0GzGJaIqXvvvu9h2AxAS8T//SCShfIT9n\nDnC/Ms7dbbdGAXDwwcm6Ed8Rng++QjDEBVkOmhZKR0e4uUYc07lFbrFF2KjUZK5hzL5CPEtcJ1XI\ni3DbLtnnu4hPUKR3zd8454M45wtyzgdyzne2VsxRs9BOnUVYZRV0G0mrAtK4R/XtW+94/fvXbcAh\nLmWm3+XoiLqHIuu1/+pX+t9j+DwLdEG2dGYJ8dtrryX7oha1zZutL/fu3WxX32ef5O+IEdk3pQhB\n1FO2l8tC3gbn9tWugtA5mbQ2+V139XOfFi6iNnPNkks2m30XWSTxQpPTZn1WfKN5+objEFTOu0ag\nNp64mE03TZ+n3CA33pg+Hx/EDT/hBOAf/6j/nlaTDxGCJs8YXSeUF53JJgJTh/XZAETwP/+T/PXV\nhtOim6AUdlF5habq+eQj5GPUXe7LurAWpvOvvx548MGwuvhis8n/4hf13+RNwmOWr/N00pmUdDZ5\n+XkCGu9xqKAVWrFNyOvgPPEGco0+XLvQyfV1he0Q+coyxGeCubJCXhWGojFefjnxKtl88/A8ZYF6\n+OF6G5jLVunq6Kq55pJLzENBtU4m1AdAlGFamGJyKbM9AOuu6762vn2BF15IPu+4I/D3v+vTHXVU\n8tfkNw/Uy5o0yZwmjWa0zDLJhJ2YX9DZ5MXf73/fnV9aIW8SGr/4BTBzpjvPUFxtJa9nuOsu+7yS\njGpKioVa35AXSGdnMkexzz7J4kT52mztoCtDN8oKWRsgzj/ooCSyq8oii9jroxPyap1+/vPG7/LL\nwMf7KWLIp+J4912zP7uNNJrIrbc2Tkhu5AjIoLPJ2+yzoUJE7hi77qrfP9Uk/EMeAJeffP/+iV1b\nh7hGnZaklifqGtNcI5s/bNfcv3/iaWUziYRMoplQlYQ83O5sXHZZ4/aFP/yhPp3uGuR5mjT3KMvi\nwYUXNnvX3HNPeF106JQsH48e9fiIEfr5MpOp7uGHk5GzbBrzDcDXNkLetvmzfJGyu5LoUIstlnjT\nyOhssz7IAt6nw+qEY8xhrgxj+uH/hhvq65B1sUboA2uzRasPV15tlMZf2nXvFlkkEZxyGjWdPEoo\nIty1694stph7+7giw3Lr5mTU31wRW/Ooh/qbOrfT3d24S5iur/tMug8f3pgvUB9BmCK6inxl+ecT\nWryS5ppBg+qCysWvftU85P/oo+YFKr6TOCppJ1PkKJqh5bqGsmmFtU3gxRSypoiHan1s5Z96KoyL\nvXR5hUxC60xfJnT5rrlm4jUk0E1Yr78+MHFi8jltLJS0TJ/ebEpjLPHSsZF3vQDzCNEkDEP6+qhR\nccw1NvOs+uJR53l8ypQJsckLZHNSywr5//yn0U1R95YX9OrV7PfuO4PvI3zShFN47jn/l5QP8vWH\naOQh16K2xVlnNefh2xaqcA8RqoLzzvNfmGbL0ye+vg7bCEMub+ZMfZwf+aEvWpPv6Eg3Z+UixuIx\n072yPeN5EmqukcOHmM4PQZ40FkLe1e9DzTWVFPJZcblj2W5IqF+27nw5nourPB/UByDt6MJ03nnn\nAWef3fibmOyRz5G1Bh8bdIhw83kBq/Vx5TNzZmN8oDzQrVgW311aXl6kFZghXkJpMcU5kp0LZI+e\nrG6JLnSjTdt1qpq8y1wDNHorqcjPlGniVc1XFvI+i9Uqa5NPQ+hb1Zbu1FOBCy7Iv5O52Hvvxkmz\nLJiu5bTT3OcutFBjRE8bPuYalTy0uBh+8Gq9/vY3YOWVw/KIJeTluiyxBPC739W/q0pFVYW8aaFe\nWX3kqqsS85vNA86nTjYhP3Bg43fZLVYn5F0ypy1s8lnJqsn//e/A8ccn39MOUWPsSAUkdkZ515qQ\nvLKOSuRzVI8QWz3SDGFjjHbSpA21ye+5p99y+zw0ebmuffvWd+UCku0GxRqGKgh50ySvqb3zaCOV\n7bZr/D5yZH3L0BCPGhkfTd6Gaq7p6HD3r7awyev48Y/Nrl8CcYN1Ni1dxzfdjN13r2vPodsQijoM\nHdq8aCMGab1kgGyjkhC3vzQTr1lXJOc14ko7KjQN/135iSiNoeWtu25idhPHsmzTZyKNIDbdZ5dX\ni+58n53TbKjuz8OHu23yunrImF7kqvlTdw7QvFhv3jxzyGWduUY2bZloGXPN8OF1tyMT4q1m8qH3\neWDVm+kb/6JPn8Q+Js7v6Ki7WqnxLUJm3HXnpJ14zSIIzzwTeO89fX1UQsw0AlG3115LPFNCyHJd\nrjqmHWH42nhVBg3yy9PmKdLRka5NirTJu/q4mh7QxzLKglyeHAQx5DpVhWbvvZO8bJ4ypnkugcvB\nQeQ9ZgzwySeJWdlaR/vh1sI2CZF2OGaKxzF2bKOW/9Zb5jwWWMAcjz6NEEnb0dN4ComyjjwSuPBC\nvzqo2lHIy9XlNunSqEJtu1lWXtrOS2uuCXEKMJ3LWGIHll/KWcoWZqE8Nz9J6+Kssuuu9QikLuR7\ntN56wIQJyWef61RHJOKcUaOS50RW7Gz3TQh5WZaI9KaRizDX+LZXWwl5WzClEHON4OuvzUGD1l8/\n2U5MIDT+vCdqy7LJpyHNQ5smql+ebZ7mGp57LrGZi3N1W9HFLE89V/wV9uasZYtVmXl617jS+7Lt\ntkloah9MddDtQRFqrrFp8nK5Nk3+ttv054X2kbYS8i5NPrSRZCGuQ7dazcQRR+iDFYmIgy7SmGtU\nu+x222UL8GaqT8gxE+Ka+vULX9CSB1k8tVRfat2m0j75hByTj8duI9uIJNRO7gqXUSTq6F5n83Zh\nMk3KeYhAfTp0cydCyJv85UVZvqOfygl53SIi3w4RusFByASLjhAhv8ACeg3BNkHjU64NtQPddVcS\n4K0KpB1liHSm4XBWc40trS9FvIhUtzygfr9jmZxUdEL+iy/Cygh9gef5MpB989MogXJamya/557m\nfIUmn6ZPei9O9EtWHP/8Z/pzXULedvNcnjsu1A2SY5OmA/q4V7lwDVNjsdhizZs721Bj6vi2z4kn\nhvm5FzVyCEW3r2xWjxrTtcZ0BTXZm222eqC+oUZMTJp8CKaFf7rRgG71tdiBTMYkvNX70LKafJal\n07ZOrntTyr+JhyatJp/2AdAt2HFpO2k1+TTEeGn5aCm9ewOPPJK9LBcXXxy2zWGaB9+0CtaWX2i/\n14WwzXq/daMDmViLoTgHll7anEbXR0ybz8iE3ivbS23kyLA8fIS8EN5y8LqdNVsniXSul67vs1k5\nIZ8FXyFvI6u5JlQoLrpo2DkHHVSskDcR2yYfyrXX1j2W8nKrA4BDDvEb5e21V+LKtu669aXvIe0Q\nIy5M1pFbkZp8SPkx8naVp5prfIPjmTR5nQOBeMGJrSVN+JprOAe22gr47W8ddbQfLhdXaFSV0CiL\nWTvV0KH11WkhQj6L8PUNcTBggDv2vQ9VmCDTcfjhwNFHJ5/lB3T//eOWc/HFetOIyqKLAqecAowb\nl66cKgj5UA+SNHmGuN660mdFlDNlStL+via/Rx+thyYwtc022zRPuPva0F3mGlnIDxgA/OY39vwK\nE/KMsYsZY+MZY68xxu5hjC3qOscVGlXFV8jH6jj33gv861/J55CHNI2/emj9p0+vb2Kw9NLJGz8N\nadrKxzsh1qrdRx9NNgkRv8fyHopBiBIR4tFhyjvPkRtQF2RiX9Q0qPvZClw2eR982/v11xvTiz0p\nZCFqK3fHHeu++CZNftAg4L77Gn/zHcW45EOVzTWPAfgfzvn6AN4FcKrviWooYRNFa/IdHc0LE/LS\n5NPUVZTz6af1WDxFIK8ezJsdd0znW18ESy3VvHmNCZ+djlTzgkqMiXZbuUKQPfec/z1W62nadjDW\nYigdW28NvP128+8u85SMbS4J8Bvl+Cp3IeYaHwoLa8A5Hy19fQnACL/zQsoIq1OsPN5+O39NXsa3\njlnLScvqq9dXDwLV9VAB8vUe6tsXeOyx+vfjjgMOOECfdu21w/LWhdvIqsn7mmt69fIfeWRdgOd7\n/sCByapVHYw12sF9zDK+5Qoh7/OshZprXC8h37haZcWuORyA5/y1P6ETr7GEj2siRSWvhzF2Obay\nXBOvLk3PNyaQL1dcAfz3v3HzjM3vf1//fPPNjbtLhWC6JzpNfqed/EcTLmJFVjWRJc9PPjEf8+3D\ntvJNoyRxjm21vcDX0mBKJ1/H5Mn6zWp0RBXyjLHRAHSOWKdxzu+vpTkdwBzO+e26PLq6ur773NnZ\niU55Q0UHoS6UOrKOBnzOL0rDtgW7youddvJ76a2wgn0f31BiT7jmjTp3EGM/0733bh5Rhrilhky8\nhgrkZZaxn7PwwmH5ZcGkydueXZcQnz3bXa7vc7/CCu4048d3449/7PbKL6qQ55xbdQbG2KEAfgBg\ne1MaWciHUrRNXuWOO8yxbmSy2uR9X0SrrprfS8vUdiFCJY1gkzduj0GsAGWhqCEz+vfPfq8WWshs\nDopBFhfK5ZZLoibq+OCDJBREXite04xG1XNc8x2ul8BGG/nPHZ1zTrJoT0wQ6+qkKsBnW5bOF2au\nYYwNB3ASgG045x7vvXDKsskL9tvPL11RNvkYVM2F8oILGvf/bVVWXRX497/LrkUjeWryNsTCqKLD\nGoRiEtJnnOHeU/eFF8LK0blKp22LIm3yVwLoA2A0S+7mC5zzn8UsINSFsqwJwaI0+RissUZxZfnQ\nr585/HMaypwU9l1wUzY6If/b3wLjx8fJP8YagRB8nh9dGtNK6XPOcefnuyl9HhTpXeNhyMhGWd4k\nobjqWRXtZc6ccP9tHUVvZB1CVdq6CrjaQr6Pe+8dlteRR/rlG1onH2KF2A4Jh5EHaduiwo9fOKE2\n+aFDG+NIAMU89HkvWolF795xNN0BA8KGq4SeskYdolxX6G0bP7OM2UM3wchKWk0+T208zznDltn+\nz4dQIS9vkA0ktt4f/ShunXQUseI1b0I7oNhkulWoom9/3vfdZZPPIuRNDBxY7ZGeTKtq8m0l5G34\nCEnTarzYFOUnTxAxSSPks4YlIHNNHTLXOKiSJtxK3jUm0mq6sRdBZaUKbelL3qOLMjR5XTkyedyf\ntHFyypw8zUKP1OTLpqjYNXmSxl/9q6+SaI1VpmrtLFPWC0m0SVGabOzrDImEaaNsIU/mGgdVenjP\nOisJrGVCdzOrNBKZNEm/laGLqgv4no6rX+Ut5MvW5F1pyp54JSFv4aabgB/8oOxa1Nl44+RfWsoW\n8r5RQYm4+Kymzou8+lwRK9F9TYRyubpdssq2yaelbWzyL74I7LOP/thhhzXe6LKFJFEdhgzJvr9v\nURx4oF+MlLSU8Vy4ysxap3feScKN+OQp/77mmo1zZx0dwCabZKtLVnq8Jr/55sWvnCsLeknFY5ll\n/HZ+qgKMta426UMeJtUsK7bl+nz9dZyFgT5lmejxQh5onRWvWdltt/TbzBF2xMO24ILl1qMMylYe\nRCTKNPHk8yZPe/zxx9s3Ns9KWwn5449PNlJ2UZWOE4L8pl95ZeC660qrSluz1VbAjBnA4MH63YSI\n/LjsMuCkk4A998y3HNPzX9airN/9Tv+7OpdAmjySVZWttrKSqBbyTk6hm8G0OmV5sggWXhhYa63y\nwn4MGgQ8+2w5ZevYdNNkh7XLL0++02KoNofcD4shZGu7dmObbaoRdVSO3V70iyfthvd5sdpqyabh\nvrtA6eiRQr7q5hpd/YYOBT76qPi6ED2Hq69OvFFikmUDeqD6z2oR/PCHyXZ/pMn3AHy2BSOIKpHG\nRdK1C1NWWvXFQUKeIIi2gDR5PWlWmQMk5AmCqBiykN9sMxrBCpZbLt1Lr0cKedIOCKK6yOaawYPj\nz0X1tOe/MCHPGDuHMfYaY2wsY+wJxtigospW6Wk3mSCKRrcRtS952+R7GkVq8hdxztfnnG8A4F4A\nZxVYdktBLyGilfn002wBAWniNS5FbuQ9Xfo6AMBnRZVNEERxyEv00wjUvBdDkZDPEcbYeQAOAjAL\nQGlrUxdfvKySCYJwcd99+Qr6nhLIUBBVyDPGRgPQRGLGaZzz+znnpwM4nTF2CoBLARwWs3wf5swB\nevcuulSC6Jm4FkPptOqdd86nLoKNN65W+IK8iSrkOec7eCa9HcBDugNdXV3ffe7s7ERnZ2fmesmQ\ngCeI4qiiaYSx6oUvCKW7uxvd3d1eaRkv6C4wxlbnnE+off4FgM045wcpaXhR9akyxx8PXHppNR8Q\nggjhoYeAXXYxb2m5zDLAlCnF16vdYIyBc64dNxVpk7+AMbYmgHkAJgL4aYFltxQk3ImeQpX2Xm5X\nivSuMWzORxAEQeRFj1zxShAE0VMgIU8QBNHGkJAnCIJoY0jIEwSRG2niyRNxISFfQbJs9UUQBCFD\nQr6CnHgi8MknZdeCIIh2oIduWVxtevUCBuqCQxBEGzFiBLDiimXXov0hIU8QRCncfXfZNegZkLmG\nIAiijSEhTxAE0caQkCcIIjeGDAGOOqrsWvRsCotC6QNFoSQIggjHFoWSNHmCIIg2hoQ8QRBEG0NC\nniAIoo0hIU8QBNHGkJAnCIJoY0jIEwRBtDEk5AmCINqYQoU8Y+wExth8xtgSRZZLEATRUylMyDPG\nBgHYAcCHRZVJ2Onu7i67Cj0Kau9iofZOKFKT/z2AXxdYHuGAHoJiofYuFmrvhEKEPGNsDwAfc85f\nL6I8giAIIiFaPHnG2GgAuq0uTgdwKoAd5eSxyiUIgiDM5B6gjDG2LoAnAMyq/bQCgEkANuOcf6qk\npehkBEEQKTAFKCs8CiVj7H0AG3POvyi0YIIgiB5IGX7ypK0TBEEURKXiyRMEQRBxsWryjLFBjLF/\nMMbeZIy9wRg7tvb7Eoyx0YyxdxljjzHGFpPOOZUxNoEx9jZjbEfp940ZY+Nqxy63lKlNxxgbxhh7\nlTH2LWNshOX8voyxO2vnv8gYW0k69ghj7EvG2P3upimeyO19HmPsP4yx6Y4yTe19KGPsv4yxMbV/\nhxvON94Xxtg86fx707ZLHsRqa8bYgoyxBxlj42v5XGAp09TWKzHGnmCMvVar0/KG84+v1fc1xtjj\njLEVa78zxtgVtWNv2Z6vsojZt6Xj9zHGxlnKNLW3th0159tkyYW1vMcxxvZN2y6FwDk3/kPiLbNB\n7fMAAO8AWBvARQB+Xfv9ZAD/t/Z5HQBjAfQGsDKA91AfLbyMZLIVAB4CMNxQpjYdgJUADAZwC4AR\nljr/DMDVtc/7AbhDOrYdgF0B3G+77rL+RW7vzWr5TXeUaWrvQwBc4VFn431xld0ObQ1gQQDb1NL0\nBvB0ir79VwAH1T5vC+DPhvM7AfSrff6J6Nu135+t1acDwPOiTlX5F6m9O6T89gbwFwCvp+jb2nbU\nnHSjg48AAAZcSURBVK+VJQB2AfBYra0XqpWzcNltbPpn1eQ555M552Nrn2cAGA9geQC7I3moUfu7\nZ+3zHgBGcs6/5Zx/ULsxmzPGlq01wsu1dH+WzvkOWzrO+Yec83EA5tvqrNRtFIDtpet5EsAMx/ml\nEau9a+e/zDmfbCvPcV8YPFxdA+5LpYjV1pzzrznnT9Xy+RbAq7V8GnC09doAnqx97q6VpatzN+d8\ndu3rS0g81QDgUwB9APRF8tLpDcB674smUntvBgCMsQEAjgNwLgx91CFLTO2oYpIlawN4mnM+n3M+\nC8DrAIZ7NEMpeE+8MsZWBrAhkkb5Hud8Su3QFADfq31eDsDH0mkfI7mR6u+ToHkQar/5pLOxPICP\nAIBzPhfAVNaCsXIytrcvtvbmAEYwxl5njP2VMWZ6EGz0Y4y9whh7gSUL4ipJrLaumRp2Q+IyrGJr\n69cACFPXXgAWZowt7qj2/0GinYJz/hYSzfKTWr6PcM7fcZxfGhnae7na53MAXIK6W7YOX1nyXTsa\n8tDJktcADK+Z6pZCMvpK83wUgpeQr705RwH4Jee8wcbLk/ELzd5GJGN7x7oX9wNYiXO+HoDRqGs0\nIazIOd8YwI8AXMYY+36kukUjVlszxnoBGAng8prmGcKJALZhjL0KYBgSgTTPUucDAWwE4OLa92FI\nBM3ytX/bM8a2DqxDIWRsb8YY2wDA9znnf0fGRZVqO/rCOR+N5MXwPIDbAbyACo9knUKeMdYbyU25\nlXMuJs+mMMYG1o4vi2S4CCSdc5B0+gpI3qaT0PimWwHAx4yxDsbY2NrEXFctrZpukqZa8sN1bu38\nV6U6iAmpXgAW5Y0++ZV+IUVob117ibwX8G1vzvkXNfMDANwIYONaHucp7S3T0Lac809qf99HYobY\n0HLphRO5ra8D8A7n/IrauSFt/QnnfATnfCMAZ9R+m6Zra8bY/wI4DcDu0v3ZAsDDnPNZnPOZAB4G\nMCR9y+RDJFmyBYBNWLLe5hkAazDGngyVJbp2DJElnPPzOecbcs53RPKyqezIyTVZwpDYsi5Vfr8I\nwMm1z6egebKkD4BVAExEfSLwJST2Ygb7xKs1HYA/wT3x+sfa5/2hTKogmXSp6sRrtPaWznVNvGrb\nG8BAKc1eAJ535NNwXwAsBqBv7fNSAN4FsFbZbZxT3z4XwN1q2we09ZKoTSoCOA9Al+H8DZHYpldV\nft8dyWhrAST2+McB7FJ2GxfQt1cCMC5Fe2vbUXO+VpYgUY6XrH1eD8A4SJPCVfvnujFbIxmGjAUw\npvZvOIAlah3pXSS2wMWkc06rNeDbAHaSft+41hjvweK1YUoHYFMk9rEZAD4z3Vwkk093AZgA4EUA\nK0vHnkGiKcyq5bVD2Tcgx/a+qHaNc2t/zwxs7/MBvFGryxMA1jCcr70vALZEMiE1tvb3sLLbN4+2\nRqIhzgfwppTP4YFtPaJW3jtIRgS9DeePRmJ3F+XcKx27tHa/3gRwSdntm2fflo6vDLt3jam9je2o\nnK+VJQD61dr5TSQmm/XKbl/bP1oMRRAE0cbQ9n8EQRBtDAl5giCINoaEPEEQRBtDQp4gCKKNISFP\nEATRxpCQJwiCaGNIyBOEBGOsizF2guX4HoyxtYusE0FkgYQ8QTTiWjiyF5LVmATREtBiKKLHwxg7\nHcDBSFZDfwTgFQBTARyFZFn9ewAOQrIc/v7asalIYpp3APgDgKWRrKQ+klc4AiTR8yAhT/RoGGMb\nA7gZSazy3kjiwf8RwJ94LRgVY+wcAFM4539gjN2MJPbRPbVjTwA4mnP+HmNscwDnc86315VFEGXQ\nq+wKEETJDAVwD082kZjNGLsPSUCrwYyxcwEsimQno0ekcxjwXdjcIQD+yth3UW/7FFVxgvCBhDzR\n0+HQxyW/GcAenPNxjLFDkEQvlc8BElPNV5zzSoVQJggZmnglejpPA9iTMdaPMbYwkp2dAGBhAJNr\nMdAPRF2wTwewCJDEfAfwPmNsH+C7DbXXK7T2BOGAbPJEj4cxdhqSjcs/BfAhErv8LAC/BvBfJHHJ\nB3DOD2eMbQngegCzAeyDRPj/EcCySGz6Iznn5xZ+EQRhgIQ8QRBEG0PmGoIgiDaGhDxBEEQbQ0Ke\nIAiijSEhTxAE0caQkCcIgmhjSMgTBEG0MSTkCYIg2hgS8gRBEG3M/wd+sjOEdLAM0gAAAABJRU5E\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110b75dd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.A.compute().plot()"
   ]
  }
 ],
 "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": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}