Stock_Prediction_Model_using_Attention_Multilayer_RNN_LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Titile: Evaluation of Stock Prediction Models using Fully Connected Feed Forward Neural Network and Adaptive Variants of Recursive Neural Networks\n",
    "\n",
    "Author: Soumyendu Sarkar : http://www.linkedin.com/in/soumyendusarkar\n",
    "\n",
    "Source Code: https://github.com/shoumo95\n",
    "\n",
    "Data Source: https://quantquote.com/historical-stock-data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Project Objective\n",
    "Evaluate models to Predict future market behavior of equities from historic data using Machine Learning techniques.\n",
    " \n",
    "Methods :\n",
    "\n",
    "Multiple models will be attempted for this project, both requiring the use of Python and Machine Learning Packages and Frameworks.\n",
    "\n",
    "[Soumyendu] Normalize data for zero mean and unit variance with logarithmic scaling of returns and data frame construction for optimized Neural Network training with back propagation.\n",
    "\n",
    "·       Develop Python Code to Implement, Train and Test Models for Prediction: \n",
    "\n",
    "[Soumyendu]   Recurrent Neural Network implementation using Tensorflow framework APIs in Python and a Baseline model of 3 Layer Feed Forward Neural Network for comparison.\n",
    "\n",
    "·       Evaluate Model Performance \n",
    "\n",
    "[Soumyendu]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data Preparation\n",
    "- Data Source research and identification (free part of QuantQuote database )\n",
    "- Importing decade long stock data of the S&P 500 companies\n",
    "- Cleaning and Normalizing data with zero mean, unit variance and logarithmic scaling for normal distribution\n",
    "- Processing and Separating data into Input Data (the intra-day price fluctuations) and Expected Output Data (discrete categorized classification values for price gain over consecutive days )\n",
    "- Forming Data Frames for Training and Testing for Neural Network\n",
    "\n",
    "Code for Two Distinct Neural Networks for Comparison\n",
    "- Code for Fully Connected Feed Forward Neural Network classifier using low level Tensorflow Framework API\n",
    "- Code for Attention Adaptation of Multilayer (2x) Recurrent Neural Networks (100x NN lookback) with LSTM / GRU Memory Optimized Cells using low level Tensorflow Framework API. This follows latest publication and enhancement in Recurrent Neural Networks.\n",
    "- Accelerated Linear Algebra optimization for faster code execution with XLA JIT (Just in time compilation) directives\n",
    "- Both these codes uses low level Tensorflow API to facilitate usage of advanced Tensorflow framework features with embedding and model structure refinements\n",
    "\n",
    "Model Visualization and Tensorboard Embeddings\n",
    "- Tensorboard code embedding for Graphical Model Visualization and Data visualization \n",
    "\n",
    "Analysis and Result Visualization\n",
    "- Diagnostics and Evaluation with Graphical presentation of the effectiveness of the Model\n",
    "- Confusion Matrix and Accuracy, Precision, Sensitivity and Specificity\n",
    "- The Graph demonstrates the effectiveness of both the Neural Networks in making daily trading decisions, scored against random buys and sells\n",
    "- Several measures of effectiveness in decision making\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data Source :\n",
    "\n",
    "QuantQuote Free Historical Stock Data\n",
    "This collection of daily resolution data goes back to 1998 for all symbols currently active in the S&P500. It is updated quarterly, the last update was 07/31/2013. \n",
    "\n",
    "QuantQuote Free Daily Data has several advantages not found elsewhere:\n",
    "\n",
    "Full split/dividend adjustments for OHLCV\n",
    "Single zip file makes downloading ~500 symbols a one step process\n",
    "QuantQuote data quality means errors in many free sources (such as Yahoo) are not present.\n",
    "For format information, please consult our minute resolution data documentation.\n",
    "\n",
    "Reference :\n",
    "https://quantquote.com/historical-stock-data\n",
    "\n",
    "\n",
    "Data Preparation\n",
    "\n",
    "1. Calculate the log return between two points. Assumption is returns are log-normally distributed. \n",
    "2. Using zscore  to map a pandas series to make it have mean 0 and variance 1, to normalize the data.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "table_qep.csv\n"
     ]
    }
   ],
   "source": [
    "print(os.listdir('./daily')[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "datapath = './daily'\n",
    "filepath = os.path.join(datapath,os.listdir('./daily')[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./daily/table_qep.csv\n"
     ]
    }
   ],
   "source": [
    "print(filepath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./daily/table_qep.csv qep\n"
     ]
    }
   ],
   "source": [
    "import re\n",
    "ticker_regex = re.compile('.+_(?P<ticker>.+)\\.csv')\n",
    "get_ticker =lambda x :ticker_regex.match(x).groupdict()['ticker']\n",
    "print(filepath,get_ticker(filepath))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "ret = lambda x,y: log(y/x) #Log return \n",
    "zscore = lambda x:(x -x.mean())/x.std() # zscore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "D = pd.read_csv(filepath,header=None,names=['UNK','o','h','l','c','v']) #Load the dataframe with headers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "scrolled": 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>UNK</th>\n",
       "      <th>o</th>\n",
       "      <th>h</th>\n",
       "      <th>l</th>\n",
       "      <th>c</th>\n",
       "      <th>v</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>20100701</th>\n",
       "      <td>0</td>\n",
       "      <td>30.2711</td>\n",
       "      <td>30.7971</td>\n",
       "      <td>29.0305</td>\n",
       "      <td>29.1397</td>\n",
       "      <td>2407928.706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20100702</th>\n",
       "      <td>0</td>\n",
       "      <td>29.2786</td>\n",
       "      <td>29.7352</td>\n",
       "      <td>28.6633</td>\n",
       "      <td>28.8320</td>\n",
       "      <td>2997759.606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20100706</th>\n",
       "      <td>0</td>\n",
       "      <td>29.3878</td>\n",
       "      <td>29.3878</td>\n",
       "      <td>27.7998</td>\n",
       "      <td>27.9090</td>\n",
       "      <td>2136222.897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20100707</th>\n",
       "      <td>0</td>\n",
       "      <td>28.2960</td>\n",
       "      <td>29.2687</td>\n",
       "      <td>27.6906</td>\n",
       "      <td>29.1198</td>\n",
       "      <td>2069802.765</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20100708</th>\n",
       "      <td>0</td>\n",
       "      <td>29.7252</td>\n",
       "      <td>30.8567</td>\n",
       "      <td>29.6855</td>\n",
       "      <td>30.0031</td>\n",
       "      <td>1988857.584</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          UNK        o        h        l        c            v\n",
       "20100701    0  30.2711  30.7971  29.0305  29.1397  2407928.706\n",
       "20100702    0  29.2786  29.7352  28.6633  28.8320  2997759.606\n",
       "20100706    0  29.3878  29.3878  27.7998  27.9090  2136222.897\n",
       "20100707    0  28.2960  29.2687  27.6906  29.1198  2069802.765\n",
       "20100708    0  29.7252  30.8567  29.6855  30.0031  1988857.584"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "D.head() #Lets peack at it"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The processed data will be the fluctation of prices. The prices are normalized to log return of the various prices followed by zscore. \n",
    "Ticker will be extraccted from the file and added to the column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n"
     ]
    }
   ],
   "source": [
    "def make_inputs(filepath):\n",
    "    D = pd.read_csv(filepath,header=None,names=['UNK','o','h','l','c','v']) #Load the dataframe with headers\n",
    "    D.index = pd.to_datetime(D.index,format='%Y%m%d') # Set the indix to a datetime\n",
    "    Res = pd.DataFrame()\n",
    "    ticker = get_ticker(filepath)\n",
    "\n",
    "    Res['c_2_o'] = zscore(ret(D.o,D.c))\n",
    "    Res['h_2_o'] = zscore(ret(D.o,D.h))\n",
    "    Res['l_2_o'] = zscore(ret(D.o,D.l))\n",
    "    Res['c_2_h'] = zscore(ret(D.h,D.c))\n",
    "    Res['h_2_l'] = zscore(ret(D.l,D.h))\n",
    "    Res['c1_c0'] = ret(D.c,D.c.shift(-1)).fillna(0) #Tommorows return \n",
    "    Res['vol'] = zscore(D.v)\n",
    "    Res['ticker'] = ticker\n",
    "    return Res\n",
    "Res = make_inputs(filepath)\n",
    "print(type(Res))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "scrolled": 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>c_2_o</th>\n",
       "      <th>h_2_o</th>\n",
       "      <th>l_2_o</th>\n",
       "      <th>c_2_h</th>\n",
       "      <th>h_2_l</th>\n",
       "      <th>c1_c0</th>\n",
       "      <th>vol</th>\n",
       "      <th>ticker</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-07-01</th>\n",
       "      <td>-1.903707</td>\n",
       "      <td>0.190696</td>\n",
       "      <td>-1.780643</td>\n",
       "      <td>-2.994340</td>\n",
       "      <td>1.836564</td>\n",
       "      <td>-0.010616</td>\n",
       "      <td>1.197635</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-02</th>\n",
       "      <td>-0.758547</td>\n",
       "      <td>0.057368</td>\n",
       "      <td>-0.348498</td>\n",
       "      <td>-1.174928</td>\n",
       "      <td>0.376646</td>\n",
       "      <td>-0.032537</td>\n",
       "      <td>1.942951</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-06</th>\n",
       "      <td>-2.586059</td>\n",
       "      <td>-1.119977</td>\n",
       "      <td>-2.733060</td>\n",
       "      <td>-2.720119</td>\n",
       "      <td>1.606555</td>\n",
       "      <td>0.042469</td>\n",
       "      <td>0.854305</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-07</th>\n",
       "      <td>1.462571</td>\n",
       "      <td>1.451465</td>\n",
       "      <td>-0.375473</td>\n",
       "      <td>0.739043</td>\n",
       "      <td>1.598386</td>\n",
       "      <td>0.029882</td>\n",
       "      <td>0.770376</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-08</th>\n",
       "      <td>0.485175</td>\n",
       "      <td>1.722356</td>\n",
       "      <td>1.034700</td>\n",
       "      <td>-0.967330</td>\n",
       "      <td>0.506006</td>\n",
       "      <td>0.006266</td>\n",
       "      <td>0.668093</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               c_2_o     h_2_o     l_2_o     c_2_h     h_2_l     c1_c0  \\\n",
       "2010-07-01 -1.903707  0.190696 -1.780643 -2.994340  1.836564 -0.010616   \n",
       "2010-07-02 -0.758547  0.057368 -0.348498 -1.174928  0.376646 -0.032537   \n",
       "2010-07-06 -2.586059 -1.119977 -2.733060 -2.720119  1.606555  0.042469   \n",
       "2010-07-07  1.462571  1.451465 -0.375473  0.739043  1.598386  0.029882   \n",
       "2010-07-08  0.485175  1.722356  1.034700 -0.967330  0.506006  0.006266   \n",
       "\n",
       "                 vol ticker  \n",
       "2010-07-01  1.197635    qep  \n",
       "2010-07-02  1.942951    qep  \n",
       "2010-07-06  0.854305    qep  \n",
       "2010-07-07  0.770376    qep  \n",
       "2010-07-08  0.668093    qep  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Res.head() # Checking data frame"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Processing all tickers to generate data framest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This will take a minute\n"
     ]
    }
   ],
   "source": [
    "Final = pd.DataFrame()\n",
    "for f in os.listdir(datapath):\n",
    "    filepath = os.path.join(datapath,f)\n",
    "    if filepath.endswith('.csv'):\n",
    "        Res = make_inputs(filepath)\n",
    "        Final = Final.append(Res)\n",
    "print('This will take a minute')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "scrolled": 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>c_2_o</th>\n",
       "      <th>h_2_o</th>\n",
       "      <th>l_2_o</th>\n",
       "      <th>c_2_h</th>\n",
       "      <th>h_2_l</th>\n",
       "      <th>c1_c0</th>\n",
       "      <th>vol</th>\n",
       "      <th>ticker</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-07-01</th>\n",
       "      <td>-1.903707</td>\n",
       "      <td>0.190696</td>\n",
       "      <td>-1.780643</td>\n",
       "      <td>-2.994340</td>\n",
       "      <td>1.836564</td>\n",
       "      <td>-0.010616</td>\n",
       "      <td>1.197635</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-02</th>\n",
       "      <td>-0.758547</td>\n",
       "      <td>0.057368</td>\n",
       "      <td>-0.348498</td>\n",
       "      <td>-1.174928</td>\n",
       "      <td>0.376646</td>\n",
       "      <td>-0.032537</td>\n",
       "      <td>1.942951</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-06</th>\n",
       "      <td>-2.586059</td>\n",
       "      <td>-1.119977</td>\n",
       "      <td>-2.733060</td>\n",
       "      <td>-2.720119</td>\n",
       "      <td>1.606555</td>\n",
       "      <td>0.042469</td>\n",
       "      <td>0.854305</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-07</th>\n",
       "      <td>1.462571</td>\n",
       "      <td>1.451465</td>\n",
       "      <td>-0.375473</td>\n",
       "      <td>0.739043</td>\n",
       "      <td>1.598386</td>\n",
       "      <td>0.029882</td>\n",
       "      <td>0.770376</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-08</th>\n",
       "      <td>0.485175</td>\n",
       "      <td>1.722356</td>\n",
       "      <td>1.034700</td>\n",
       "      <td>-0.967330</td>\n",
       "      <td>0.506006</td>\n",
       "      <td>0.006266</td>\n",
       "      <td>0.668093</td>\n",
       "      <td>qep</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               c_2_o     h_2_o     l_2_o     c_2_h     h_2_l     c1_c0  \\\n",
       "2010-07-01 -1.903707  0.190696 -1.780643 -2.994340  1.836564 -0.010616   \n",
       "2010-07-02 -0.758547  0.057368 -0.348498 -1.174928  0.376646 -0.032537   \n",
       "2010-07-06 -2.586059 -1.119977 -2.733060 -2.720119  1.606555  0.042469   \n",
       "2010-07-07  1.462571  1.451465 -0.375473  0.739043  1.598386  0.029882   \n",
       "2010-07-08  0.485175  1.722356  1.034700 -0.967330  0.506006  0.006266   \n",
       "\n",
       "                 vol ticker  \n",
       "2010-07-01  1.197635    qep  \n",
       "2010-07-02  1.942951    qep  \n",
       "2010-07-06  0.854305    qep  \n",
       "2010-07-07  0.770376    qep  \n",
       "2010-07-08  0.668093    qep  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "pivot_columns = Final.columns[:-1]\n",
    "P = Final.pivot_table(index=Final.index,columns='ticker',values=pivot_columns) # Make a pivot table from the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"10\" halign=\"left\">c_2_o</th>\n",
       "      <th>...</th>\n",
       "      <th colspan=\"10\" halign=\"left\">vol</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ticker</th>\n",
       "      <th>a</th>\n",
       "      <th>aa</th>\n",
       "      <th>aapl</th>\n",
       "      <th>abbv</th>\n",
       "      <th>abc</th>\n",
       "      <th>abt</th>\n",
       "      <th>ace</th>\n",
       "      <th>acn</th>\n",
       "      <th>act</th>\n",
       "      <th>adbe</th>\n",
       "      <th>...</th>\n",
       "      <th>xl</th>\n",
       "      <th>xlnx</th>\n",
       "      <th>xom</th>\n",
       "      <th>xray</th>\n",
       "      <th>xrx</th>\n",
       "      <th>xyl</th>\n",
       "      <th>yhoo</th>\n",
       "      <th>yum</th>\n",
       "      <th>zion</th>\n",
       "      <th>zmh</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1998-01-02</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.017649</td>\n",
       "      <td>6.991394</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.047250</td>\n",
       "      <td>1.845516</td>\n",
       "      <td>-0.601609</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.911471</td>\n",
       "      <td>-0.375461</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.676802</td>\n",
       "      <td>1.391497</td>\n",
       "      <td>-1.045626</td>\n",
       "      <td>-1.147952</td>\n",
       "      <td>-0.817355</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.368713</td>\n",
       "      <td>-0.678763</td>\n",
       "      <td>-0.724482</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-05</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.373488</td>\n",
       "      <td>-1.224062</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.160945</td>\n",
       "      <td>0.031659</td>\n",
       "      <td>-0.234577</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.090086</td>\n",
       "      <td>0.791331</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.621025</td>\n",
       "      <td>1.019607</td>\n",
       "      <td>-0.615370</td>\n",
       "      <td>-0.556684</td>\n",
       "      <td>-0.757660</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.064841</td>\n",
       "      <td>0.980283</td>\n",
       "      <td>-0.735275</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-06</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.718569</td>\n",
       "      <td>6.857266</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1.102853</td>\n",
       "      <td>-0.473164</td>\n",
       "      <td>0.452262</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1.239961</td>\n",
       "      <td>1.451409</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.634169</td>\n",
       "      <td>0.020195</td>\n",
       "      <td>-0.527923</td>\n",
       "      <td>-1.131583</td>\n",
       "      <td>-0.629732</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.150071</td>\n",
       "      <td>1.697887</td>\n",
       "      <td>-0.683202</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-07</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.980669</td>\n",
       "      <td>-3.007557</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.553079</td>\n",
       "      <td>0.746779</td>\n",
       "      <td>0.137372</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.878292</td>\n",
       "      <td>-0.534781</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.616242</td>\n",
       "      <td>1.023705</td>\n",
       "      <td>-0.134130</td>\n",
       "      <td>-0.329006</td>\n",
       "      <td>-0.504232</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.463651</td>\n",
       "      <td>0.515268</td>\n",
       "      <td>-0.635858</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-08</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-1.828950</td>\n",
       "      <td>1.388888</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.488170</td>\n",
       "      <td>-0.221620</td>\n",
       "      <td>-0.484264</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.300492</td>\n",
       "      <td>0.538778</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.655003</td>\n",
       "      <td>0.645285</td>\n",
       "      <td>-0.907375</td>\n",
       "      <td>-1.066107</td>\n",
       "      <td>-0.560549</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.157777</td>\n",
       "      <td>0.235907</td>\n",
       "      <td>-0.605092</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 3500 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           c_2_o                                                             \\\n",
       "ticker         a        aa      aapl abbv       abc       abt       ace acn   \n",
       "1998-01-02   NaN -0.017649  6.991394  NaN  0.047250  1.845516 -0.601609 NaN   \n",
       "1998-01-05   NaN  0.373488 -1.224062  NaN -0.160945  0.031659 -0.234577 NaN   \n",
       "1998-01-06   NaN -0.718569  6.857266  NaN -1.102853 -0.473164  0.452262 NaN   \n",
       "1998-01-07   NaN  0.980669 -3.007557  NaN  0.553079  0.746779  0.137372 NaN   \n",
       "1998-01-08   NaN -1.828950  1.388888  NaN  0.488170 -0.221620 -0.484264 NaN   \n",
       "\n",
       "                               ...       vol                                \\\n",
       "ticker           act      adbe ...        xl      xlnx       xom      xray   \n",
       "1998-01-02 -0.911471 -0.375461 ... -0.676802  1.391497 -1.045626 -1.147952   \n",
       "1998-01-05  2.090086  0.791331 ... -0.621025  1.019607 -0.615370 -0.556684   \n",
       "1998-01-06 -1.239961  1.451409 ... -0.634169  0.020195 -0.527923 -1.131583   \n",
       "1998-01-07 -0.878292 -0.534781 ... -0.616242  1.023705 -0.134130 -0.329006   \n",
       "1998-01-08  1.300492  0.538778 ... -0.655003  0.645285 -0.907375 -1.066107   \n",
       "\n",
       "                                                            \n",
       "ticker           xrx xyl      yhoo       yum      zion zmh  \n",
       "1998-01-02 -0.817355 NaN -0.368713 -0.678763 -0.724482 NaN  \n",
       "1998-01-05 -0.757660 NaN  0.064841  0.980283 -0.735275 NaN  \n",
       "1998-01-06 -0.629732 NaN  0.150071  1.697887 -0.683202 NaN  \n",
       "1998-01-07 -0.504232 NaN -0.463651  0.515268 -0.635858 NaN  \n",
       "1998-01-08 -0.560549 NaN -0.157777  0.235907 -0.605092 NaN  \n",
       "\n",
       "[5 rows x 3500 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "P.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Flattened pivot\n",
    "source http://stackoverflow.com/questions/14507794/python-pandas-how-to-flatten-a-hierarchical-index-in-columns\n",
    "P will be a flattened dataframe of all the entries for each stock, one day per row"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "mi = P.columns.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "new_ind = pd.Index(e[1] +'_' + e[0] for e in mi)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "P.columns = new_ind\n",
    "P = P.sort_index(axis=1) # Sort by columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "scrolled": 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_c1_c0</th>\n",
       "      <th>a_c_2_h</th>\n",
       "      <th>a_c_2_o</th>\n",
       "      <th>a_h_2_l</th>\n",
       "      <th>a_h_2_o</th>\n",
       "      <th>a_l_2_o</th>\n",
       "      <th>a_vol</th>\n",
       "      <th>aa_c1_c0</th>\n",
       "      <th>aa_c_2_h</th>\n",
       "      <th>aa_c_2_o</th>\n",
       "      <th>...</th>\n",
       "      <th>zion_h_2_o</th>\n",
       "      <th>zion_l_2_o</th>\n",
       "      <th>zion_vol</th>\n",
       "      <th>zmh_c1_c0</th>\n",
       "      <th>zmh_c_2_h</th>\n",
       "      <th>zmh_c_2_o</th>\n",
       "      <th>zmh_h_2_l</th>\n",
       "      <th>zmh_h_2_o</th>\n",
       "      <th>zmh_l_2_o</th>\n",
       "      <th>zmh_vol</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1998-01-02</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.017390</td>\n",
       "      <td>0.670259</td>\n",
       "      <td>-0.017649</td>\n",
       "      <td>...</td>\n",
       "      <td>0.246401</td>\n",
       "      <td>0.710284</td>\n",
       "      <td>-0.724482</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-05</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.022601</td>\n",
       "      <td>0.462520</td>\n",
       "      <td>0.373488</td>\n",
       "      <td>...</td>\n",
       "      <td>0.127875</td>\n",
       "      <td>0.710284</td>\n",
       "      <td>-0.735275</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-06</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.001693</td>\n",
       "      <td>-0.088471</td>\n",
       "      <td>-0.718569</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.677287</td>\n",
       "      <td>-1.264514</td>\n",
       "      <td>-0.683202</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-07</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.042290</td>\n",
       "      <td>1.002099</td>\n",
       "      <td>0.980669</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.677287</td>\n",
       "      <td>-1.175835</td>\n",
       "      <td>-0.635858</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-08</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.024094</td>\n",
       "      <td>-2.074267</td>\n",
       "      <td>-1.828950</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.202411</td>\n",
       "      <td>-2.278200</td>\n",
       "      <td>-0.605092</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 3500 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            a_c1_c0  a_c_2_h  a_c_2_o  a_h_2_l  a_h_2_o  a_l_2_o  a_vol  \\\n",
       "1998-01-02      NaN      NaN      NaN      NaN      NaN      NaN    NaN   \n",
       "1998-01-05      NaN      NaN      NaN      NaN      NaN      NaN    NaN   \n",
       "1998-01-06      NaN      NaN      NaN      NaN      NaN      NaN    NaN   \n",
       "1998-01-07      NaN      NaN      NaN      NaN      NaN      NaN    NaN   \n",
       "1998-01-08      NaN      NaN      NaN      NaN      NaN      NaN    NaN   \n",
       "\n",
       "            aa_c1_c0  aa_c_2_h  aa_c_2_o   ...     zion_h_2_o  zion_l_2_o  \\\n",
       "1998-01-02  0.017390  0.670259 -0.017649   ...       0.246401    0.710284   \n",
       "1998-01-05 -0.022601  0.462520  0.373488   ...       0.127875    0.710284   \n",
       "1998-01-06  0.001693 -0.088471 -0.718569   ...      -0.677287   -1.264514   \n",
       "1998-01-07 -0.042290  1.002099  0.980669   ...      -0.677287   -1.175835   \n",
       "1998-01-08 -0.024094 -2.074267 -1.828950   ...      -0.202411   -2.278200   \n",
       "\n",
       "            zion_vol  zmh_c1_c0  zmh_c_2_h  zmh_c_2_o  zmh_h_2_l  zmh_h_2_o  \\\n",
       "1998-01-02 -0.724482        NaN        NaN        NaN        NaN        NaN   \n",
       "1998-01-05 -0.735275        NaN        NaN        NaN        NaN        NaN   \n",
       "1998-01-06 -0.683202        NaN        NaN        NaN        NaN        NaN   \n",
       "1998-01-07 -0.635858        NaN        NaN        NaN        NaN        NaN   \n",
       "1998-01-08 -0.605092        NaN        NaN        NaN        NaN        NaN   \n",
       "\n",
       "            zmh_l_2_o  zmh_vol  \n",
       "1998-01-02        NaN      NaN  \n",
       "1998-01-05        NaN      NaN  \n",
       "1998-01-06        NaN      NaN  \n",
       "1998-01-07        NaN      NaN  \n",
       "1998-01-08        NaN      NaN  \n",
       "\n",
       "[5 rows x 3500 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "P.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "clean_and_flat = P.dropna(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "target_cols = list(filter(lambda x: 'c1_c0' in x, clean_and_flat.columns.values))\n",
    "input_cols  = list(filter(lambda x: 'c1_c0' not in x, clean_and_flat.columns.values))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "388"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(target_cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "InputDF = clean_and_flat[input_cols][:3900]\n",
    "TargetDF = clean_and_flat[target_cols][:3900]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>aa_c_2_h</th>\n",
       "      <th>aa_c_2_o</th>\n",
       "      <th>aa_h_2_l</th>\n",
       "      <th>aa_h_2_o</th>\n",
       "      <th>aa_l_2_o</th>\n",
       "      <th>aa_vol</th>\n",
       "      <th>aapl_c_2_h</th>\n",
       "      <th>aapl_c_2_o</th>\n",
       "      <th>aapl_h_2_l</th>\n",
       "      <th>aapl_h_2_o</th>\n",
       "      <th>...</th>\n",
       "      <th>yum_h_2_l</th>\n",
       "      <th>yum_h_2_o</th>\n",
       "      <th>yum_l_2_o</th>\n",
       "      <th>yum_vol</th>\n",
       "      <th>zion_c_2_h</th>\n",
       "      <th>zion_c_2_o</th>\n",
       "      <th>zion_h_2_l</th>\n",
       "      <th>zion_h_2_o</th>\n",
       "      <th>zion_l_2_o</th>\n",
       "      <th>zion_vol</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1998-01-02</th>\n",
       "      <td>0.670259</td>\n",
       "      <td>-0.017649</td>\n",
       "      <td>-0.934572</td>\n",
       "      <td>-0.735743</td>\n",
       "      <td>0.435162</td>\n",
       "      <td>-0.761999</td>\n",
       "      <td>1.019696</td>\n",
       "      <td>6.991394</td>\n",
       "      <td>6.649433</td>\n",
       "      <td>8.707782</td>\n",
       "      <td>...</td>\n",
       "      <td>0.416891</td>\n",
       "      <td>-0.322652</td>\n",
       "      <td>-0.855900</td>\n",
       "      <td>-0.678763</td>\n",
       "      <td>0.224681</td>\n",
       "      <td>0.419702</td>\n",
       "      <td>-0.312078</td>\n",
       "      <td>0.246401</td>\n",
       "      <td>0.710284</td>\n",
       "      <td>-0.724482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-05</th>\n",
       "      <td>0.462520</td>\n",
       "      <td>0.373488</td>\n",
       "      <td>-0.732608</td>\n",
       "      <td>0.061561</td>\n",
       "      <td>0.900313</td>\n",
       "      <td>-0.566965</td>\n",
       "      <td>-0.984079</td>\n",
       "      <td>-1.224062</td>\n",
       "      <td>2.242381</td>\n",
       "      <td>-0.763862</td>\n",
       "      <td>...</td>\n",
       "      <td>1.360326</td>\n",
       "      <td>-0.644971</td>\n",
       "      <td>-2.369968</td>\n",
       "      <td>0.980283</td>\n",
       "      <td>0.673049</td>\n",
       "      <td>0.719822</td>\n",
       "      <td>-0.392667</td>\n",
       "      <td>0.127875</td>\n",
       "      <td>0.710284</td>\n",
       "      <td>-0.735275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-06</th>\n",
       "      <td>-0.088471</td>\n",
       "      <td>-0.718569</td>\n",
       "      <td>-0.720530</td>\n",
       "      <td>-0.966978</td>\n",
       "      <td>-0.014696</td>\n",
       "      <td>-0.665787</td>\n",
       "      <td>-3.590776</td>\n",
       "      <td>6.857266</td>\n",
       "      <td>13.046850</td>\n",
       "      <td>12.876073</td>\n",
       "      <td>...</td>\n",
       "      <td>0.539951</td>\n",
       "      <td>-0.942530</td>\n",
       "      <td>-1.652489</td>\n",
       "      <td>1.697887</td>\n",
       "      <td>0.122250</td>\n",
       "      <td>-0.486186</td>\n",
       "      <td>0.393348</td>\n",
       "      <td>-0.677287</td>\n",
       "      <td>-1.264514</td>\n",
       "      <td>-0.683202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-07</th>\n",
       "      <td>1.002099</td>\n",
       "      <td>0.980669</td>\n",
       "      <td>-0.569841</td>\n",
       "      <td>0.386474</td>\n",
       "      <td>0.996909</td>\n",
       "      <td>-0.708534</td>\n",
       "      <td>-4.732173</td>\n",
       "      <td>-3.007557</td>\n",
       "      <td>3.126659</td>\n",
       "      <td>0.308601</td>\n",
       "      <td>...</td>\n",
       "      <td>0.954036</td>\n",
       "      <td>1.362584</td>\n",
       "      <td>0.219573</td>\n",
       "      <td>0.515268</td>\n",
       "      <td>-0.524524</td>\n",
       "      <td>-1.069686</td>\n",
       "      <td>0.333468</td>\n",
       "      <td>-0.677287</td>\n",
       "      <td>-1.175835</td>\n",
       "      <td>-0.635858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-08</th>\n",
       "      <td>-2.074267</td>\n",
       "      <td>-1.828950</td>\n",
       "      <td>0.917882</td>\n",
       "      <td>-0.503345</td>\n",
       "      <td>-1.501386</td>\n",
       "      <td>-0.784476</td>\n",
       "      <td>-0.790227</td>\n",
       "      <td>1.388888</td>\n",
       "      <td>2.631329</td>\n",
       "      <td>2.667391</td>\n",
       "      <td>...</td>\n",
       "      <td>0.768144</td>\n",
       "      <td>-0.263050</td>\n",
       "      <td>-1.233393</td>\n",
       "      <td>0.235907</td>\n",
       "      <td>-2.687743</td>\n",
       "      <td>-2.603061</td>\n",
       "      <td>1.400707</td>\n",
       "      <td>-0.202411</td>\n",
       "      <td>-2.278200</td>\n",
       "      <td>-0.605092</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2328 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            aa_c_2_h  aa_c_2_o  aa_h_2_l  aa_h_2_o  aa_l_2_o    aa_vol  \\\n",
       "1998-01-02  0.670259 -0.017649 -0.934572 -0.735743  0.435162 -0.761999   \n",
       "1998-01-05  0.462520  0.373488 -0.732608  0.061561  0.900313 -0.566965   \n",
       "1998-01-06 -0.088471 -0.718569 -0.720530 -0.966978 -0.014696 -0.665787   \n",
       "1998-01-07  1.002099  0.980669 -0.569841  0.386474  0.996909 -0.708534   \n",
       "1998-01-08 -2.074267 -1.828950  0.917882 -0.503345 -1.501386 -0.784476   \n",
       "\n",
       "            aapl_c_2_h  aapl_c_2_o  aapl_h_2_l  aapl_h_2_o    ...     \\\n",
       "1998-01-02    1.019696    6.991394    6.649433    8.707782    ...      \n",
       "1998-01-05   -0.984079   -1.224062    2.242381   -0.763862    ...      \n",
       "1998-01-06   -3.590776    6.857266   13.046850   12.876073    ...      \n",
       "1998-01-07   -4.732173   -3.007557    3.126659    0.308601    ...      \n",
       "1998-01-08   -0.790227    1.388888    2.631329    2.667391    ...      \n",
       "\n",
       "            yum_h_2_l  yum_h_2_o  yum_l_2_o   yum_vol  zion_c_2_h  zion_c_2_o  \\\n",
       "1998-01-02   0.416891  -0.322652  -0.855900 -0.678763    0.224681    0.419702   \n",
       "1998-01-05   1.360326  -0.644971  -2.369968  0.980283    0.673049    0.719822   \n",
       "1998-01-06   0.539951  -0.942530  -1.652489  1.697887    0.122250   -0.486186   \n",
       "1998-01-07   0.954036   1.362584   0.219573  0.515268   -0.524524   -1.069686   \n",
       "1998-01-08   0.768144  -0.263050  -1.233393  0.235907   -2.687743   -2.603061   \n",
       "\n",
       "            zion_h_2_l  zion_h_2_o  zion_l_2_o  zion_vol  \n",
       "1998-01-02   -0.312078    0.246401    0.710284 -0.724482  \n",
       "1998-01-05   -0.392667    0.127875    0.710284 -0.735275  \n",
       "1998-01-06    0.393348   -0.677287   -1.264514 -0.683202  \n",
       "1998-01-07    0.333468   -0.677287   -1.175835 -0.635858  \n",
       "1998-01-08    1.400707   -0.202411   -2.278200 -0.605092  \n",
       "\n",
       "[5 rows x 2328 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "InputDF.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>aa_c1_c0</th>\n",
       "      <th>aapl_c1_c0</th>\n",
       "      <th>abc_c1_c0</th>\n",
       "      <th>abt_c1_c0</th>\n",
       "      <th>ace_c1_c0</th>\n",
       "      <th>act_c1_c0</th>\n",
       "      <th>adbe_c1_c0</th>\n",
       "      <th>adi_c1_c0</th>\n",
       "      <th>adm_c1_c0</th>\n",
       "      <th>adp_c1_c0</th>\n",
       "      <th>...</th>\n",
       "      <th>x_c1_c0</th>\n",
       "      <th>xel_c1_c0</th>\n",
       "      <th>xl_c1_c0</th>\n",
       "      <th>xlnx_c1_c0</th>\n",
       "      <th>xom_c1_c0</th>\n",
       "      <th>xray_c1_c0</th>\n",
       "      <th>xrx_c1_c0</th>\n",
       "      <th>yhoo_c1_c0</th>\n",
       "      <th>yum_c1_c0</th>\n",
       "      <th>zion_c1_c0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1998-01-02</th>\n",
       "      <td>0.017390</td>\n",
       "      <td>-0.015503</td>\n",
       "      <td>-0.016078</td>\n",
       "      <td>0.004586</td>\n",
       "      <td>-0.006547</td>\n",
       "      <td>0.047809</td>\n",
       "      <td>0.012268</td>\n",
       "      <td>0.038082</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.012420</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.006149</td>\n",
       "      <td>-0.010796</td>\n",
       "      <td>0.004856</td>\n",
       "      <td>-0.019672</td>\n",
       "      <td>-0.011189</td>\n",
       "      <td>-0.029680</td>\n",
       "      <td>0.015005</td>\n",
       "      <td>-0.054224</td>\n",
       "      <td>-0.023887</td>\n",
       "      <td>0.019158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-05</th>\n",
       "      <td>-0.022601</td>\n",
       "      <td>0.168686</td>\n",
       "      <td>-0.018823</td>\n",
       "      <td>-0.016672</td>\n",
       "      <td>0.007916</td>\n",
       "      <td>-0.005666</td>\n",
       "      <td>0.035934</td>\n",
       "      <td>-0.033379</td>\n",
       "      <td>-0.017727</td>\n",
       "      <td>0.002139</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.012078</td>\n",
       "      <td>-0.012148</td>\n",
       "      <td>-0.003001</td>\n",
       "      <td>-0.006644</td>\n",
       "      <td>-0.038411</td>\n",
       "      <td>-0.002427</td>\n",
       "      <td>-0.008338</td>\n",
       "      <td>0.016891</td>\n",
       "      <td>-0.040999</td>\n",
       "      <td>-0.010841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-06</th>\n",
       "      <td>0.001693</td>\n",
       "      <td>-0.082508</td>\n",
       "      <td>0.010072</td>\n",
       "      <td>0.005546</td>\n",
       "      <td>0.005239</td>\n",
       "      <td>-0.022684</td>\n",
       "      <td>-0.016370</td>\n",
       "      <td>-0.043925</td>\n",
       "      <td>0.062917</td>\n",
       "      <td>-0.013562</td>\n",
       "      <td>...</td>\n",
       "      <td>0.026259</td>\n",
       "      <td>-0.008895</td>\n",
       "      <td>-0.017745</td>\n",
       "      <td>-0.044161</td>\n",
       "      <td>0.030221</td>\n",
       "      <td>-0.006268</td>\n",
       "      <td>-0.034613</td>\n",
       "      <td>-0.001095</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.029190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-07</th>\n",
       "      <td>-0.042290</td>\n",
       "      <td>0.034934</td>\n",
       "      <td>0.005435</td>\n",
       "      <td>0.015691</td>\n",
       "      <td>-0.017187</td>\n",
       "      <td>0.020888</td>\n",
       "      <td>0.010469</td>\n",
       "      <td>0.002260</td>\n",
       "      <td>-0.016930</td>\n",
       "      <td>-0.004169</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002141</td>\n",
       "      <td>-0.014475</td>\n",
       "      <td>0.005005</td>\n",
       "      <td>-0.020771</td>\n",
       "      <td>-0.006658</td>\n",
       "      <td>-0.020850</td>\n",
       "      <td>0.006869</td>\n",
       "      <td>-0.040050</td>\n",
       "      <td>-0.083267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-08</th>\n",
       "      <td>-0.024094</td>\n",
       "      <td>0.010464</td>\n",
       "      <td>-0.031078</td>\n",
       "      <td>-0.011075</td>\n",
       "      <td>0.005301</td>\n",
       "      <td>-0.056066</td>\n",
       "      <td>-0.026867</td>\n",
       "      <td>-0.060529</td>\n",
       "      <td>-0.014034</td>\n",
       "      <td>-0.017884</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.083382</td>\n",
       "      <td>-0.008964</td>\n",
       "      <td>-0.018331</td>\n",
       "      <td>-0.037006</td>\n",
       "      <td>-0.022243</td>\n",
       "      <td>-0.049740</td>\n",
       "      <td>-0.040273</td>\n",
       "      <td>-0.039685</td>\n",
       "      <td>0.007296</td>\n",
       "      <td>-0.037270</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 388 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            aa_c1_c0  aapl_c1_c0  abc_c1_c0  abt_c1_c0  ace_c1_c0  act_c1_c0  \\\n",
       "1998-01-02  0.017390   -0.015503  -0.016078   0.004586  -0.006547   0.047809   \n",
       "1998-01-05 -0.022601    0.168686  -0.018823  -0.016672   0.007916  -0.005666   \n",
       "1998-01-06  0.001693   -0.082508   0.010072   0.005546   0.005239  -0.022684   \n",
       "1998-01-07 -0.042290    0.034934   0.005435   0.015691  -0.017187   0.020888   \n",
       "1998-01-08 -0.024094    0.010464  -0.031078  -0.011075   0.005301  -0.056066   \n",
       "\n",
       "            adbe_c1_c0  adi_c1_c0  adm_c1_c0  adp_c1_c0     ...       x_c1_c0  \\\n",
       "1998-01-02    0.012268   0.038082   0.000000   0.012420     ...     -0.006149   \n",
       "1998-01-05    0.035934  -0.033379  -0.017727   0.002139     ...     -0.012078   \n",
       "1998-01-06   -0.016370  -0.043925   0.062917  -0.013562     ...      0.026259   \n",
       "1998-01-07    0.010469   0.002260  -0.016930  -0.004169     ...      0.000000   \n",
       "1998-01-08   -0.026867  -0.060529  -0.014034  -0.017884     ...     -0.083382   \n",
       "\n",
       "            xel_c1_c0  xl_c1_c0  xlnx_c1_c0  xom_c1_c0  xray_c1_c0  xrx_c1_c0  \\\n",
       "1998-01-02  -0.010796  0.004856   -0.019672  -0.011189   -0.029680   0.015005   \n",
       "1998-01-05  -0.012148 -0.003001   -0.006644  -0.038411   -0.002427  -0.008338   \n",
       "1998-01-06  -0.008895 -0.017745   -0.044161   0.030221   -0.006268  -0.034613   \n",
       "1998-01-07   0.002141 -0.014475    0.005005  -0.020771   -0.006658  -0.020850   \n",
       "1998-01-08  -0.008964 -0.018331   -0.037006  -0.022243   -0.049740  -0.040273   \n",
       "\n",
       "            yhoo_c1_c0  yum_c1_c0  zion_c1_c0  \n",
       "1998-01-02   -0.054224  -0.023887    0.019158  \n",
       "1998-01-05    0.016891  -0.040999   -0.010841  \n",
       "1998-01-06   -0.001095   0.000000   -0.029190  \n",
       "1998-01-07    0.006869  -0.040050   -0.083267  \n",
       "1998-01-08   -0.039685   0.007296   -0.037270  \n",
       "\n",
       "[5 rows x 388 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "TargetDF.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "corrs = TargetDF.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setting Targets\n",
    "We now have an our inputs and targets, kind of. \n",
    "InputsDF has all the inputs we want to predict. Targets DF has the return of each stock each day. \n",
    "For starters, lets give a simpler target to predict than the reuturn of each stock, since we don't have much data. \n",
    "\n",
    "\n",
    "We're going to label the targets as either up (1) down (-1) or flat (0) days.\n",
    "The top chart shows what would happen if we bought 1 dollar of each stock each day\n",
    "The bottom chart shows what would happen if we longed the whole basket on (1) days, shorted it on down days (-1) and ignored it on  (0) days. \n",
    "You can see that this is a valuable target to predict."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "num_stocks = len(TargetDF.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "TotalReturn = ((1-exp(TargetDF)).sum(1))/num_stocks # If i put one dollar in each stock at the close, this is how much I'd get back"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "def labeler(x):\n",
    "    if x>0.0029:\n",
    "        return 1\n",
    "    if x<-0.00462:\n",
    "        return -1\n",
    "    else:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Labeled = pd.DataFrame()\n",
    "Labeled['return'] = TotalReturn\n",
    "Labeled['class'] = TotalReturn.apply(labeler,1)\n",
    "Labeled['multi_class'] = pd.qcut(TotalReturn,11,labels=range(11))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(-0.00301, 0.00119], (0.00741, 0.0999], (0.00119, 0.00741], [-0.108, -0.00877], (-0.00877, -0.00301]]\n",
       "Categories (5, object): [[-0.108, -0.00877] < (-0.00877, -0.00301] < (-0.00301, 0.00119] < (0.00119, 0.00741] < (0.00741, 0.0999]]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.qcut(TotalReturn,5).unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "def labeler_multi(x):\n",
    "    if x>0.0029:\n",
    "        return 1\n",
    "    if x<-0.00462:\n",
    "        return -1\n",
    "    else:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    1301\n",
       " 0    1301\n",
       "-1    1298\n",
       "Name: class, dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Labeled['class'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Labeled['max_return'] = Labeled['class'] * Labeled['return']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>return</th>\n",
       "      <th>class</th>\n",
       "      <th>multi_class</th>\n",
       "      <th>max_return</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1998-01-02</th>\n",
       "      <td>-0.000101</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-05</th>\n",
       "      <td>0.008144</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.008144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-06</th>\n",
       "      <td>0.005501</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.005501</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-07</th>\n",
       "      <td>0.006905</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.006905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-08</th>\n",
       "      <td>0.031674</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0.031674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-09</th>\n",
       "      <td>-0.002883</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-12</th>\n",
       "      <td>-0.017040</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.017040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-13</th>\n",
       "      <td>-0.007591</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.007591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-14</th>\n",
       "      <td>0.004450</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0.004450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-15</th>\n",
       "      <td>-0.009007</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.009007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-16</th>\n",
       "      <td>-0.016143</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.016143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-20</th>\n",
       "      <td>0.005261</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.005261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-21</th>\n",
       "      <td>0.011348</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>0.011348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-22</th>\n",
       "      <td>0.006010</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.006010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-23</th>\n",
       "      <td>0.001350</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-26</th>\n",
       "      <td>-0.006524</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.006524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-27</th>\n",
       "      <td>-0.012866</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.012866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-28</th>\n",
       "      <td>-0.008553</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.008553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-29</th>\n",
       "      <td>0.002832</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-30</th>\n",
       "      <td>-0.019022</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.019022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-02</th>\n",
       "      <td>-0.007555</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.007555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-03</th>\n",
       "      <td>-0.004857</td>\n",
       "      <td>-1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.004857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-04</th>\n",
       "      <td>-0.000357</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-05</th>\n",
       "      <td>-0.006568</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.006568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-06</th>\n",
       "      <td>0.000050</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-09</th>\n",
       "      <td>-0.008998</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.008998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-10</th>\n",
       "      <td>-0.002157</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-11</th>\n",
       "      <td>-0.003066</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-12</th>\n",
       "      <td>0.000292</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-02-13</th>\n",
       "      <td>0.001665</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-05-22</th>\n",
       "      <td>0.001382</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-05-23</th>\n",
       "      <td>0.002729</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-05-24</th>\n",
       "      <td>-0.008017</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.008017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-05-28</th>\n",
       "      <td>0.007752</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.007752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-05-29</th>\n",
       "      <td>-0.003864</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-05-30</th>\n",
       "      <td>0.012106</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>0.012106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-05-31</th>\n",
       "      <td>-0.001223</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-03</th>\n",
       "      <td>0.004919</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.004919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-04</th>\n",
       "      <td>0.014113</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>0.014113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-05</th>\n",
       "      <td>-0.008522</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.008522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-06</th>\n",
       "      <td>-0.014675</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.014675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-07</th>\n",
       "      <td>0.001853</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-10</th>\n",
       "      <td>0.010318</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>0.010318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-11</th>\n",
       "      <td>0.008233</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.008233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-12</th>\n",
       "      <td>-0.018200</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.018200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-13</th>\n",
       "      <td>0.005591</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0.005591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-14</th>\n",
       "      <td>-0.007812</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.007812</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-17</th>\n",
       "      <td>-0.007354</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.007354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-18</th>\n",
       "      <td>0.013241</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>0.013241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-19</th>\n",
       "      <td>0.025786</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0.025786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-20</th>\n",
       "      <td>-0.002306</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-21</th>\n",
       "      <td>0.011943</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>0.011943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-24</th>\n",
       "      <td>-0.011590</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.011590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-25</th>\n",
       "      <td>-0.009281</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.009281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-26</th>\n",
       "      <td>-0.008712</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.008712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-27</th>\n",
       "      <td>0.001504</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-06-28</th>\n",
       "      <td>-0.004822</td>\n",
       "      <td>-1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.004822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-07-01</th>\n",
       "      <td>0.001017</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-07-02</th>\n",
       "      <td>0.001468</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-07-03</th>\n",
       "      <td>-0.010142</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.010142</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3900 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              return  class multi_class  max_return\n",
       "1998-01-02 -0.000101      0           6   -0.000000\n",
       "1998-01-05  0.008144      1           8    0.008144\n",
       "1998-01-06  0.005501      1           8    0.005501\n",
       "1998-01-07  0.006905      1           8    0.006905\n",
       "1998-01-08  0.031674      1          10    0.031674\n",
       "1998-01-09 -0.002883      0           4   -0.000000\n",
       "1998-01-12 -0.017040     -1           0    0.017040\n",
       "1998-01-13 -0.007591     -1           2    0.007591\n",
       "1998-01-14  0.004450      1           7    0.004450\n",
       "1998-01-15 -0.009007     -1           2    0.009007\n",
       "1998-01-16 -0.016143     -1           0    0.016143\n",
       "1998-01-20  0.005261      1           8    0.005261\n",
       "1998-01-21  0.011348      1           9    0.011348\n",
       "1998-01-22  0.006010      1           8    0.006010\n",
       "1998-01-23  0.001350      0           6    0.000000\n",
       "1998-01-26 -0.006524     -1           2    0.006524\n",
       "1998-01-27 -0.012866     -1           1    0.012866\n",
       "1998-01-28 -0.008553     -1           2    0.008553\n",
       "1998-01-29  0.002832      0           7    0.000000\n",
       "1998-01-30 -0.019022     -1           0    0.019022\n",
       "1998-02-02 -0.007555     -1           2    0.007555\n",
       "1998-02-03 -0.004857     -1           3    0.004857\n",
       "1998-02-04 -0.000357      0           5   -0.000000\n",
       "1998-02-05 -0.006568     -1           2    0.006568\n",
       "1998-02-06  0.000050      0           6    0.000000\n",
       "1998-02-09 -0.008998     -1           2    0.008998\n",
       "1998-02-10 -0.002157      0           4   -0.000000\n",
       "1998-02-11 -0.003066      0           4   -0.000000\n",
       "1998-02-12  0.000292      0           6    0.000000\n",
       "1998-02-13  0.001665      0           6    0.000000\n",
       "...              ...    ...         ...         ...\n",
       "2013-05-22  0.001382      0           6    0.000000\n",
       "2013-05-23  0.002729      0           7    0.000000\n",
       "2013-05-24 -0.008017     -1           2    0.008017\n",
       "2013-05-28  0.007752      1           8    0.007752\n",
       "2013-05-29 -0.003864      0           4   -0.000000\n",
       "2013-05-30  0.012106      1           9    0.012106\n",
       "2013-05-31 -0.001223      0           5   -0.000000\n",
       "2013-06-03  0.004919      1           8    0.004919\n",
       "2013-06-04  0.014113      1           9    0.014113\n",
       "2013-06-05 -0.008522     -1           2    0.008522\n",
       "2013-06-06 -0.014675     -1           0    0.014675\n",
       "2013-06-07  0.001853      0           6    0.000000\n",
       "2013-06-10  0.010318      1           9    0.010318\n",
       "2013-06-11  0.008233      1           8    0.008233\n",
       "2013-06-12 -0.018200     -1           0    0.018200\n",
       "2013-06-13  0.005591      1           8    0.005591\n",
       "2013-06-14 -0.007812     -1           2    0.007812\n",
       "2013-06-17 -0.007354     -1           2    0.007354\n",
       "2013-06-18  0.013241      1           9    0.013241\n",
       "2013-06-19  0.025786      1          10    0.025786\n",
       "2013-06-20 -0.002306      0           4   -0.000000\n",
       "2013-06-21  0.011943      1           9    0.011943\n",
       "2013-06-24 -0.011590     -1           1    0.011590\n",
       "2013-06-25 -0.009281     -1           2    0.009281\n",
       "2013-06-26 -0.008712     -1           2    0.008712\n",
       "2013-06-27  0.001504      0           6    0.000000\n",
       "2013-06-28 -0.004822     -1           3    0.004822\n",
       "2013-07-01  0.001017      0           6    0.000000\n",
       "2013-07-02  0.001468      0           6    0.000000\n",
       "2013-07-03 -0.010142     -1           1    0.010142\n",
       "\n",
       "[3900 rows x 4 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Labeled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3900"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(Labeled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<matplotlib.axes._subplots.AxesSubplot object at 0x7f87ca432780>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f87ca360588>], dtype=object)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAADyCAYAAABZCO4fAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIdyMJHYhhHAz/w98jDPQXONu3wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f87d0718eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Labeled[['return','max_return']].cumsum().plot(subplots=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Making a baseline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "from sklearn import linear_model\n",
    "from sklearn.metrics import classification_report,confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "logreg = linear_model.LogisticRegression(C=1e5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "test_size=600"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report,confusion_matrix\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Baseline Fully Connected Feed Forward Neural Network\n",
    "Baseline feed forward network for benchmark."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from  tensorflow.contrib.learn.python.learn.estimators.dnn  import DNNClassifier\n",
    "from tensorflow.contrib.layers import real_valued_column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "test_size=600"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Labeled['tf_class'] = Labeled['multi_class']\n",
    "num_features = len(InputDF.columns)\n",
    "dropout=0.2\n",
    "hidden_1_size = 1000\n",
    "hidden_2_size = 250\n",
    "logpath2 = '/tmp/fc_logs/example'\n",
    "num_classes = Labeled.tf_class.nunique()\n",
    "#NUM_EPOCHS=200 Just speeding up\n",
    "NUM_EPOCHS=100\n",
    "BATCH_SIZE=50\n",
    "learning_rate=0.0001"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "train = (InputDF[:-test_size].values,Labeled.tf_class[:-test_size].values)\n",
    "val = (InputDF[-test_size:].values,Labeled.tf_class[-test_size:].values)\n",
    "NUM_TRAIN_BATCHES = int(len(train[0])/BATCH_SIZE)\n",
    "NUM_VAL_BATCHES = int(len(val[1])/BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3900"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(InputDF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "class Model():\n",
    "    def __init__(self):\n",
    "        global_step = tf.contrib.framework.get_or_create_global_step()\n",
    "        self.input_data = tf.placeholder(dtype=tf.float32,shape=[None,num_features])\n",
    "        self.target_data = tf.placeholder(dtype=tf.int32,shape=[None])\n",
    "        self.dropout_prob = tf.placeholder(dtype=tf.float32,shape=[])\n",
    "        self.summaries = tf.summary.merge_all()\n",
    "        with tf.variable_scope(\"ff\"):\n",
    "            droped_input = tf.nn.dropout(self.input_data,keep_prob=self.dropout_prob)\n",
    "            \n",
    "            layer_1 = tf.contrib.layers.fully_connected(\n",
    "                num_outputs=hidden_1_size,\n",
    "                inputs=droped_input,\n",
    "            )\n",
    "            layer_2 = tf.contrib.layers.fully_connected(\n",
    "                num_outputs=hidden_2_size,\n",
    "                inputs=layer_1,\n",
    "            )\n",
    "            self.logits = tf.contrib.layers.fully_connected(\n",
    "                num_outputs=num_classes,\n",
    "                activation_fn =None,\n",
    "                inputs=layer_2,\n",
    "            )\n",
    "        with tf.variable_scope(\"loss\"):\n",
    "            \n",
    "            self.losses = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=self.logits,labels=self.target_data)\n",
    "            mask = (1-tf.sign(1-self.target_data)) #No credit for flat days\n",
    "            mask = tf.cast(mask,tf.float32)\n",
    "            self.loss = tf.reduce_sum(self.losses)\n",
    "        \n",
    "        with tf.name_scope(\"train\"):\n",
    "          opt = tf.train.AdamOptimizer(learning_rate)\n",
    "          gvs = opt.compute_gradients(self.loss)\n",
    "          self.train_op = opt.apply_gradients(gvs, global_step=global_step)\n",
    "        \n",
    "        with tf.name_scope(\"predictions\"):\n",
    "            self.probs = tf.nn.softmax(self.logits)\n",
    "            self.predictions = tf.argmax(self.probs, 1)\n",
    "            correct_pred = tf.cast(tf.equal(self.predictions, tf.cast(self.target_data,tf.int64)),tf.float64)\n",
    "            self.accuracy = tf.reduce_mean(correct_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tb_writer = tf.summary.FileWriter(logpath2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "if tf.gfile.Exists(logpath2):\n",
    "   tf.gfile.DeleteRecursively(logpath2) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From <ipython-input-49-49e7fe1d783e>:11: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02.\n",
      "Instructions for updating:\n",
      "Use `tf.global_variables_initializer` instead.\n",
      "step - 66 loss - 8476.85726928711 accr - 0.26\n",
      "step - 132 loss - 6955.656211853027 accr - 0.2\n",
      "step - 198 loss - 5977.432769775391 accr - 0.26\n",
      "step - 264 loss - 5110.788524627686 accr - 0.4\n",
      "step - 330 loss - 4245.8157024383545 accr - 0.68\n",
      "step - 396 loss - 3607.078588485718 accr - 0.74\n",
      "step - 462 loss - 3002.383544921875 accr - 0.8\n",
      "step - 528 loss - 2545.9910039901733 accr - 0.82\n",
      "step - 594 loss - 2186.544906616211 accr - 0.9\n",
      "step - 660 loss - 1765.569980621338 accr - 0.86\n",
      "step - 726 loss - 1636.0608968734741 accr - 0.92\n",
      "step - 792 loss - 1484.5583610534668 accr - 0.88\n",
      "step - 858 loss - 1364.3384790420532 accr - 0.72\n",
      "step - 924 loss - 1345.0335240364075 accr - 0.88\n",
      "step - 990 loss - 1055.7422652244568 accr - 0.94\n",
      "step - 1056 loss - 898.1392440795898 accr - 0.86\n",
      "step - 1122 loss - 685.99458360672 accr - 0.92\n",
      "step - 1188 loss - 549.8827750682831 accr - 0.94\n",
      "step - 1254 loss - 454.62357687950134 accr - 0.92\n",
      "step - 1320 loss - 395.6857932806015 accr - 0.96\n",
      "step - 1386 loss - 341.3021819591522 accr - 0.94\n",
      "step - 1452 loss - 308.39223194122314 accr - 0.98\n",
      "step - 1518 loss - 249.54367098212242 accr - 0.94\n",
      "step - 1584 loss - 225.93377578258514 accr - 0.96\n",
      "step - 1650 loss - 219.61940878629684 accr - 1.0\n",
      "step - 1716 loss - 199.0495092868805 accr - 1.0\n",
      "step - 1782 loss - 165.52649900317192 accr - 1.0\n",
      "step - 1848 loss - 144.46484792232513 accr - 0.98\n",
      "step - 1914 loss - 124.52791276574135 accr - 1.0\n",
      "step - 1980 loss - 113.99635702371597 accr - 1.0\n",
      "step - 2046 loss - 104.07442286610603 accr - 1.0\n",
      "step - 2112 loss - 93.1643833592534 accr - 1.0\n",
      "step - 2178 loss - 83.90173250436783 accr - 1.0\n",
      "step - 2244 loss - 75.42272339761257 accr - 1.0\n",
      "step - 2310 loss - 66.30829929560423 accr - 1.0\n",
      "step - 2376 loss - 62.693609081208706 accr - 1.0\n",
      "step - 2442 loss - 57.511074259877205 accr - 1.0\n",
      "step - 2508 loss - 47.96602101624012 accr - 1.0\n",
      "step - 2574 loss - 43.750628527253866 accr - 1.0\n",
      "step - 2640 loss - 40.626658007502556 accr - 1.0\n",
      "step - 2706 loss - 40.862501472234726 accr - 0.98\n",
      "step - 2772 loss - 36.453553944826126 accr - 1.0\n",
      "step - 2838 loss - 40.108197174966335 accr - 1.0\n",
      "step - 2904 loss - 32.8252830170095 accr - 1.0\n",
      "step - 2970 loss - 31.781498232856393 accr - 1.0\n",
      "step - 3036 loss - 29.16536554135382 accr - 1.0\n",
      "step - 3102 loss - 26.81434302404523 accr - 1.0\n",
      "step - 3168 loss - 26.31245904043317 accr - 1.0\n",
      "step - 3234 loss - 23.597407203167677 accr - 1.0\n",
      "step - 3300 loss - 23.54064904153347 accr - 1.0\n",
      "step - 3366 loss - 20.757203612476587 accr - 1.0\n",
      "step - 3432 loss - 20.403653709217906 accr - 1.0\n",
      "step - 3498 loss - 19.45376329123974 accr - 1.0\n",
      "step - 3564 loss - 18.737640742212534 accr - 1.0\n",
      "step - 3630 loss - 17.313241884112358 accr - 1.0\n",
      "step - 3696 loss - 16.359520316123962 accr - 1.0\n",
      "step - 3762 loss - 16.3890398144722 accr - 1.0\n",
      "step - 3828 loss - 17.54691051132977 accr - 1.0\n",
      "step - 3894 loss - 16.91192803159356 accr - 1.0\n",
      "step - 3960 loss - 19.25345172919333 accr - 1.0\n",
      "step - 4026 loss - 22.967422127723694 accr - 1.0\n",
      "step - 4092 loss - 21.353355914354324 accr - 1.0\n",
      "step - 4158 loss - 24.28590626269579 accr - 1.0\n",
      "step - 4224 loss - 14.716766377910972 accr - 1.0\n",
      "step - 4290 loss - 96.8187497407198 accr - 0.98\n",
      "step - 4356 loss - 363.2208235114813 accr - 0.96\n",
      "step - 4422 loss - 808.2296719551086 accr - 0.88\n",
      "step - 4488 loss - 1044.0656297206879 accr - 0.88\n",
      "step - 4554 loss - 1380.471583366394 accr - 0.82\n",
      "step - 4620 loss - 1312.5821695327759 accr - 0.84\n",
      "step - 4686 loss - 697.7750639915466 accr - 0.94\n",
      "step - 4752 loss - 305.68601059913635 accr - 1.0\n",
      "step - 4818 loss - 212.19604140520096 accr - 1.0\n",
      "step - 4884 loss - 97.63792329281569 accr - 0.96\n",
      "step - 4950 loss - 32.9873710013926 accr - 1.0\n",
      "step - 5016 loss - 22.23895899578929 accr - 1.0\n",
      "step - 5082 loss - 15.98843104019761 accr - 1.0\n",
      "step - 5148 loss - 14.92642168328166 accr - 1.0\n",
      "step - 5214 loss - 13.859583123587072 accr - 1.0\n",
      "step - 5280 loss - 11.590983875095844 accr - 1.0\n",
      "step - 5346 loss - 10.680404931306839 accr - 1.0\n",
      "step - 5412 loss - 10.47627243027091 accr - 1.0\n",
      "step - 5478 loss - 10.593862196896225 accr - 1.0\n",
      "step - 5544 loss - 9.822078425437212 accr - 1.0\n",
      "step - 5610 loss - 10.851491765119135 accr - 1.0\n",
      "step - 5676 loss - 8.054766088724136 accr - 1.0\n",
      "step - 5742 loss - 8.2694038040936 accr - 1.0\n",
      "step - 5808 loss - 7.437652374617755 accr - 1.0\n",
      "step - 5874 loss - 7.6057448741048574 accr - 1.0\n",
      "step - 5940 loss - 14.255520699545741 accr - 1.0\n",
      "step - 6006 loss - 13.055462911725044 accr - 1.0\n",
      "step - 6072 loss - 29.328807862475514 accr - 1.0\n",
      "step - 6138 loss - 29.727606911212206 accr - 1.0\n",
      "step - 6204 loss - 10.01085755508393 accr - 1.0\n",
      "step - 6270 loss - 7.216032577678561 accr - 1.0\n",
      "step - 6336 loss - 6.681037124246359 accr - 1.0\n",
      "step - 6402 loss - 5.639110388699919 accr - 1.0\n",
      "step - 6468 loss - 6.4367835177108645 accr - 1.0\n",
      "step - 6534 loss - 6.1414301516488194 accr - 1.0\n",
      "step - 6600 loss - 5.710682256147265 accr - 1.0\n",
      "done training\n",
      "0.1\n",
      "0.02\n",
      "0.2\n",
      "0.14\n",
      "0.1\n",
      "0.24\n",
      "0.1\n",
      "0.06\n",
      "0.12\n",
      "0.18\n",
      "0.12\n",
      "0.12\n",
      "Run the command line:\n",
      "\n",
      "--> tensorboard --logdir=/tmp/fc_logs --port=6008\n",
      "\n",
      "Then open http://0.0.0.0:6008/ into your web browser\n"
     ]
    }
   ],
   "source": [
    "with tf.Graph().as_default():\n",
    "    model = Model()\n",
    "    input_ = train[0]\n",
    "    target = train[1]\n",
    "\n",
    "    config = tf.ConfigProto()\n",
    "    jit_level = tf.OptimizerOptions.ON_1\n",
    "    config.graph_options.optimizer_options.global_jit_level = jit_level\n",
    "\n",
    "    with tf.Session(config=config) as sess:\n",
    "        init = tf.initialize_all_variables()\n",
    "        sess.run([init])\n",
    "        epoch_loss =0\n",
    "        summary_writer = tf.summary.FileWriter(logpath2, graph=tf.get_default_graph())\n",
    "        for e in range(NUM_EPOCHS):\n",
    "            if epoch_loss >0 and epoch_loss <1:\n",
    "                break\n",
    "            epoch_loss =0\n",
    "            for batch in range(0,NUM_TRAIN_BATCHES):\n",
    "                \n",
    "                start = batch*BATCH_SIZE\n",
    "                end = start + BATCH_SIZE \n",
    "                feed = {\n",
    "                    model.input_data:input_[start:end],\n",
    "                    model.target_data:target[start:end],\n",
    "                    model.dropout_prob:0.9\n",
    "                            }\n",
    "                \n",
    "                _,loss,acc = sess.run(\n",
    "                    [\n",
    "                        model.train_op,\n",
    "                        model.loss,\n",
    "                        model.accuracy,\n",
    "                    ]\n",
    "                    ,feed_dict=feed\n",
    "                )\n",
    "                epoch_loss+=loss\n",
    "            print('step - {0} loss - {1} accr - {2}'.format((1+batch+NUM_TRAIN_BATCHES*e),epoch_loss,acc))\n",
    "            # Write logs after every epoch\n",
    "            tb_loss = tf.Summary()\n",
    "            tb_loss.value.add(\n",
    "                tag='ep-loss', simple_value=epoch_loss)\n",
    "            tb_writer.add_summary(tb_loss, (e))\n",
    "            tb_acc = tf.Summary()\n",
    "            tb_acc.value.add(\n",
    "                tag='ep-accuracy', simple_value=acc)\n",
    "            tb_writer.add_summary(tb_acc, (e))\n",
    "            tb_writer.flush()\n",
    "                \n",
    "        \n",
    "        print('done training')\n",
    "        final_preds =np.array([])\n",
    "        final_probs =None\n",
    "        for batch in range(0,NUM_VAL_BATCHES):\n",
    "            \n",
    "                start = batch*BATCH_SIZE\n",
    "                end = start + BATCH_SIZE \n",
    "                feed = {\n",
    "                    model.input_data:val[0][start:end],\n",
    "                    model.target_data:val[1][start:end],\n",
    "                    model.dropout_prob:1\n",
    "                            }\n",
    "                \n",
    "                acc,preds,probs = sess.run(\n",
    "                    [\n",
    "                        model.accuracy,\n",
    "                        model.predictions,\n",
    "                        model.probs\n",
    "                    ]\n",
    "                    ,feed_dict=feed\n",
    "                )\n",
    "                print(acc)\n",
    "                final_preds = np.concatenate((final_preds,preds),axis=0)\n",
    "                if final_probs is None:\n",
    "                    final_probs = probs\n",
    "                else:\n",
    "                    final_probs = np.concatenate((final_probs,probs),axis=0)\n",
    "                tb_acc = tf.Summary()\n",
    "                tb_acc.value.add(\n",
    "                    tag='test_accuracy', simple_value=acc)\n",
    "                tb_writer.add_summary(tb_acc, batch)\n",
    "                tb_writer.flush()\n",
    "\n",
    "        prediction_conf = final_probs[np.argmax(final_probs,1)]\n",
    "        \n",
    "        print (\"Run the command line:\\n\")\n",
    "        print(\"--> tensorboard --logdir=/tmp/fc_logs --port=6008\")\n",
    "        print(\"\\nThen open http://0.0.0.0:6008/ into your web browser\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Result = Labeled[-test_size:].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Result['nn_pred'] = final_preds\n",
    "Result['mod_nn_prod'] = list(map(lambda x: -1 if x <5 else 0 if x==5 else 1,final_preds))\n",
    "Result['nn_ret'] = Result.mod_nn_prod*Result['return']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f878be415c0>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZoAAAD6CAYAAACYhYGNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niLJCY4wxJqL+D1W/PEkGhwGMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f878be21588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Res = (1+Result[-test_size:][['return','max_return','nn_ret']]).cumprod()\n",
    "Res[0] =0\n",
    "Res.plot(secondary_y='max_return')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 73  31  99]\n",
      " [ 59  32 109]\n",
      " [ 76  19 102]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.35      0.36      0.36       203\n",
      "          0       0.39      0.16      0.23       200\n",
      "          1       0.33      0.52      0.40       197\n",
      "\n",
      "avg / total       0.36      0.34      0.33       600\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(confusion_matrix(Result['class'],Result['mod_nn_prod']))\n",
    "print(classification_report(Result['class'],Result['mod_nn_prod']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f87705452e8>"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAADOBJREFUeJzt3XGonXUdx/HPJ2ekuzGV2WFN6fqHGMOhtkNZQpybFiuj\nGYQoJrOM2x9qFoNY/qMQwf5QKySCpeag5UXUmKhYtrxJENKuSne6RLGpW3NXmS6vCLb69sd9Llw3\n5zn3Oec8z873vF8w7jnPfZ7z+z0823vPnnueM0eEAACD70N1TwAA0BsEHQCSIOgAkARBB4AkCDoA\nJEHQASAJgg4ASRB0AEiCoANAEkuqHGz58uUxOjpaatu3335bS5cu7e2EjnHs83Bgn4dDN/s8NTX1\nekSc2m69SoM+OjqqHTt2lNp2cnJSrVartxM6xrHPw4F9Hg7d7LPtlzpZj0suAJAEQQeAJAg6ACRB\n0AEgCYIOAEkQdABIgqADQBIEHQCSIOgAkESld4oCONL03oO6auNDlY+7e9PFlY+J/uIMHQCSIOgA\nkARBB4AkCDoAJEHQASAJgg4ASRB0AEiCoANAEgQdAJIg6ACQBEEHgCQIOgAkQdABIAmCDgBJtA26\n7dNtP2b7WdvP2L6+WH6K7UdtP198Pbn/0wUAHE0nZ+iHJG2IiFWSzpd0je1VkjZK2h4RZ0raXjwH\nANSkbdAjYl9EPFk8fkvSLkkrJa2TtKVYbYukS/o1SQBAe4u6hm57VNJ5kp6Q1IiIfcW3XpXU6OnM\nAACL4ojobEV7RNKfJf0kIu63/WZEnLTg+29ExBHX0W2PSxqXpEajsWZiYqLURGdnZzUyMlJq20HF\nPg+HmQMHtf+d6sddvXJZ9YMWhvE4d7PPY2NjUxHRbLdeR/+nqO3jJd0naWtE3F8s3m97RUTss71C\n0sz7bRsRmyVtlqRmsxmtVquTIY8wOTmpstsOKvZ5ONy2dZtuma7+v/fdfUWr8jHnDeNxrmKfO3mX\niyXdIWlXRNy64FsPSFpfPF4vaVvvpwcA6FQnpwUXSLpS0rTtp4tlN0jaJOke21dLeknSpf2ZIgCg\nE22DHhF/keSjfPvC3k4HAFAWd4oCQBIEHQCSIOgAkARBB4AkCDoAJEHQASAJgg4ASRB0AEiCoANA\nEgQdAJIg6ACQBEEHgCQIOgAkQdABIAmCDgBJEHQASIKgA0ASBB0AkiDoAJAEQQeAJAg6ACRB0AEg\nCYIOAEkQdABIgqADQBIEHQCSIOgAkARBB4AkCDoAJEHQASAJgg4ASRB0AEiCoANAEgQdAJIg6ACQ\nBEEHgCQIOgAkQdABIIm2Qbd9p+0Z2zsXLLvJ9l7bTxe/vtLfaQIA2unkDP0uSWvfZ/lPI+Lc4tfD\nvZ0WAGCx2gY9Ih6XdKCCuQAAutDNNfTrbP+9uCRzcs9mBAAoxRHRfiV7VNKDEXF28bwh6XVJIenH\nklZExLePsu24pHFJajQaayYmJkpNdHZ2ViMjI6W2HVTs83CYOXBQ+9+pftzVK5dVP2hhGI9zN/s8\nNjY2FRHNdustKfPiEbF//rHtX0l68APW3SxpsyQ1m81otVplhtTk5KTKbjuo2OfhcNvWbbplutQf\nxa7svqJV+ZjzhvE4V7HPpS652F6x4OnXJe082roAgGq0PS2wfbeklqTltvdIulFSy/a5mrvkslvS\nd/s4RwBAB9oGPSIuf5/Fd/RhLgCALnCnKAAkQdABIAmCDgBJEHQASIKgA0ASBB0AkiDoAJAEQQeA\nJAg6ACRB0AEgCYIOAEkQdABIgqADQBIEHQCSIOgAkARBB4AkCDoAJEHQASAJgg4ASRB0AEiCoANA\nEgQdAJIg6ACQBEEHgCQIOgAkQdABIAmCDgBJEHQASIKgA0ASBB0AkiDoAJAEQQeAJAg6ACRB0AEg\nCYIOAEkQdABIgqADQBIEHQCSaBt023fanrG9c8GyU2w/avv54uvJ/Z0mAKCdTs7Q75K09rBlGyVt\nj4gzJW0vngMAatQ26BHxuKQDhy1eJ2lL8XiLpEt6PC8AwCKVvYbeiIh9xeNXJTV6NB8AQEmOiPYr\n2aOSHoyIs4vnb0bESQu+/0ZEvO91dNvjksYlqdForJmYmCg10dnZWY2MjJTadlDNHDio/e9UP+7q\nlcuqH7TAca4Ox7la3ezz2NjYVEQ02623pNSrS/ttr4iIfbZXSJo52ooRsVnSZklqNpvRarVKDTg5\nOamy2w6q27Zu0y3TZQ9RebuvaFU+5jyOc3U4ztWqYp/LXnJ5QNL64vF6Sdt6Mx0AQFmdvG3xbkl/\nlXSW7T22r5a0SdIXbT8v6aLiOQCgRm3/nRcRlx/lWxf2eC4AgC5wpygAJEHQASAJgg4ASRB0AEiC\noANAEgQdAJIg6ACQBEEHgCQIOgAkQdABIAmCDgBJEHQASIKgA0AS1X+qPnAMGt34UG1jb1hd29BI\nhjN0AEiCoANAEgQdAJIg6ACQBEEHgCQIOgAkQdABIAmCDgBJEHQASIKgA0ASBB0AkiDoAJAEQQeA\nJAg6ACRB0AEgCYIOAEkQdABIgqADQBIEHQCSIOgAkARBB4AkCDoAJEHQASAJgg4ASSzpZmPbuyW9\nJem/kg5FRLMXkwIALF5XQS+MRcTrPXgdAEAXuOQCAEl0G/SQ9EfbU7bHezEhAEA5jojyG9srI2Kv\n7Y9JelTSdRHx+GHrjEsal6RGo7FmYmKi1Fizs7MaGRkpPddBNHPgoPa/U/24q1cuq37QQl3HeXrv\nwcrHnNc4QUN3nPm9vThjY2NTnfyMsqugv+eF7JskzUbEzUdbp9lsxo4dO0q9/uTkpFqtVrnJDajb\ntm7TLdO9+DHH4uzedHHlY86r6ziPbnyo8jHnbVh9aOiOM7+3F8d2R0EvfcnF9lLbH51/LOlLknaW\nfT0AQHe6+SuyIel3tudf57cR8UhPZgUAWLTSQY+IFyWd08O5AAC6wNsWASAJgg4ASRB0AEiCoANA\nEtW/EXQA1fUe5Q2raxkWwIDiDB0AkiDoAJAEQQeAJAg6ACRB0AEgCYIOAEkQdABIgqADQBIEHQCS\n4E5RYEjV+7801TNunft819qlfR+DM3QASIKgA0ASBB0AkiDoAJAEQQeAJAg6ACRB0AEgCYIOAElw\nYxGOUO8NJ4d0VY3jA4OMM3QASIKgA0ASBB0AkiDoAJAEQQeAJAg6ACRB0AEgCYIOAEkMzI1F03sP\ncsMJAHwAztABIAmCDgBJEHQASIKgA0ASBB0Akugq6LbX2n7O9gu2N/ZqUgCAxSsddNvHSfqFpC9L\nWiXpcturejUxAMDidHOG/mlJL0TEixHxrqQJSet6My0AwGJ1E/SVkl5Z8HxPsQwAUANHRLkN7W9I\nWhsR3ymeXynpMxFx7WHrjUsaL56eJem5knNdLun1ktsOKvZ5OLDPw6Gbff5ERJzabqVubv3fK+n0\nBc9PK5a9R0RslrS5i3EkSbZ3RESz29cZJOzzcGCfh0MV+9zNJZe/STrT9hm2PyzpMkkP9GZaAIDF\nKn2GHhGHbF8r6feSjpN0Z0Q807OZAQAWpatPW4yIhyU93KO5tNP1ZZsBxD4PB/Z5OPR9n0v/UBQA\ncGzh1n8ASGIggj5sHzFg+3Tbj9l+1vYztq+ve05VsH2c7adsP1j3XKpg+yTb99r+h+1dtj9b95z6\nzfYPit/TO23fbfsjdc+p12zfaXvG9s4Fy06x/ajt54uvJ/dj7GM+6EP6EQOHJG2IiFWSzpd0zRDs\nsyRdL2lX3ZOo0M8lPRIRn5R0jpLvu+2Vkr4nqRkRZ2vuzRSX1TurvrhL0trDlm2UtD0izpS0vXje\nc8d80DWEHzEQEfsi4sni8Vua+4Oe+i5c26dJuljS7XXPpQq2l0n6vKQ7JCki3o2IN+udVSWWSDrB\n9hJJJ0r6V83z6bmIeFzSgcMWr5O0pXi8RdIl/Rh7EII+1B8xYHtU0nmSnqh3Jn33M0k/lPS/uidS\nkTMkvSbp18VlptttL617Uv0UEXsl3SzpZUn7JB2MiD/UO6vKNCJiX/H4VUmNfgwyCEEfWrZHJN0n\n6fsR8e+659Mvtr8qaSYipuqeS4WWSPqUpF9GxHmS3laf/hl+rCiuG6/T3F9mH5e01PY3651V9WLu\nrYV9eXvhIAS9o48YyMb28ZqL+daIuL/u+fTZBZK+Znu35i6pfcH2b+qdUt/tkbQnIub/5XWv5gKf\n2UWS/hkRr0XEfyTdL+lzNc+pKvttr5Ck4utMPwYZhKAP3UcM2Lbmrq3uiohb655Pv0XEjyLitIgY\n1dzx/VNEpD5zi4hXJb1i+6xi0YWSnq1xSlV4WdL5tk8sfo9fqOQ/CF7gAUnri8frJW3rxyBd3Sla\nhSH9iIELJF0padr208WyG4o7c5HHdZK2FicqL0r6Vs3z6auIeML2vZKe1Nw7uZ5SwjtGbd8tqSVp\nue09km6UtEnSPbavlvSSpEv7MjZ3igJADoNwyQUA0AGCDgBJEHQASIKgA0ASBB0AkiDoAJAEQQeA\nJAg6ACTxf+5eVHkgnIMrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f877054ec18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = pd.DataFrame(confusion_matrix(Result['multi_class'],Result['nn_pred']))\n",
    "#sns.heatmap(cm.div(cm.sum(1)))\n",
    "Result[Result.multi_class==6]['nn_pred'].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.36      0.15      0.22        52\n",
      "          1       0.12      0.10      0.11        50\n",
      "          2       0.13      0.18      0.15        60\n",
      "          3       0.18      0.05      0.08        62\n",
      "          4       0.15      0.12      0.13        50\n",
      "          5       0.07      0.12      0.09        51\n",
      "          6       0.13      0.37      0.20        60\n",
      "          7       0.00      0.00      0.00        53\n",
      "          8       0.08      0.09      0.09        54\n",
      "          9       0.08      0.02      0.03        56\n",
      "         10       0.18      0.15      0.17        52\n",
      "\n",
      "avg / total       0.14      0.12      0.11       600\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(Result['multi_class'],Result['nn_pred']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f87705057b8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f87703c6eb8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f8770533b70>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f877030e7b8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f87702ef828>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f87702ef860>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f87701cacf8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f877018e240>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f8770119438>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f877012a588>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f8770068048>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f876a1835f8>]], dtype=object)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEcCAYAAAAvJLSTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAHPVJREFUeJzt3X+QXXWd5vH3QyKrAjJgQswQQusYcRAFnBTjjoxAKYrO\nrqBuObCjwq5l3EHcYmdqlozFlDvK1IatXV1rxB9RGHBHQHaEJQuIP6hyXacESRARBgIpBjAxJJ0d\nUHAcNPDsH+d0uN10ujt97/nR5zyvqq6+9/TNuU/lk3z6nO/9nu+RbSIiol/2azpARETUL80/IqKH\n0vwjInoozT8ioofS/CMieijNPyKih9L8IyJ6KM1/DiQdKuk6ST+X9LCkf910pqiOpPMkbZT0lKTL\nm84T1ZP0zyRdWv7/fkLSnZLe2nSuKi1uOsACcQnwS2AZcBxwo6Qf2r6n2VhRkZ8AFwFvAV7QcJao\nx2Lgx8BJwCPA24BrJL3a9kNNBquKcoXvzCQdADwGHGP7/nLbl4Cf2F7baLiolKSLgBW2z2k6S9RP\n0l3An9v+atNZqpBhn9m9Atg90fhLPwRe1VCeiKiYpGUU//c7e3af5j+7A4GfTdn2M+CgBrJERMUk\nPQ/4MnCF7fuazlOVNP/ZPQm8aMq2g4EnGsgSERWStB/wPyg+4zuv4TiVSvOf3f3AYkmrBrYdS4dP\nByP6SJKASykmdrzL9q8ajlSpNP9Z2P45cC3wMUkHSDoReDvF0UF0kKTFkp4PLAIWSXq+pMyM677P\nAr8J/Evbv2g6TNXS/OfmXIopfzuBK4E/zDTPTrsQ+AWwFnhP+fjCRhNFpSQdCXyQYir3o5KeLL/+\noOFolclUz4iIHsqRf0RED6X5R0T0UJp/REQPpflHBJIuk7RT0t0D2w6V9E1JD5TfD2kyY4xWmn9E\nAFwOnDZl21rgFturgFvK59ERtc72WbJkicfGxmp7v5hs06ZNu2wvrfM9U/Nm7UvNJY0BN9g+pny+\nGTjZ9nZJy4Fv2z5qtv2k5s2aa81nvXBF0hHAlyiuejOw3vanJB0KfAUYAx4C3m37sZn2NTY2xsaN\nG2dPH5WQ9HDd75maN2vImi+zvb18/ChFD5hVat6sudZ8LsM+u4E/tn008DrgQ5KOZgGcEo6tvXHP\nV/RX6j88F0MEex0mkLSmvAHOxvHx8RqTxXzN2vxtb7d9R/n4CeBe4HDgdOCK8mVXAGdUFTIiGrGj\nHO6h/L5zby+0vd72aturly6tdWQx5mmfPvAtxwSPB25jnqeEEbFgbADOLh+fDVzfYJYYsTk3f0kH\nAl8Fzrc9aX37mU4JczoY0X6SrgK+Bxwlaauk9wPrgFMlPQC8qXweHTGnlQrLmxt8Ffiy7WvLzTsk\nLR+YCTDtKaHt9cB6gNWrV2choYgWsn3WXn70xlqDRG1mPfIfWOP6XtufGPhRTgkjIhaouRz5vx54\nL/AjSXeW2z5CcQp4TXl6+DDw7moiRkTEqM3a/G1/F9BefpxTwoiIBSjLO0RE9FCaf0RED+W+pLFP\nJD0EPAE8Dey2vbrZRBExH2n+MR+n2N7VdIiImL8M+0RE9FCaf+wrA9+StEnSmqbDRMT8ZNgn9tWJ\ntrdJOgz4pqT7bH9n8AXlL4U1ACtXrmwi4x5Z0TNiejnyj31ie1v5fSdwHXDCNK/JCo8RLZfmH3Mm\n6QBJB008Bt4M3D3zn4qINsqwT+yLZcB1xXJPLAautH1zs5EiYj7S/GPObD8IHNt0jogYXoZ9orPy\nYW+/5RauM+vckX+KHRExuxz5R0T0UJp/REQPpflHJ2X4L2Jmaf4RET2U5h8R0UNp/hERPdSZqZ4Z\n442ImLsc+UdE9NCCPvLP0X5ExPzkyD8ioofS/CMiemjBNP8s0hQRMTpDNX9Jp0naLGmLpLWjChXt\nlZr3T2reTfP+wFfSIuAS4FRgK3C7pA22/24++xs8qn9o3e/N6XVRr1HXPNovNe+uYY78TwC22H7Q\n9i+Bq4HTRxMrWio175/UvKOGmep5OPDjgedbgd/elx3MdhSfo/zWGbrmseCk5h1V+Tx/SWuANeXT\nJyVtnuZlK4FH9vyZiyvIMfd9TsrSoCpyHDni/U1rPjWvPNPe69+WekNqXolpap+aM1zz3wYcMfB8\nRbltEtvrgfUz7UjSuO3VQ2QZmbZkaUuOKTpX87bkgHZlGZCaV6jJLMOM+d8OrJL0Ukn7A2cCG+a5\nr8eHyDFqbcnSlhyDuljztuSAdmWZkJpXq7Es8z7yt71b0nnA14FFwGW275nn7n463xwVaEuWtuTY\no6M1b0sOaFcWIDWvQWNZhhrzt30TcNMIcsx4uliztmRpS45JOljztuSAdmXZIzWvVGNZZLup914w\nJP018CbghcCjwH+x/cVmU0XVJK0CfgT8je33NJ0nqiXp28DrgN3lpm22j2ouUbUWzPIODVsHvMz2\ni4C3AxdJ+q2GM0X1LqEY847+OM/2geVXZxs/pPnPie27bf/jxNPy6zcajBQVk3QmxYdxtzSdJaIK\naf5zJOkzkv4RuA/YzmjGQKOFJL0I+BjwR01nidr9Z0m7JP2tpJObDlOlNP85sn0ucBDwu8C1wFPN\nJooKfRy41PbWpoNErS4AXkZxVfN64H9L6uwZfpr/PrD9tO3vUlzo8odN54nRk3QcxYf7n2w6S9TL\n9m22n7D9lO0rgL8F3tZ0rqos6Ns4NmgxGfPvqpOBMeARSQAHAoskHW37tQ3mivoZUNMhqpIj/1lI\nOkzSmZIOlLRI0luAs8gHgV21nuIX+3Hl1+eAG4G3NBkqqiXp1yS9RdLzJS2W9AfAG4Cbm85WlRz5\nz84UQzyfo/hl+TBwvu35XuIeLVbO6pqY2YWkJ4F/sj3eXKqowfOAi4BXAk9TTOw4w/b9jaaqUC7y\niojooQz7RET0UJp/REQPpflHRPRQmn9ERA+l+UdE9FCtUz2XLFnisbGxOt8yBmzatGmX7aV1vmdq\n3qzUvH/mWvNZm7+ky4B/Aey0fUy57VDgKxRXQj4EvNv2Y7Pta2xsjI0bN872sqiIpIfn+LrUvCPm\nWvNRSs2bNdeaz2XY53LgtCnb1gK32F5FcaXr2n1KF213Oal5dNDY2hubjtAaszZ/298B/mHK5tOB\nK8rHVwBnjDhXNCg1j+i++X7gu8z29vLxo8CyEeWJ9krNIzpk6Nk+LtaH2OsaEZLWSNooaeP4eJZH\n6YLUPGLhm2/z3yFpOUD5fefeXmh7ve3VtlcvXVrrpIMYrdQ8okPm2/w3AGeXj88Grh9NnGix1Dyi\nQ2Zt/pKuAr4HHCVpq6T3A+uAUyU9QHHXo3XVxow6peYR3TfrPH/bZ+3lR28ccZZoidQ8ovuyvENE\nRA+l+UdE9FDvm3+u+IuYmaSHJP1I0p2Ssm5DR+QevhExF6fY3tV0iBid3h/5R0T0UZp/RMzGwLck\nbZK0pukwMRoZ9omI2Zxoe5ukw4BvSrqvXPxvj/KXwhqAlStXNpEx9lGO/CNiRra3ld93AtcBJ0zz\nmizpscCk+UfEXkk6QNJBE4+BNwN3N5sqRqHXwz6Z5hkxq2XAdZKg6BdX2r652UgxCr1u/hExM9sP\nAsc2nSNGL8M+ERE9lOYfEdFDaf4RET2U5h8R0UNp/hERPZTmHxHRQ2n+ERE9lOYfEdFDaf7RS7m6\nO/ouzT8ioofS/CMieijNPyKih9L8IyJ6KM0/IqKHhmr+kk6TtFnSFklrRxUq2is175/UvJvm3fwl\nLQIuAd4KHA2cJenoUQWL9knN+yc1765hjvxPALbYftD2L4GrgdNHEytaKjXvn9S8o4Zp/ocDPx54\nvrXcFt2VmvdPat5Rld/GUdIaYE359ElJm6d52Urgkaqz7I0unvS00SwDqshx5Ij3N62FUHPYU/fG\ncwxIzWugi9uRo9RYzYdp/tuAIwaeryi3TWJ7PbB+ph1JGre9eogsI9OWLG3JMUXnat6WHNCuLANS\n8wo1mWWYYZ/bgVWSXippf+BMYMM89/X4EDlGrS1Z2pJjUBdr3pYc0K4sE1LzajWWZd5H/rZ3SzoP\n+DqwCLjM9j3z3N1P55ujAm3J0pYce3S05m3JAe3KAqTmNWgsy1Bj/rZvAm4aQY4ZTxdr1pYsbckx\nSQdr3pYc0K4se6TmlWosi2w39d4LiqQzgY9SfEDzKHCO7f/bbKqogqQnp2x6AfAZ2x9uIk/UQ9IY\n8BngnwNPAX8DnG97d4OxKpPlHeZA0qnAxcC/AQ4C3gA82GioqIztAye+gJcAvwD+Z8OxonqfAcaB\n5cBxwEnAuY0mqlDlUz074s+Bj9m+tXz+nNkO0VnvAnYCOcvrvpcCn7b9T8Cjkm4GXtVwpsrkyH8W\n5eXtq4Gl5domWyV9WtILms4WtTgb+JIzPtoH/x34fUkvlHQ4xZIWNzecqTJp/rNbBjwP+FfA71Kc\nDh4PXNhkqKiepCMpTv2vaDpL1OI7wDHAzyiuZN4I/K9GE1UozX92vyi//6Xt7bZ3AZ8A3tZgpqjH\ne4Hv2v77poNEtSTtR3GUfy1wALAEOITis75OSvOfhe3HKI4CBk/7MwTQD+8jR/19cSjFTL5P237K\n9v8D/ooOH+Sl+c/NXwEflnSYpEOA/wDc0HCmqJCk36FYwCyzfHqgPKP/e+DfSVos6dcoPu+5q9lk\n1Unzn5uPU1zmfj9wL/AD4C8aTRRVOxu41vYTTQeJ2ryT4kPecWAL8CuKA71OykVeERE9lCP/iIge\nSvOPiOihNP+IiB5K848IJF0maaekuwe2HSrpm5IeKL8f0mTGGK00/4gAuBw4bcq2tcAttlcBt5TP\noyNqne2zZMkSj42N1fZ+MdmmTZt22V5a53um5s3al5qXSxrfYPuY8vlm4GTb2yUtB75t+6jZ9pOa\nN2uuNa91Vc+xsTE2btxY51vGAEkP1/2eqXmzhqz5Mtvby8ePUqxzNavUvFlzrXmGfSKAsbU3Nh2h\n1cpVTfc6TCBpjaSNkjaOj4/XmGw4fa57mn9E7M2OcriH8vvOvb3Q9nrbq22vXrq01pHFmKc0/4jY\nmw0Uy1xQfr++wSwxYmn+EYGkq4DvAUeVNyx6P7AOOFXSA8CbyufREbmNY0Rg+6y9/OiNtQaJ2uTI\nPyKih9L8IyJ6KM0/IqKH0vwjSn2e8x39k+YfEdFDaf4RET2U5h8R0UNp/hERPTTURV6SHgKeAJ4G\ndttePYpQ0V6peUQ3jOIK31Ns7xrBfmLhSM0jFrgM+0RE9NCwzd/AtyRtkrRmuhcstHW+M9d7Vp2r\neUQfDdv8T7R9HPBW4EOS3jD1BVnnu3NS84gOGKr5295Wft8JXAecMIpQ0V6peXRNX8/25938JR0g\n6aCJx8CbgbtHFSzaJzWP6I5hZvssA66TNLGfK23fPJJU0VapeURHzLv5234QOHaEWaLlUvOI7shU\nz4jovT6O+6f5R0T0UJp/REQPpflHRPRQmn9ERA+l+UdE9FCaf0RED6X5R0T0UJp/REQPpflHRPRQ\nmn9ERA8N1fwlnSZps6QtktaOKlS0V2reP6l5Nw2zpPMi4BKKm3ocDZwl6ehRBYv2Sc37JzXvrmGO\n/E8Atth+0PYvgauB00cTK1oqNe+f1Lyjhmn+hwM/Hni+tdwW3ZWa909q3lHD3MxlTsqbfE/c6PtJ\nSZunedlK4JGqs8zRSl3ciixV/J0cOeL9TWuB1XxSDl3cYJLUvC7T5mio9o3VfJjmvw04YuD5inLb\nJLbXA+tn2pGkcdurh8gyMm3J0pYcU3Su5m3JAe3KMiA1r1CTWYYZ9rkdWCXppZL2B84ENsxzX48P\nkWPU2pKlLTkGdbHmbckB7coyITWvVmNZhrmN425J5wFfBxYBl9m+Z567++l8c1SgLVnakmOPjta8\nLTmgXVmA1LwGjWUZaszf9k3ATSPIMePpYs3akqUtOSbpYM3bkgPalWWP1LxSjWWR7abee8GQ9JsU\nc51/CxgH/sT2dc2milEpj2zPAV4NXGX7nIGfvZGi9iuB24BzbD/cQMyIkcryDrOQtBi4HrgBOJRi\nRsNfS3pFo8FilH4CXARcNrhR0hLgWuDPKGq/EfhK7ekiKpAj/1lIOga4FTjI5V+WpG8At9n+s0bD\nxUhJughYMXHkX05fPMf275TPDwB2Acfbvq+xoBEjkCP/+RFwTNMhonKvAn448cT2z4Et5faIBS3N\nf3abgZ3An0h6nqQ3AycBL2w2VtTgQJ47G+NnwEENZIkYqcqv8F3obP9K0hnAXwIXUIz7XgM81Wiw\nqMOTwIumbDsYeKKBLDFPkl5JsR7RxLIU24ANtu9tMNOJFOsm3W37G01kyJH/HNi+y/ZJtl9s+y3A\ny4DvN50rKncPcOzEk3LM/zfK7TEDSacNPD5Y0qWS7pJ0paRlNea4gGIxOlH8n/1++fiqOpenlvT9\ngccfAD5NcQb50aaWyc4HvnMg6TXA/RS/LM8FPgS80naO/jugnNG1GPgoxfIFHwB2A4dQjPH/W+BG\n4GPAG2y/rqGoC4akO2y/tnz8ReBR4AvAO4GTbJ9RU477gVfZ/tWU7fsD99heVVOOH9g+vnx8O/A2\n2+PlAcWttl9dR45BOfKfm/cC2ynG/t8InJrG3ykXAr8A1gLvKR9faHsceBfwF8BjFKfpZzYVcgFb\nbftC2w/b/iQwVuN7PwP8+jTbl5c/q8t+kg6R9GJgUflva2ISwe4ac+yRI/+IGDlJW4FPUAyxnAe8\nbGCq9F22X1NTjtMohlge4NmlqVcCLwfOs31zTTkeovhlI8DA621vl3Qg8F3bx9WRY1A+8I2IKnyB\nZ2dFXQ4sAcYlvQS4s64Qtm8uL8g8gckf+N5u++kac4zt5UfPAO+oK8egHPlHROeVHzLvaf62d/Q5\nB6T5R0RF2jDFUtJxwOcopuhupRh2WUGxlPK5tu+oKcfxwGfLHBP3Q6g9x6RMaf4xlaQjgC8ByyjG\nJ9fb/pSk/0QxE2a8fOlHyhUfIyYpp1ieRTHNcmu5eQXFB+ZX215XU447gQ/avm3K9tcBn7d97PR/\nsps5Jr13mn9MJWk5sNz2HZIOAjYBZwDvBp60/V8bDRit16Iplg/s7b0kbbH98j7lGFTrB75Llizx\n2NhYnW8ZAzZt2rTL9tLZXmd7O8XUVmw/Iele5nnT7tS8WXOteQUmplhOXf667imWX5N0I8WZ7MRs\nnyOA9wG1zPRpWY49am3+Y2NjbNy4sc63jAGS9nkdekljwPEUa9m/HviwpPdRLHPxx7Yfm+nPp+bN\nmk/NR+R84BZJ006xrCuE7X8v6a0897OHS+ocsmxLjkGZ6jlHY2tv5KF1v9d0jFqVc5C/Cpxv+2eS\nPgt8nOJzgI8D/43i6tepf24NxX0PWLlyZX2BKzS29sZJz/v2b2FftWWKZZnla8DX6nzPNueYkCt8\nY1qSnkfR+L9s+1oA2ztsP237GYp53CdM92dtr7e92vbqpUubGHGINrD9jO1bbX+1/Lq17sY/k/Ig\npXFN5Ujzj+eQJOBS4F7bnxjYvnzgZe8A7q47Wyx8km5oOkNJTQcoNZIjwz4xnddTrGf0o3KKGsBH\ngLPKedMGHgI+2Ey8WOA+UOebzXC9wef7mGNCmn88h+3vMv3RSOb0x9DK2WS1mHK9wcSyyisolnSu\n83qDVuQYlOYfESMn6WDgTymuDzmM4mxxJ3A9sM724zVFeT/TX2/wCYr7MtTVdNuSY4+M+UdEFa6h\nWAb7ZNuH2n4xcEq57Zoac7RlSee25NgjR/4RUYUx2xcPbrD9KHCxpOdMD65QK643aFGOPdL8I6IK\nD0v6j8AVEytXlitansOzza9ybbneoC05BqX5R0QVfp/izmj/R9Jh5bYdwAaKNaJqU16Xcmud79nm\nHBPS/CNi5MplPy4ov5D0dtsbmk0Vg/KBb0TU4aKmA8RkQx35l/elfAJ4Gthte/UoQkVE57Tlatoo\njWLY5xTbu0awn4jortw4pGUy7BMR0UPDNn8D35K0qS0r5EVExOyGHfY50fa2cirXNyXdZ/s7gy/o\nwtruE2u593FN/5je1PX9IWv8z2JH0wFisqGO/G1vK7/vBK5jmvXds7Z7RNg+tekMMdm8m7+kA8qb\neyPpAODNZH33iIgFYZhhn2XAdcV9P1gMXGm7kRsRR0TEvpl387f9IHDsCLNERERNMtUzIqKH0vwj\nInoozT8ioofS/CMieijNPyKih9L8IyJ6KM0/IqKH0vwjInoozX8fTbegV0TEQpN7+EaMyNQDg6zy\nGW2WI/+IiB5K84+I6KFhb+B+GvApYBHwRdvrRpIqWis1n7sMA0Wbzbv5S1oEXAKcCmwFbpe0wfbf\njSpctEtqPpz8Mog2GWbY5wRgi+0Hbf8SuBo4fTSxoqVS84iOGGbY53DgxwPPtwK/PVycaLnUfIRm\nmzY825lBziRiGJVP9Ry8gTvwpKTN07xsJfBI1VnmaNYsurgdOebhyBHvb1oLrOZtyQFTsuzrv7O9\nvL6WmsfCM0zz3wYcMfB8RbltEtvrgfUz7UjSuO3VQ2QZmbZkaUuOKTpX87bkgHZlie4bZsz/dmCV\npJdK2h84E9gwz309PkSOUWtLlrbkGNTFmrclB7QrS3TcMPfw3S3pPODrFNP+LrN9zzx399P55qhA\nW7K0JcceHa15W3JAu7JExw015m/7JuCmEeSYcYigZm3J0pYck3Sw5m3JAe3KEh0n201niIiImmV5\nh4iIHkrzj4jooTT/iIgeSvOPiOihXt/MRdIrKdamObzctA3YYPve5lKBpBMp1tG52/Y3mszSNW2s\neeodTaj9yL9cEnji8cGSLpV0l6QrJS2rMccFFAuTCfh++SXgKklr68pRZvn+wOMPAJ8GDgI+WneW\nKqTmz8nR6XrHwlD7VE9Jd9h+bfn4i8CjwBeAdwIn2T6jphz3A6+y/asp2/cH7rG9qo4c5Xv+wPbx\n5ePbgbfZHpd0AHCr7VfXlaUKqflzcnS63rEwND3ss9r2ceXjT0o6u8b3fgb4deDhKduXlz+r036S\nDqE4E1tkexzA9s8l7a45S9VS837VO1qqieZ/mKQ/ojjdPliS/OzpR53DUOcDt0h6gGeXKV4JvBw4\nr8YcAAcDmyj+Tixpue3tkg4sty10qflkXa93LABNNP8vUIxvAlwOLAHGJb0EuLOuELZvlvQKig/a\nBj/8u93203XlKLOM7eVHzwDvqDFKVVLzyTnG9vKjrtQ7FoDeL+9QfuC4pxHY3pEs3daWv+e25Ih+\naqT5t2G6naTjgM9RnIJvpTjdXkGxrO65tu+oMcvxwGfLLBPr4zeSpSqp+aQcna93tF8Ts30uAM6i\nmHK3tdy8gmJt+Kttr6spx53AB23fNmX764DP2z62jhxty1KF1LydOaLfmmj+bZlu98De3kvSFtsv\nryNH27JUITVvZ47otyY+8G3LdLuvSboR+BLPzvw4AngfcHONOdqWpQqpeTtzRI81ceR/GsUVjdNO\nt7Nd2z9+SW9l+nHoUdysZMFmGbXUvL05or+a+sB3P1owxTLqk5pHtEvvp3pOR9Ia2624pV6bsnRZ\nW/6e25Ijuq9VSzpLuqHpDKU2XWXZpiwjl5o/R1tyRMe16sh/4jL3Gt+v8bnnbcxSp77WvC05or9a\ndeRfcxNoxfK+bctStz7WvC05ot+amO1zMPCnwBnAYYCBncD1wDrbj9eUoxVzz9uWpQqpeTtzRL81\nceR/DfAYcLLtQ22/GDil3HZNjTkm5p5P1cSSzm3KUoXUvJ05oseaOPLfbPuoff1ZBTnaNPe8NVmq\nkJq3M0f0WxPN/xvAt4ArJlYxLFc3PAc41fabaszSmrnnbcoyaql5e3NEfzXR/A8B1lLMdDis3LwD\n2ABcbPsfag0UlUvNI9qn8amekt5ue0OjIaJWqXlE89rQ/O+y/ZpGQ0StUvOI5rVhnn+uaOyf1Dyi\nYW1o/u25xDjqkppHNKwNzT8iImqW5h8R0UNtaP47mg4QtUvNIxrW+GyfiIioXxuO/CMiomZp/hER\nPZTmHxHRQ2n+ERE9lOYfEdFD/x/4Qfi+0XAiXwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8770497cf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Result.hist(by='multi_class',column='return',sharex=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Attention Recurrent Neural Network with MultiRNNCell LSTMs\n",
    "In this section we'll make an rnn model that learns to take the past into account as well\n",
    "This model is based on AttentionCellWrapper a novel approach described in the following publication https://arxiv.org/pdf/1409.0473v7.pdf titled \"NEURAL MACHINE TRANSLATION BY JOINTLY LEARNING TO ALIGN AND TRANSLATE\"\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Defining an rnn Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "from tensorflow.contrib.layers.python.layers.initializers import xavier_initializer\n",
    "\n",
    "# Parameters\n",
    "learning_rate = 0.0007\n",
    "display_step = 1\n",
    "logpath = '/tmp/rnn_logs/example'\n",
    "rnn_num_units=100\n",
    "multirnn_num_cells=2\n",
    "attn_win_size=30\n",
    "first_layer_size=1000\n",
    "batch_size=50\n",
    "num_of_train_batches = int(len(train[0])/batch_size)\n",
    "num_of_test_batches = int(len(val[1])/batch_size)\n",
    "#NUM_EPOCHS=200\n",
    "NUM_EPOCHS=20\n",
    "beta=0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "class RNNModel():\n",
    "    def __init__(self):\n",
    "        global_step = tf.contrib.framework.get_or_create_global_step()\n",
    "        self.input_data = tf.placeholder(dtype=tf.float32,shape=[batch_size,num_features], name='InputData')\n",
    "        self.target_data = tf.placeholder(dtype=tf.int32,shape=[batch_size], name='LabelData')\n",
    "        self.dropout_prob = tf.placeholder(dtype=tf.float32,shape=[], name='DropoutData')\n",
    "        self.summaries = tf.summary.merge_all()\n",
    "       \n",
    "        def lstm_cell():\n",
    "            return tf.contrib.rnn.BasicLSTMCell(rnn_num_units, forget_bias=0.0, state_is_tuple=False,\n",
    "            reuse=tf.get_variable_scope().reuse)\n",
    "\n",
    "        \n",
    "        def makeGRUCells():\n",
    "            at_cell = lstm_cell\n",
    "            stacked_cell = tf.contrib.rnn.MultiRNNCell([at_cell() for _ in range(multirnn_num_cells)], state_is_tuple=False)\n",
    "            attn_cell =tf.contrib.rnn.AttentionCellWrapper(cell=stacked_cell,attn_length=attn_win_size,state_is_tuple=False)\n",
    "            return attn_cell\n",
    "        \n",
    "        self.gru_cell = makeGRUCells()\n",
    "        self.zero_state = self.gru_cell.zero_state(1, tf.float32)\n",
    "        self.start_state = tf.placeholder(dtype=tf.float32,shape=[1,self.gru_cell.state_size], name='StateData')\n",
    "\n",
    "        with tf.variable_scope(\"feedfor\",initializer=xavier_initializer(uniform=False)):\n",
    "            droped_input = tf.nn.dropout(self.input_data,keep_prob=self.dropout_prob)\n",
    "            \n",
    "            layer_1 = tf.contrib.layers.fully_connected(\n",
    "                num_outputs=first_layer_size,\n",
    "                inputs=droped_input,\n",
    "                \n",
    "            )\n",
    "            layer_2 = tf.contrib.layers.fully_connected(\n",
    "                num_outputs=rnn_num_units,\n",
    "                inputs=layer_1,\n",
    "            )\n",
    "        \n",
    "        split_inputs = tf.reshape(droped_input,shape=[1,batch_size,num_features],name=\"reshape_l1\") # Each item in the batch is a time step, iterate through them\n",
    "        split_inputs = tf.unstack(split_inputs,axis=1,name=\"unstack\")\n",
    "        states =[]\n",
    "        outputs =[]\n",
    "        with tf.variable_scope(\"rnn\",initializer=xavier_initializer(uniform=False)) as scope:\n",
    "            state = self.start_state\n",
    "            for i, inp in enumerate(split_inputs):\n",
    "                if i >0:\n",
    "                    scope.reuse_variables()\n",
    "                \n",
    "                output, state = self.gru_cell(inp, state)\n",
    "                states.append(state)\n",
    "                outputs.append(output)\n",
    "        self.end_state = states[-1]\n",
    "        outputs = tf.stack(outputs,axis=1) # Pack them back into a single tensor\n",
    "        outputs = tf.reshape(outputs,shape=[batch_size,rnn_num_units])\n",
    "        self.logits = tf.contrib.layers.fully_connected(\n",
    "            num_outputs=num_classes,\n",
    "            inputs=outputs,\n",
    "            activation_fn=None\n",
    "        )\n",
    "            \n",
    "        with tf.variable_scope(\"loss\"):\n",
    "            self.penalties =    tf.reduce_sum([beta*tf.nn.l2_loss(var) for var in tf.trainable_variables()])\n",
    "            self.losses = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=self.target_data, logits=self.logits)\n",
    "            self.loss = tf.reduce_sum(self.losses + beta*self.penalties)\n",
    "            tf.summary.scalar('mloss', self.loss)\n",
    "        \n",
    "        with tf.name_scope(\"train_step\"):\n",
    "            opt = tf.train.AdamOptimizer(learning_rate)\n",
    "            gvs = opt.compute_gradients(self.loss)\n",
    "            self.train_op = opt.apply_gradients(gvs, global_step=global_step)\n",
    "        \n",
    "        with tf.name_scope(\"predictions\"):\n",
    "            probs = tf.nn.softmax(self.logits)\n",
    "            self.predictions = tf.argmax(probs, 1)\n",
    "            correct_pred = tf.cast(tf.equal(self.predictions, tf.cast(self.target_data,tf.int64)),tf.float64)\n",
    "            self.accuracy = tf.reduce_mean(correct_pred)\n",
    "            tf.summary.scalar(\"maccuracy\", self.accuracy)\n",
    "            \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def variable_summaries(var):\n",
    "   \"\"\"Attach a lot of summaries to a Tensor (for TensorBoard visualization).\"\"\"\n",
    "   with tf.name_scope('summaries'):\n",
    "     mean = tf.reduce_mean(var)\n",
    "     tf.summary.scalar('mean', mean)\n",
    "     with tf.name_scope('stddev'):\n",
    "       stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean)))\n",
    "     tf.summary.scalar('stddev', stddev)\n",
    "     tf.summary.scalar('max', tf.reduce_max(var))\n",
    "     tf.summary.scalar('min', tf.reduce_min(var))\n",
    "     tf.summary.histogram('histogram', var)\n",
    "\n",
    "train_writer = tf.summary.FileWriter(logpath)\n",
    "test_writer = tf.summary.FileWriter(logpath)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "if tf.gfile.Exists(logpath):\n",
    "   tf.gfile.DeleteRecursively(logpath) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training the RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:<tensorflow.contrib.rnn.python.ops.core_rnn_cell_impl.BasicLSTMCell object at 0x7f86a08baf28>: Using a concatenated state is slower and will soon be deprecated.  Use state_is_tuple=True.\n",
      "WARNING:tensorflow:<tensorflow.contrib.rnn.python.ops.core_rnn_cell_impl.BasicLSTMCell object at 0x7f86a08bae48>: Using a concatenated state is slower and will soon be deprecated.  Use state_is_tuple=True.\n",
      "WARNING:tensorflow:<tensorflow.contrib.rnn.python.ops.rnn_cell.AttentionCellWrapper object at 0x7f86a08ba668>: Using a concatenated state is slower and will soon be deprecated.  Use state_is_tuple=True.\n",
      "WARNING:tensorflow:From <ipython-input-76-b82cfa8620ab>:13: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02.\n",
      "Instructions for updating:\n",
      "Use `tf.global_variables_initializer` instead.\n",
      "step - 0 loss - 7776.2886962890625 acc - 0.16\n",
      "step - 1 loss - 7627.6904373168945 acc - 0.22\n",
      "step - 2 loss - 7631.982322692871 acc - 0.2\n",
      "step - 3 loss - 7607.545478820801 acc - 0.18\n",
      "step - 4 loss - 7590.102462768555 acc - 0.14\n",
      "step - 5 loss - 7487.799835205078 acc - 0.18\n",
      "step - 6 loss - 7460.214218139648 acc - 0.2\n",
      "step - 7 loss - 7389.6628494262695 acc - 0.2\n",
      "step - 8 loss - 7374.651885986328 acc - 0.16\n",
      "step - 9 loss - 7291.082458496094 acc - 0.2\n",
      "step - 10 loss - 7264.15608215332 acc - 0.24\n",
      "step - 11 loss - 7266.151580810547 acc - 0.24\n",
      "step - 12 loss - 7230.570152282715 acc - 0.2\n",
      "step - 13 loss - 7181.594749450684 acc - 0.24\n",
      "step - 14 loss - 7107.202896118164 acc - 0.22\n",
      "step - 15 loss - 7135.000396728516 acc - 0.2\n",
      "step - 16 loss - 7039.515487670898 acc - 0.2\n",
      "step - 17 loss - 7101.137413024902 acc - 0.28\n",
      "step - 18 loss - 7046.2463455200195 acc - 0.28\n",
      "step - 19 loss - 7044.586639404297 acc - 0.26\n",
      "Test: step - 0 acc - 0.08\n",
      "Test: step - 1 acc - 0.1\n",
      "Test: step - 2 acc - 0.3\n",
      "Test: step - 3 acc - 0.22\n",
      "Test: step - 4 acc - 0.1\n",
      "Test: step - 5 acc - 0.06\n",
      "Test: step - 6 acc - 0.14\n",
      "Test: step - 7 acc - 0.22\n",
      "Test: step - 8 acc - 0.02\n",
      "Test: step - 9 acc - 0.1\n",
      "Test: step - 10 acc - 0.08\n",
      "Test: step - 11 acc - 0.06\n",
      "Run the command line:\n",
      "\n",
      "--> tensorboard --logdir=/tmp/rnn_logs \n",
      "\n",
      "Then open http://0.0.0.0:6006/ into your web browser\n"
     ]
    }
   ],
   "source": [
    "# Construct model and encapsulating all ops into scopes, making\n",
    "# Tensorboard's Graph visualization more convenient\n",
    "with tf.Graph().as_default():\n",
    "    model = RNNModel()\n",
    "    input_ = train[0]\n",
    "    target = train[1]\n",
    "# XLA\n",
    "    config = tf.ConfigProto()\n",
    "    jit_level = tf.OptimizerOptions.ON_1\n",
    "    config.graph_options.optimizer_options.global_jit_level = jit_level\n",
    "\n",
    "    with tf.Session(config=config) as sess:\n",
    "        init = tf.initialize_all_variables()\n",
    "        sess.run([init])\n",
    "        # op to write logs to Tensorboard\n",
    "        summary_writer = tf.summary.FileWriter(logpath, graph=tf.get_default_graph())\n",
    "        loss = 2000\n",
    "        \n",
    "        for e in range(NUM_EPOCHS):\n",
    "            state = sess.run(model.zero_state)\n",
    "            epoch_loss =0\n",
    "            for batch in range(0,num_of_train_batches):\n",
    "                start = batch*batch_size\n",
    "                end = start + batch_size \n",
    "                feed = {\n",
    "                    model.input_data:input_[start:end],\n",
    "                    model.target_data:target[start:end],\n",
    "                    model.dropout_prob:0.5,\n",
    "                    model.start_state:state,\n",
    "                    }\n",
    "                (_,loss,acc,state) = sess.run(\n",
    "                    [\n",
    "                        model.train_op,\n",
    "                        model.loss,\n",
    "                        model.accuracy,\n",
    "                        model.end_state\n",
    "                    ]\n",
    "                    ,feed_dict=feed\n",
    "                )\n",
    "                epoch_loss+=loss\n",
    "                # Write logs at every iteration\n",
    "                tb_loss = tf.Summary()\n",
    "                tb_loss.value.add(\n",
    "                    tag='loss', simple_value=epoch_loss)\n",
    "                train_writer.add_summary(tb_loss, (e*num_of_train_batches+batch))\n",
    "                tb_acc = tf.Summary()\n",
    "                tb_acc.value.add(\n",
    "                    tag='accuracy', simple_value=acc)\n",
    "                train_writer.add_summary(tb_acc, (e*num_of_train_batches+batch))\n",
    "                train_writer.add_summary(model.summaries, (e*num_of_train_batches+batch))\n",
    "                train_writer.flush()\n",
    "                \n",
    "            print('step - {0} loss - {1} acc - {2}'.format((e),epoch_loss,acc))\n",
    "            # Write logs at every iteration\n",
    "            tb_loss = tf.Summary()\n",
    "            tb_loss.value.add(\n",
    "                tag='ep-loss', simple_value=epoch_loss)\n",
    "            train_writer.add_summary(tb_loss, (e*num_of_train_batches+batch))\n",
    "            tb_acc = tf.Summary()\n",
    "            tb_acc.value.add(\n",
    "                tag='ep-accuracy', simple_value=acc)\n",
    "            train_writer.add_summary(tb_acc, (e*num_of_train_batches+batch))\n",
    "            train_writer.flush()\n",
    "#            print('step - {:4d} loss - {:.9f} acc - {:.9f}'.format((e),epoch_loss,acc))\n",
    "        final_preds =np.array([])\n",
    "        for batch in range(0,num_of_test_batches):\n",
    "            start = batch*batch_size\n",
    "            end = start + batch_size \n",
    "            feed = {\n",
    "                model.input_data:val[0][start:end],\n",
    "                model.target_data:val[1][start:end],\n",
    "                model.dropout_prob:1,\n",
    "                model.start_state:state\n",
    "                }\n",
    "            (acc,preds,state) = sess.run(\n",
    "                [\n",
    "                    model.accuracy,\n",
    "                    model.predictions,\n",
    "                    model.end_state\n",
    "                ]\n",
    "                ,feed_dict=feed\n",
    "            )\n",
    "            print('Test: step - {} acc - {}'.format(batch,acc))\n",
    "#            print(acc)\n",
    "            assert len(preds) == batch_size\n",
    "            final_preds = np.concatenate((final_preds,preds),axis=0)\n",
    "#           summary = tf.summary.merge_all()\n",
    "            tb_acc = tf.Summary()\n",
    "            tb_acc.value.add(\n",
    "                tag='test_accuracy', simple_value=acc)\n",
    "            test_writer.add_summary(tb_acc, batch)\n",
    "            test_writer.flush()\n",
    "        \n",
    "        print (\"Run the command line:\\n\")\n",
    "        print(\"--> tensorboard --logdir=/tmp/rnn_logs \")\n",
    "        print(\"\\nThen open http://0.0.0.0:6006/ into your web browser\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RNN Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Result['rnn_pred'] = final_preds\n",
    "Result['mod_rnn_prod'] = list(map(lambda x: -1 if x <5 else 0 if x==5 else 1,final_preds))\n",
    "Result['rnn_ret'] = Result.mod_rnn_prod*Result['return']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[13  8  7  2  7  2  2  2  0  1  8]\n",
      " [ 3  5  5  2 18  4  1  5  2  3  2]\n",
      " [ 2  7 10  6 25  2  1  2  1  4  0]\n",
      " [ 4  3  8  7 28  4  1  0  0  3  4]\n",
      " [ 0  4  4  7 20  6  4  2  0  3  0]\n",
      " [ 2  2  6  1 29  2  1  1  0  4  3]\n",
      " [ 0  6  6 10 26  6  2  1  0  1  2]\n",
      " [ 2  2  5  2 27  6  2  2  1  2  2]\n",
      " [ 1  6 10  4 21  3  6  0  0  3  0]\n",
      " [ 2  2 10  9 20  4  2  1  0  4  2]\n",
      " [ 8  5  5  3  9  4  2  1  1  5  9]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.35      0.76      0.47       203\n",
      "          0       0.37      0.08      0.13       200\n",
      "          1       0.38      0.21      0.27       197\n",
      "\n",
      "avg / total       0.37      0.35      0.29       600\n",
      "\n",
      "[[154  11  38]\n",
      " [153  16  31]\n",
      " [139  16  42]]\n"
     ]
    }
   ],
   "source": [
    "print(confusion_matrix(Result['multi_class'],Result['rnn_pred']))\n",
    "print(classification_report(Result['class'],Result['mod_rnn_prod']))\n",
    "print(confusion_matrix(Result['class'],Result['mod_rnn_prod']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f86a1938128>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd8AAAFKCAYAAABcq1WoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f86a19224a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = pd.DataFrame(confusion_matrix(Result['multi_class'],Result['rnn_pred']))\n",
    "#sns.heatmap(cm.div(cm.sum(1)))\n",
    "Result[Result.multi_class==6]['rnn_pred'].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f8691dfe2e8>"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIUAAAIaCAYAAAC+rP0BAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xd8VfX9x/HXufve5GbfJCTsvQUH4gJE0LpnLe6qdbRq\nq7Z2+Kuj1lHbat111qq11aoo4kYRB4hM2bIhQPa+N7kr957fH4FoSkICBC4k7+fj4UPuPd/zPZ+b\nb0B55zsM0zRNRERERERERESkS7EkugAREREREREREdn/FAqJiIiIiIiIiHRBCoVERERERERERLog\nhUIiIiIiIiIiIl2QQiERERERERERkS5IoZCIiIiIiIiISBdkS3QBO5SV+Tusr/R0D1VV9R3WnxwY\nNK6dk8a189GYdj4a085J49o5aVw7H41p56Gx7JwOhnH1+bytXuuUM4VsNmuiS5B9QOPaOWlcOx+N\naeejMe2cNK6dk8a189GYdh4ay87pYB/XThkKiYiIiIiIiIjIrikUEhERERERERHpghQKiYiIiIiI\niIh0QQqFRERERERERES6IIVCIiIiIiIiIiJdkEIhEREREREREZEuSKGQiIiIiIiIiEgXpFBoL9XV\nBZg3b26iyxARERERERER2S0KhfbS6tXfKhQSERERERERkYOOLdEFHOjee286c+fOoby8jCOPPIq5\nc2djGBaOO24CF1xwMQ8++Gfq6+vo0aMny5cvZcKEEzjmmOOYPfsLZs36hCuuuJq77roNt9vDueee\nzyOPPMAZZ5zNnDlfEolEePjhJ/B4khL9MUVERERERESkizloQqH/zlzH/G9L29XWajWIxcw22x0x\nOJvzJ/Zvs11JSTF33HE39913F0888RwAP/3plRx//CQuvPASNmxYz5lnnsPy5UtbvH/t2tW88cY7\npKam8be//Zlevfpw0UWXcccdv2PBgvmMGzehXZ9LRERERERERKSjHDShUCINGTKUVatWsHXrFm64\n4RoA6uvrKC4ubNf9+fndSU1Na3p9yCGjAfD5cqirC3R8wSIiIiIiIiIibThoQqHzJ/Zv16weAJ/P\nS1mZv8OebbPZsdnsHHXUMfz61//X7Fph4bamXxuG0fTrhoaGZvd/n9Vqbfq1abY9o0lERERERERE\npKNpo+l2GjRoCIsWLSQUCmGaJg899FfC4RCGYRCLxQDweJKoqCgHYOnSbxJZroiIiIiIiIjILh00\nM4USLTc3l/PPv4DrrrsKi8XCuHETcDpdDBo0mCeffBSfL5sf/OAU/vCH3zNr1kwGDBiY6JJFRERE\nRERERFplmAfI+qWOXO7V0cvH5MCgce2cNK6dj8a089GYdk4a185J49r5aEw7D41l53QwjKvP5231\nmpaPiYiIiIiIiIh0QQqFRERERERERES6IIVCIiIiIiIiIiJdkDaaFhEREREREZEuIx6NEN6yhdDG\nDUSKi0mfNBlHTm6iy0oIhUIiIiIiIiIi0qmFCjZT9MRjAESrKiEW++5iPEbOJT9OTGEJplBIRERE\nRERERDo1/9dziZaXYU324urZC1efPrh696Xk5RcJrl2b6PISRqHQAeCbbxbRq1dv0tMzEl2KiIiI\niIiISKfhXzAfa1ISwTWrwWqlz/1/xeJ0Nl2v/Wo29atWEgsEsCYnJ7DSxNBG0weAd999m6qqykSX\nISIiIiIiItJpREpKKHrqCQqfeJTQ5k24evVuFggBuPoPACC4rnG2UDwUpODeuyh+7hligcB+r3l/\n00yhNrz33nSWLv2GqqpKtmwp4MILL+Gll/7JGWeczZw5XxKJRHj44SfweJJavP/666+mb99+AFx7\n7fXce+8f8Pv9xGIxbrzxFqqqKvjii1ls3LiBu+/+M7m5XXNzKxEREREREZGOVPn+u2CaxINBANyD\nBu/UxjNwEJU0hkLJo0bjX7iA0IYNhDZsoG7VCnIvu4KkESP3c+X7z0ETCk1d9w6LS5e1q63VYhCL\nm222G509gnP6n9Zmu/Xr1/Hkk/9g69Yt3HHHrcRiMXr16sNFF13GHXf8jgUL5jNu3IRW7+/btx9n\nnXUe//znsxx55NGcfvpZbNy4gYcf/isPPfQE/fsP5Oabf61ASERERERERKQDREqKqf1qNnafj4aa\nGsxIBM+gQTu1c/XpC1Yr1Z/OxLDbCa5dA0Da5JOonvkx2x5+kNRxE/Cd/yMsLvf+/hj73EETCiXS\n8OEjsVqt+HzZ1NU1Th875JDRAPh8OU3vtWbIkOEALFu2lOrqKj788D0AwuHQPqxaREREREREpOsx\nGxooeuYpiMXIOveHhLdtwz9vLu7+A3dqa3G5yP3xlZS99gqV06cB4Orbj+wfXUDq0cdQ9OzT1Hw+\ni/qVK8i96hrc/fp3WJ2x+nqKn30Kz9BhpJ0wGcMwOqzv9jpoQqFz+p/Wrlk9AD6fl7Iyf4c922q1\nNv3aNM1W32uN3W5r+vdNN93C8OGdd+qZiIiIiIiISCJVvP0W4U0b8R51NN7Dx+A9HLLOPLvV9ilH\nHU3SIaPY9sjfCK1bS8pRRwPg7NGTnr+/g8rp06h8/122Pfw3et1xF/bMzL2qr3rmxwTXrcXVfwB1\nS5dQt3QJkaIisi+8GON7WcP+oI2m96OhQ4fz+eezANi4cQOvvPIvACwWC7FYLIGViYiIiIiIiOy5\nWDDItscfIbB4UULrqP92FZXvv4vd5yP7wkvafZ/V46H7Tb8i/8abSR1/fNP7FrudrHPOI/viy4jX\n11H87FNtTgxpS+m//4V/3teUv/E6GAaOvHxqPvuUbY8+TDwU3Ku+d5dCof3ovPN+xLZtW/jZz37C\n/fffzahRhwIwatSh/P73v2HDhvUJrlBERERERERk91VOn0bd4kVUfzYrIc8PF26joaaG4ueeBsMg\n96prsbp3bw8gi9NJ0vCRGJado5LUcePxDB9BcO0aosVFe1ynGY9/9+twCM/gofS89fd4ho+gfvlS\nttx/LzF/x618astBs3wsUU455fSmX3s8Hl5/fXqz69dff+Mu73/ssae/d38S99zzl53aXHHF1Vxx\nxdV7WamIiIiIiIjI/hcpKqTqkxkAhLcU7Pfn186dQ/GzT2M4HJiRCJlnnYN7+yngHcUwDJIPGU39\n8mUE167F0S1vj/qJlpc3e51y9NFYXG7yb7iRkn+9QO0Xn1P9+SwyTz29lR46lkKhDlBcXMzdd9++\n0/ujRx/GlVdek4CKRERERERERPY90zQpfeXfEIth9aYQq6mmoaYGW2rqPn922WuvYjgc1C9fBoaB\nGY3iHjyEjFPatx/x7nIPGABAcN0aUseN36M+IsWFAGScejrOnr1IHt24gsiwWsk68xxqv/ic0H5c\nRaRQqAPk5uY2mxEkIiIiIiIi0hXULfmG+hXL8QwdhqtvXyrfmU5465Z9HgoF166h6sP3m14njTyE\nnEsvx5KU1OLyr47gyMvH4nYTXLt2j/uIFDUuPXP27IX3sMObXbOlpWHLyCC0YQOmae6X08j2+Ct1\n77338qMf/YgpU6awdOnSZteKioq44IILOO+887j99p1n0IiIiIiIiIjIwS0ejVD26r/BasU35SKc\nPXoBEC7Y90vIKt5uPD7ecDgASB1/PLa0NCx2+z57pmGx4O4/gGhZKdsefYiK6dOIhcO71Udk+35E\njm7dWrzu6tOXmL+WhsqKva63PfYoFJo3bx6bN2/m1Vdf5Z577uGee+5pdv1Pf/oTV1xxBa+//jpW\nq5XCwsIOKVZEREREREREEqOhtpa6lSua/il/43WiZWWkT5yEMy8PZ8+ewL7dV8g0TcqnvUn9qhV4\nhg4j/+c3kXHKaSSNGLnPnvl97oGDgMYZUhXT3mTRz35O7by57T6RLFpcDBYLdl92i9dd2/dCCm3Y\n0DEFt2GPlo999dVXTJo0CYB+/fpRU1NDIBAgOTmZeDzOwoULefDBBwG44447Oq5aEREREREREUmI\nbQ89QLhgc7P3rN4UMk4/EwB7ZhYWt5vQxn2z/CkejVDy/D/wz5uLPctH9kWX4MjJxTN4SIc+Z1fS\nJk7C4nTi6j8A//x5VM/4kOKnn6R65ifk/fT6NpfNRYqKsPt8rc5ocvXpC0Bow3q8R4zp8Pr/1x6F\nQuXl5QwbNqzpdUZGBmVlZSQnJ1NZWUlSUhL33XcfK1as4PDDD+eXv/xlhxUsIiIiIiIiIvtXg7+W\ncMFmHHl5eMeMbXo/afgIrB4P0Li8Kmn4CPzz51G/cgVJw4Z36PMLH3uE0Pp1uPr1J++6n2NLSemw\n/tvL4nSSNrFxkoyrZy/6nHkKq//+DHVLvqH6kxlknXNeq/eGC7cRC/hxDxrUahtXr96NbbdubXdN\npmkSWDgfDAN7Ria2zCysXm+7QrkO2Wj6+9OkTNOkpKSESy+9lPz8fK6++mpmzZrFhAkTdtlHeroH\nm83aEeUA4PN5O6yv/amwsJDy8nJGjtw/U98ONgfruMquaVw7H41p56Mx7Zw0rp2TxrXz0Zh2Hgfz\nWFasWwlA7vHj6XF+68GHe8p5LJk/D/+H75Lua33WjDs/H0faztfrC7ZQPucrfMcdizs/j3gkQmD9\nBjY/+DDh0lKyxh3LgBuuw7J9L6HE8zL81zfx9UWXES/csssx3vzRYgDyJhy3i3ZeNqWkYNZWtfv7\npeTjTyh68olm71lcLnImT6LnlB/u8t49CoWys7MpLy9vel1aWorP5wMgPT2dvLw8em5fS3jUUUex\ndu3aNkOhqqr6PSmlRT6fl7Iyf4f1tz/NmDGLYLCebt36JLqUA87BPK7SOo1r56Mx7Xw0pp2TxrVz\n0rh2PhrTzuNgH8vSBd8AEM/vvevPkZqNZ8gwaleuYPmtt7XazJ7lo/fd92HYbJjxeONMm5kfU7+q\nMXyqWrMeV6/elE99vfEGwyDzzLNJP+0MKmrCwO5t8Lyv+HxeqoImdp+P2rXrWPHok9SvXEGv2+/C\nsH0XuZimSclnX2A4HMT7Dt7l19CamkaotITS0tpdzvYpf2sqwdXfEikuwnA4yDzzbBqqqmiorCC0\ncSNF098hUFLOyP+7pdU+9igUOuaYY3j00UeZMmUKK1asIDs7m+Tk5MYObTZ69OjBpk2b6N27NytW\nrODUU0/dk8ccEN57bzpz585h+fKlOBwORo4cxbp1axk4cBC//e1t3HPPnWRmZrFmzbeUlBRz++13\nM2jQ4Bb7eu65pygs3EZRUSGPPvoUzz77JEuXfkM8HuOcc87n8MOP5B//eBqbzUZOTi7HHjt+P39a\nERERERERkZ0F16zGsNlw9Wl7AkP2pZdRO2c2mPEWr4c2bKB+5QpK//0SdcuXEfP7MaNRANyDhxAp\nKqR+1UpCmzZiOBx4hg0nffJJeAa2vuwq0Zy9+hBYMI/qT2dCLEakuAhn9x5N14OrvyVaUoJ3zJFY\nnM5d9mVLTye8pYB4sB6rJ6nFNvFwmKqPPsCMRADIOuc8Mk46+bvr0Sibb7uVumVLdv2s9n7A7zv0\n0EMZNmwYU6ZMwTAM7rjjDqZOnYrX62Xy5Mnceuut/Pa3v8U0TQYOHMjEiRP35DHNlL32Cv4F89vV\ndrPVQizW8jff93kPPwLfD6e02a6kpJjHHnuayy6bwjXXXEd6egZnn30Kfn9jsheNRnnwwcd4663X\n+eCDd1sNhQAaGqI88cSzLFmymJKSYh5//BkikQhXXHEx48ZN4OSTTyMtLU2BkIiIiIiIiCRcpKiQ\nmtlfEt5SgLv/ACz2tpdtOXzZZJ15dqvXo5UVbPzdr6n5/DMMmw1HXj6uPn1IO/4EnN17UPLSC9R8\n9inx+nq8R46l21XXduRH2idcvXoTWDAPYjEAQps3N4VC0coKip7+OxgGacef0GZftvQMABqqqloN\nhepXLseMRPAeORb3wEGkHnNcs+sWux3PiBHUfDpz189qs5pW/OpXv2r2evDg74KQXr168Z///GdP\nuz7gDBkyFMMwyM/vQWZmFgBZWT7q6gIAHHLIaAB8vhxWrlzRRl+NG3QvW7aEFSuWcf31VwNgmvFm\nS/JEREREREREEiWwdAmV771DaN1aACyeJNJP7phVQPaMTFLHTaDms0/J/ck1eA8/otn1pOEjqPns\nUwCSDz2sQ565r7l69272OlywGY45lng4TOFjjxCrrcU35ULcAwa22ZctY0coVIkzv3uLbQKLFgGQ\ndsKJuPv2bbFN0rB9GArtb74fTmnXrB7o+LWaNlvjUXFWa/ONsHdssP3997+/6XZL7NuPnbPb7Zx2\n2plccsnlHVaniIiIiIiIyN5qqK2l8PFHIB7HM3QYqceOI2n06HbNEmqv7AsuIvOMM7F5dz5BzDNk\nCFit208zOzgOYXL26gWGgS0jg4bKSsIFmzFNk+LnnyNcsJmUY8eRdsLkdvVl3zFTqLKqxetmQwOB\nJd9gS0/fKYz6Ps/gwWDd9YFelnZVJB1u6NDhzJ79BfF4nHA4zN/+9mcALBYLse3TzURERERERET2\nF7OhgeCGDfjnzoFYDN/5U+h+8y2N++B0YCAEjcfXtxQIAVhcbrKnXET2hRe3uf/OgcLqSaLbtT+j\n2zXX4ejWjVBBAZXTpxFYMA/3gIHkXHxpu46Ih8Y9hQCiVZUtXg+uXUO8vo6kUYdiWFqPdSwuN94j\nxuz6We2qSDrciBGHMHr0YVxzzeWAydlnNx4TN3z4CO6++07S0tI58cSTd9WFiIiIiIiISIcpfeXf\n1MyaiWG3g9VKytijE1ZL2vF7vzfx/uY9rHEZnLNHLyKFhVS8/Ra2zEy6/fT6ZieRteW7PYVaDoUC\nixc2Pq8dS+u6/eSaXV43zLbWO+0nHbnc62A/6k9apnHtnDSunY/GtPPRmHZOGtfOSePa+WhMO48D\nfSwbamvZ+OubMRsaAEg+7HDyfnp9gqs68LU0rtWffUrpSy9g92WT/4ubcOR2260+4+Ew6667Bs+w\n4XS/qfl+zqZpsvHXvyQeDtPvwYfbFTb5fN5Wr2mm0D5w6623UFtb0+y95ORk/vSnBxNUkYiIiIiI\niEjrqj/9BLOhgYzTzsCMRkgdf/DN1DlQpB43Hmf3Hrh698FoY0+fllicTiweD/UrlrP5D7eTd8Mv\nsGdkAhBav46GqkpSjjpmt2YftUah0D5w771/SXQJIiIiIiIiIu3m/2oOFrebjB+cgsXlSnQ5BzXD\nYsHdr//e9gJAeEsBgUULSZ90IvFwmJIX/wlAytHH7GX/jbTRtIiIiIiIiEgnF49GqF+zusVr0bIy\nouVleAYPVSB0gMg840w8w4YDjRtLA5S9+h8ihdtIm3gCniFDO+Q5CoVEREREREREOrnyN15n65/v\no27F8p2u1a9eBYB78OD9XZa0In3SieTf+Ets6ekE16zBP38eNZ/PwtmjB1k//FGHPUehkIiIiIiI\niEgnFg+HqZ39BQC1c2bvdL3+28ZQyDO4Y2afSMcwDAP3gIHE/LUU/+MZDKeTbtf8DIvd0WHPUCgk\nIiIiIiIi0kk1VFdRPfNj4sEg0HiceTwUIh4OE962lcDiRdSvXIHVm4IjLy/B1cr/cg8YCIAZjZJ9\n4SW7fZJZW7TRdBseeeQBVqxYjmEY/OIXv2TIkGGJLklERERERERkl0zTpOqjDyif+jrEYmAYpBx1\nDLVzvmTDr39JvL6uWXvvUUdjGEaCqpXWeIYOA6uVlCPHdtjm0t+nUGgXFi9eyNatW3jqqefZtGkj\n9913F0899XyiyxIRERERERFpVayujuLnn6Xum8VYU1PxHnYEzl698AweQv2qFWC14urVC7svu/Gf\nbB+eocMTXba0wJGTS9/7/4o1JXWfhHYKhXZh4cL5HHfcBAB69+6D319LXV2ApKTkxBYmIiIiIiIi\n8j/CWwoIFxVSPvV1GsrLcQ8eQrerrsWWmtrUpu9f/pbACmVP2NLS913f+6znDjZn5no2fFvarrYW\nq4V4LN5mu76Dszl6Yr9Wr1dUVDBo0He7r6elpVNRUaFQSERERERERA4otV/PpfiZJxtfGAYZp59J\n5ulnYli0lbC07qAJhQ4EpmkmugQRERERERE5yJmmSc3ns6hbtrRd7S0OJ8mHHUby6MN2CnnCWwqo\nW76cyvffwXA6yTztDNwDB+Hu139flC6dzEETCh09sd8uZ/V8n8/npazMv9fPzMrKoqKioul1eXk5\nWVlZe92viIiIiIiIdE3xaJTip58ksHjhbt3nnzeXjFNOI+uc85reM+NxCp94lGhZGQA5l15O6rjx\nHVqvdG4HTSiUCGPGjOW5557irLPOZfXqb8nKysLjSUp0WSIiIiIiInIQMmMxip9pDITcgwaT8+Mr\nsCa1/XfMhopKtj70V6o//YSMU07D4nIBUP/tKqJlZSSNGk3m6Wfi6tV7H38C6WwUCu3CiBGHMGjQ\nEK699goMw+Dmm3+T6JJERERERETkIGTG45S88DyBRQtxDx5C/i9uwmJ3tOteqyeJtAkTqZj2JlUz\nPiR51GgAqmd+DEDGyacqEJI9olCoDT/96Q2JLkFEREREREQOYqZpsvG556md8yXO3n3Iv/7n7Q6E\ndkgdN4GKd96mYtqbVEx7s+l9R14err7t22pF5H8pFBIRERERERHZhyrfeZuKd97DkZdP9xt/icXl\n3u0+bKmpdLvqGoJr1373pgHeMWMxDKMDq5WuRKGQiIiIiIiIyD4S2rSRirffwpntI//mX2FNTt7j\nvryHj8F7+JgOrE66OoVCIiIiIiIiIh0sUlpKcPUqqmZ8BKZJ/xuuI5qWnuiyRJpRKCQiIiIiIiLS\ngcpee5WqD99vep06bjxpI0dQVuZPYFUiO1MoJCIiIiIiItKBAosXYXG5yDr3fJy9euPq0yfRJYm0\nSKGQiIiIiIiISAeJ1dcTLS3BM2QoacdPTHQ5IrtkSXQBIiIiIiIiIgeimN9PQ3XVbt0T3lIAgLNn\nr31RkkiHUigkIiIiIiIi8j9M02TLA39m0223Eikpafd94c2bAXD2UigkBz6FQiIiIiIiIiL/I7xp\nI5GtW4gHgxQ+8SjxcLhd94UKNgHg6tl73xUn0kEUComIiIiIiIhs11BdRdWMD6ma8SEArj59iWzb\nSskLz2Oa5i7vNU2T8KZNGE4X9uzs/VGuyF7RRtMiIiIiIiLSpcXDYfzz5xHatBH/vLnE6+sBsHpT\n6P6r37D1wb/gnzcXV58+pE8+qcU+Yn4/xS8+T6S4CM+w4RgWzcGQA59CIREREREREemSIqWl1Hz6\nCTWzv2gKgiwuF+kn/oC6pUtIHT8Bi9NJt2uvo+CPd1D22qs4e/bCM2hws37qViyn+B/PEqupxj1w\nEDmXXZGIjyOy2xQKiYiIiIiISJcRD4fZ+uBfaKisoKG6GkwTqzeFjNPOIHn0oThyu2FxOvGdP6Xp\nHnt6Ot2uvY6tf72foiefoOftf8Dm9VLyrxeIlpQQXLsGrFayzv0h6SedrFlCctBQKCQiIiIiIiJd\ngmmaVH8yg9D6dVhTUnAPGEjq+AkkH3o4Frt9l/d6Bg7Cd/4Uyl75N0VPPk7mGWdR++UXADjy8sm9\n4ipcvXvvh08h0nEUComIiIiIiEin0VBbi8XlwuJwNL1nNjRQ/MI/qFuyBOIxLJ4ket/9J6wez271\nnXbCZOpXf0vd4kVUvDUVgPwbb8YzbASGYXTo5xDZHzSnTURERERERDqFBn8tm/7vN5T+64Wm98x4\nnKJnnsT/1RzioSDxUIiMH5y824EQgGEYpJ8wGYDQxg0YDgfuQYMVCMlBSzOFRERERERE5KAVCwTA\nasXqdhNYuIB4MEjgm28w43EMi4Wy114lsHAB7oGDyPvp9YQLt+EeMHCPn+ceOAhbRiYNlRV4Bg/B\nYne0fZPIAUqhkIiIiIiIiBx0omVlVH7wLrWzv8SWmUmv2+/CP+9rAOL1dYS3FBBav47qGR/iyMsj\n7/qfY/Uk7XRy2O4yLBZSjjqaynenkzRiZEd8FJGEUSgkIiIiIiIiB7x4OMy2hx7A4nJh9XqpnfsV\nxONYXC6iJSUUP/c0wbVrMGw2zIYGKt6aSt3yZVi9KeT//CasnqQOqyXj5FOxpaaScuy4DutTJBEU\nComIiIiIiMgBr2rGh41Hv2/n6JZHxqmnkTzqUDb/8U4CixYCkHHq6VRMe5O6ZUsx7HbybrgRe5av\nQ2uxuFykTZzUoX2KJIJCIRERERERETmgNVRXUfn+u1i9Xrpd8zPMWAzPkKEYlsazk/J/fiN1y5bh\nyMvDM3gI/gXziRRuI/cnV+Pu2zfB1YscuBQKiYiIiIiIyAEh5vcTKtiMxenE4nRhuJxY7A4K//44\nZjhM5vkX4Bk8ZKf7HDm5OHJym153u/paYrW1eIYM3Z/lixx0FAqJiIiIiIhIwpmmybbHHia0fl2L\n171jjiR13Ph29eXM7w75HVmdSOekUEhEREREREQSLrR+HaH163D27kPS0GHEQ6HGf8IhbKlpZJ37\nQwzDSHSZIp2KQiERERERERFJuKqPPwLAd975LS4RE5GOZ0l0ASIiIiIiItK11Xz5OYEF83H26Il7\n0OBElyPSZSgUEhERERERkYQJbthAyQvPY0lOJvfKq7RETGQ/UigkIiIiIiIiCRNYMA9Mk9wfX4mz\ne49ElyPSpSgUEhERERERkYSpW7oEw+HAM2xYoksR6XK00bSIiIiIiIjsd7G6OiLFRUSKi0gaNRqL\n3ZHokkS6HIVCIiIiIiIist+Y8TjVn8yg/M03MCMRAJJGHJLgqkS6JoVCIiIiIiIisl+Et22j5IXn\nCG3YgCU5GUduNxqqqkgeNSrRpYl0SQqFREREREREZJ+IlpdR9NTfsft8APgXLoBYDO+YsfguuBCb\nNyXBFYp0bQqFREREREREZJ+ofP89Qhs3ENq4AQBbRibZF11C8iGaGSRyIFAoJCIiIiIi8j1mPE68\nvh5rcnKiSzko1M6ZTWjzJmzp6aRNmIjF5QIg5vdT+9VsbFlZ5N9wE4bNht3nw7DoEGyRA4VCIRER\nERERke3i4TDbHn6QcMFm+vzlb1jd7kSXdEBrqK6m+PlnwTQBqPr4I3znno/3yLFUfvg+ZiRC+gmT\ncebnJ7hiD+neAAAgAElEQVRSEWmJQiERERERERHAbGig8O+PE1yzGoBocRHWPn0TXNWBLbB4EZgm\nGaedDoaFqg/eo/i5p6n88H0iW7dgy8oi9bhxiS5TRFqhUEhERERERLo8Mx6n+LmnqV++FIsniXh9\nHZHSUlwKhXYpsGgBAKnjJmDPyCT12OMoe+2/BBbMw7DZyPvp9Vhcmm0lcqBSKCQiIiIi0sXFAgGi\nFeW4evVOdCkJYZompf96Ef/8ebgHDCRt8kkUPfEo0dKSRJd2QAp8s5i6ZUvAhPrV3+Lq0xd7RiYA\n9sws8q79GcENP8CwWLrs95TIwUKhkIiIiIhIF1f0zJPUr1xBr9v/gLNHz0SXs0vxUAj/gnm4+vTF\nmd99r/sz43HKXvk3NZ/PwtmjJ3k33EgsEAAgWlq61/13NvFolOJ/PEO8vr7pPe9RR+/Uzt1XM6xE\nDgYKhUREREREurDQxg3Ur1gOQOUH79HtqmsTXFHrwtu2svWv9xPz+3EPGkyPW367132Wv/Yq1TM/\nxpGXT/5Nv8Lq8WBxOsFqJaKZQjupX76UeH09qeMnkD7pRLDZsGf5El2WiOwhnQUoIiIiItKFVb73\nLgDWZC/+eV9Tt2wp5vaTpA4kO5Z4xfx+AEIb1mPG43vVZzwUovqzT7FlZdHj17/DlpICgGG1Ys/M\n0kyhFtTO/QqA1PHH4+iWh8OXjWEYCa5KRPaUQiERERERkS4qHokQWPoNju49yL70x2CabHv4QTbf\n8X9Uz5pJPBwGoG7ZUsqnvk71rJmYsdjePTMUpKG2drfv8381h+DaNSSPPgzvmLGY0SjRivK9qiWw\n9BvMSISUo47Bmpzc7Jo9O5uYv5ZYMLhXz+hMYsEgdUu+wdEt74BfZigi7aPlYyIiIiIiXVR4SwHE\nYngGDcZ76GHYbr2d6k9m4F8wj9J/vUj5m2+QPvkkKt6a2nRP7VdzyL/+F1i93t16lmmalP77JWq/\n+BwzFiNp1GhyLr4MW2pqm/dGSoop/fdLGE4nvh9NwT9/Pv55cwkXFODwZe/2597BP38eAN7Dx+x0\nzZGdTT0QLS3Bqs2SgcaTxsyGBrxHjtXsIJFOQjOFRERERES6qNCG9QC4tm8K7O7bl25XXUPf+x8g\n47QziIdCjYGQ1Uq3a39G8uFjCK1fR8nLL+32s8JbCqj5dCbW1FScvXpTt3gR2x56gNj3NixuSTwa\noejJJ4iHQuRcchn2LB/Onj2b+txToY0bqF+2FEdePs78/J2u27NzgMZAqrOomD6N4uefI1pexqY7\nb8O/aOFu3e//ei4A3iPH7ovyRCQBFAqJiIjIASNaWUFww/pdLk+JlJYSKti8H6sS6bxCGzcA4Ord\n/KQoW1oaWWedQ/7Pb8KWmUn2hZfgPXwM3a6+Fle//gQWzCOweNFuPcs/72sAfOdfQM//u53UcRMI\nbymg8PFHiEcjrd5X9sq/CW8pIHXceFLGNp5y5ezeA9jzUCi4fh1bH/wLZixG5plntdhmx/Ko8MaN\ne/SMA039mtVUTHuT2tlfsPVvfyWydQvVMz5s9/0NNdXUr1qJq2+/vZqdJSIHFi0fExERkf3CjMXA\nNDFsrf/vx7aHHiBSWIjF48EzZChJw0aQfMQYrG430HgU8pb77yFWU4OrX3/STzqZ5FGjMSz6OZfI\nnght3IAlORl7dst/yU8aNpy+9z/Q9NqwWMi57AoK7rqdkpdfxD1oEFZPUpvPMeNx/PO+xuJ2kzRy\nJIZhkH3xpcQCfgKLFlL8zFN0u/a6nX4v1379FTWfzcLZowe+KRc1vW9LTcWamkpo00ZigcBO+wHt\nSnDtWrY9/ADxSITcq67Be9gRLbZz9e4DFgvB9eva1W+kuKgxZDMMMCyNy6ssBvYsX2NfCWQ2NFD6\nrxcaa7NYiJY0nqoWXLeWhppqbKlpbfbhn/c1mCbesUft63JFZD9SKCQiIiL7VKy+ntDGDY2nBgXr\nyTjxB6RNnITF5WrWrqGmmkhhIbb0DLBaCCxcQGDhAmrnzaXHr34DQGDhfGI1NdgyMgmtX0fRE49i\n92WTc9nleAYPScTHEzloNdTUEC0rwzN85G7tD+PMyyPjtDOoeGsq5a//l5xLL2/sr7oKw+HE6vE0\nf05tLVUfvk9DZQUpRx+Dxe4AGgOm3KuuYdtDDxJYtJDSl18k++LLmmqJlBRT8uI/sbhcdLv2OiwO\nR7N+k0YcQu2Xn7Pptt/R/eZf4+zRo83a61d/y7ZH/obZ0EC3a37aaiAEYHE6cfbsRWjzJuLRSFPd\nLYnV17Pl/vuI+VveQDvlmOPIueSyXYbi+1LVRx8QKSwkdfzxWJxOqj76gKQRI6lbtpTA4kWkTZjY\nZh+1X88Fi6XF/ZdE5OClUEhERET2mbplS9n2yN/ANMEwsDidlE99naqPPiT9pJPxjj0KW2oqhsVC\ncH3j3iap4yeQcerpREuKKf7nPwh+u4rwtq0487tT/elMMAy63/IbzGgDVTM+oHbObEpffonef7w3\nwZ9W5OARj0YpfvYpAJKGDt3t+zN+cAr+BfOp+fwzvGPG4urXn8133o7hdJBz6eXUr1pJeEsB4S0F\nxLafNGbP8pFxymnN+rHYHeRd93O2/uVP1Hw2C+/hYzAcDmxpaVR+8B5mOEzOT67GkZO7Uw05F1+K\nIzub8qmvs+2xh+j5+zuweVNarbl+1Uq2PfoQZixG3rXXkTz60DY/p7t/f8KbNhLetBn3gAHNrsWj\nERoqKwGo/mQGMX8tKcceh6tPPzDjYJqY8Ti1X35B7ewvcPXtR9r4CW0+s6NFykqpmD4NqzeFrHPP\nw+JykzpuPIbdwcbf/JKqjz/CPWBQi/sqNfVRUkx400Y8w0dgS2n9aywiBx+FQiIiIrLPVH/2KZhm\n414gRx+LIy+P6o9nUDXjQ8rf+C/lb/wXu89Hr7vuIbR+LQDu/gMwDANHbjcyTvoBhevWUvPZp7j6\nDSC0fh1JI0Y27WeRe9kVxGpqqFu6hGhFOfbMrER+XJEDUv2qlQSWLCbr3POx2O0E166l5F8vENm2\nlaTRh5J2wuTd7tOw2cj98RUU3HMXJS88T9Z55xML+CEA2/7216Z2tsxMkkaNxtWnL+knTN5phiCA\n1eMh++JL2XLf3ZT99z+Et27FnuUj5q/FlpGBd0zLmxobNhsZp5yGGYtRMe1Niv7+ON1vvqXF2Th1\ny5dR+PgjYJrkXXcDySNHtetzuvsNoPrjGQQWL8Se7SMWCFC/cgV1K5YTXLMaM/LdXki2rCyyL7pk\npxlFyaNGs/E3vyKwcP5+D4VM06T05Zcwo1F8P76iaamfI7cbAKnHn0DNp59Q8Mc7yDz7XNInn9Ti\nctwdG0ynHKmlYyKdjUIhERER6XCV778LFgt1y5bi7NGjaXkJQOYZZ5E2aTI1sz7Fv3AB4c2bqF+1\niuC6dWCxNNt7I2nkKKxpaVTP+hQ+m4XF7Sbrh1OaPStpxEjqli6hbtnSdi2BEOlK4tEIxc8/S0Nl\nJWYshhltoPbLzwFIHTce35SLMKzWPerb1bsP6ZNOpGrGh5S+9AIA3rFHEQvUkTZ+PO5BQ3ZaStZq\nX3374ezdh/Cmxk2do2WlAKSdMLnNPcMyTj2d8NYtBBYuoPSVf5Nz8aXNrtetWE7hYw+DYZB3/S9I\nGj6i3Z/RPWAAGAZVH31A1UcfNLvmyO+Oq1dvsFowDAspxx7X4hIze2YWzt59qF/97U77H8Xq6qie\n+TFmrIHMM8/p8GPeAwsXUL98GZ4hw1oM13IuuoSkocMoefGflL/2KnVLl5B7xU92Ctjrv10FhkHS\nqNEdWp+IJJ5CIREREelQwQ3rKX/jtabX3hZ+smz1JJFxymm4+g9g65/vI7BgPuHNm3B279FsJoFh\ntZJ7+U8of/MNosVFdLv6pzjz8pr1lTR8JIBCIZEW1Hw6s3GJk2FQ8+lMABzde5Bz8aW4+w9o4+62\nZZx6OjVffk4s4Mea7CX38p/sUchkGAbpk0+k+Jmn8B45lvqVKxuXYx19TNv3WizkXnEVW0pLqJk1\nE2f37k1/FoQLt1H05OMA5N1wI0lDh+1WXba0dHr8+ncEFi8iWlmBxeHAPWgwnqHDsaent7sf76GH\nEd60kapPZpA++UQAqmZ8RPXHHxEPBhvbHHEkzvzuxAIBIkVFgIlpmtSUuqmvqsfu82HPyNyt+mu+\n+AyA7IsuaTVwSh59KK7+/Sl58Z/ULV7E5jtvI+OU03D26IkjNxdbZhbhLQXYc3KaNv0Xkc5DoZCI\niIh0qMr33wXA6vUSD4Xwjjmy1bbufv2xJCdTO+dLADzDhu/UJmnYcJJaeH8Hu8+HI7cb9StXENq0\nCVfv3nv3AUQ6kaqPP8LidtPt2uson/o6KUeObZx9s4ezg/6XNTmZtEknUjl9GkkjD9mrfr1jxmJL\nTcPdfwChggIaysta3EuoJRank7zrf0HBH/9A6X9exjN0OPUrl1P++n+Jh0LkXnXNbgdCO7gHDMQ9\nYOAe3btD8uFjqHj7LSqnT6Py3ekYdjtmOIw12UvyYYcTWLiAuiXfEPP7KXziMeL1dU33bt3+b8Nm\n2+VpabFAgEhJMfYsH7bUVMx4nNCG9dhzcnDk7vrraPOmkPezG6id/SVlr7zcLNhPP/lU4sEgnmHt\nn2ElIgcPhUIiIiLSYcKF26hbvAhX3750v/kWGmr9u/zJtmG1kjRiJP6v5mDLyiLz1NNabbsrmWef\nQ9GTT7D1wb/Q6867dvun6SKdUTwcpqGyEs+QYW2Gq3sj46STMaNRUo8bv1f9GIbRdIqgu29f6Nt3\nt+63Z2bhu+BCip95iuLnnia0fh0WTxI5l1+Z8L1wHNnZ9Pz9HfgXLqB+5QpiNTWknnY8acdPxGxo\nILB4EdWzZtIw7U2A705oNAw8SU7q/EGqZ35C0ZNPYPzshmabZJvxODWfz6L8jdeIB4MYNht5P78J\nmzeFeDBI8ujD2lWjYRikHnscScOHU796NdGSYiqmT6N6xocA7TrdTUQOPgqFREREpMNUffAeABkn\nn4bF5cbhanupQdq4CYTWrSX3iquwtKN9S7yHHUH07HMpn/o6dUuWkHa8lpGJRMvLgcbZdPuSxeXC\nd975+/QZ7eU97AjK33iN0Pp1jScV3nzLATN70Nm9B87uPeDMs3e65h4wkODqb8FiIf8XNzcL8Hw+\nL2VlfpJHH8bWv/6Jomefpuf/3YYzL5/w1i2UvPRCYwDmdpM6/nhqPp9F6csvkXb8CY197+YyQVta\nOilHNu4/VLd8GaENjSdDKhQS6Zx2vWubiIiISDtFKyqo/Xoujrw8kg5p38k+0PiXoT73/WWvl2ck\njTwEgND2jWpFvi9aUU7ZG68R/95pUYnU4K8l8M1izHh8nz1jx2bN9qyucyqfYbM1naaWOm7CARMI\ntcV7+BgAfD/8Uaszutx9+5L74ysxwyGKnvo7le+9w+Y/3klo/TqSDx9D7z/eR84ll5F2/ESiJcWU\nvfIyAK7+/fe4rh1/rgI4u/fc435E5MClmUIiIiLSIao++gBiMTJ+cGqbpwXtC45ueRgOxy5DoWhZ\nGVWfzCDr7HOxOJ37sTpJtKoPP6B65sc4cnJJPfa4hNbSUFPNlj//iWhJMUkjRpJ59rk4u/fo8N83\n380Uyu7Qfg906SdMxpaWRvKoQ9tufIBIHT8Bz9Chbe6h5B1zJPWrV1Hz2SzKp76ONS2NnEsvJ/l7\n4U3mWecQXLuW8JYC4Lvj5/dE0shDqHhrKhZPErbd2FhbRA4eCoVERESkRbG6OqxJSe1q2+CvpeaL\nz7BlZO5yY+l9ybBacfXqTXDdWuLhcIuhT/WsT6j++COcPXqQesy+DQbqVq6g6sP36XbNz9p9LLfs\nO8G1qwGoW7Yk4aFQ0VN/J1pSjD07h7plS6lbthSr14tn6DA8g4dgz/Lh6t+/xePNd0e0vOvNFILG\n2UKJ3kNodxkWS7s31fadfwGhDesxY3Hyf3Ez9szme6hZPUn0uPX3VH3wPvYs316Fjc4ePXH17Ycj\nP7/V08tE5OCmUEhERER24l84n6K/P07OZZfjO6f1zZ8b/LVsvvM2zGgUMxIh/aQfYNgS978Xrt59\nCK5dQ2jzJjwDB+10Pby18Ryf4OrV+zwUqnznbYJrVhNcvardG73KvhGrr2sa+/qVKzAbGhL2fRop\nLSW4ZjXuwUPofvMtjadOLVtK3crl+L+ei//ruUDjRsPZF168V8/qqjOFOjuL00nP398JFkurQY3F\n7iDz9DP3+lmGYdDz1tv2uh8ROXApFBIREZFmzIYGyl9vPI647L+v0HPisUDLx0xXffQhsZoaAKxp\naaQeO25/ldkiZ58+AIQ2bmg5FNq2PRRas3qfPD8WCFDx9ps4e/YiuHZN4zO3blUolGDBdevANDFs\nNuLBILVfz8XZY/tyLcMChoFhsWBLT2/XssJ4OEzN7C9IHjV6t0+6CyyYB0DK2KMxLBa8R4zBe8QY\nTNMkUlhIcP1ayl9/jcDihWSefS7hzZtwDxzUbLaHaZoEV3+Lq3cfzFis8ft92PCdAoJoWRkWlwtL\nO2f8ycHDsLb8Z7KIyO5SKCQiItIJBdeuxZaZsUdHs9fM/oJoWSn27ByipSVsev4F0i++Yqd2sUCA\n6pmfYE1Npddtd2LY7Anfp8fdfyBYLNTM+pS0iSc0W34TCwSIVVcDEC0vI1pZsVdH15vxOPUrl+Pq\nNwCr202oYDOFTzxKw/bZGTuEt27Z42d0RfFQkNo5s0kaeQj2rI45NWtHQJc6YSLVH39EyfPPttjO\n7sum973373KZTDwUZNsjDxFcs5rqT2bQ83e3YU1Obt4mEiFaUtLiaU3+BfPBaiV51Ohm7xuGgTM/\nH2d+PsE1q/HP/Yot991NpHAbzt598Awegi0lFWtqKvUrllE7Zzbeo44mHgxS981i0iaeQOaZ54DP\nCzQGR9HyMhzZ2Vr2IyIirVIoJCIi0snUzP6Ckuefw5aeQa877trpL6zQ+Jfk+m9X4R40uNmMmng4\nTMX0aRgOB91v+S2Fjz9C2azPcR5+FJ7BQ5r14Z8/DzMcIv2MM7GlHRgbkNozMkg7YTLVMz6k6oP3\nmy2f2DFLyOJ2Ew8GKbjrTtyDBtHt2uv26C/NFdOnUTl9Gu5Bg0k5+lhK//UCZjSKd8xY/PPmNs4+\ncTgIb1Eo1F7R8jK2PfowkW1bMV7/LzmXXb7Xe8PEIxECCxeAxULmGWdhS02loaYGTLPx5C/ThHic\n+lUriZaVEgv4sXlTWu2v6uMZBNesxu7zES0pofCJR+l+8y1Ny9Gi5WVse+wRIlu3kDZxEuknnYwt\nLQ3DaiWweCHhgs14ho9o8fflDsmHjMY/9ysihduwJCcT3rSRcAsbqAfmz2s6vax65idUz/yEzcnJ\nWLN82DMyMMNhbB0UrImISOe0x6HQvffey5IlSzAMg1tvvZWRI0fu1OaBBx7gm2++4aWXXtqrIkVE\nRKR9wlsKKHnhebBaaaiqpPCJR8k6+zxc/fs3BR+1c+dQ/OzTjTcYBhmnnIY9KwvD6SS0bi2x6urG\n99LTybn4UgruuYvSl1+i1x/ubraEZcfsi+RDRu9URyJlnnEW/nlfU/neO3jHHoVj+34qO0Kh1AkT\nqfrgPWIBP4GFC/DP/5qUMWN36xl1K5ZTOX0aAMHV3xJc/S0Wt5tu115H8iGjcA8ajBkOE1j6DcHV\n3xIPhbC4XB37QTuBeDhM3YrlJI0YSXjTJgqfeISY30/yoYdRv2olJS/+E8+gIdjS0vb4GeVvvEa0\ntIS0SZOxejxknHxqi+3KXnuVqg/fJ1pcsstQKLL9+6j7r35L2X//Q2DhAkpefJ6cH19J5XvvUPne\nO5iRCFavl+qZH1M982MwDGxpacT8fgyHg6xzzttlzZ5hw8BqhViM/Ot+gS0ri4bKChpqahqXa5px\nYnV1VEx7E2g8bSoeChEp3Ea8spzQ1i1NIZKz+86zlURERHbYo1Bo3rx5bN68mVdffZX169dz6623\n8uqrrzZrs27dOubPn4/dbu+QQkVERKRtdcuXQTxO7pVX4583l7plS9ly/z3YfdmkHH0M3iOOpPzN\nNzBsNnwXXEz5m69T+e70Zn1YPB7STzoZaNy4OWfSCZTM+Bj/vLmkjD26qV1w3RqsXi/2nJz9+hnb\nYnW78f1oCsVPP0nZf14m74YbMQyDyPaNhlPGjCXlqGMwoxG23Hc3Zf99heSRh2Bxudv9jMr33gEg\n/+ZbKH35JQy7nbyfXtd0elDa+AkARCvLCX67ivC2rbj79e/YD9oJ1H41m9J/vYglKQkzHMaMx8m+\n8GLSJk6i+rNZlL70T8rffIPcy6/co/6jlZVUz/wYe24uWef8cJdtd3wfR0qKcQ8Y0Gq7SFkZhs2G\nLT2d3CuuYmtlJbVzZhMpLiK0YQPWlBSyLrqU5EMPo/rjj4gUFdFQVUm0ogKr10v2JT/G1bPXLmux\nepLIPPV04tFoUy32/zkOvMFfS8U7b2NYraRNnNR0wp3P56W0pIaG6ioaqmtaXMImIiKywx6FQl99\n9RWTJk0CoF+/ftTU1BAIBEj+3jTYP/3pT9x000089thjHVOpiIhIF2eaJpGiQhzd8lpd7hTaPjvA\nPXAQ3iPHUr9qJbVfzSawaCEV095smlmQNulE0sZPIHnUKILr1mKGI8QjYeLhMO4BA5sdRd/9h+dS\nMvNTKt55G++YsRgWC9GKChoqK0kefdgBuV+J94gjqfn8M+qWLqFuyTckjxrd+LWxWrHn5mLZ/kOr\n9JNPpXL6NCqmv43vhz9qV9+RkhKCq7/FPXgISUOH0fuue1o9BWjHLI0tf74Pd99+5Fx2BY7c9h07\n3RVESxuPTI/X1WHxeMi75mckDRsOQOpx46j+9BNq53xJ2vEn4Orde7f7DyxeCKZJ+sRJWBy7Pt59\nR6AXLS3Zdc3lZdgyszAsFgynk7zrf07BPX8ktGEDdl82PW+9Dau3cV+fvTn9KfOMs3Z53eZNIffK\nqzAs1qZAaAfDYsGekblXe2aJiEjXYGm7yc7Ky8tJ/95PKzIyMigrK2t6PXXqVMaMGUN+fv7eVygi\nIiLEw2GKn/47m2//P6o/mUGsvo7Qpk3EI5Fm7UIbN2L1pmDLyMCwWEgaNpxuP7mGfg8+TM6Pr8Q9\ncBCObnlknNJ4zLwtNQ3vYUeQcvQxpE2YSMZJJ+Pu269Zn66cxllG0eJi/PMbT04KrmtcOrarGRWJ\nZBgG2RdeAlYrpf/5F5GSksa9XAYObgqEADJOPhVbVhZVH39EpKiwXX3Xzv4CaAwtoPEUoNaCMc+Q\nYdjSM7BnZBBcu4bNd91O9aczMU1zLz9h59Cw/eS6rHPPp+dtdzYFQtAYbGT/6AIwTcpe/fcefc0C\nixYCkHxo26e/7QiFIiXFrbaJBYPEAwHsvu/26bGlppF/4y9JOXYc+Tfe3BQI7Q8pY8biPfyI/fY8\nERHpfDpko+nv/0e6urqaqVOn8vzzz1NSsuuftHxferoHm63jjlb0+fbff5Bl/9G4dk4a185HY7rn\nGgJ1WJyOZsFFLBxm5V1/wb98BQDBRfOJrFxGzdJlYLHg6dmD5H79SBs1kobKCtIPP4zs7P/dE8VL\nTo9T4OxT9qiuAZdMYdGc2dS8P52+J0+kpqBxRlK3MaPxHqjj7RtEw5mns23qW5Q88TAAuceN3en7\n03b1lXx77/1Uvf4Kw/5we5sznwq+WYTV7ab35PFY2zptzecl75/PAFD+5WzW//1pSl9+kciqZQz4\nxfU49mKvnI6Q6N+rxXW1AAy44Jxm3/NNfEdS/+URVH49H8vaFWQd0/5Np6M1NaxZuwbvoEF0G9Cz\nzfZmVjKbXC7i5aXNvi5VixZT9N4HmNEoKcOGApDSI6/51843iO6HDPrfLhMm0eMqHU9j2nloLDun\ng3lc9ygUys7Opvx7x62Wlpbi2/4Tk7lz51JZWclFF11EJBKhoKCAe++9l1tvvXWXfVZV1e9JKS3y\n+byUlfk7rD85MGhcO6f/Z+++w9s6yMWPf4/2tC15j0xnOHGSZjVNJ510EUopbQMFSqGFQgsULvy4\nl9tLgEuBC7fAZZXVTWlLJ917p2l24sTZie3E25ZlWXud8/tDthI3juN4ys77eZ4+laWjc15FSRS/\nfoe8rxOPvKeDp6kqB777bRSdQuH1NxD3dBCtqyG0exfxlhYcS5aS7OoisCdVpWMsKETvdBI+dJBQ\nbR2tr78BgK5k0rC+B/n5Tvw6G87Tz6DrvXc58MJrtH+wFp3DQdiZRySD32/r+ZdgePMdwg2pKiCt\nfM5RvzbatAps8+bj21pFzctv4lxy7MqLZCBApLER29xKOrpiQOyYxx5l9gIm//C/ab73bjo3bWb7\nHb+g7Lv/Pmbtd5nwZzXclpqz4+mMAJE+j8n6+FV0bNjEgXvuJzltFjpj/21gAMlQiMY//BZUFcsp\niwb8Oo0FhYSbmtIzedoeTQ2S7tG5tQqAhCNnzH/tjiUT3lcxvOQ9nTjkvZyYxsP72l/SalDtY2ee\neSYvv/wyANXV1RQUFKTnCV1yySW88MIL/POf/+T3v/89lZWVx00ICSGEEALi7e0kfZ0kvF4afvMr\nWh+8D987b5Po6MB5+hkU33QzjiMSFvlXX8Pk/7idGb+7i5JbvgHdyQXz1GkjEp/78hWg09H60IMk\nfT6ci5ei6Ievynck6CwW8ld+GgBT2SSMuXlHHaMoCgWfvg7FYKDt0YdRo1EglaT7cMtSpOYAAJYP\ntdgNlCHHRelt/4Z94SLCe3bT9Oc/0njX7wnt2T2o8413ic7O424WMxUW4brgIuLtbXQ892yfx4T2\n7Kbux6uINjSQ6PRy6Bc/I7x7F47FS8g+7/wBx2MsKESLxWj758PU3v4fBDZuwFI+g8k/+BG2efNT\n6+sBo6x5F0IIMUEMqlJo8eLFVFZWsnLlShRFYdWqVTz55JM4nU4uuuii4Y5RCCGEOCn0rLq2TC/H\nMq6EOkwAACAASURBVHUa5rJJWKZNSw2WNqQ+sh2LFtH2yEMYXG7sCxYCqZk2jkWLyf/0dXS9vxrb\nrFkjEp8pv4Cs5WfQ9f57ADhPXTYi1xlujsVLKbjuc5gnHbuFyFRYhOujl9DxwnO0/v0BFIsF3ztv\npTZhfeS89HHhA/sBsEyfPuh4FEWhYOVnqN1RTWDDegACGzdQ/NVb+q1SmmiS4TBaNIIhx3XcY90f\nW4F/43o6nn8WU1ExWaef0evxzjdeI3qwjrof/wBDTg4Jj4fsc8+n4DOfRdEN/Geg5rIyAhvW0fna\nq+idTvKu+xxZp5+JotPhWLiI0PZtAL1mCgkhhBDj2aBnCn3nO9/p9XVFRcVRx5SVlfHggw8O9hJC\nCCHGIf+evdTf+yCaqqa+GdPpIJlESyYpvOFLmPILxjrEjBXtTgq5L1+B45SFfR5jzM2j8IYbMebn\nH1Wl4zr/QlznXziiMbovX0HXB++jdziwzj76sz8TKYpCznkXHPe4ntfWtWZ1+r6u997tlRSK9CSF\npg0+KQSpSpPSr99G3OPBkJ1Nw//9Cv+6tSdXUqjTC4B+AHOV9DY7pd/8Nod+9hOa77sbg8uFrWLO\n4XN1dXXfSJLweMj9xCdxX77ihFvzXBddjKm4BDQN25y5vbbw2U9ZBH9/AJBKISGEEBPHsAyaFkII\nIXo0Pvs8oR3VfT4W2laFaYSTFuNZrDE198Z8nO2d2WeeNRrh9MlUWEjJ176OzmY7oQqM8UBnNlP2\nb98jvG8POouVztdeIbx3DwmfD0N2NglfJ5GaGoz5+RicHx7kfeJsc1JDizVNw+ByE96zG03TxmzG\n0GhLdHYCDKhSCMBcUkrJLd+g/tf/S+Mff8ekf78dc0kJmqYRbahH53Bgq5iDY8FCss44c1Ax6cxm\nnEuW9vmY0eXCNqeShLfjqBXwQgghxHglSSEhhBDDRtM0fNu2oc/OZvr/3ImmqaBqRA7sp/7OXxDv\n6KBr3QeowRA5JzDn42QRbahHMZsxuHPHOpR+ORYuGusQRoypsBBTYSEA8fY2wnv30PnWGyR9Prre\nfw8tkSBr+cA3YA2EoihYZ83Gv3YNsaZGzCX9JwUnikR3pdBAk0IAtoo5FH3hizTf/Vca/u9O8q68\nCsvU6ajBII7FSyi5+ZaRCheAkq9/Mz1XSAghhJgIJCkkhBBiSJLBIN5XXyb7nHNRI2Hi3k6cpy1H\nMRjoqXcwFqRaxhJeL/71a0l0dpJ99jnpOTkCtESCWHMTlslTJlwFznjlOGUR7Y89Ssez/wLAmF+A\n6+JLyDrz7GG/lm12Bf61awjv3nXyJIW83Ukh1/Hbx46UdfqZxNvb8fzrKZr/9hd03ctOTKVlwx7j\nh+lMx998JoQQQown8q9xIYQQQxLYtIGO556h47lncF16OUCvWR8AhuwcUBTiba0kOjpA04g1N2Eu\nmzQWIY+qWFMjXWvex73iCnRG4zGPi9TVQjKJ6SRJCIwHpqIi7AsXkezy4broEhxLlo5Ywq5nPpN/\n3Vqyzjx72JMPvvfeQdHryTp9cG1VI+FE28eOlLviCmxzK2l/6gnCu3YCqSHRQgghhDgx8qNIIYQQ\nQ6KGw+nb3hefBw7PSumhGAzos7JTiY/u1oueocoTnee5Z+h44bn0lqm+hHbtpP7OXwBgmzPnmMeJ\n0Vd66zeZ/P0f4Dx12YhWcBkLCrBVziO8dw/1v/x5OmEyHII7qmm57x6a7/4r3ldfHrbzDoV/wzoC\nmzcC3UnjQbCWzyD/U9ekvzaXTvwksxBCCDHcJCkkhBBiSJLdSSHnaadjmTYd9/LT+tzMY3S7IZlM\nfx2tn/hJIU1VCVWnhm77N6w75nHe119Fi8Uo+uJNOE8b3nk1YnxQFIWSW79J1hlnEqk5wME7fkzk\nYN2Qz6tGwrTcfw/odOidWbT98xESvuFLOA1GwtdJ05/vItHVRfY556LPGvzQbsvUaTiWLsOYl59u\nUxVCCCHEwEn7mBBCiCFRQyEAXBdfgmXyFPLznbS1+Y86zuByQc3hr2MnQaVQ9OBBkoHUr0Vw+zYi\nB+sw5uWhtx1ec62pKuHduzHk5Q16Y5KYGHRGI4U33IipuIT2Jx/n0P/8lNKv33ZUO+aJaHv8MRIe\nD+7LV6Alk3hfeoF4e/ugq3OGQ2DLZtA08j91Da6LLh7y+Yq/fDOAzOISQgghBkE+PYUQQgyJGk4l\nhfTW/lc0G9zuXl+faKVQwt9FtKHhxIIbY8HqbQCYJ02GZJKDP15Fw6/vRDtie1G0/hBqKIhttrSN\niVTFkPvSyyn52q1o8Thtjz7c6/fLiQjt2onvrTcwlZTi/tjHMWRnA4e3fo2VwMYNADgW9736/UQp\nOp0khIQQQohBkkohIYQQQ5LsrhTS2Y6TFHIdTgqZp04jWltDMhRCf5zn9Wi5926C27dR9u3vDqly\nYrip8Tjel17APGUKCa+X4NYtJHw+kl0+Ej4fKApFX7yJ5nv/RjLgJ1JzgMi+vVhnzgJID8m1VVSM\n5csQGcaxaAmORYsJbNzQ6/fLQKmRCC333ZP6/XfDl9AZjemBzgmfbyRCHpBEp5fQ7l2Yp07DmJs7\nZnEIIYQQIkWSQkIIIYakZ9C0zmrt9zhjd1JIZ7Njq5hDtLaGwKYNZJ91znGvoSUShHbtBFWl6c93\nMeVHP8EwhDkkw8n3zlt4/vVUr/sUkwlDVjaWqdOwz1+AedIkpvzgR4T27Kb+Fz/D88zT2BcuItHh\nIbB5M3B4+5QQPXLOv5DAxg14X3/thJNC7U8+Rry9Ddcll2GZNh0AfXelUHIYh1ifCP/G9bT+/QFI\nJslafsaYxCCEEEKI3iQpJIQQYkjUUAidxXLc9o2e9jFjQQE5519A5xuv0f7EYzgWLUFvt/f73EjN\nAbRYDL3TSdLfhX/dB7gu/OiwvYbB0pJJvK+8hGI04jx1GTqbDdfFl2HIyUFRlKOOt86chXnSJEI7\ndxDauSN9v2XGTIxuqZoQvVlnzcaYX0CoehuapvX5e6ovod276HzjdUxFxeRe8Yn0/T1zhIZzs9lA\nJAMBWv/xIP51a1GMRvKvWUnO+ReMagxCCCGE6JskhYQQQgyJGg6jO848IQBjfgHodJhLSjG6c8ld\ncQXtTzxG+9NPUHjd5/t9bqi7xcr9sY/T9vBDBLdvy4ikUNea1SQ8HrLPO/+4rwFS82KKbvoqoe3b\nMLhcGNxuDO7c9KwXIY6kKAqmsjKCmzeR7Ooa0O8TTVVpuf9eUBQKb/gSOqMp/ZghpzspNIrbx9Ro\nlLr/XkXC48EyvZyiL96Iqah41K4vhBBCiP5JUkgIIcSQJEOh1Gax4zBkZzPpe99PJYcA10UX07X6\nPXxvvUn2medgmTr1mM8N7d4FikLWaafje+dtwnt2o8Zi6EymYz5npEVqa2n9x99RzGbcH710wM8z\nl5RgLikZwcjERGIqLCIIxFqaB5QUijU1Em9twXnacqzlM3o9pjOb0Vmto1opFK0/RMLjwXnqMopu\nulkGQgshhBAZRj6ZhRBCDJqmaajh0HHnCfWwls9IzwJSDAYKrvscaBqtDz2Apqp9PkeNx4js24u5\nrAy9w4G9ch5aLEZ4755hex0nKu710vD736DF4xTfdDPG/Pwxi0VMbD1VNbHmJiDV+hVtbCDZPcvr\nw8L79wGp1rO+GLJzSI7ioOlYc3Mqnoo5khASQgghMpBUCgkhhBg0LRoBTRvwBrEPs82Zi3PZafjX\nrcX37jvkfOTco46J7N+Plkhg7V7Zbquch/eVlwhuq8JeOW8o4Q+KGo3S+Pv/I9nZSd7V1+JYuGjU\nYxAnD1NREQDxlma0RILaVf+JGgwCqeHuBncueVdelf59GDmwHwDr9Bl9nk+fk0OsuQktkUjfp2ka\nkQP7MZeWobNYhjX+nmSWqbBoWM8rhBBCiOEhP7IRQggxaMnQwDaP9Sfv6pUoZgvtTz6GGo8d9Xho\n9y6A9Bp666zZ6KxWAps2oGnaoK87GJqq0nzPX4nW1ZJ11tm4PnrJqF5fnHx6kimx5mbi7W2owSDG\noiLs8xdgcOcSb2mm6a9/ItbSAqSSQorZjKm0tM/z9bSgHbmWvuOF5zj0s5/Q8fKLwx5/vLtSSOYI\nCSGEEJlJkkJCCCEGTQ2HANANslIIwOhykX322ajBIJH9qSqHjhefx/PcMwCEu+cJ9azk1hmNOBYt\nJtHRka6KGClaIkHHyy8Sa2sFwPPMUwQ2bsA6azaFn71+wNughBgsvdOJzm4n1tycTvxkn3EWpd/8\nNlN/9BMKb7gRLRql4df/i++dt4k1NWGZNv2YrVrpYdOdXgB8776N56kngNQ8ouHS8fKLHLzjx0Tr\nD6KzWNDLMHUhhBAiI0n7mBBCiEFTuyuF9APYPtYfW8VcOl97ldDuXaixGO1PPAZ6PTnnX5hqa5k0\nudfaesfSZXS9vxrP009iLCyCD1UM6axWcldcgc5sHlJc3ldeov3Jxwnv3kXW8jPoeO5ZjPn5lHz1\nVhSDfISK0WEqKiZSW0OssQEAY0Fh+rGs05YTa2yg48XnaXngXgCs08uPea70Wnqfj45162l54D50\nDgdqOEzC4xlQPJ3vvEVg4wZKbv0mOqOxz2PaH3s0fds8dZokUIUQQogMJf+iFUIIMWjJcM9sk6El\nhayzZoGiENy6Bd+7b3efPInv3bfREol061gP+9xKdHY7oZ07YOeOPs9pKi4h+8yzBhWPpmlE9u1N\nVysFq7YS3rsHxWyh5OvfQu90Duq8QgyGqbCIyP59BLdVdX9d2OvxvCuvIuuMs/CvX0u8tZXscz5y\nzHMZ8lJD0b0vPU9LfT2K0Ujp12+j+e6/Em9vH1A8na+9QqyxkWhtTbqC73jxCyGEECIzSVJICCHE\noKmhnvaxwc8UAtDb7JgnTyFaVwuAqbSMWEM9na+/CoC1oqLX8YrBQOk3vpVqlZkyBfSHP87iLU00\n/uF3RPbvG1RSSFNVGv/4O4JbNgPgWLKUwMYNqOEw+dd8WtbJi1FnmTGDrvffI7xnN9C7UqiHqbCQ\n3I99/Ljnciw4BfspCwlu3QI6HSW3fgNr+QyMubmEdu5AjcXQmUzHfH6is5NYY6rNLLx3T59JIU1V\nUQyG9DBrU7HMExJCCCEylSSFhBBCDJoaHp72MQBbRQXRulrMU6ZS8JnPcuhnPyHR0dE9T+jo9drW\n8hlYy4/esGQqKEAxmdKruU+U9+UXCW7ZjKV8BrkrrsA2Zy61q/4TncVKzgUXDuqcQgyFfW5l+rbB\n5RpSW6RiMFB88y10PPsvChbOQ52eSrgacnMBSHR4+h0KHdp1uDIvvG9vn8ckA4Fe2836+nMqhBBC\niMwgSSEhhBCDlhymSiGA7LPOIdbURP41K1PfoOp0oKqYp0xFfwLbzRSDAcvUaYT37qHticcwZGXh\nuujiAT03fGA/7U8/iT47h5Jbv4HBmQXAlFU/BhQUvX4wL02IITHm5WMsLCTe0pKaoTVEOqORvE9+\nitx8J21t/tQ1cvMAiHtSSSE1HuPgj1ZhX3AK+desTD83tHMnAIrRSHjfvlRV0IeGWie8HQDkXHgR\n7ksuw5DjGnLMQgghhBgZsn1MCCHEoPVUCg11phCkZgCVfuNbmIqK0RlNmIpTbVofnic0ENYZM0HT\n8L74PO1PPYGWTB73OclQiKa/3AWqSvGNX04nhAB0RtMxB+oKMRps3dVCpj5ax4bD4aRQaq5QrL6e\nWHMT3ldeSs3uIjVrK7RzBzq7HcfipaihYJ8byxLe1GYzg8stCSEhhBAiw0lSSAghxKAlfT5geJJC\nH2aZMgVItZWd8HOPaFfRYrH01qZj0TSN1gfvI9HejvvSy7HNmXvC1xRiJDkWLgbAPGXqiJw/3T7W\nPWw62lCffqzlgXtRo1Hira0kOjzYKuZgm5v6M9L4x98TPXSo17kSHaktZgaXJISEEEKITCftY0II\nIU6IpqoENm3E+8pLRA7sB0XBkJV1/CeeIPdlKzAWFmGbO++En2ubXYFtbiWK0Uhw6xbC3Wvtj6Vr\n9bv4169LzRH6+CeGErYQI8JeOY8pq36MqaR0RM5v7E4KxbvX0kfrU0kha8Ucwrt24nn6SYxFqdY1\nW8VcspafQayhAe8rL9Hwu98wZdWP0dvtqXN0VwoZXe4RiVUIIYQQw0eSQkIIIQYs2thA429/Q7y9\nDQD7KQtxX3I5eodj2K9lKioi9/IVg3quzmKh7NvfJdpQT3DrFiL798NHzut1TMLfRfPf/oK5rIzO\nN99AZ7VSfNNXUAzy0SgyU3+JzaEy5LhAUQhWbaHpz38k0dkJQPGXv8qhn9+B97VX0gkp25y5KHo9\n+desRGe14vnXU7Tcfw/FX70VRVHSM4UMbkkKCSGEEJlO/uUrhBBiwDpff414extZZ5yF+7LL+91S\nlAlMxSXoLBYiNQeOeqz9iccIVW8nVL0dgKKbb8GYlz/aIQqRERSDAceSpQS3bMa/fh2QGnBtyMqi\n8PobqP/lz4k11GNwuTEWHp5r5L58BaGdOwhs2ojvnbfI+ch5qZlCioIhO2esXo4QQgghBkiSQkII\nIQYserAO9HoKPnf9uBi8rOh0WKZNJ7RzB3U/+gGK0Zj6z2AgVL0dU2kZ9sp56LOycC49dazDFWJM\nldx8C7HWVmr/83ugaZjKyoBUO2b2R87D9/ab2ObMQVGU9HMUnY6iG79C3Y/+i7ZH/oF1xkwSXi/6\nrCypuhNCCCHGAfm0FkKIDKWpKsBR657HipZIED10EHNp2bhICPXIOvMsoo0NxNvbUGMx6NlEptdT\ncN3nsM2aPbYBCpFBTAUF2OcvIFi1FXNpWfr+/KuvQWcykXX2OUc9x+h2U/SFL9H4h9/S8Jtfkej0\nYplePpphCyGEEGKQJCkkhBAZyPvaK3Q8/xyKwcC0n/0iI37iHmtqREskMHdvBRsvspafQdbyM9Jf\na6qKFo+DoqAzmcYwMiEyk/vyFUTq6rAvOCV9n85iJf/aTx/zOY5Fi8k+7wJ8b76OzuHo91ghhBBC\nZI6x/y5DCCFEL5qq0vbPR6C7Uiju8WA6YobHQJ6fDAYwOId3I1ikrhYAywitxB4tik6HYjaPdRhC\nZCxr+QzK7/zNCT8v/5prMRUXY5+/AFN+wQhEJoQQQojhlhk9CUIIIdLUUCidEAJIdHhO6PnN9/6N\nA9/5Fgmfb1jjitTVAeM/KSSEGBk6ownX+RdKQkgIIYQYRyQpJIQQGSYZCKRudA9zjZ9AUqhr3Qf4\n17wPyWR6bfxwidbVgl6fHj4rhBBCCCGEGN8kKSSEEBkmGUwlhcyTJgOQ6OgY0PPiXi+tf3/giPME\nhy0mLZkkWn8Ic0kJOqPM4RFCCCGEEGIikKSQEEJkmJ5KIfPkVFIo7jl+pZCmqrTc+zfUUAhL+QwA\n1O6kUMLXSbSxYUgxxZqb0GIxzNI6JoQQQgghxIQhSSEhhMgw6aRQT6WQ9/iVQp1vvk5oRzX2Bafg\n+ujFqfN0Vxy13H8vB3/yI5Lh8KBjitTWAjJPSAghhBBCiIlEkkJCCJFh1O5kjtHtRudwkOijUkjT\nNFoevB/v668Sa26i/fF/onM4KLz+BvR2B3C4fSza2IAWixHeu+eocwxUtHvzmHny+FpHL4QQQggh\nhDg2WUkvhJjQYm2tGLJz0JnGzxycnkohnd2B0Z2bat3SNJTuwdMAaiCA7+03AQjMrkCLxyn60pcx\nZOeQ7PKnjgkG0JLJ9Eyi8K6dOBacQqLTS/N99xJrqGfKD/8bvd3ebzyapqXW0et06eolIYQQQggh\nxPgnlUJCiAkr7vVSe/t/0PHCs2MdygnpafvSOxwY3G60WCw9H6hHrLUlfTu8exeW8hk4liwFQNed\n5EkGgyQ6ven19qFdO/FvWE/tqtsJba8i4e0gUlvTbyyR2lr23fIVIvv3YSouGVfJNSGEEEIIIUT/\nJCkkhJiw4m2tkEwSa2oa61BOSE+lkN7uwOh2A0evpY+3tPT6OveKK9OVRPojkkLxtsNr6aMH62j6\n0x/Q4nEcS08FINZQ328s4b270WIxdFYrWcvPGMKrEkIIIYQQQmQaaR8TQkxYSX9X9//9YxzJiemZ\nBaS32zG4cwGIt7VhOWKeT6y1GQD35SvQO53Y5sxNP6aYTCgGA2owSNzTDoCxsIh4SzPmKVMpvukr\noGkENqwnWt9/Uqhn81nZt7+LZdr04XuRQgghhBBCiDEnSSEhxITVkwwad0mhQACd1YpiMKS3fYX3\n7sHZ3R4GhyuFss85F2Nubq/nK4qCzu5IVQq1p5JCBZ/+DCg6bLMrUAwGtGQSxWA47qr6nqSSITdv\nuF6eEEIIIYQQIkNI+5gQYsLK9KRQpLaGA9/7t6Pm+qjBQHqDmGXGTBSTiVD19l7HxFpaUIxGDC5X\nn+fW2+0kgwES3UkhY0ER9sp5KIbUzwIUvR5TcTGxxga07plDfUl4PCgmE3qnc9CvUwghhBBCCJGZ\nJCkkhJiw0u1jwUC/iY+xEtxWRcLjwfvaK73uTwYC6ByppJDOaMQ2u4JYU2N6rpCmacRbWzDmF6Do\n+v5rXG+3o4ZCxNvbQFHSs4mOZCopQ4vFes0d+rC4px2jO7fX5jMhhBBCCCHExCBJISHEhJWuENK0\n9EavTBJrTs0FCmzaiBqJAKDGYmjxeK818bbKeQCEdlQDkOzqQo1EMBYWHvPcOrsdNI3ooYMYclzp\nCqEjmcvKUnE09j1XSI2EUYNBDB9qTxNCCCGEEEJMDJIUEkJMWIkj2sYysYUs1pJKCmmxGIFNG4Ej\nNo91VwoB2OamkkLB7akWsnj3OnpTwbGTQj3tZ2okgjGv73lApuKSVBzd29lCO3dQ+8P/IrRzR+o6\n3UOmj/V8IYQQQgghxPgmSSEhxLgXbWjA8/yzR7WIJTM4KaRpGvGW5nSbWNea1UBqnhAcTuoAmIqL\nMbjdhHZWo6kqse4h0/1VCh1ZaWSdOavPY4z5+QCpFjPAv2kjsfpD1P/mTnyr3zu8uUyGTAshhBBC\nCDEhyfYxIcS4psZiNP7ht8RbW7BOL++1mr1nplDqdmYlhZJ+P2o4jH3RYpJ+P6FdO4l3dJDw+YDe\nlUKKomCbO4+u994hWlc7oEoh3RFJIfvCRX0e05Psibelkj+x7k1kOrOFlnv/hnnyFABpHxNCCCGE\nEGKCkkohIcS41vHcM+kkSXjvnvT9mqqmW7Eg85JC8e7WMVNhEVmnnwGahn/tmvQmMlPZpF7H2+d1\nt5BVb0+3nRkLi455fr3Vmr5tmTqtz2N0Fgt6Z1a6UijW2IgxL5/J378dY34+0YN1qetIpZAQQggh\nhBATkiSFhBDjmu/dd9ItWEcmhdRQCFQVxWwGIBnIrKRQLJ0UKsS5dBmKwUDXmtVEDuwHwFpe3ut4\nW8VcUBRC1duJt7agmEwYcnKOeX41HgdAZ7Uec0MZgDE/j3iHh0RXF0l/F6aSEkxFxUz6j//CMr0c\nxWjEVFQ81JcrhBBCCCGEyEDSPiaEGLcSPh9Jfxf2hYuIt7YS3r8PLZFAMRjSrWOm4hKitTW9Wsky\nQc/mMWNhEXq7HfspCwls3ECspQVDXh6G7N4JH73DgWXqNMIH9qPo9RgLCvtdE5991jlEa2vIveLK\nfuMw5uUTOXCAUHVqiLWppBQAQ1YWk/79P1GDwV6tbEIIIYQQQoiJQyqFhBDjVrT+EADmsklYZ85C\ni8WI1NUChzePmUtSG7Yyrn2sZy5Q97DorNPPTD2QTGKdPqPP59gq50EyiRaLpZ93LHq7neKvfO24\nVT7GvNSw6WDVllQ83RvJABSdDr3TefwXI4QQQgghhBiXJCkkhBi3jkwK2WZXANDwmztpvv+ew5Uv\nxanKl0SmJYXa21FMJvRZ2QDY581H70glYCwfah3rYa+cl75t7GfI9InoSQoFqqoAMJeWDst5hRBC\nCCGEEJlP2seEEONWrL4eAHNZGcaCQtxNjXS99y5d776TPsbgykFntWZepVCHB4PbnW4BUwwGss44\nE++rL2OrmNPncyzTpqOzWFAjkeNWCg1Uz1p6LRoBRZH5QUIIIYQQQpxEJCkkhBi3ovWHUIzG1Hwd\nnY68K64kd8UVhHfvomvNaqKHDmGdNRtjYRHRg3XE2lox5ReMddiokQhqIIBlytRe9+d98lNknXU2\n5pK+q3UUgwHrnLkEN28atkohQ97hzWI5F1yIzmIZlvMKIYQQQgghMp8khYQQ45KWTBJrasRUNqnX\ndi1Fp8M2Zy62OXPT97kuupjmv/4J74vPU/j5G8Yi3F7iHg9w9Kp3xWA4ZkKoR97Hr8RUWIR1xsxh\nicXozsU8dRqWyZPJv+bTw3JOIYQQQgghxPggSSEhxLgUa2lGSyQwl5Ud91jnqcvwPPM0vtXvkXvl\nVRicWaMQ4bHFPW0AGPPyjnPk0cyTJpE/adKwxaLo9Uy5fdWwnU8IIYQQQggxfsigaSHEuHTkkOnj\nUXQ6HIuXQDJJvHsV/FhKdFcKGXJzxzgSIYQQQgghxMlMkkJCiHEpPWS69PiVQgAGlwuAhNc7YjEN\nVLy9HTi6fUwIIYQQQgghRpMkhYQQGav5/nto/NMf+nysp1LINID2MQBDTndSqPPopJCmaTT9+Y+0\nPPTAICM9Pt/q99h3263EPe3pmUIGSQoJIYQQQgghxpAkhYQQGUmNxeh6fzWBzZvQVPWox6P1h9Bn\n5wx4PpCxu1Io3kelUGTfPvzr1+F7842hBd2t/V9PceC730aNRtP3+d5+AzUQILh1CwlPO+j1GLKz\nh+V6QgiRyd5vXMce7/6xDkMIIYQQfZBB00KIjBSprYFkEoCk398rgZIMBUl0dGCrnDfg8/XXPuZ9\n/dX0bTUWQ2cyDTZsAIJVW0l4O4i3tmKeNIlEp5fIgQMAhPbsJt7ehtGd22trmhBCTEQtwVYe2vU4\nAGeVnMaVMy7HYrCMcVRCCCGE6CHfkQghMlJk39707Q+3fEV75gkNYMh0D31WNuh0R50r4fMRoIzo\n3wAAIABJREFU2LTh8NfejsGEm6apKrHGBgDiHak2scDmzenHA5s3kfT7sUyfPqTrCCHEeNAcagXA\nqDPwXuNafrru1+zx7iOpJtnaVs3WtuoxjlAIIYQ4uUmlkBAiI4WPTAp5vTBlavrr0PZtQGo9+0Ap\nOh2G7JyjkkKxxgY4oj0t4fViKiwaZNQQb21Fi8dT5+pIJZgCWzYBYJ05i/DePQDknHvBoK8hhBDj\nRUuoDYDPz11Jvb+RV+re5P82/6XXMb88+0fYjNaxCE8IIYQ46UlSSAiRcTRV7Z0UOiKRE2trxfvK\nS+hzcnAsXHxC5zW4cojU1aGparp1K96duDFPnUa0tiadyBmsaMOhw3F7O0iGgoR27cQ8eQrO5acT\n3rsH8+QpWGbMGNJ1hBBiPOhJCpXai1hcsIBT8it5oeY1IskI+zprAOiM+oYlKdQW8vCHrX/DEzn8\nmWHQGfjK/OupcM8kGA+xpXUbp5ecik6RYnkhhBACpH1MCJGBIrW1qKEQhrzUdq4jk0Jtjz6MlkiQ\nf/VKdJYTm0thyHFBMkkyEEjf19MuZp1eDkB8kO1jmqrS+c5bhKq3p++Ld3gIbquCZBLH4iU4Fi7G\nPHUaeZ+8CkVRBnUdIYQYDw7661nduJbWUBs6RUeeNReAKVmT+OopN/CtxV/l8mkXAeCLdg35egk1\nwUO7HqMt7GGSo5SpWZOZ7CwjloyxunEtAK/WvcU/dj/B5taqIV9PCCGEmCikUkgIkXH861P/gHed\nfyFt/3yEhLcTgOC2KoJbNmOdNRvnstNO+LwGlxtIJZkMWamtZT1JIcuMGfDGa30Ooh6I8O5dtD5w\nX6/7Eh0dBDanWsccixZjyM5myu2rBnV+IYQYT3696U/EkjEACmx56HX6o47JMacWCHTGhpYUemjn\nY7zftB6ABXmVfHn+51EUBU3T+OEHv2C7ZxfxZJy9namB/9s9u1hSuHBI1xRCCCEmCqkUEkJkFE1V\n8a9fi85mI+vMs4FUEkeNx2l95CHQ6Sj49GcHVWljyOneQHZEi1hPEshaPqP7Mc+g4j6y3Q1An51N\nvK2V4LZtGAsKMZWUDuq8QggxHvUkhAAKrPl9HpNtTiXnfVHfoK/TGGjm/ab1ZJucnFq4mJWzP5n+\nfFAUhYX584glY2xtr+agP7WkYIdnN6qm9ndaIYQQ4qQhSSEhREYJ791DsrMTx+Kl6O12dFYric5O\nOl99mXhLCznnXXBCA6aPZHCnkkI928EgNVNIMVswuHNRzOZBVwpFalI/gbZMLyf7vPMxuNwkvF60\naATHokXSLiaEOKmY9ab07UJb30mhnkqhwbaPtYU8PHPgRQA+XXEVX6hcSbbZ2euYhfnzAHh2/0uo\nmoqCQiAe5JC/4ajzCSGEECcjSQoJITKKf12qdSzrtOUAGFwuYg31eJ57Br3TSe4Vnxj0uW0Vc9BZ\nrXief5ZYa2pNcsLbgdHtRlEUDC7XoGYKaZpG+MB+DHl5TP7+f1F43ecxut3pxx2Llgw6ZiGEGG8i\niSjRIyqFCu3HqBQypSqFOgeRFHq3YQ0//OB/2Na+k0mOEublzunzuClZk5jsLKM9kvq7fVlRakHB\new0fnPA1hRBCiIlIkkJCiIyhJRL4N65Hn5WFdXYFcLjlS4vFyLvqavQ2+6DPb8jOoeCzn0eLRmm+\n+y8kw2HUYBBDdwLH6MpFDQRQY7HjnKm3eGsraiCQHlYNpM+pz8rCcsT9Qggx0fW0gy3Mn8dl0y5i\naeGiPo+zG20YFP0JVwpta9/Bo7ufxml0cNWMj3HLwhuPWY2pU3RcP/dajLrUGM2Pl19Csb2Q95vW\n81b96hO6rhBCCDERyaBpIUTGCO3agRoIkHP+hemV8WokAoDBnZueMTQUWaedTnDrFvzr1tL2j7+n\nzu1ydf8/B0jNGTIVFh7zHJqmEd67h4THQ9LvJ3xgPwCWadPTxxjdqU07joWL069FCCFOBt7upFCJ\nozi9YawviqKQbc6i8wRmCh3squee7Q9h0Bm4+ZQvMDVr8nGfU2Qv5MZ5n6Mz6iPHnM1XF9zALzf+\nnsf3PEOexc28vL6rjNKvJ9LJB00biKsJlhYupMRRNOB4hRBCiEwnSSEhRMbwr021jh25WSznwovw\nPBOk9OvfGra5PAWf+RzhvXvoWpP6KXHPVjK9M9XKkAz4oZ+kUHDbVhp/+5vedyoKtjlz01/aKudh\nLCoi+9zzhiVmIYTIRAk1gV7R9/r7uSfJ4+qeGdSfbHM2Nb46VE1Fp/SfQPeEvdxVdS9xNcFN8z8/\noIRQjyMTP7lWNzcv+AK/2fRn7ql+iG8v/hplzpKjnpNQE7xx6F1erHmNmBoH4ICvltsW3zzg6woh\nhBCZTpJCQoiMoMZjBDZvxJCbi6V7ExhA1rLlZC1bPqzX0jscFH7hSzT8+n8BMPYkhRypAaXJQCB9\nbKLTi+/dd3BfejmKIfVXZrytDYDs8y7ANmcuBqcTQ25erzlC5tIypv3k58MatxBCZApN03i/cR1P\n7HuWhfnz+dyca9KJIW8klRTKGVBSKAsNja6Yv9/jY8kYf6y6h66Yn0/N/Din5FcOKf6pWZO5fu5K\n/rb9Qe6qupdlRYs5tXARreF2Ht71BMuLl7KtfSctoVYcRjufmvlx3m38gP2+WoLxEHajbUjXF0II\nITKFJIWEEBkhuG0baiRC9rnnj8qmLnvlPHIu/Cidr72CqawMAL2zJynkTx/nee5ZfG+9gXnSZBwL\nF3U/nkoaOZcsxVbRf9uBEEJMJGuaNuAw2lA1jX/sfgKAtc0bmZEzjTNKlgHQGe0EwGXJOe75ctJr\n6bv6TQq9evBtmoMtnFN6OudNOmuoLwOARQXzuXLG5Ty173leqXuTxkATDqODQDzIawffRkHhnNLT\nWTH9YmxGG8F4iEP+Bra37+S0YlkgIIQQYmKQQRdCiIzgX5faBHNk69hIy7/200y94+fpAdF6hwOA\npD+VFNI0jeC2rQDE29vSz+t5vKfdTAghTgbhRJiHdj7GP3Y9QbVnFwA3zP00VoOVf+75F83BFuBw\n+9hAKoV6jlnbvAlVU/s8xhvp5NW6t8gyObmi/LLheClpF07+CD86/Xs4jHaagq20hdsBuHzaRfy/\npV/n2tlXYuuuCpqfn2oR3ta+Y1hjEEIIIcaSJIWEEGMmWL2dbf9xO7HWVoJVWzEWFWGeNPAZEUOl\nKAqmwsMDQw9XCqUqgWJNTSQ8HuBDSaHuSqKeJJIQQpwM9ngPpFu9NrVWYdIZWVSwgOsqPkVcjXNP\n9T+IJ+N4oz4sejNWg+W45zy1cBH51lzerl/N3dv/Tix59PbHf+1/ibga5+Pll2IxmIf9deVZcym2\nF9IR8dIUbCHP4uayaRcxOaus13FFtgLyLG52duxF07Rhj0MIIYQYC5IUEkKMmbaHH6Jrx07qVv0n\nWiyG89TTRqV17FjSlULdSZ/Q9qr0Y/H29vTtnqSR3m4fxeiEEGJs7fbuTd8OJ8JMzZ6CXqdnUcF8\nzio5jYZAEw/vfhJPuGNAVUKQmin0naW3MjNnOlvatvPrTXf12kZW4zvI+pZNTHKWclrR4mF/TT0K\nbfloaIQSYQps+X0eoygKJY5iIskIgXhwxGIRQgghRpMkhYQQY0afk5o3ocVTW12yRrF1rC8frhQK\nbkslhRSDIT1cGlLtYzqbLT14WgghTga7O/b12hBWnj01ffuqmSsotheytnkjkWSUpYULB3xeh9HO\nrQtv5PTiUznob+CXG37PQX89mqbxxN5nUuefseK428mGotBekL6db8s75nF51tRCgfZwx4jFIoQQ\nQowmSQoJIcaMGomkb5snTcZUfPRK4NGks9pApyPp96NGIoT37sE8eQrGwiLi7e3pdoFkIJDeVCaE\nECeDjoiX5lArFe6ZuC0uoHdSyKQ38cXK6yh1FHPljMu5ZOoFJ3R+g87AdRWf4hPll+GLdvHrjXfx\n0K7Hqek6yKL8+cx0TR/Ol3OUQtvhpFBBP0mh3O6kkCciSSEhhBATg/yYWwgxZhKdXiBViZN97nlj\nHE2qNUDvcJAM+Ant2omWSGCfN59oYwOxhnrUQABd9+PG3NyxDlcIIUbN0/teAGBh3jw6nJ2sbdrI\ntOwpvY4pcRTx/WXfGvQ1FEXhoinnUmDL54Edj7CmaT0GRc8nZlw+pNgHouiIlrFCa9/tYwB5FqkU\nEkIIMbFIUkgIMSY0VSXp8+GcU0Hh17+NYjSOdUgA6B1OEp2d6a1j9vmnoMaiQGrYtFGvh2RShkwL\nIU4aVW3VbGzdyrSsyZxecio6RceK6ReP2PVOya/k9tP+jRdqXmVa9tR0y9ZIcllyMOoMxNXEgNrH\nPGHPiMckhBBCjIZBJ4V++tOfsnXrVhRF4fvf/z4LFixIP/bBBx/wq1/9Cp1Ox7Rp07jjjjvQ6aRT\nTQhxWLKrCzQNk9uFzmQa63DS9E4nscYGglVV6Gw2LNOnE6mrASDuaUdnd3QfJ+vohRATXzgR5pHd\nT2FQ9Fw35+oRnetzJJclh+vmXD0q1wLQKTpK7MW0hFpxW3KOeZxbKoWEEEJMMIP6ZF+3bh11dXU8\n+uij3HHHHdxxxx29Hv/BD37Ab3/7Wx555BGCwSDvvvvusAQrhJg4Er5OAEzukf8J8InoqQBKeDuw\nV85D0esx5qVaCeJt7bKOXghxUnlq3wv4Yl1cMvUCiu2FYx3OiLp+7rXcuvCmfhNfJr2RbFOWzBQS\nQggxYQyqUmjNmjVceOGFAJSXl+Pz+QgEAji6v0l68skn07fdbjder3eYwhVCTBSJzu6kkMs1xpH0\nduQAadu8VAWkMS/VShBvb8NUWnLUcUIIMRHt8e5jdeNaSuxFXDTl3LEOZ8QduYGsP3lWNwd8dSTV\nJHqdfoSjEkIIIUbWoJJC7e3tVFZWpr92u920tbWlE0E9/29tbWX16tV885vfPO45XS4bBsPwfbDm\n58s3bBORvK8TRzIZBsDkdmXU+xoqzMXXfXvyuadjynGSsE+jDlC6vNiVBAA5JXkZFXemkV+biUfe\n04npWO9rUk3y6LqnUBSFW0+/nuLczErgj6XSnEL2+2pRrRGKsorGOpw+TZQ/r0k1SW1nPdNdk1EU\nZazDGVMT5T0V8l5OVOP5fR2WQdM9a5qP5PF4uPnmm1m1ahWuAVQCeL2h4QgFSL0hbW3+YTufyAzy\nvk4s3kPNQKp9LJPe16jODIB5ylR8cT10x6ZzOAg2NmNobAMgpBkzKu5MIn9WJx55Tyem/t7XhkAT\nzYE2Ti1cRLaaK+//EYrMxQD877t/5esLb8Jhso9xRL1NpD+vr9a9xdP7X+CK8kv56JSx31I6VibS\ne3qyk/dyYhoP72t/SatBzRQqKCigvb09/XVrayv5+YfXdwYCAW666SZuu+02zjrrrMFcQggxwSUz\ndKaQITsbAPv8+b3uN+blk/C0pwZkIzOFhBATW2MglbifkjVpjCPJPGeXLuesktOoDzTyp6p7iSXj\nYx3ShKRpGqsb1wLw7IGX2ddZM8YRCSHExDSopNCZZ57Jyy+/DEB1dTUFBQXpljGAn//851x//fWc\nc845wxOlEGLCSc8UcmdWS4J94ULyrroG10cv7XW/MS8PLZEgWn8IkJlCQoiJrTGYSgqVOjKzPWos\n6RQdK2d/klMLF1HTdZAHdz6KqqljHdaEs99XS1vYw2RnKQD3bH8IfyyAqqlUe3YRjA9fl4EQQpzM\nBtU+tnjxYiorK1m5ciWKorBq1SqefPJJnE4nZ511Fk8//TR1dXU8/vjjAHzsYx/j2muvHdbAhRDj\nW9zjQTGZ0NtsEAqMdThpOqMJ96WXHXV/zway0K6dKEYjhtzc0Q5NCCFGTWOgCYASe/EYR5KZFEXh\nujlX0xHxsqm1igJrHivKLxnrsCaU9xvXAXBF+WUc9Nfzr/0vcv+OR5iVU86/DryIRW/ms3OuYVHB\n/OOcafA0TaMx2EyJveikn2kkhJi4Bj1T6Dvf+U6vrysqKtK3t2/fPviIhBATnhqJEGtqxDJt+rj5\nR1bPBjKSSSwzZqIzGsc2ICGEGEENgWayTc6Mm5eTSYw6A1+efz2/3Ph7Xqp7g3xbHksLF7KmaQOa\npnF26fJx8xmXaQ75G1nXvIlCWwGzXOXMcpWzv7OG7Z5d7OzYg81gJZqM8UrdGyOaFNratp2/bn+Q\nG+d9bkSvI4QQY2lYBk0LIURf4l4vdatuxzx5MnlXXoW1fAYAkdoaUFWsM2aMcYR9iydUjIbe3bU9\nlUIA1pmzRjskIYQYNaF4GG+0kzlu+bvueBwmO19bcAO/3PgHHtr1OI/vfZZwIrVdc1v7Dgrt+XRF\n/XijnXREOtE0lc9UfIp5eXOOe+6qtmqSmjrgZIQn3EFTsIWIN0htWyNtYQ9dsS4+Nv0SKnNnD+l1\njhZ/LMA/9zxNXdchNDSunvVxdErq8/hzc6/l5+v+D2+0k0/MuIz3GtbSGGwmqSbR64Zvg/GRDvob\nANjZsUeSQkKICUuSQkKIERM9dBA1FCS8ayeHfvYT7KcsJO8TVxHetxcAS/nMMY7waLXNXfz3fRu4\n9ZPzWTTrcCJIkkJCiJNFzzyhErvMExqIQnsBX5l/PU/sfYZoMsayokXU+5vY0bGbHR27gdQcomxT\nFsFEmLu3/51vLf4qk7PKjnnOuJrgvh0PE1cTrHJ+lzxr/y3LL9W+zrMHXu7zsY0tW0Y1KRRXE4Ti\nYbLNJz57781D77GptQqAUwsX9UpMOox2vrHoy+zrrGF58RJqfYc46K+nJdRGyQjNvmoLpxbr1Pjq\nRuT8QgiRCSQpJIQYMUl/ajVj9kfOI9bYQHDrFoJVW9FnZQFgLS8fy/D6VNfsRwPer27ulRQy5OaC\nooCipCuehBBiIqr27ALoN2khepvpms6/L7st/bWqqTQGmlE1FafJQbY5C52iY2tbNX/Zdj/PHniZ\nWxZ+6Zjn29d5gGgyBsDd2x8iGA8R6/76wzQ0AvEgLnMOZ5WexvTCUsxxOy5LDt9/7ye0hzuG98X2\nIxQP8bstf+WQv5H5eXOZ5SqnzFFMqaMYm9HW73PjyTirG9diN9j44en/r8/jC2x5FNhS7dylzmJo\ngvpA4wgmhTwANAVbCMXD2IzWEbmOEEKMJUkKCSFGTDKQSgrZ582n4LOfJ7R9G01//TNJnw9jfj6G\n7JwxjvBo/lBqtfCO2g4SSRWDPlW2rjMasZTPQO9woLNYxjJEIYQYMfFknPcb12E32FiQVznW4Yxb\nOkVHmbPkqPtPya9kkrOUXd69BOMh7MdIlGxr3wmAzWDloL8es95EjvnYn5lljhKum/Mp3BYX+flO\n2tpSn7855mw8kdFJCmmaxp+q7uegv4FsUxZV7dVUtVenHy+05XPVzBVU5lb0+fy1zRsJxINcNPnc\n4yaQIPWaAer9jSwrWjw8L+IImqbRFkolhTQ0arsOMnectOEJIcSJkKSQEGLE9FQK6Z1OFEXBPn8B\nJV+7lYbf3Ilt7rzRj0dNsrF1K/Pz5mI19J3YCYRTSaFwNMn+Bh+zJ7vSj0363vdB00YlViGEGG0b\nW7aysXVr+htzk14G6o+ExQULOORvYGtbNWeUnJq+v2fT1UF/A1Vt1Vj0Fm5bfDPb2ndyZskynCbH\nCV8rz+pmX2cN8WQc4wi/n43BZvb7aqhwzeSWhV+iJdRGvb+RhkAT9YFG9nj388et9/DFys+wpHBh\nr+fu66zh8b3PYNIZ+UjZGQO6XqkjtRnvUKBx2F8LQCAeJJKMYNFbiCQjvNe4lqSWRKH38HC3xTVi\nlUpCCDEaJCkkhBgxyUBq1bzecXiugK1iDtP+50709tHfaFPt2cX9Ox5hQV4lX57/+T63wvRUCgFU\nHfD0Sgop3e1jQggx0WiaxkO7HiOajGHQGTi7dPlYhzRhLS5YwL/2v8j6ls0sL17C5tYqqj272dmx\nh66Yv9dxpd2tV4OVa3Wzt/MAHREvhfaC4Qj/mKradgCwvHgpOkVHsb2QYnshp7IIgINd9fxiw+94\np2FNr6RQQ6CJP1XdS1JT+eqCG3BZBlZFbDVYyLPmcrDrEDW+g0zLnjysr6endWxxwQI2tW5la9t2\ntrYdvWHZoDPw0zNvP2bVlxBCZDpJCglxEmryBMnNsmAyjsy2jh497WN6R++fbhqys0f0usfiiXgB\nqGqvZk3Tes4oWXbUMT2VQnqdwrb9HVx97mhGKIQQYyMQDxJNxpjlmsHn51wz4G/MxYnLs+YyI2ca\ne7z7+NGaX9De3d7lNDo4tXAx5TlTSKoqp+QPvX0vz5IaUN0+Gkmh9mp0iu6Y7WGTs8ooz5nK/s5a\nvJFOXJYcPOEO/rDlb4QTEa6fu/KE27POLl3OU/ue586Nf+DGeZ9l4TBuCGsLtafjvqL8UnZ799IR\n6ex1zB7vfnZ07Oagv1629Qkhxi1JCglxkvH6o/zg7nV8dNkkrj53ZAcmJ/1+0OnQ2TLjp2feaOof\ncwoKj+19hpk55eTbem908YdiGA06Zk3Kobqmg46uCO4smSEkhJjYOrqT5qX2IkkIjYIvzfssv950\nF62hdublVrBi+iWUOIrS69eHS641Ve3q6a56GSneSCcH/fVUuGb2O4x5ScFC9nXW8EHTRhYWzOMv\nVffji/m5auaKQc0FunDyR5jsLOWuqvu4f8cj5NvyKHUUp1vxSuxFfVYFD0RPpVC+NReHyX5UyxtA\nvi2PHR27OdTVIEkhIcS4NbyfPEKIjNcZiJJUNfbW+0b8WslAAL3dgaLLjL9qfNEuAFZMv5hYMsYD\nOx9B1dRexwTCcRxWIwump5JF22tGb2uLEEKMlZ4KCLckhEZFlsnJvy2+hZsXfIGvLPgCZc6SYU8I\nAelV9u0jPGz6jUPvAqlWq/4sKpiPQdHzXM3L/GTtnbSG2/nolPM4f9LZg772LNcMrp9zLTE1zp+q\n7sMfC7C+ZTM/XfdrXqp9Y1DnjCZjbG9PtcP1bDvry2RnKQB1/vpBXUcIIQYilozzwI5H2ePdNyLn\nl0ohIU4ysXgSgPrWAJqmoSgKqqZx11PbqZji4oIlw7eCOOn3Y8jJnG8wOqM+FBQunPwR6gONbGqt\n4pW6t7hk6vnpY/zhOIU5VuaX5/Lw63vZtt/DOaccvUFmpHm7v0GTn9gLIUZDT6WQ2+I6zpFiuDhM\ndubnzR3Ra+Ra3AB4RnAtvS/axbsNa3CZc1hWvKTfY50mB7cuvIl3G9YQSUY5s2TZsGy5W1gwn8um\nXcQLNa9y9/a/E4yHAHi57nUag03E1QRfrPwMJr3puOeKqwn+uu0BDgUaOa1oSb9/JlzmHBxGO4ck\nKSSEGEE7O/awtnkj29t38h/LbsNlyUHTNJ7c9xytoTYArAYr18y6YkDbGz9MkkJCnGTiiVRlTCSW\nxOOLkJdjJRRJsHFPG8FIfEBJoX31Pv755j5mT87h8tOnYDEd/VeJlkyihoLoy4YvyTRUnREfTpMD\nvU7PytmfZH9nLc/XvMLigvkU2PKJJ5JEY0mcNiOFLisFOVaqP7SafjRomsYvNvyOrpifhfnzuXHe\nZwdd/i6EEAORTgpZJSk0kWSZHBh1RpqCLcSSMTa0bKUyt4Jss/P4Tx6g5w68TFxNcOnUCzDqjv+t\nxUzXdGa6pg/b9XtcOvUCGgPNbGnbBqQSnB0RL5taqwC4f8ejfGnedf1WZKmayv3VD7OzYw/z8+Zw\nXcWn+r2moihMdpaxo2M3gXgQh3H0l2gIISa+/Z01AAQTIe6t/gffXPQV6gON6SrNHhaDhZWzrzzh\n80tSSIiTTDR+uF3qUGuAvBwrwZ417LHkgM7xfnUz+xp87GvwEU+orLxg5lHHpDePOYfvH55DoWka\nnbEuSuyFANiNNi6aci6P732G2q5DFNjy05vHHDYTiqIwf3our2+qZ1+9j4opo/eNUjQZTW+g2dK2\nDV+sixzz2AznFkKcHA63j0lSaCJRFIV5eXPY3FrFnRv/SH2gEavBworpl3B26fIht6zt6tjL+03r\nKXUUs7x46TBFPTg6Rcfn5lxDW7idhkATN877LLs69pJtzuKDpg1sadvGP3Y9wWcqrurzdWuaxsP/\nn733Do/jPM+9f7O9YlEWvRBEIwrBLlIU1XuzmtUs2ZKLXGMn8Ul8Uk6+fF/KiZPjJLaPuyXZlmzL\nkmxLsipNVUsUxV5B9N7bYnex2L478/0x2AVAFIIU2IT3d128LmLKOzNLYnbmfu/nfhp/z6GRY5Sn\nlvDZmk+i1Zy8IUdRiioKvdT+R24vufG0ZukFAoFgIdq8nckg/2Oj9bzW+QbK5LrPVH+C6oxK/uvA\nD9jZt5tgLIgn7MUd8hCIhbgoex0VaWVcn3nJvOOfH0EfAoHgrBGJTQk/PSOqcOMPxQAIhWOLGqNv\ncj+tRqKxyz3nNnO1oz+R/YOH+OKLf8vwZIePM4k/GiAmx0g1TpVjOc2qrd4TVvOVEp3HbGY9ALWl\n6vpj7Wc2oPNEErb3BCd2OzmRIf8wT9Y/k7wOgUAgOFVcoTEMWgNWnXih/ajxsZXXo5E09E70JzOG\nnm1+gW/t/z5tns7kJMSpEolH+E3j75GQ+GTlPYsSUM40Jp2Rv1j/Rf5q41dYkVLIDcVXc3HuJr5Q\n+xBF9gI+GNjHLxuenZUnKCsyzzS/wK6BfRTZ8/nimk9j0OoXdczNORtIMdjZ2bebx+t+jaIoJ99J\nIBAIFsFocIzGsRa6fb0U2Qt4qOo+MkxpbO98i/f79qDT6FjtrMaiN3NPxe0oKOwfOkybp5OYHMeg\n0fNu3wc8VvfLBY8jnEICwTIjkSkEaq4QQCC0eKeQoij0jvjJzbBgtxho6fUQCMWwmGbeTpLt6Bdw\nCv2hfTvukJfnWl/iS2s+c8rXslgURUl2Hks1piSXJwQid0gVU94f/ADj2vcxmx4AYFVRGnqdhqPt\nLu656sx2aptOQhTSSBpkRcYdcoNjxbzbP9f6MnWuRtxhL19b98gZCSsVCATnjgZXMztyeGb4AAAg\nAElEQVT7d/NQ9f0YF5GJshCyItPq6aDkhHvKWMhDuilNlKp+BMm2ZnF5/lZ2D+znS2s+jUVn4fnW\nl9k3dIj/PvhDAO6puJ0rC7ad0rgvt+9gNDSmdgBLOX9Kxa16CyWO4hnLLHoLX1v3eX5w5HH2Dh4k\nLsd5uPr+pJD1XMvLvNf3Afm2XL6y9nOYdYvvOpptyeSft/4tPzn2BA1jzRwdrWdt5ofPSRIIBCrH\nXU283vU2Oo2Oqwsvozpj1bk+pbPGrxt/lwyXLk0txqI385maB/jvgz/CF52gJqMSk84IQGV6Of94\n8TfQShpSjQ50Gh1xOc7hkbqTiv/izUEgWGZETigfA5iYFIUW4xRy+8IEwzHyM22UFzhQFGjrn+1Q\nifsmRaEFnEJZZrWjR8NYCzF5cS6l0+FHR3/Ov+/7LsCMMqw0k/r3hMOmc6IdjTEIRj8ARr2WVUWp\n9I34GRsPnbHzO5GEKFRoU7uauBdwAHWN91DnakRCotndyp96d52VcxQIBGeP9wf2cnikjnpX04ce\n67irke8e+gnf3PsdJsJ+QrEQ3vA4wVhQdB77CPPx8o/xzUv/kVxrNg6jnU/XfII/X/cFtuSowdDH\nXY2nNF7XeA9v9bxHpjmDW1ZedyZOecmx6M18dd0jlDiKOTB8hJ8ff4q4HCcUC7GzfzcZpnT+cv0X\nsRtspzy2Xqvn7vLb0EgaXmrffgbOXiBYvrza8TotnnYaxpr5wZHHeaH11VluvwT+aIDtnW/R5uk8\nuyd5hpjebWxVmhrXsdKxgttKbgRmd3zMtmTiNGegm8x302q0bMxey1WFly54HOEUEgiWGYnyMUmC\nYXeQcCROYLJ8LBKTicsy2gVayPeOqIJJgdNKcW4K0EVLr4fayRbuCRbjFPJH1bFicoyjo/UnbWV7\nukx/2J0uCll1FvQaXdJF5IuqLeu1himBqqoojbr2MToGxklPWfzM4Ych8bkU2HPp8vUsWD72Wueb\nADxcfT+/a3mRF9pepTK9nNzJ7CSBQHDhk+gs0jDWzPqs2g85llquOxgY5rMv/PWMdSJP6KOLRtJg\nOKFhwqr0Mlall9Hq6aBrvCfZkfRkxOQYv2r4LQoKD1TevaiOXucLZp2JP1v7OX589OccGjmGfPzX\nrHFWE5VjbMnZ8KHygHKsWVSkltLobiEYC52S2wggFAupLiZFZl3matF9VCBA7cbbOd7NqrQy7iy7\nhZ8d/zWvd7/DcGCEh2s+McM92z8xyHcO/hh/LIBZZ+Z/bf76Bf17JCsyGklDliWTe8tvZ1X6VNXC\ndSuuZE1mTXKC/cMinEICwTIj4RQqzLShAL2jE8mgaVC7ki1E36jqLsrPtFGW70ACWnqmnCwThw4Q\nGRqc5hSaf8bNExlP/n3f4KFTvZRFcaIDyTGtfEySJNKMqcn27wFZvTZ0keQ2malmAEa9Z88pNBFT\nnUIFCafQPKJQj6+PY6P1lDhWsCl7HQ9U3k1MjvHE8d+cUeeVQCA4e8iKnBRyGseaP3ReScJCXuIo\nZkNeLaszqlidUcXazNVcmrflQ5+v4MKjOKUQfzTA6Ena1o8EXBwcPsr3Dz9Gv3+QbXlbqEgrPUtn\nuXSYdEa+vPazVKSVcWSkjqcafw/Ahuy1H3rsxITMoH/olPd9s+c9nml+gd+1vMi/7/su7/XtZt/g\nIfYPHuL97n0cGDpM38QA/miAXxz/Dd8++CNebt8xr2NCIPgocGTkOADrMldTaM/nGxu/qv7ujh7n\n2wd/NCNP843uP+GPBah1VhOMBXmi/ukL+vfDF5lAVmTyrNkzBKEE2ZbMJSv5Fk4hgWCZkXAKleQ7\n6B6eoHd4Ihk0DRAMx7Ca5g9X7B2edLFkWrGYdBRk2WgfGCcWl1HGvfT/4HtY165Llo3pM6YcRK7g\nGMdGG9iSuwGDxsBExE9VZjnegI96VyOBaBCL3ryk1zsx6bpJkGWZqainmlIZdo8SiAaJogo/sjac\nXJ8UhTxnv3wsy+LEoNGrmUJzsH3SJXRT8bVIksTazBq25l7EBwP7eK3jDT5WeuNZO2eBQHBm8IS9\nRGVVuHeF3IwEXbPuY6eCN6yK8Z+u/gSVRUWMjJxeyLDgo0Oxo4gDw0foHO8m0zL1ne0N+9jZv5se\nXy9bczfz8+O/Jjo54bDWWcNdZbecq1P+0Bi1Br685jP8dDIHKM+asyQO21ybOsaAf4iVC2QBzsWx\nkePoJC03rbyOVzp28HTTc7O2kZDItmQyGBgGoNXTQYe3C6c5nbgiE5PjyEocg9bAXWW3iE5oggue\nwyPHAFgzmdNl0Vv46trP8XTTc+wa2Me39n+fr657BIfBzsHhI2SaM/hC7UM8XvcrDo/UsaPrHW4s\nvvpcXsJpkxC8zkYHYiEKCQTLjIRTqDQvhXcO9dE77Cc8LXz6ZE6hYXcArUZKiiXlBQ56hifoGvSR\nM9ajHqO/D60jFTQa9M5MvOFxNSW/fw9xJc7R0eM8UHk3CgppZgdl9lJeat/OkZE6tuZdtKTXOxGZ\nLFEbLuCB9dfPKo9Im7zRdo53J5fJmilRyJmq2r9HvMElPa+FSIhCVr2VNFMaY+HZTqFIPMKRkeMU\n2PKoSq9ILv94+cdodrfyx663uaxgq2hlLxBc4CRcQg5DCt7IOE3u1g8nCk06hVKM85f2CpYXxSmF\ngPo9eFHOerrGe3i7ZycHh48SV9RngmOjDQDcWHwNRfYC1jirL/hQcoNWzxdrH2ZH9zusSluaZhK5\n1hxAFYVOBXfIQ89EP1XpFdxYfDU1GZV0j/egoKAANpsRz7if7Z1vMhgYptZZxaeq7uMnR39Bo7sF\n5pg70mq0rM+sxaq3UGjPX4KrEwiWFlmRiSsyes3ckkQgGqTN20lxStGM51mtRssDlXeTZcnkhbZX\neabpeWony0Avzb8YjaThgcq76Rzv4ZWOHaxKK2Olo+hsXdaSkagUSBOikEAgWGoSTqHiHDuSBD3D\nPqzmKWdQKLywKBQIq53GNBr1YbC8IJW3DvbR3Oshzd0LQHR0lNjEBFGHlWfbXuaDgX1E5ShOcwap\nxhSa3K080/Q8AOkmBxvT1/JS+3YODB9ZclHIG1ZfgJSIiV37/FxeMTMzIXGj7fB2JZdFlSlXkNWk\nx2zUndXysUSmkE1vIc3oYCgwTCQemZHbMBJ0oaCwIqVwxvWYdSY2ZK3l9e53cIc8QhQSCC5whibz\nhDZkr+Htnp2n/LJ5IuPhcax6y7wP4YLlR4EtH42kocXTzpGROn567EkAcixZXFGwjagc5fnWV7iq\n8FI+VnLDOT7bpUWv1S9pUHauNQs4dVHo2Gg9AGuc1QAU2vMotOcl12dm2hkZ8VGTUcnh4WNcVrAV\ns87EX274EkOBESRAK+nQajRoJA3fO/QoO/t2s7NvN2nGVP51298vzQUKBCchJsd4rfNN1jirWZFS\nyKB/iO8c+gm3rLyOy/K3JrebiPj574M/xKg18o1NX52zc26TuxVZkefsNiZJEtetuJJmTxv1riY6\nvF2YtEYuztkEqF0IH66+j/976FF+cfwp/nbzX55yzte5JtFoJvUs5CKJJwKBYJmRcArZLAay0yz0\njPgpzLQm1wcjC2fRBMIxzMapW0d5gSo6tPR4We/tUxcqCkowSG+agXf7dpFqdHBT8TVszb2IqBzj\nb3f+M/VjahedNHMqmZYMVqQU0uRuxReZOK3OH/Mx6ldLJZSYgZZhL8PuINnpU3bq1MkOZB3TnEIn\nlpxlOkwMugOLDuH8sEw5hSzJbkDukIfsyYdNmHIPzOUYSJTgJcYRCAQXLglRaHVGFW/37GRk8nf/\ndPFGxkkzXrjBm4Klx6DVU5tRxZHR4zzV+HskJL605tPUZFQmv/O25l605OXdH0XMOjOpRgf9E4OL\n2j4QDfJ2z3u81bMTgNXOqgW3z7I4ub74quTPGkkzZ9nbfavu4HuHH0NWZNxhD+6QZ8kDd0eDLjrH\ne5CQ0Gl0VKVXYNDOHz8gWB680/s+2zvfZMA/xBdqH2L/0GF8kQmeaXqB7vFe9JMTnB3ezuT328vt\nO2gcayEQC1BkL+CKgm2UOFYkO27WLNCC/ubi66h3NRFT4jxUeT82w9Q7TUVaGdetuJIdXW/zasfr\nfLz8Y2fwypceUT4mEAjOGJHJUjGDTkNBlo3BxuFkRzFQM4UWIhiOkWo1Jn9OTzHhdJho7fMS9vbO\n2NZj1/LlNZ+hIq0s+aCg1WhZkVJAq6cDgDSzeqPblLWWrvEeDg0f5fKCSz78hU7iSohCUfVLaHQ8\nNEMUSrwctU9zCp0oCjlTzXQPTzAeiOKwnrkuK13jPbzfvxdveBy9RodBa0g+xI2dIAolXgwz5+g6\nYNWp1xeInb2SN4FA8OHZM3AAu8E2Y1Y00XlsRUohdr0t+fPpEIlHCMZCFKeknHxjwbLi+uKrODJ6\nnImon03Z62aJE0IQWjy51mwaxpp5reMNtBotGkmD05TO2szVSZHNE/byfv9e3u55j2AshE1v5YHy\njy9ZB8CKtDL+5ZK/Y/fAAV5q3067t5ONpnVLMjaAoih87/BjjAZdyWU3FV/LrSXXL9kxlgJZkZEV\nOdmeW3Bm8YZ9vNbxBgC9vn4A6lyNaCUtBq2eXQP7Zmy/Kq2MZncbf+x6CwkJq97CgeEjHBg+QqEt\nD3fYi1VvocheMO8xVzqKuHnldRg0ejbOERZ/y8rr2Dt4kJ39e7ix+BqsF1DOVkIUEuVjAoFgyYnE\nVKeQQa+hMMvG/sZhAtOEoIUyheKyTCQqYzZqZywvL3Cwu26AcH+/2ut+sjtOINU856xXiaM4KQql\nT4pCG7LX8lzrK+wfOrykopAnNJmfYbDiAdzj4RnrnWY1VDMcn1p+osPG6VDtpqOe4BkVhf7Q9hpN\n7lZgalYg26IKQd2+XqoyprKDhoMLOYXULzzhFBIILhwmIn5+2fAsCgq3ldzI9SuuIqbE6fH14TCk\nYNaZyLQ46fB2EZNjp/WSkyinTTGIPCHBTIpTiqhKr6BxrIXriq4816dzQVOcUkTDWDMvd+yYsbzU\nsZJsi5OO8e5keZlNb+WO0pu5LH8rJp1xruFOm1Sjg/LUEgDavF1szF46UahjvIvRoIuK1FLWZdXy\nfOsrHBw+uqAoFIlHafN24I/4STenU3KKQdyLQVZk/NEANr0VSZL4Wd2v6Rzv4X9s/PKSCW6C+Xmx\n7TVC8TB6jR5XaIxB/zA9vj5WpZXxyOpP4Z6WkamRNORYsnii/mn2DR3ivlV3cmneFlo87fyp932O\njBxHQWFT9ro5S8ums1AJqE6j4+rCy3iu9WXe7d3FTSuvXbLrPdO4Q14kpLPynS1EIYFgmRGJxtFq\nJLQaDYWZs8u0Qgs4hYKTeUPTy8dAzRVqPNQCsSjmilUEm1W7p5SZOec40x8EUs0OCKkPL2WpK2nx\ntC+pzXk8rLaZz3Wk4RmK4Z6YKQplWzIpsufT7VNL3/SYZpePTYZqj3iDlOYvTq2PxKPsHtjPxbmb\n5rRT7x7Yz7HRBj5b8wBajRZ3yEOzuy25PjGTsSqtDAmJhrFmbpjWPWE4MIqEhNOUPmvsxL4BIQoJ\nBBcMrZ52FBQkJF5s385QYISy1BImon6uLrwMUEXgdm8nruDYDOfgYvFGVOekwyicQoLZfKbmAVzB\nMQqmZdkITp0biq+mIq2UmBxLOlXe799DnauRNm8Heo2e6vRV1GRUcnHupiUXg6ZTZM9HJ2np8HYu\n6bj7Bg8DcN2KK6nOWEWTu5UjI3U8euyXHB09Puc+J7YG35S9jrvLbzutyIC9gwd5teN1grEQVxRc\nwpUFl2LRm/lVw2/ZM3gAm97K51Y/yDFXAzE5xo+O/Jz/sfErp50p0+vrJxQPU5a68rT2Xw50eLvZ\nPbiffFsuq9LKeKvnPV7vegeAmoxKLHrznI7DT1bdw43FV5MzWQZZkVZKRVopYyE3R0aOs8ZZ86HP\nbVveZrZ3vsk7ve9zTdHlMzI6T6TH14cn7GV1RtU5D9P3hD04jCloNdqTb/whEaKQQLDMiMRkDHr1\n5lKQNVV3q9VIxGWF4AJOoURpmWWWKOQgM6K2vrDWriHY2Q6RKObcuR8sV6ZMiUJpJgcTIbXd8sbs\ndbR42jkwfIRri65g0D9EVI7PCFs8VSaiqii0wplBQ/MQHt9MUUiSJK4suJQnG54BwK5NxRMdnpEf\nNOUUWnzY9IGhwzzT/DzheJjrVlw5a/3ugf20eNoZCY6SY81m39ChyR4jKokSMJvBSqE9n3ZvF6FY\nCNPkA81ocJQ0Uyr6OQSnhCjkF+VjAsEFQ7OnHYDPrn6QN7r/xJ7BA+wdPIiExBWT7smsyXLR4eDo\naYlC45OdxxwGIQoJZmPVWy6o0orzFb1GR0Va6Yxltc5qXKExYnKMDFP6nN/dZ+RctHoK7QV0+Xrw\nRwOn/e/b6unAojOTZ8shLsc5OHwEu96W7Nq2LnM1R0bqODxyjFSjY05XjkaSWJFSSKY5g90DB9g/\ndJiGsWbuLr+NFfYC3GEvlenlJz2Xdm8nT9Q/jU6jw6DR80rH67zZ/R61zmr2DR0kxWBnPOLjVw2/\nIybHsOmt9PsHebzuV3x5zWdO+QVbVmR+dPTnjEd8/MX6LwphaA5kRea3zX8A4J7y25NlT3sGDwAL\nZ2XpNLqkIDSddFMaVxVeuiTnZ9KZuLzgErZ3vsmugX1cWbBtzu1+0/h7dvbvAeD6FVdxW8mN50wY\nkhUZT3j8rHUOXNiLJRAIPnJEonEMevVXPyPFlHT9pKeoM1WhBYKmE6KQ6QRRKNdppTim1pWbVpYQ\nz8siYJRIyyqccxybwUquNRu73oZZPzVrsz6zFo2k4cCQOgP1k6NP8OOjPz+dy5w651gARYGSbPVl\nyn2CKATgVFR7tRwyY9NbkRWZYGxKAEqKQidpS98/MUhcVkW10dAYAEdG6ubcdiykimjDgVEURWHP\n4EF0khaDRn1QnD6bVZVeQVyJ0zL50hiKhfFGfMkXxBOx6NSZGOEUEgjOf+JyHH80QIu7Db1GT62z\nmr9c/yU2Zq1FQWG1szJZ5pplUd2Xw6cZNu0Nq04h0Y5eIDi7SJKE05xBjjX7rAlCCdZlrUZWZF5s\n335a+4diIb53+FGeqH8agEZ3CxNRPxuy1yYFllpnFXqNDpveyl9v/DP+auNXZv35+oYvc1fZrVyW\nv5W/2vgVPl7+MaLxKE/UP80/7/lPvnf4UbrHexc6FRRF4bmWlwH42rrP8y+X/D13lt2CXqtj35D6\nHPXn679Ahikd1+Rz2Cer7mF1RiUNY8082/wCijI1ATfkH6be1bRgMHjTWCuesBdZkXms7pdJwWNx\nn12YVk8H9a4m3u3dhSvoXvS+FxJ7Bg7Q5ethY9ZaytNKkm5DBYWajEqyLXNXDpxNrizYhl6j583u\nd5PP6tMZj/jY2b+HLLOTLLOTHV1vc2jk2CkdwxV088fOt5iI+E++8UkYDbqIK3Gc5tkVAWcC4RQS\nCJYZkZiMUad+iUuSRGGmleZeLxkpJkY8oWSJ2FwkRKETy8c0kkRJ1EUciZAznwPb1tLgDXKvbf6Z\n7M/XPkQkHp2xzGawUplWTv1YE83utmRuTiQePe2OFiElCDE9uRk2DDrNLFFocCzA/37yEOiuAiQc\nV49BSA2bTthcnY7J8rEFnEK9vn6+ue87rM9aw+dqHsQdUuumO8a78YbHZ5RrxOV4ss3kcHCUnok+\nBv1DrMusxRfx0ebtxDNZ5gFQlV7OH7veomGsmVpnNSOTwY6Zc+QJwXSnkBCFBILzmf2Dh3i+7dXk\nS8aqtLJkq/jP1DzAltyNFKcUJbdPZIi1ezup8VeSZkrFuIAN/kQSolCqKB8TCJYNVxZs44OB/bzf\nt4eLczay8hSzfBrdrcTkGP3+QaJyjH2DhwC4KHt9chuzzszXN3wZs868qPJ/jaTh6sLLWOOs4YXW\nVxgKjNDvH+ToaD1FKfOHCr/X9wEd492sz6xNOnauLbqCy/MvYd/gQVJNDnKt2WzMXsuOrrfRSBrK\nU0spTy3l2wd/xM7+PTjNGVy34kqGAiP8773fJq6oz70ljmKuKLiEdZmr0Wl0dHi7eL9/LyOTz6IX\nZa9n39AhHq/7FX+x/ouLynX7deNvOTh8NPlzoW0v/6fg70+634WEoii81P5HDBo9d5bdAqiuVr1G\nR1SOnTcZZXaDjU3Z6/hgYB/dvj5WOopmrG/3dAJwce4m1mTW8K97/os9A/vZkLXmpGOH4xGeOfYS\nLzbuICrHCMXD3F5606LOS1Zk+iYGZjmCOsd7AGY8A5xJhCgkECwzItE4VtPUr35Blk0VhSbdMAs7\nhdQvzhPLx+RwGMf4EAPGDHwjQfZ7Rog6dKQbM+Yda75Zg03Z66gfa+J3LS8ml3nD42Ra5h9rIaJK\nECVmwG7Rk2Y3zsoUGnRNCicx1SmVZraDRxWFslBfwIwGLSkWPS7v/KJQs0fNAzo0fJSdaaVJ0Qfg\n6OhxLsvfOnU9kfFkbf1wYARPSN12S84GGsaaafN2Ep0mmK10rMCoNdAw1gyQfEDJMs/9meg1enQa\nnQiaFgjOYwb9wzzZ8CxaSUO2JZOhwMiMrmOSJFGTUTljn0xzBlpJy+GROg5PuhCtegt3lN7CJXkX\nLXi8sZCb/UOHkZBE4KpAsIzQaXR8YtVdfPvgj/hN03P8zaY/P6USquOjjYD68to93suR0eM4zRkU\np8x0g69ImdsdvhBOczqP1H6KYCzE37z3Txx3NbA+q5ahwAhxOU44HiYcjxCOhwnEgvypdxc2vZW7\nym+dMY5Bq2db/pbkz5uy17Gj622KU4qSmU1fXvsZvrX/+7zQ9mryRTyuxLk8fyujoTHqXU20eztJ\nMdjJNKv5bYmy/kxzBg9X34+syBwYPsLvW17mvlV3nPT62jwdWPUWrizYRtd4L3WuBl5reYct6ZtP\n+bM6XwnEgngj49Q6q5OCoFaj5ZK8LQSiwfOq3K4yrYwPBvbR6mmfJQq1etUGOKWpK8m1ZlNoz6d+\nrJmJiH9Gm/vpxOU4+4cO82L7djxhLw6DHW/Ed1LH23Re7Xid1zrf5EtrPk2tszq5fEoUOvXfq9NB\niEICwTIjEpPR66YeBgqz1IC/hBtmcU6hmQ8ToY52JEWmx5yFvn+csMaLpEgoISuc4oT0mswadE06\n+iYGksu8kdMTheJynLgmAlErNpMqCg11e4jFZXRatYTuRJEo3aKWVfjnaEvfNegjLstISGg0M2uM\nO73dgDr79Vb3ewDJWZIjIzNFobHQVPeFAf8Qw4FRbHor1RmrKE1diTc8PqM7gm4yn+DYaAOu4Fiy\ndGQ+p5AkSVh1ZlE+JhCcp8TlOM82v0BcifO51Q9S66yma7yXopNkBxi0Br6y9rOqmzDkwR320uHt\n5umm5yi0582bPRCJR/nB4cdxhz3cXnJTsruhQCBYHpSlrmRr7kV8MLCPt3t3cm3RFYvaT1EUjrsa\nkz+/1vkGkXiEi7LXLWnWillnojR1Jc3uVr659zszMhano9PoeGT1J08qbOfbcvlk1b0U2KYyKVON\nDv5i/Rf47qGfJkvpCm153FtxB5IkMRwY4b2+3XwwsI92byd5thyuLryMOldj8nofrLqHAf8Q7/bt\nojilkC25G+c9B294HG/ER62zmptXXsdExM8/7/4WTx19gZQ1qVSlV8y774XEfG3T7624/VyczoKU\npalxES2e9ll5n22eTnSSlhV21al2UfZ6enx97B7cP+fvy3BghP976FHcYQ86jY67qm9im/MSvrnv\nO/T4+vht8x9oHGvh7zd/fV4RdiLi580e9Z3hg/59J4hC3Wgl7Yz/w2cSIQoJBMsIWVGIxmSM+qk4\nsYsqs+ka9HHZmlxefL9jQadQYJ7ysWBrCwC9pizczSNIJX6UsJlRT5iiU8xCNetMrM6oTM6CA6dU\nv52gf2KQY6P1AGgUExqNRKpdnS3yTISTItjY+Ez3j02vzgb4TqgHdjpMtPeP873fH+Nom4sffP3y\nGZ9D53g3Nr2aldTq6UAraci15QDQ5G4lEA0my9FcwbHkfu3eLgCuKNiGTqNDp9HxhTUPz7qeyvQK\njo020DDWzEgg4RSaWxQCtS39+GT7aYFAcP4wFBjhyfpn6BzvpjpjFWucNUiSNGvWcj4q08tnhLEe\ndzXxwyOP8/3Dj3Ft0RVUpVfM6h71SscOBgPDXJ6/leuLr1rS6xEIBBcGd5TdzLHRel5p38GGrDXz\nCitxOc6b3e9ydLSekeCo6py2OBkOjCYdy5umlY4tFaszKml2t2LSGbmp+FoMWgMmrRGj1oBRa8So\nM5BhSl90t7KtuZtmLcuyZPLXG/+M1zrf4NhoA3eV35oUt7IsmXy8/GPcUXozCkqyPOziaeMYtQY+\nX/sp/mPf9/hN0+/Js+Ukxfi+iQHC8TAljmJA7WIFJNfbDFYeqf0kPzzyM3569Am+tv7zyW0vZBLP\n6I4LYLIh1eggy+ykzdOJrMhoJA2yIjMUGKF3op/ilMJk5tfG7LW83P5Hnm99hdHgGHeW3TKjXPv9\n/r24wx4uztnEzSuvpbJoBSMjPors+RwcPsq7fR8gKzKDgWHybblzns8b3X8iEo+gkTTUuRqZiPqx\n6a1E5Rh9vn7ybblnLYNMiEICwTIiGlVLlhLdxwAsJh0P3aiWKJgNugWdQgnBaJYo1KI+JHjS8xj1\nj2PWR5EnUhkaOz2nysbsdRweqUNCQkHhqXeOUnBdBTnps7tmKIpCRFZLrYKxIAeHj7J34AA9E/3J\nbYyy+gCRlhCFfJGkKJTIGPrzj68hO91MQDsCQJevh0uYKsdItKU/2qbm+Qy7g6zIUV1F3rAPV8jN\n6oxKMszptHjaiSlx0o2pFNrz6fH1cdzVyKbsdewdPEjv5LnpNDpisvqZbsnZsOBnUj05o9Qw1sJ4\nxIeERMYC4XMWnYVB/3DyS08gEJw7ZEV1GL7Xt5vnW18mIkfZlL2O+yru/NCz7S0FarEAACAASURB\nVDUZq7in4nZeatvOC22v8kLbqzxcfT+bJ+8pHd4u3ux+F6cpnTsm8x4EAsHyw6a3cmfZLfyy4Vl+\n2/wiX5xjAmrIP8wT9c/Q5etBI2lwmtIpTinihuKr+K8DPwRUkSPnNLofnowtORvpnejnyoJtp1WK\ntljSTKk8UHn3vOtPVlqXZcnk0zX38+Ojv+CxY7/k/7n4r9FpdDxe92uGAsNcUXAJd5TeQo9Pfdab\n7gCtSCvj65c8wn++/1N+eORn/OX6L80S8S80EhEIJzqFzlfKUkvYNbCXXzX8Fm94nC5fL8HJbr3l\nqVNdA1ONDr6+8cs8Wf8M7/V9QMNYMw9V3UdpajGgOos0koZ7V90xQywqshdwcPhoMiaix9c3ryhU\nP9aEXqPnhhVX8XLHDr578CeYdWaicpSYEj9reUIgRCGBYFkRjqmCj0E3t0hgMmpP2SmkyDKhtlb0\nOTlk5Dhxjak1uXLIyuBpikK1GVWsz6wl06Km//vjPr71m0N86yuXoDnhBeonx55IOoISaCQNtc4q\n1jnX8PgfWsl3qPXMaTZVFBrzhQD1yyvhFKpZmY5epyEumzDrzNS7muZsS5/AGwwyGowyEZ2gcUx1\nShWnrMA+re441ZTK2szVvNyxgyOjx7HozTzZ8ExyfamjmCZ3K9mWLIrs8wcrAmSanWSY0mhyt6CV\ntGSY0hYMObTqLSgoeMPjpBjsp9yCVSD4KBOMhfj58afIMKVxTdHlye5eZ4LhwCjf3PcdtJKWYCyI\nRWfmk1X3sDF73ZId48qCbWzKWseewQM81/oyjWMtbM7ZQDQe5ZcNv0VB4ZNV95xSKLVAIPjosSVn\nIx8M7OPo6HGa3W1UpJXydNPzROIRCmy5vNj+R6JylM05G7in/PakwxnAac5gNOhi0xLeu6ZjM1h5\nuPr+MzL2UlPrrOay/K281/cBx12NVGdUMhxQJxX/1LuLJncbRo16vz2xrHdT/lo+VXUvT9Y/w/cP\nP8bXN3yJbGsWiqJwZKSOFSmFiwrrPl+YcgpdGA0MVjsr2TWwlz2DBwDVdb86o5IVKYUzXGGgCjx/\ns+nPebljB292v8u3D/6Ia4uu4PoVV9Ht66XQlj/re/XE5/keX9+scUGdLBoOjJBjyeKSvC2817eb\nAf9Qcr1Bo2dtZs1SXfZJEaKQQLCMiERVUWh6ptB0zAYdXn9k3v0TLiKzYWr/cG8PciiErayC/Ewb\nLQG17EoJWRftFIrG4ui0mqQAo9fqeaT2U3jCXnZ0vQ36MG5fmJ1HB7h87dSMSlyO0+Bqwqq3UJxS\nhEbSUJlWzsbstdgNNsb9EeJeL6k5k13Eprl9NldlA6pTKMWiRz8plGk1WqrSyzk4fJTBwDC5VnU7\nozUCkgyKBsnq4dGObyN3zHRVlaauQKeZsnmmGdUuGE5zBvWuRlJOsDxXZ6yiyd3K5pwNJ3ULSJJE\nVXoFO/v3ALO/dE4k8SD3D7v+jYtzN/GpqnsX3F4gWE7UuxqTORlHRur4x4u/gUlnOslep0eLu41I\nPIJFZ6bWWc39q+48I5k+NoOVqwov5dWON+gcVzPOXu18IzlzXZ5WepIRBALBRx1Jkrir7Fb+z/7v\nJV2F7/V9AMAewKqz8FD1fXN2XCp1FOMNe8+YKHShsS1vC+/1fcCewYNkWzJRUNiUvQ6r3sqfet8H\n1I5XDsNssWRzzgZCsRDPNL/Afx74AZ+suhdFkXm07pdkmNL5xqavLrpM7lwzX6bQ+coaZw3/uOWv\nCcsRMkzpyY6986HXql3Vap3V/LL+GV7vfoeDw0eIK/Gka2g6hfZ8JCTSTKm4Q56kY+xE3CEPUTlG\ntjULh9HOv136D0txeaeNEIUEgmVENKZaGadnCk3HZNQyOBYjHIljNMwWjuZqSZ8oHTOXV1BgtSIN\nqaJQijaNQXfwpOfU3uflr777Hg9cV86V62bOpqQY7KBISIYQRr2W373TxsZVmVhNqvAyFBghpsTZ\n7Kzhwap7Zo09HlAFLrtFVfFXr0xnRY6dXXWDrC/PZEOFkzFfmLyMmV0FqjMqOTh8lOOuRnKt2YwE\nXPyq58focsqJDZSgsY4jE6citZRCez42gxWnOYPy1FLC8ang6jRTKpIksTazhje732VX/94Zx7k0\n72K0kpZteVtYDNNFoflCphNYdVNfcokQbIFAoNLqUR2N1RmrqHc1sb3zLe4ou/mMHKvPr4bmf3Xd\nI2e0JAJUl2RxSiGN7hbqXU283vUOGaZ0bitZXGtcgUDw0WdFSiHrs9ZwaPgov2r4LQDrs9Zg1hq5\nteSGeR0f91Tczs0rrxVB9ZMU2HLJs+ZQN9rA6oyqyWV5XLfiSmoyVvF00/Oszqicd9Lv8oJL0Gn0\nPNv8Ao8eezL5ubtCY/zdzn+ZVfqfZ8vhrzZ85axlzCwWT3gcuDAyhUAVRrNPo/yxLHUlf7f56/z8\n+FPUuRoAVSg9EYvezOdWf5J0UypPNjxL70Qfvb5+JqY1sNFpdMmStfm6MZ9tRNCEQLCMiMyRKTSd\nmuJ04rLCK7u75lw/tyiklk6Zy1WnkMas3vSyrZmM+yMEQvOXowG88n4HsbjMO4f6Zq3TSBqIGtGb\nIty2rZiJYJQX3u1Iru+f7FCWN0+tri+gZg3ZLeoXqE6r4fO3VqPXaXhieyP9o36iMZn0FOOM/arT\n1bbQ9a4mAHon+lFQ0FjV2RBJp4pNNxRfzV3lt3L9iqvYkLUGSZIw6UxkTIY3phlV++9a52oAonIs\nWVecY8nCpDNyVeGlGBb5BV+RVpZ8SFgoZBrUoOkErpAbRZm7k4dAsBxp8bRj0Oj5bM2DpBlTebvn\nvWRXv6ViR+fbvNT+R3p9A0hI5FpzlnT8+UgEVj9W90sUFB6svDvZklkgEAgAbiu5AY2kod2r5qI8\nsOouHqy6Z8ESILPOdEZLbS80JElic84G4kqcN7rfAdS8IYCajEr+5ZK/475Vdy44xiV5F/EX67+I\nTqPDE/ayMWstNxVfQ3FKEUX2guSfDFMaPb4+/tS3i4moP5lXcz7gCXsx60zL4nvGpDPycPV9pBlT\n0UgaSuZwCgGsz6plRUohhbY8wvEI39z3Hb53+NHkn28f/BHPtbwMQLZl6fO5TgfhFBIIlhHhyfIx\nwzxOoZu2rOC9owNs39PFFWvzyDghRycYjiEBpkkXkaIoBFub0Toc6DMzKYjG0ZoD6BQD+anpNNLH\nkDvAyty5HzKC4RjvHuoFoHtoggGXn9xprp1wJI4cMYLVx7WbCnj36ABvHerlsrW5FGXb6fMPApBv\nm/tly3eCUwggz2nlnitLeeqNFn74gtrhLN0+8zodRjuF9nxaPR2EYiFcIbVbmDUlRl6OnT6dKjZZ\n9TMdRgkK7Pm4Qm4yzKo4tNJRhN1gwxeZoCajki/WPozhNLI9LHozxSmFtHu7TuoUskwrhYnKUSai\n/gvGiiwQnEkmIn4G/ENUppVj1pm4q/xWHq/7Fc+1vsSX1nxmUWPE5TjeyPi86+tdTfyh/TUA9Bod\nWZbMRYu/H5ZEMGU4HuHS/ItZlV52Vo4rEAguHLIsmVySt5mdfbtZlVY2YyJJsHjWZNbwQturDE3m\nCWWf5NlsLlY6ivjc6gd5s/tdbi25gSyLk1tLbpixTSAa4P/94D/4Q9trPN/6CjcVXzNrm3OFJ+y9\nYFxCS4FFb+HrG76MK+RSKxoWoCx1JfuGDpFldiabPygovNb5JsNBdSLqTIS2nw5CFBIIlhGRWBwU\nhczWg0SrUtBnzrQsGg1arttUyLNvt9LW751TFDIZdUkrbGx0lLjHg23jJjxhL880v4DGFCDfnk+6\nrO67UEbRvsZhQpE4BZk2ekcm2NswzO2Xrkyud0+EUSJGsHkJyUEevLac/372CE+93szfPLiBvoRT\nyLo4p1CCqzcWcKR1lOOdbgDSUmbPbtRkVNLj66PJ3cboZAt5rTnIV++q5e+27wTANs9D1B2lN3NR\n9vqkxVojaVjjrOH9/j1UpJUu2DXsZFyadzH+aIDik5ShJDqyJRgNjglRSLDsqRtt4J3JrIey1BIA\n1mfWUp5awrHRBo67mhiP+BgNurh15fXz2v4frXuSY6MNizqm6hA8Oy4hgGKHmq/mMKRwZ+mZKYkT\nCAQXPresvI6xkJtrC68416dywZJlduI0pTMaGlO7tZ2mk6rWWU2ts3re9Ra9hdtKb+TppufRSVre\n6d3FdSuuOufNAyLxCIFY8IyXRp9vZJjTkhO/C3FJ3mbybXkU2fNnNHzpGu+hztWIhETmSZz/Zwsh\nCgkEy4hoVCY9Ok72rldwKR5yPvf5Wdtkp6sBxS5vaNa6YDiGxTh1Uwu2TuUJ7Rs8lOwCVuuswuZT\nhZiEW2cuugZ9AHzy+gr+65nD7Kkf4rZtxckXMbcvjBJRxaWu8R5Wl1SxvtzJoZZR9tQP0TcxgMOQ\ngs0wt2Mn6RQyzxSFNJLEZ2+p5h8f34M/FCPdPpcotIrtnW9y3NWIO+QBwB8NoDfISHp13PmcQlkW\nJ1knzBbdVnojZakrqUwrn/fzWAxbcjeyJXfjSbfbmnsRHd4ubAYbO/t2MxYaS5aVCATLEUVReLrp\nedxhDxISq51qBoQkSdxTcTvf3Psdnmr8Hd7wOAoKVr0Fo9bAGmfNDEE1LsdpHGvBbrBRlV4x57E0\naFiXtZqfHnsSWZHnbUd7JrDprXxt3SOkm9LOWHi2QCC48Ekx2PmztZ8716dxQSNJEtUZlbzbt+uk\nXWE/LJflb2VLziZe73qbVzvfYO/gQS7Lv/iMHW8hIvEortBY8vn4Quk8drbRSJo5n703Zq+jztVI\nuintrLmIT4YQhQSCZUQ4Fscoq4JGuHdm+LAcCiLp9GSkqC8Ro+NziULxGfk700OmWz1vAfCvl/w9\naaZUDreotsiJYHTWOAk8E2ooc57TytoyJ/sbh+kZnqAoW7VjenxhYqP56HJ6+G3Li1SklXH/NeXU\ndYzx9J/qiVZ6k/k/c+ELJpxCs2dS0uxGPndLNc+9205F4ezWn8UpRVgmW9PrtVO3Sl9sHI0+CrL2\nlG7kNr01aR09G1j1Fj5f+xDHRuvZ2bcbV8h91o4tEJyP9E0M4A57WJe5mgcr755RLpFvy+Wy/K28\n27cLAL1Gz+9bXgKgM7eHB6vunjFOVI6xxVnDJyo/vuAxK1JLaXS3nFVRCNT8MYFAIBCceWoyVvFu\n365kntCZxKDVc2n+xWzveovftbxIi7uNLbmbqEovRyNpiMQjtHu7WJVWdtKutvOxvfMt3GEP91fc\nOe8Y3z/8GG3eqYzPRIamYHGscVZj1Vsocaw416eSRIhCAsEyIhKV0ctqrlC4vx8lFkPS6VBiMTr+\n199hW7cO5z0PArOdQoqiEIzEMButKIqCEokQbGlBMprQ5+fT3tWJ05xBmkn9YkiUbE0EFhaFdFoN\nVpOOLVVZ7G8cZk/DUFIUck+EUQIOau2bOObbx/bON7mt9EZu2FzIq8cOYYQFX7aS5WPWue2168qd\nrCuf27apkTRUpVdwYPjIjOWu0JgqCsXPrWV3saRPhl67JkvgBILlSsLJuD5rzZz5GbeWXE/9WBM1\nGasoTinixbbtBGIB6lwNyIqcDHnvmGz3nsjuWYhbSq7H1msVIo1AIBB8RFmVVsb6zFo25aw/K8dz\nGFN4uPp+Xu14gwPDRzgwfASHwc59q+6ka7yXP3a9xV1lt3JN0eUoinJK4lBcjrOj6y3C8QgXZa8n\n3ZRKmjF1xhiBaJB2bycZpjSqMyrRabRsy9t8Ji71I4tJZ+IfL/4GBs358y4hRCGBYBkRCMXQKZPd\nwOJxIoMDGAsKifsniHs9RPr7sZj0mI1aXCc4hXbXD6EoYDPrGd/1PkM/fwwAS3UNg6FRgrEQa5w1\nye1tlkT52EKiUIR0hwlJklhTmoHJoGVv/TB3X1GKJEm4faqT6Jr8q+npaObN7j9xce4mVq/M4LVW\ntfQsb4GsDt9knpHNfHq3upqMyqQopNPoiMkxXEE3ijaCErww8nkSndCEU0iw3Dk6Wo9G0lCTMbe7\n0Kq38E9b/yb58+acDTxZ/wx7Bg/Q4+tLZiZ0TopCiynHLHGsOK9mAgUCgUCwtOi1eh6p/dRZPeam\n7HVszFpL53gPewYPsHtgH882/wGdpEY8vNj2Gm92v0uqycFnax5YdNZRt6+XcFx9dv5Z3a/xRsa5\nreRGbii+OrlNu7cTBYXNORvOm7DrCxHbPBEU5wrRkl4gWCZ0D/l4aVcnBiWeXBbu7QEg7g8AagkZ\nQEaKCZc3lGxjvrdhiEdfqsds1HHz1hUEGo6rA2i12DZdRJunE4DSaa0Z7WZV/Z6vfEyWFbwTkWS5\nml6nZUNFJq7xEG39alcfz6QolOWw8/HyjxFT4vy+5UUKMq1oLKootKBTKBjFatKh1Zzera562stj\niUO9tgH/IGjiyFF9spvb+YxJZ8KqtzC2gCgUiAaJy+f/tQgEp0ubp5NuXy8VqaWYdeZF75cI/tw9\ncIAeXz89vn7aPZ2YdaazUiogEAgEAsFcSJLESkcR96+6ky25asOX0dAYWRYncUUmGAvSNd7Dv+/7\nLodH6hY1ZpO7DVC7ZiY6bL7XtxtZkZPbtHrUsrFEswbBRwPhFBIIlgnPv9tOMBzjmjVZoHZyJ9zT\nAxeD7PcDIAdVd1BGioneET+BcIyx8TA/e7UBk0HL3zywnqJsO119fUgGA2Xf/zGSRkPb8acAKHVM\ndQ4zG7VoNdK8QdO+QARZUUhPmQpC3VyVza66QfbUD1GW78A9EUarkbBb9Ky31FKRVkadq5G2iRYM\ndj+yIpG9wIuZLxCZM09osdgNNorsBcmXyWZ3K92+XgCUmAFfIILRsfgXzHNFhimNfv/QjBKYBFE5\nxv+3+z+wG+x8Y+OfzQqmDUfibN/bTdegj221uWxcJV6EBRcWcTnOM83PA3BLyXWntG9lejlaScu7\nfbuSeUMAlWnls36XBAKBQCA4F2zL3czOvt0A3FZyE4X2fFIMdg4OH+Hppud59NiTXFV4KVfkb8Nu\nsKLX6Hm7dycaSUOmOQOnOYMMUzrN7lYAHln9KfYPHSEUD3FstJ4md2uysUKrpx2NpFlUCbXgwkGI\nQgLBMmE8EEGv01CabWF4clnSKRRQRaF4wik02Yq+e2iCX7zWQCQq82d31lKUbUeJx4kM9GMoKESa\ndOC0eTqx6a0zBBpJkrCZ9cmw5xPxTKhiUfq0tvfVxWnYzHr2NQ5z39Vl9I/6yUw1o5msZb634nb+\nbe+3+V3ziygmH3LQSiAkkzJHZ3hZUZgIRslJn7tt/GJZm7mabl8vq9LL2NH1Ft2+PnVFTM9EMIrz\nAhCFCmx5dPv66PX1U5RSMGOdKziGPxrAHw3w8+NP8eW1n52x/mDzCH/Yqc4KjQciQhQSXHC82/cB\nfRMDXJy7Ken4WyxmnYmHq++nw9s1tVCCLTkn7wAoEAgEAsHZoCilgBUphYwERqnJWIVhslX9xbmb\nKLIX8Fjdr3i7Zydv9+wkzZjKx0pu4PnWV+YcK9+Wy2pnFaudVbR5Ojk2Ws/P657COpnFNxJ0UZRS\ngEk3u3Ov4MJFiEICwTIhGI5jNmhRwoHksnCv6npJOIWU0KRTaFKo+elLx/FORLhl64qkGBAdHkKJ\nxTDmq+LCWMiNO+xhrbNmVpid3aLHNR6e83wSncfSprWD12k1bFqVyTuH+3nzQC+hSJyyAkdyfa41\nmysLtvFWz3sggRyw0zc8QUpx+qzx/cEoijJ357FT4bqiK6hML6M4pYgCez7t3k4AlKhhwRDt84mK\ntDJ2Deyjyd1KUUoBiqKwb+gQ/RODlKVOubvqXI34o4HkFz/AsCeY/HvP8ARxWT7tcjyB4GzjDY/z\ncvsOLDozd5TefFpjbMxey8bstUt8ZgKBQCAQLB1fWfNZwvFwUhBKkGfL4X9u+ho7ut6mcayFLl9P\nsrvmfRV3EoqFGAm6GA26GAt7uCx/a3LfEscK1jhr6BrvJhxXn9sdxhQuzbv47F2Y4KwgRCGBYJkQ\njMQwGXXIEfWmLhkMxL0eYr7xpFNIicWQo9Fkzo93IsLqlencedlU3bC3S603NubnAyTzhEqm5Qkl\nsJn19I74icVldNqZQkJCFMpwzCxX2lKdzTuH+3nxfdWdUp7vmLH+5pXXsW/oEL7IBErATs+In6o5\nRKFEwHWKZfFt4+dCq9EmLbJlqSunRKGYnqffaiXncD92i56sVDPXby5MCiayotA16KM4x37abUGX\nioq0UgCa3W1szbuI3zQ+x+GRYwCMR9RsplSjA0/Yy4B/aIZQlOhCV1HgoLnXy8BogIKsCyNkWyB4\nvvUVQvEQ96+6E7tB/L8VCAQCwUcTm8GKjbnDi006I7eV3siWnA38857/xB8LUGTP5/KCrXNun0CS\nJL645uEzcbqC8wwx3SsQLBNC4Thmgw4lopZtmYrVF/9Iby/xSacQqGHTmalqSZTTYeILt9Wg0UyJ\nGm/t+z0A+oQoNCmSTM8TSpBw6cwVNp0sH0uZKQqVF6SSajMQDKvBx+WFqTPWm3Um7q24A4PGQHzc\nSe/wxJzXm8gysn1Ip9B0poslNr2NAZefg80j/OlwP799p4269qm274eaR/mXJ/azq25wyY5/ujiM\nKeRYsmjxtPNve/6bwyPHkl0P6lwNgNppDWDAPzRj30QXug0VqlOsc9B3tk5bIPhQNLvb2Dd0iCJ7\nAdvytpzr0xEIBAKB4JySbc2ifDIg+qKcDef4bATnE0IUEgiWAbKsEI7GMRu1yAlRqER1j4R7epLl\nYwByKERxjp0Hr6vgr+9fh80802ljd6nlRL7JrJ42Twd6jZ5Ce96s4yba0s9VZpVwCp0oCmk0Epur\nstVjWfRkp83O7NmQtYZvXfZPaMMOekbmE4WiyTGWiumtpb/ysQ08+o2r+O6fX8rnbqkCoK3fm1zf\nN6qe196GYc4HKtLKiMpRfFE/t5fcxBdq1Zkff1QtJ6x1qtcwOE0U8kUmGJCbsOUNErZ3gT5ElxCF\nBBcAarj0C0hI3L/qThEKLRAIBAIBcHvpTazLXM3FIhtPMA1RPiYQLANCkRgAJoMOJaiKQubSUtxA\nuLcbRZ5qNSkHg0iSxDUbC+YaCnsgTlQLgxo/jmiAAf8Q5akl6DSzbyf2SUFprg5kiXbz6Q4zwYnQ\njHVbqrPZsa+HioLUeUuvdFoteU4r/aN+ZFlJupniskxDp5u3D6mB0EspCk1vZW3WmdBoJOwWA+vK\nnQC0948n149NZik1dI0RDMcwG8/t7fbqwssIx8NcWbiNInsBwdjUZ67X6JMzR4N+VcTq8fXxwyM/\nI5yrikDbBw5jKM6mayj77J+84LSpdzXRMd7Nxqw15FiXz7/d2707GfQPcWneFlakFJ7r0xEIBAKB\n4LxgpWMFn6996FyfhuA8Q4hCAsEyIFGKZTZqkb2qQGMoKETS6wn39qJzTOX2yKHQnGMkMEYUwgYN\nA4Eh9Fo9Cgqlc+QJwVT52FwdyMZ8YQw6DVaTjuAJZp+VuSl85Y7VFOfYFzyXwkwb3UMTDLkDZKdZ\neGFnO+8dGcDrV68xJ91CRUHqgmOcKvdW3MHugf0zOq1ZTXpy0i2094/TP+rHZNAy5lM/x1hc4Vi7\nK+l+OldkWjJ4qPq+5M9mnYl0UxpjITcZ5nRMOhNpxlQG/IPUjTbw+PFfE41HifaXUJyWw7jjKBN2\nL92dvhkinOD8pX9ikJ8ee4KoHOPVjtdZYS9kS+5GNmavTZYPfhSZiPp5peN1rHoLt5XedK5PRyAQ\nCAQCgeC8RohCAsEyIJhwChmnMoW0JjOGvHwifb0wrZuUHAzOOQaArMgYojIho4b+iUFisjruXHlC\nQLL0zHdC+Vg4Gqd/1L9gCPOmyqyTXlci8Lh3xE/30AQv7+rCatJx5fp8LlmdQ2leypKHPF9RcAlX\nFFwya3lpXgrv1w3yD4/toSjbRjyuIAEKalv3cy0KzUWeNYexkBunKQ1Qu7vVjzXx46O/QKfRckvu\nnfx2b4Cy/CLcKV6ORo4TIcCAy09+5pkN7Y3LcSRJEmU/p0kgGuDxul8RlWPcWHwN3b5eGlzNya4j\nm7LX8YnKj6Ofw+F3Ii3uNjTWFVwojwxtnk4i8QjXFl0xo5OeQCAQCAQCgWA2F8YTnkAg+FCEEk4h\ngw45rIpCksGAsbCQcFcnkd6e5LavN+/Ap2/inorbMetm5v2E42EMUQWvTWLAP4gv4kNCothRNOdx\nE6Vbu48PUl7goChbdf50DowTlxVKT+gsdqokRKGe4YlkmPVf3ruW0rwPN+7pUDIpCgF0D01gNGjJ\ny7QSjsQ52uYiGpPR684vgSPPlkOdq4EMs9q9LSEKWfUWvrTmMwz1GoB6nA4Tdns+R0ePo7GO0zXk\nO6OiUEyO8d8HfwTA/9z0tTN2nI8q0XiUHx39BYOBYa4uvIyPldwAqO3Z9w0dYlf/PvYMHqAopYAr\nC7YtOFarp4PvHPoJBW25fGPD1+YsEz3f6B5X72clKStOsqVAIBAIBAKB4Px6QxEIBGeEhFPIbNSi\nRCdFIb0eY4GataHEYsltB8d62DN4gMaxllnjBIIT6GSI6CWGAiP/P3vvHR/HXef/P2e2r3bVe69u\nsuXe7ThOHCekECBOp4XvEUK50LmD33EX7ijHHQcccJCDCyQQEpKQhISEVMd23LstW7Ykq/e2Wq22\nt5nfH6NiWSvL3bL9eT4ePIh3PvOZz+zsrnZe+3q/3jQOtpBryxonHg0zLS+RWYVJ1HcM8tjv9vL4\nK0fp6vdR164FMpeepyiUNyRMtPV4qG0dwGiQKcg4fcnZxWJWYTK6k0qqgqEoKfFmFkxLIxCKcrzZ\nCTCSgTQVyLVp4eAZVs2VtTx7MUszF/L1RV+gKCGfvqF29CnxZvLsWrc52Tp40TuQvdG0kebBVpoH\nW0eCsAVnzuHeozS4mpifXsGHS28beTzBFM+6/DV8ecEjGHVG3mp6D194BwscegAAIABJREFUrDNQ\nURWC0RCqqmphzTUvA9A22Mm7LVsu6XmcTIOriZ8f/A3f3vEDTjgbTju2aUgUyo+PnYsmEAgEAoFA\nIBhl6v/kJxAIzht/8KSg6VAIyWBAkuURUehkjGEtdLo/4Bw/j2cAgJBB05MVVWFacumEx9XrZL56\n7zyqGvt5cUsDe473sK+6l0S7ljV0vk6h+Dgj8XFGTrQN4A1EmFWYhF53ebTujGQrP/viarZVdvLs\nRk1QS7abWDAtjbf3tnLwRC92q4F/e2oft68o4CPXlVyWdZ7M/PQ5PMQDzE0tBzSn0HDuUCgcpaqx\nH4CUBDN2u3aDLce5LmoHsubBVt5u3jTy7zZ3B9NP8xoTjGdYFLkhb1XM8ju70cYNeat5s2kj/7Dt\nO8QZrCiKQkgJER4qCZ2WWMKCjLl0eLtYkF5B42AzbzZtZHnWEhJMl1Z4VVSFJ6uexRFwIiHxm6O/\n5xuLHiV1yOE2jKqqKKpCi7uNNEuKKB0TCAQCgUAgOAOEU0gguAYIhE4Kmg6FkIyaKHOyKCSZTAAY\nw5qLJbYopDl8zLYEpiWVclvRTdxetP60x5YkidnFKfzzJxfxuQ/NxmSU6R8MkppgJsluOu9zu25u\nFt7A0I3sBQ6VPlssJj25aaMBvknxZkpzErBbDRw80cfhuj4ANh1oJxSOXq5ljiBLMosy5mHQje3Q\nNugL8Z9/OkhN6wAzC5LITo0jwWQnwRiP3u6hpdtzUdxOYSXC748/j6IqrMpZBkCrp/2CH+dqp9nd\niizJI06wWNxScAO3Fq6jwJ6LVW/BbrKTHZdFWWIx6ZZUagfqefHEXzHIBjaU3cld5bcSViK827L5\n0p3IEMf7a3EEnKzIWsK90z+MN+zjpbrXxo17pvpFHt38TXwRv+g4JhAIBAKBQHCGCKeQQHANMOwU\nshj1qKEgslETY3Q2G/qkJCJOJ4aUVEId7dgUTSDoDwyMmyfkHcQAxMen8sX5D5/VGiRJYtGMdMxG\nHT99oZJZhUnnd1JD3LGikL3VvXT3+yjLu7yiEEBO+mjWTrLdhCxLzCtNZWtlJ5sOagKHNxBh97Fu\nVs+d+Kb9ctHV7+Onzx+mZ8DPslkZPHTrTOShsO78+ByOhI4TVH1s3N/G9fOzMeh1F+zYuzv30eXt\nZnXOcm7IW8229l20uoUoNBH7ug9xqOcIObZs8uNzyB8SeFrdHWTFZWDUGSfc16AzcFvxem4rHi/q\n9vj6+O7u/yKshLkx7zoSTHbWZi3nxaNvsLV9FzcVXE+88dK4hQZDbt5r2QrA6pxl5Nlz2N6+i8re\nKpyBAZLM2ns+rETY33NoZL8CuygdEwgEAoFAIDgThCgkEFwDjJSPmfQooRCyeTQDyJiTR8TpRJ+S\nQqijHbtqwCAbcMZwCoW8HgyAzmI557XMLk7h3x9ZNtKu/nwx6HX8/UfmcLiuj+n5l18UirdqJW2D\n3hDJ8drzvGBaGlsrO3H7wiTZTQx4gmw/2jXlRKH2Pi///vR+vIEIt68o5MOri8Z0b8uz5XCk7zh6\n+yDPbjzB6zubuGFBLtcvyCH+DK+nL+zHpDOik8eLScM5VjfkrSbVkoxZZ6bV3XFBzu1qIxwN80Lt\nK3jCXg72Hhl5PN+eQ1gJU2A/d6dMujWV9QXXs7vrADcVXA9oItL6grU8V/sy77Zs4SOlt+MOeejz\n91M0QdD8udLl7WZ31wGOO2po9WjXvyg+fyQj6Lrclfyx+gXeb9/JnUMt5+sHGglGQ2TGZeANe5mV\nMuOCrkkgEAgEAoHgakWIQgLBNcDJ5WORUAg5Pn5kmykvD9/RSnTJmnPHFFZJNifFdAqF/R4AdNa4\ncdvOhtSEcxeVYpGdGkd26vmt6UKSmxbHMW+I5HjNkTWrMAmTQUcwHGVBWRqVDX1090+9AOXtlZ14\nAxHuX1fGTYvGiwrDN+W3XJ+A0lnA5oPt/GVbI+/sa+U/P7cCs/H0f1L8ET//vPMHrMxeOiYAGbQ8\nmBMDDSSaEkizpCBJErn2LOoHmghGQ5hO43q5FjnYewRP2MvqnOXMSp5Gi7udKkc1Le42AArOM2T5\n9uKbuX2oa9kwy7MX81bze2xt28lN+dfzx+o/U+Wo5l+X/+OIY+dC8Hjlk/T6HeglHdOTSpmVMp1l\nmYtGti/KmMdf6l7n7eZNNLlauDH/OmqcdQBsKL2DmSnTLthaBAKBQCAQCK52RKaQQHANcHL5mJYp\nNJrlYy4s1P4jLRXQMoWSzYl4Iz4CkeCYeSI+LwCG8xSFrnZuW17IbcsLSE/UxC+DXsecYi0Ud3p+\nIqkJFlzeEOHI5c8VOpn2Pu36Li/PjLl9uANZX7CbDdeX8KPPr2B2cTLeQIQBT2jS+Xt8ffgjARpc\nzeO2dXq78YS9lCWWjLiTMsxZqKg0u1rP9ZSuWt5v24mExLr8NVSklXN78Xq+OP8zI53kihIufDt2\ng6xnfcFaQkqYV+vfpMpRjaIq1A00XrBjDARd9PodTEss4YerH+PR+Q+zLn8NNuPoZ45RZ+Bz8z7F\njKQyagfq+VXl73ivdStG2UBpYtEFW4tAIBAIBALBtYAQhQSCawD/kFPIpAeiUWTjqOvCNn8hWZ/9\nAtKyhURkMIQVks2aa+jUsOmIT3O3GKw2BBMzsyCJu9aUjCm9unN1MesW5TK3NIWUobIyx2Bwoiku\nCx19XhLijNgshpjbE4zx2I02WoZyfsxGPTlDDq1h4fF0DAS1oPI+v2PctlpnPQDTkka7svW2a6La\nrpZjZ3EWVwd7ug7wh2PP817L+9QNlUYN0+tz0DjYzIzksjEduMx6E4/O/zT/b/ZHybFlXZR1rcha\nTIIxnh2de1BUrVNhvavpgs3f6GoBYEZyGWb9xEH0hfH5/P38T/PNxV9iaeZCZElmbtqccaHpAoFA\nIBAIBILTI8rHBIJrgMBwphDaTZx0kigkyTL2hYtwDrYRNkjoQ9ExolC2bdQ1ovj9AJjiRsvPBGdG\nTmocD6zTylpSEzRRqM/lJzN5arTNDoQiOAYDzCyYOABckiTy7blUOapHQn4tJu3PiC8wuSjkDGii\n0GDIPaYkTFVVDvRUAjAtqXhk/ECXDTUbaodKg64VXq1/k7ea3xvzmIRERVo5n5x1P5V9VQDMT58z\nbt9EUwIL0isu2tqGs4VeOPEKAHpJR/0FcAp5wz62te+iz98PQPEZOp1y7dl8fNa93DPtQxhk8ZVG\nIBAIBAKB4GwR36AEgmsAfyiCTpbQqdqNu2QYn88SiAYI6SWsoSjJQ/kgp+YKqYEAIESh8yVlSBRy\nuAKXeSWjdDo0F9hk2UyliUVUOar5t90/YkX2EgwGzdlzJk4hZ3D09dTnd4y4WQ70HKbe1ci0hDLe\n2d7PrcvtxJn1tHeF0SXacVq7CUfD14QLJBAJ8nbzJpLNSXxy1v04A06a3W3UOOs43HuU/zv6B3xh\nHxISc1JnXZY1rsxewqa2baSak1FUhRMDDfjCPqyGcxc432vdyptNGwGQJfmsW8qfzlUkEAgEAoFA\nIJgYIQoJBFcxUUVhW2Un/YNBzEYdhMIAY8rHhvFHgoQMMnZ/ZMQp1DTYwnUsHxkjDYlCZiEKnRej\nTqGpIQo1d7mpbtFKBScThW7IWw0qbG7bxqbWbUhsxzg9mY19jeR57yAjLn3CfYfLx2BUFApEArx4\n4jX0sp5UzyLe2ddKV7+PDdeXEIkqSO5k5Dg3tf2NlKdd/QHCXb5uVFQqUmdRklgIFLIocz5hJcKv\nK5+iylENaE6aS9UW/lQMOgP/35IvI0sybzZtpHagnhMDjcxNKz/nOY/0jZYI5tqyMYpgcYFAIBAI\nBIJLgsgUEgiuYnZVdfPUmzU43UEsQ+3oASRTDKdQJEDAJCGFQmSHLGRY09ndtZ99XQdPGqTtb7Ff\n/tbvVzKjmUJnLwptOdTOl3+xjUHf5MHOZ0IgFOE7T+7lhU1apk/OJKKQXtazvnAt/7rim3x85r0k\nGVLRJThoCdXw3wf/l/878gd+d+D5mPs6T3Ke9Q7lCr3e+A6u0CDr86+npVUrbzzS4OBPG7X29Aa/\nJjLtbb02coU6PN0AZMeNDfs2yHoervgEK7IWA7Akc8ElX9vJGHVG9LJ+xK20uW37Oc/l8Dtp93SS\na8vGbrAxP218WZxAIBAIBAKB4OIgRKEJeO9AG9/41Q48/vDlXopAcM4crh8N9HX7w6ghLdg4llMo\nGA1SVWxBUmHgzy/wmTkfx6wz8XT1n2n3dGr7BbX3g150HzsvkuJNyJI0rnyso8/Li1vqiSrKhPu+\nf7gTlydEU+fgBVlL78DYNUzmFBpGL+tZmrWQB/P/H/596yiTl+MKuTnYe4Q3TmzCFRy/vrFOoX46\nPF1sbttOqjmZ67Kvo6FjkPQkCxaTjuPNmnNpVUk5qgq1Aw3ncZZXDp3eLgCybOM7wBlkPQ/M2MB3\nV3yLVdnLLvXSYlIQn8fM5GnUOuv41eHf8XztK6iqesb7u4KDbG3fCcCqnKV8f9U/sb5w7cVarkAg\nEAgEAoHgFIQoNAEHa3vpcwU41tR/uZciEJwTkahCVaMDnax1wAqGoqNOoViZQpEgNYUmlMJcPAf2\nk9Dn4+Oz7iWshPl15VP4wj50wTBRGSTD1Z/tcjHRyTJJduO48rHnN9Xx+s5mqhqdMffz+MMjYlB3\nv/+CrKVvQJsnzqxn7YKcCTuPTYTVbABFT1q4nEfnPcx1OVq5Yc0p4dCKqjAQHBxxwPT6+niu9mUU\nVeHuaXfS0uUjqqgsKEvjvhvKAJAkuHlhCaovHpfaQyh6YdxRU5lOr+YUypqgDE+SJJLMiWM6211u\nPlC4DoCjjuNsadt+Ri3qHf5+nqn+M/+04/u807IZCYnZKTORJfG1RCAQCAQCgeBSIjKFJqC9zwtA\nbesAS2ZmXObVCARnT12bC38wytr5OYSjCmU5CajhPmCCTKFoACQJ3eL5qE1tBNtambtyNbcU3MCb\nze/xu2PPsigUJWLQTakb0iuVlHgzJ9pdRKIKep2MyxviaIMmQte3u6goSRm3z/FmJ8MejG6n74Ks\no3dImPr4LTNYPGPiPKCJsJ7UfWx68iwsBjPvt++k1lk/psTJHfIQVaNkxKXjDnuocdahojI3tZzZ\nqTN5sUorX5tRkMic4hTq2jVXUZLdhC2aiU+qpcbRyJz06ed7ylOaDk8XSaZELHrL5V7KGVOSWMgj\nFZ9kIOjiTzUv83bLJspO6iJ3Ms7AAG80bWRn514UVSHDmsbC9LkUJhSQZBZlqQKBQCAQCASXGiEK\nxcAbCDPg0X6RrmkdmGS0QDA1qRwqHZtbmjoiMLj3dwAgGcd36glGtdIyU2oaASDi0Pa/rXg9Le52\njjlqWB6KEjaJj40LQVqihdo2Fw5XgIxkK7uPdaMMld3Ud7hi7nO0YbQcsNs5uVMoHIli0OtOO6Z3\nyCk0HH59tljN2uthuPtYri2bOKN1XBv54dKxJFMCSzIXsLtzP3GGOO4q+yC+QIStlZ0Y9TJluZoL\n5qFbZ47sWxxfzFFq2dN67JxEoVA0RP1AE0ElRDgaJqxEiChhrQQucyE6+fTP0aXgYM8Rnq/9C4Mh\nN7OSrzzhazhbaF/3IY45amh1d5BnzwbAH/GzpW0nESXMe61bCUZDpFtT+UDhOhZlzBPuIIFAIBAI\nBILLiLi7i0F7r3fMf7t9IexW0QlFcGVxuL4Po15mRv7or+/qUPlYLKdQIKKJQubUdAJAeEgUkiWZ\nh8rv56cH/xdjuBev9fLfQF8NZCRr7bu7+n1kJFvZebQLWZKIjzPQ0DGIoqjI8qgjS1VVqpr6iTPr\n0ckS3f2ndwq1dLt57Hd7+ej6adywIHfCccPlY2mJ5+ZMMRuHnEJDopAsyZSnT2NP2yEe2/lDDLIB\ng84wUvqVZErghvzr+Ejp7SNzPPNOLYPeEB++rhhLDNFxcd4MjrS8RZ3r3HKFXql/Y8IgZG/Yxw15\nq1FR0cuX70/itvZdDIbcAGTFXbnu1PUFN1A38ATvNG/iU7MfBGBjy1beaHoXAIvezIMzNkwZMU4g\nEAgEAoHgWkeIQjHoGCodS0+00DPgp6ZlgEXnUFYhEFwuegf8dDp8zC1JwWgYvfGarPsYgDUtiwEg\n7Ogb2WY1WPnK/M/SEnkEfULaxV38NULmkCjU3e+jrddDc7ebuSUp2KwGth/posPhJTfNNjK+0+Gj\nfzDI4hnpuDzBMaVnsRh2ij39du3pRSFXAItJR5z53P4cyLKExaQbcQoB3FSymkZHK8FoCE/YR0TR\n3DkGWU9RQuGY/Vu63Ww80EZGkoVbluTHPEZ5fjrq8XgG43oIRIKY9SYc/n4GQ54J12XRm8gcEldO\nDDRgkA3cWfIB9LIeg6xHL+t5ofYV/tb0Lhtb3yfBGM9XF37usrRCjygR6l1NpJiTmZc+m7W5qy75\nGi4Us5KnkWvL5kBPJbf7bibNksLe7oMYZQMfm3UvJQmFJJjiL/cyBQKBQCAQCARDCFEoBsNOoVuW\n5vP7t2p4a28LC6eniRwVwRXDsCBQUZo65vFwj9bZKFbQtH+ofMxitqGLjyfSPxqyrkajDPzxjwAk\nZsa+cRecHSNOIacf11DHqRVzsvD6w2w/0kV9u2uMKFTVqF2P2UXJnGh3UdvmonfAT1ZK7G5hJuOo\nGOgPRmI6cFRVpdflJyPJel6fbxaTHl9gVBSamzmLx5b/w7hjqagMuEMEw1FMBh2KqvL027WoKjy4\nfhoGfWyBy2LSE69k4ZFcHO+rJz8hi8d2/QeKOnGXNoAvzf8MefZcOjxdFCcUsjZvrNgSjAR5puZF\nQtEQ7pCHP5/4Kw/MuCvmXMFoCFfQhd1ou+B5P02DrYSVMLNTZ4xxUF2JSJLE+oLr+W3VM2xs2cKy\nrMX0+R0sypjHgvSKy708gUAgEAgEAsEpCFEoBu192q/Py8szOdrYz4HaXg7U9rFwunBICK4MDtdr\nLp+K4tGw4sHdO3G+9Sa6hAQsZWXj9glGghhkAzpZhz45hVBbK6qiIMkyvX9+nsFtWzEVFpG64Z5L\ndh5XM+lJmrDQ2eelZ8CPxaRnXmkKTV1aCdGpncmqhjohlhclM+jTHF/d/ROLQl5/eHTfxv6YbsdB\nX5hQWDnnPKFhrCY9/YPB046RJAmvP8K3frOL6yqyeeCmaWw/0kldu4tF09OYXTQ+WPtkihOKqVSr\n2dt2HHdkAEVVmJM6i0zr+PMaCLrY232QamcdKqCiUpQwXsxcnr0Yk95EdlwmTx57lu0du5mRXDZO\nvIgqUb6/5yf0+R2YdEb+ft7DMec7U9whD82DrSSYEkg2J3LCqYVsT0ssOec5pxLz0uaQaklhV+e+\nkSypxRnzL/OqBAKBQCAQCASxEKLQKdS0ODnR5iIrxYrJqOOuNcUcqO1l08E2IQpd4SiqCipjclqu\nRoKhKNXNA+SmxZFy0s2+6/0tAOR+5Rvo7ePLNwLRAGadFkBtSEkh2NRIdHAQX81xBt55C2NmFrlf\n/QY6y5XTFWkqYzLoSI43Uds6gApcNzcbg1430hLec5KoE44oVLc4yUqxkhxvJiNpqPTsNB3IvCc5\ndw7V9cUUhc43T2gYq0lPe9CLoqrIp3Ec9Tj9hMIKjV2DeANhXthUj8mg474bx4uUp7IkfzqHG9+g\nYbABxagJZ3eXfZAUS/K4sb6wj73dB2lwNWOSNVdcUfx4EUeWZBZlzAPgU+UP8sO9/80z1X+mwJ47\nZt5DvUfo8zvIt+fQ6u7gf488yTcW/T3J5qRJ1x2L31Y9My6IG6A0MXbHrisNnazjpvw1PFvzEkcd\n1eTaspmZPO1yL0sgEAgEAoFAEAPR8uMkup0+fvHSEQA+tl7r/pKVEkdqgpnWnomzKwRTB0VV2V/T\ny57j3bg8Y50L//WnQzz631t5a08LqqpOMMOVz/FmJ5GoQkXJ2NIxxe9HMpkx5eTE3C8QCWDWD4tC\n2r6ew4fofup3yGYz2V94VAhCF5iMJOtIi/kVszMBiIshCtW1DRAKKyNummGX0ek6kJ3sFGrrjf35\n1eO8MKKQxaRHRRMkT0f/oOZ+6nX62bivDY8/zB0rC0mOn9ypNCsvHdWbiFvtpbr/BOnW1JiCEGgZ\nWJlxGTQPtlDvagKgcBJnT2ZcOndP+xD+SIDfVT1LVBk9l02t25CQeKj8Ae4quwN3yMPjlU+OhLOf\nDe2eTmqddeTYsrguZwVzUmeSY8tiVc4ybMbYrq8rkaWZC8m0ppNvz+Xv531ahEoLBAKBQCAQTFGE\nU2gIXyDMz/5ciTcQ4ZMfmMGMgtFfgPPSbRw80YfLGyIhTnQhm8pUNfbzPy9rwl5uWhyPPbQEWZaI\nRBWONzsBeO69OiwmPdfNzb6cS71oVA6XjpWMLcdR/H5ky8Q334FokPihAFh9irZvzx9/D4pC1ucf\nxZiZdZFWfO2SkWzleLOT1AQzZbkJACOBzyeLOkdPKh0DRp1Cp+lA5glo+6fEm+kbCMQc0z4Uqp+T\nen5ixHBbel8gdnbRME63JqIM+sIj78dVFWf2ujIZdaQH59Jr20xYCU/atr04Pp8d3m6O9deQZEok\n0ZQw6TGWZy2iur+W/T2Heb3xHT5Ycgs1/XU0DrYwO2Um6dY00iypdPl62Na+iyePPcvDcz5+Ri3V\nfWE/z9W+TI+vF4Dbi9ZTkVZ+Rud+JWLQGfjmki+hk3Qij08gEAgEAoFgCnPFOIVUVeWdva18/Zc7\n+OyPt7C/pve85uvo8/LZH2/hYG0vUUXhV385SqfDx/rFeePEgpyhsNc24Raa8jR2DAKQlmimrdfL\nziotwHfYUVFelIzFpOO5906MuBauJlRV5XC9gziznpKcsSViSsCPzhzbEaKoCsFo6KTysSGXkaKQ\nfPsd2OYvuKjrvlYZ7kC2YnbmyI2zTpaxmPRjnEJVDf3odRLT8xIBTSBJtBnpOV35mD+CXieRnRqH\nLxjBFwiPGzPsIMpJOz9RaFgIOrkDWSz63aPvudq2AeLjjMRbz1xov2/pciJtWhnSjIRZpx073OVM\nURXuKL75jOaXJIn7Z3yEFHMybzdv4nBvFc/V/gUJiVuL1o2MuafsTqYnlXKk7xiv1L9xRnPv6TrA\nvu5DtLjbSbWkMDt15hntdyWjl/VCEBIIBAKBQCCY4lwRopA/GOFXr1Tx7MYTeAJhgqEoB2p7xo17\na08LL26pR1EmLw2qbnESDEXZcriDl7Y0UNXkZG5JCvesLR03Ni9dE4VOLiHr7vddlaLClc7wNXr4\njnL0Opm/bG0gHInSOeSIKC9M5u61pfiDUd7Y1XI5l3pRaO3x4HQHmV2cgk4e+/bWnEKxRaHhdvRm\nveYkMmbngCRhnV1Bygc/fHEXfQ2zck4mt68oZP3ivDGP2yyjopDLG6Klx0NZbuKYjmIZSVb6B4OE\nI7FLtryBMHFmA6mJ2jU9NbgatE6LCXFG7GchzMRi2CnU0DmIyxuacNywUwhAVTU339kwsyCJ20pu\nxL//RrpbT19yNi2pBFmSmZ9ewZLMMxc1LXoLn5r9AJIk8esjT9Ht62F1zjIK4kevkU7W8XezP0q6\nNZV3W7aws2PvpPMe6DmslaDNup/Pz/1/Z+QuEggEAoFAIBAILjZT/ltpR5+X7/5+H/uqe5iWm8D3\nP70Ms1E30qFnmPZeD8+/V8frO5t57t3aSeftdGi/sB9rcrJxfxsp8WYe/mB5zBDi4RuX9l4P/YMB\nnnjtGN/69S7++Yk9NJ+yDsHl4Vd/Ocp/v3CYlh43NouB4ux41i3KxTEY5L0D7XQ4NFEoO9XK6oos\nkuwmth/tJBCK7WzodHj55q93Uds6cClP47w5eEIrHZtfdkqeUDiMGokgT+AU8kU0J1WcXnOuGNPT\nKfjO98j5wqNI8pT/mLhiiTMb+Mh1xVjNhjGP2ywGPP4IqqpyvHm0Ff3JZCRbUBnNBToVrz+MzWIY\n6Sx2rMnJb/56jMZOzU3nD0ZwDAbO2yUEo06hJ9+o5qu/2M5vXjkSc9ypHcpyh1yYZ8PqiiyIGjja\n2D/hmMbOQZ5/s52vzf0yD826/6zdKoXx+Xx69sdYmD6XldlL+WDJLePGWA1WPlvxEFa9hWdrXqLD\n0zXhfANBF/WuJkoTi1iUOZ90a+qEYwUCgUAgEAgEgkvJlMsUUlWVwe3bMKSmclRN5nd/qyYYjrJ+\ncR4bri9Br5PJz7BzonWAQCiC2aidwqvbm1ABi0nHM29VM6cwkYwkK9XNThq7BrllSf6YG4OuIZEg\nElUAWLcod8IsjPQkC3qdzMETfeyp7iEcUchIttLT7+NHfzrIDx9ZPu6m7nLwXutWjvYd55GKT2LU\nXTvZR81dbvZWjzrHZhYkIUkSty4r4P1DHby2o4nibC1PJDslDp0ss2ZeNn/Z2sjOqm7Wzh8fvLz7\nWDfd/T7+tquZaUMlO1cCB2t70esk5hSPzRNSA5pLZKJMIW9YE0mthlHRyJR9dWYuXQnEWQxEogqh\nsEJ3vyb65Gfax4wZzhX6xctHMRt1/OODCzAZNCeRoqr4AhFyUuNIS9Cu6V+2NhCKKOw53s2dq4qY\nnq+9rnNSz16YOZXhz2GARLuRV99vYG1FFgk205hx/e4AEoyEa5+LIJUcbyYrxUp1i5NwRMGgHyta\nevxhfvHSEZzuIDMLkihYcG4BxxVp5ZNm/qRb03hwxgZ+c/QPvNuyhY/OvJtDvUep6qtmMOTGHXIz\nGHLjG3LiLUife05rEQgEAoFAIBAILhZTygKgRiJ0/u+v6H7yCZr+99c8/koVSPDZD83mvhvL0Ou0\n5RZm2lGBlm6tVKitx8O+6h4KMu18cGURwIiD5+l3anlhUz3vHWgfc6zOfh+6IVeQyaDTfn2eAJ0s\nk5um5XLYLAY+detMvvd3S7ltRQHeQITKBseFfirOiR0de6hx1vFOy5bLvZSLyjPv1vLzFyt5bUcT\nx5r6eWvP2DKw/AztJtdmMXDrcu0aHWlwYDTIJA+5Jq6bm41OlniNYMohAAAgAElEQVRjV3PM8pvq\nFs0hdKTBMabkZSrTN+CnpcfDzILkcQJnNKAJC5M5haxDTiHB5eXktvSOoTLVlFM6dKWfFDbd3OVm\n00mfcf5gBBVNXEoZes2HIgpGg0x8nJGX3m/g5y+OBrKfL7OLkplTnMK3PraQGxfkAnC8xTlmjKKo\nDLhDI+W42rHPTZCaXZRCKKxQ1zbWyaeoKv/32rGR92xdu+uc5j8bKtLKybSms7f7IP+x92c8cfRp\ndnXt41h/Dd2+XgyygXRrKjOSyliUIUQhgUAgEAgEAsHUYko5hRyv/xXPvj0A6N1OSFf59scXkX1K\nZ5zCoV/Mm7rcTMtL5JXtjajAh1YVjZR/dTp8dDt9dAxlyTy/qY7p+YnkptkIhqL0D2q/IqfEmynI\ntE/q9PnYzdNp7nazojwT49Cv8Yump/PajmYq6xwsm5V5IZ+KsyYUDdM91NXmnebN5NtzmJ0y86oL\n+ewfDPDuvjZgtFQKtGBpfzCKxx8mP33UUbFuYS4b97fhdAfJTLYiDz0fiTYTNy7M5e29rbyzr41b\nlxWM7BMKR2nocGmOBhV2HO3ktuWFl+T8zod3hp6X+dPGl6Yo/iFRaIJMIV94SBQyiJbzUwGbeVQU\nGs4uS7aPdd1kJI9eKwn4265mrp+fjdmoH+lcFmceLR8DmJ6XxKfvmMUf3qoZcddlXwBRKC3Rwpfv\n0QQPvU57j1U3O8d8Lrq8IRRVJTPFissXYtATGvfZfqbMLk7mnX2tHGnoZ2bhaFndW3taqKx3MKsw\nieYuN/Xtg+dxVmeGLMmsy1/D09Uv0ObpZFHGPG7Kv55USwpmvWnyCQQCgUAgEAgEgsvIOTuFvv/9\n73Pvvfdy3333UVlZOWbbjh072LBhA/feey//8z//c0bzBdvb6P/ba+iTkrHM0LqyZFjUmDcNBUOi\nUHPXIC3dbvbX9FKUFU9FSQpZKdqv550OLwdrNdFg4fQ0whGF/321inAkStdQG+fMFCufum0mNy7M\nnXR9RVnxXD8vZ0QQAi2AOslu4kiDg6iinNF5Xiw6vJ0oqkKGNY2wEubxyif5VeXv8Edi541cqdQM\nOXhuW17A5z88h1uXFVBRksL9N05j5ZxMZEka03XLaNBx5yrNPXZq2+07VhZisxh4bUcTLs+oG6iu\n3UUkqrJ6bhYmo4639rSO6QQ1FTna4OCdfa1kJFtZHkOgVIbLx8xmnIHxOUm+yFD5mF6IQlOBEadQ\nIIxjMIjdahjz2QOQnjh6rW5fUYjHH+bFLQ0AeANaVlacRY/NYhgJqJ5ZkITNYuCRO8v5zAfLuW15\nAUWZY7vUnS/56XbiLAaONY11Cg2LW0l2E+sX53HzkvyRcrezZUZ+Imajjn01PaiqVoxW1+7ixc0N\nJNiMPHxHOSU5CfQM+PnvFw7z+7dqzu+kJmFp1kIenLGBby75Eg+VP0CuPVsIQgKBQCAQCASCK4Jz\ncgrt2bOH5uZmnnvuOerr6/nWt77Fc889N7L9u9/9Lk888QQZGRl89KMf5eabb6a0dHxXr5M58scX\niY9GSX/wY3iPV+GvPk6eMXbZTkaylXirgUN1jpFON3euKkKSJKzOLm50HqS2dyUDnhAS8NH107Fb\njWw+2M4Lm+opHhINspLPr1RGkiTmlqSw+VAH9e2DmAw6nO4g88oufojokb5j9Pn7WZW9FIPOQKu7\nA4B1+ddTEJ/LSydeo8pRzY/2/Q+fqfjkVRNsWj1UkrJwehqFmfEsnJ42sm12cTJrF+SOuVkGWDUn\ni0AoypzisUG9cWYDH15dxB/eruWl9xt46NaZ9A34eWO3Vo42vyyNrJQ4nnuvjhc21fHQrVOzhXQ4\nEuWpN2vQyRKPfLB8TIeqYYadQjv7K3lzx26+tvALFCXkj2wfdQqJ8rGpQNyQKOQdcgrFEseNBh1f\nv28eCTYTqQlm9tX0sHF/G/NKU0eEkjizAUmSSE0w097rZWZBEqB9di2dlcFSMi742mVZYk5JCruO\ndvH1X25HUbXSseH8tmS7mZtO6bZ2thj0OuaVpbKrqpumLjdpiRYef+UoKiqfuaOc+DgjJdnxVNY7\nOFzvQCdL3H9jKQb9uYlQkyFLMiuyl1yUuQUCgUAgEAgEgovJOTmFdu7cybp16wAoKSnB5XLh8Wj5\nPq2trSQkJJCVlYUsy6xZs4adO3dOOqe+7hhyfAJxFXMJ27Qblyw5tigkSxK3LC3AHwhTX9/NjDQj\nszJMBDs6aPvud1jsOILc1siJtgFKchJIiDNy7w2lZKVYeXd/G+/s1cpsslLOv2xiwZAo8cLmOv7j\n2YP87MVKXt3WOHJTdjFQVZU/HHueP594le/v+QnV/SdodWt5Inn2HHJsWXxu7qdYm7eKLl8PP9r3\nC2qddRdtPZeSmtYBLCbdmBKxYfQ6eZwgBNpN6vrFeTGv93XzsslJjWNbZSd1bS5+8McDVDX2U5hp\nZ0ZBEusW5ZKbZmNrZScn2qZmJ7K397biGAywblHuiIvuVJShTKG2iNaxqT8w1sXhFU6hKcWwU6ir\n30c4oowrHRtmZmEy2alxGA06Hr6jHJ0s8cTrx+ge6kg2LC4tKEtjWm7CmDyfi8nNywpHytZ0soTJ\nqMNs1JFgM44EXJ8vS2ZogtbOqi5++/px+geD3LmqiBlDwldpTsLI2Kii0jyUQScQCAQCgUAgEAhG\nkdRzUC++/e1vs2bNmhFh6IEHHuB73/seRUVFHDhwgCeeeGKkbOyFF16gtbWVr3zlK6ed8z8efRbF\nYMQUbyccCILXQ9RoxmyPLdyogN/pwqDEbinu1xkJyQZsFgPmoRKFiKIyMFQmJEsSiTYTMTrQxySq\nKkTVKAbZwMm7qIDbFyY0FFYsIaGiYjHqsZr1XIxEn6iqMBB0gSqBpJ50XEg2J445ZiAaGuksFWew\nYNZdwSUNkkSfy49RryPeeuG6vYUiCoO+0Mi1Mxv1xJ107cJRFZc3iE6WSbQZL8o1HT6OTpbO+DUJ\n2uuvfzAIEiSd5vWsBAJEvV78JomwXsJmiMN0Uoc6T9hHMBok0ZSATrq0+fOyTkaJXt7yy6nG8GvS\nqNcRikQxG/XYzJMbO33BCL5gZOS1bLcYMRkufT+BS3FNh1/76lAvM4NOC9GWTtruC2iB24FQhDiz\nAUsMF53gzBDv06sTcV2vTsR1vfoQ1/TqQVzLq5Mr4bp++V/WT7jtggRNXyhXTFjWY9HJqLKMBMio\nyLrYNzRqJIJBiaDKEjrDSQKBJKEEgsiqioSExWRgOGvZqNMChhVFxWTUczYZzIMBD+FoGL2sI85o\nHdPy3R5nxDkYwKCXsVuNON1B/KEIkgQ26/m1hldV8Pi1EjmTQYfRoCMY1v6tRg2gykj6EKqkoJf1\n6E55vqw6MwadjsGgRxOHpCs3TDgQ0oQ3o0E34eviXDDrZILhKMFwFAkJm9UwEkgNYNKBJazHH4oQ\nDCtYz+Dm/GyJKioubwCjXkfiBK6QWCiKqglZej16/cTPiSqN/X9kxj6HEe09rNPpxpz7peJCXs+r\nAf3QR+pwVpleJ5/Rc2SzGglHFMJDf5R0OumyPbeX4riJdhO+QBhVhfg440ijgWHscUaiikogFCES\nVZB1F05MvhYR79OrE3Fdr07Edb36ENf06kFcy6uTK/m6ntPdbXp6On19o52fenp6SEtLi7mtu7ub\n9PT0Seec2/MSj+ffw399bBWbd9Yz/dnfopTOYsZnvhFzfPsv/htv80FyvvgV4uZUjDyuBPzUfeGz\nNFqy6Lz1Ezz4gRmj24JBWn7wXSKOPhLWrCX1IxuQ5MkvnqIqfHXLtwEIKxFUVMoSi9lQ9kFy7dmA\n1rHKoJeRJIkBT5Af/ekQHX1ebpmXxT1rT5+ndCqBSJCoGiUc1PHzFytpHBgKOQ6GSYgzklZxgvZo\nLQvUu7AoSWw80IqU2M2Ns6dz/8qFMed0+Pv58YFf4Q17+bcV38JujF1G4gv7MegMGOQp1ZgOgOc2\n17NtVzPfvmcORVkXNhy3q9/H936/j1uW5sfsNObxh/nWr3cRjkT43v0LST6lPfj5sq2yk01/Ow6R\nCF+7eTazCpMn3wmoaXHyw2cOctuibO5aUzLhuL6XX6T/9b/ywrpEOtKNfKT0dm7MXzqy/acHHufE\nQAM/u/4H6ORL66ZIS7PT2+u+pMec6vQN+PnG4ztB0dShz95cxuIZk3+OAnQ7fTz2270Ew1H+ZcOc\nCUsKLyZT6ZqqqsqXfr4No17mO59YQCiiaMJZREGvj11yKhjPVLqmggvDoDdEfm4SA07v5V6K4AIj\n3q9XH+KaXj2Ia3l1cqVf13OSs1auXMlbb70FQFVVFenp6dhsmsiQm5uLx+Ohra2NSCTCpk2bWLly\n5aRzPntLMmpiHy+/30BPQCUgG7W29DGI+v14Dx/CVFiEdfacsSdktmBMSSFf9nBD3ds4Xv/ryLbe\n558l1NaKGg7jfPNvBJubzuh8+/z9hJQwc9Nm860lX6Y8ZQYnBhr4rwO/pH5Am8No0I20f0+0mfjG\nA/NJTTDz7r42fIHYJW6xcIc8/GDPT/in7d/nsT+/TmOnm5WzM/n6ffNYVZGFyxuizdeKGjGwrLiU\nB26axrc/sRiTL5eDlbEzmABSLMncVHA9YSXClrbt47aHo2Feb3ibb277V56sembMNlVVCUfDo/+b\noGTvYnO0vg+zUUd+xoXPRclMtvKzL66esPW8zWLg7rUlBMNRXt3edMGOGx4qE6ppHX2tv/R+wxm7\n73oHtI5OaZPc2A4HTYcM2ms0FB3bTc0X8WPWmS+5ICSIzXAW0DDJ8WfuHstIsvJ3t89kXmnqObd8\nv5qQJInirHgcg0G+8NOtfOUX2/mHx3fyT/+3m398fCcHa3sv9xIFgktOZb2Dr/1yB1//+fv4g5fn\nb7pAIBAIBIKpwTnZQRYsWEB5eTn33XcfkiTxL//yL7z00kvY7XZuuukmHnvsMb761a8CcOutt1JU\nVDTpnD6LTFymg82H2pGTuplmMmNxOlBVdURsGSbY0gyqinX69HHbAKx5uYQOHSZw+ACh2uMk3/wB\nvEcO49qyGWNuHgkrVmkCUVcn5qLiSdfW4ekEIMeWRbYtk8/N/RQHeir5XdUz/PLwb/nOin/AZhh7\n8xVvNTJ9doidhx3sr+1hdUX2hPNX9laxu2s/qqrS5euhL9CPqgJ5u1hYsoqHVs5AlmVmFCRR292O\n2+Qn6kynNEcLbC3MjKckO56jjf14A2HizLFLJJZnLeZvje+wuW0HK7OXkmTW9q911vFszUv0+DSH\n16Heo3R5e8iMS0dRFX68/5c0DraMmevGvOv4SNntkz53F4oBT5D2Xi8VJSnozsDddS7Eei2dzMrZ\nWby6rYndx7q594ZSLKbzd1P94e0a9hzvxqjXEWfWMy0vkYMn+jhc5zijLna9A5rYM5koFB0RhbTn\nLqSExmz3hf1XbFnh1YjZqMNmMeDxa+Jdsv3snGkLp6ezcPqZOYuuBW5Zmo+KFnpt0MsY9DJ6ncy2\nyk6eefcEO6u6SIgz8cBNZZN+DggEVzodfV5+8dIRIlGF+jYX//PyEb5091z0V7DtXSAQCAQCwblz\nzne1X/va18b8e8aM0TKtxYsXj2lRfyakW1Lpl3owlVYiJ3cy0BEloz1EzzNPk37v/Uj60aUOO3xM\nBYUx57Lk5jBw6DCgOSTce3bT8/yzSAYDWZ9+hKh7EIBQZ+cZra3d2wVAti1z5LEF6RV0e3t4rfFt\nqvqqWZo1tmzrjcaNHAi/hbHMwJbDeditRnQJvfQG+lDU0RAqd8jDxpb3R8JSAZS+HFRHAbaZhzgW\n3sZzJyLcU3YnOllHRlkv7hAkK4VjWo/nZ9g52thPS5ebmROUHpl0Rm4vupnnal/mN0f/wAPT7+K9\n1q3s7tqPhMTa3FXk2LN5+vjzbG7bzn3TP8yuzn00DraQYU0neUhEanQ1s7trPx8qvRX5NKHEYSVC\nRIlg0Z9/qdVwK/rpeRemc9G5IMsS183L5uX3G9hV1cXaBbnnNV8kqrC/ppdQWCEUVphflspHrivm\n0Ik+Xnq/gYrSlEnzfXpdQ6JQwumf46hfCxs3WmyAn/A4p5CPdMvkIpTg0iBJEv/08YVsPtQxFIp/\nftlk1zrT85OYnp807nGLUc+be1pwDGqOu6xUKzec5/taIJjq1LYNEIkq3H9jGQ1dbnZXdfHE68f5\n9B2zLkumnEAgEAgEgsvLlAmOWZAxlzebNiIna0LNjoVm5pCIa9NGgs1NZD3yeQzJmtgRGBKFzAWx\nHUjWvDwAJJMZNRig66nfQjRK6gMP8kfnZjyObm4Bgl2d+Gpr0CckYszIGDNH2Okk0NiAfcFCOjya\nKJRjyxozZm7abF5rfJvKvmN0+3opTiigPGUGf214i7ea39PWYAjT5Gng52+0YS7fFXO9cQYrj1Q8\nRLwumX97ah8ej8qX755LdvYaHq98km3tuzjhrGdB+lw6lGrkqImbypaMmWO4pKq52zOhKASwOmcZ\nTYMt7O7azw/2/hSAPFs298+4i4L4PKJKlNcb3mZ35z5uyFvNaw1vYZANPDr/0ySatBbPTx9/gZ2d\ne2lxt1EYnz/hsX595CmaXC08UvEQJYmFY7apqorLG8JmMUz662Rdm4tn3z0BcMZZOxeL1RVZvLqt\nka2VnectCtW1ufAHI+hkiaiiMj0vkZw0G8vKM9hZ1c2+6h6WzMw47Ry9A350skTSJOVFEb+WGWG2\nxYPfP8YpFFWiBKMhLAbreZ2P4MKSnmQ96zwywdlxx8pC/KEIOalxvLq9iT9trKMsN5G89AtfoioQ\nTBVcHu3zPzstjrtums4//nwru491E281ct+NpcItJxAIBALBNcaUEYVuKbyRaYlaUO7bzZupppbk\nr38V77PP4969i5Z/+xeyHv4s1pmzCDQ1IVssGCYIsI6fXY5kMNC5fj4Z7xxACgSJq5jLvmId+xsO\ng6qy1iAh1dXiPXxIm+ernxkTOq1/+U18O3Zg+Ofv0OHpxKq3kGAcG26cFZdBijmJQ71HADDIBham\nz2VX1z7SLCl8sOQDPHH0afKmD9LjdgFwZ9HtZNnHOjIK4vOIN9p5Y3czbrfKnauKmF2cAsBXFnyO\n52v/wv6ew7zR9C4A64vXsrZkrBhTkKGFybb0nD7gSpIkHpyxgdLEIqocNZQkFrImZ8VIloxO1nFz\n4Vr+VPMyP9r3C7wRHx8oXDciCAGUp8xgZ+deqhw15Ntz2dy6jYq02aRaRgWbYDREdf8JFFXh54d+\nw7eWfAk1EMehuj7q2l3UtbtweULMKU7hixsqxnUNGiYUjvKLl4/g9Ud45CMVlyU092QSbSam5ydy\nrMmJyxMkwXbmWS8ALk+QbUc6OVDbB0PusE/dOpNel5/Vc7USwztXFbHneA8vb21k4fS005bL9Q4E\nSIk3T1pSF/X7iejAbkkAfzeh6GiGhC+iuY3i9KJ8THBtYTHp+cQtmss1NcHCz16s5PFXjvLPn1g8\nxokpEFxNDHi0/MFEmwmTQcejGyr49z8e4J19rRRm2lk+O3OSGQQCgUAgEFxNTBlRyCDrmZ6s/Spe\n2VdFtbOWAcVH3t99BktJKT3PPUvbj/+T5A/cRri7C8uMmRP+mmXNzcH4g2/z3IFfcGOJmfl9iYTv\nuoW/1j5FgtHO0qxFOONfJNOhCSjhzk5efu0ntGSP3uB/vNZNEtB35AC9FgeliUXjjidJEuUpM3m/\nfQcG2UBYCbOrax+Z1nQenf8w8UY7qZYUev31YFeJulJ4+w0daQkB8jPsFGTaKMiwYzPEEYkqvLuv\nDZNRx02LRh0oZr2Jj8+6l3um3UmVo5pWdwc35a8Zd85pSRbMRh3NXZOnnutkHSuyl7Aie0nM7Suy\nlvBey1Z6/H3k2LK4ufCGMdtnJJciSzLHHDXkxGXyYt1r7Os+zNcWfX6knKx+oBFFVciwptHt62VH\nxz42/c2G26eVLSXEGclIsnCkwcGr2xv50OrY2U5bKzsZ9Ia4dVkBt60smhKp7nOKUzjW5ORoYz8r\n52RNvgNal7B397VxqK6PqDJaKmg0yCyakYZBP3oDmp5kZVVFFlsOdfCvT2od0ZaXj/+SHgxFGfSG\nyCscXxZzKkrAT8ggjXSdC5/kFPKGtdIykSkkuJaZV5bKuoW5vLu/jWc31vLJD8y83EsSCMbgD0Z4\neWsDtyzJP68OmMNOoeGyVJvFwMN3zOKx3+3leLNTiEICgUAgEFxjTBlR6GRSzNpNriPgJD8+l8Qb\n1mEqKKTz8V/S/7fXADAXnj68envXXgA2zjUxe97f8WzNyyiqwsdm3YuiqtTZ9WQ6Rt0Sy+oVpi3T\nxJZDXZXYnD0AtFXuRF2qhTTHYknmAnZ27uWjM++mebCVDk8Xnyy/f+Tm+/ai9bzdvAlZ0hHqKccR\nVqhtG6CmdWBkjqKseEpy4nG6g6xfnIc1RlC0WW9mYcY8FmbMi7kOWZLIz7Bzom0AhytAyiQZM6dD\nJ+u4e9qdvFz3Op+Ydd+49vQWvYXShCJqB+p5c6hMrtndypa2HazJXYEsydQ66wG4s+RWfn/sT+zp\nPIDbt4w5xal8bP00UhLMeAMR/vXJvby2o5klMzPGdUqKRBXe2N2MUS+zfkkenpCXJ44+zfqCG8iz\nTxzcfT4owSCuLZsJ9ztI23DPmCyrYWYXp/Dce3UcaXCckSjk8Yf5j2cPoqqQl25jzbxspucl8tL7\nDRRk2scIQsN8aFURXQ4ftW0D/GVrQ0xR6ExDpgHUQICQXibeqDmtTu4+NuwUsupF+Zjg2ubutaXU\ntg7w/uFOZhUmT1q+KRBcSnZWdfHuvjYsRj0fvm7yJhkTMeAJYtDLWE9qlpCVon3+D+drCQQCgUAg\nuHaYkqJQ8lAZkiPQP/KYpaSUgn/+Du4D+4i63SSsvm7C/f3hAPu6DyJLMoqq8JujT+OL+Lkx7zpm\nJk/D4e9nb4J2I+5LseGUg+S0upnXFU/CqtXQ148hWgOAtbWXnBvmsDhzfsxjFSXk8+M1/4YsySyK\nIdgszpw/uu+QMScYitLa46G5282RBgeV9Q4aOwdJT7Rw67KCs36+hlldkUVt6wB/fKeW/Awbs4tS\nKM1NmHzHGMxKmc6slOkTbl+bt4ragXpa3e0km5PwhDz8+cSrvNm0kTmps6h3NaKTdMxILmNBwkx2\n9B9Ctg0wt3Q6qUMihs1i4P51Zfz8xSO8sKmOL949l6jPh6TTIZtM7DjaRf9gkJsW5RFvNbKr9SAH\neiox60w8OPPuczqv06EqCq0//L7W3Q6wzZ2HdeasceOyU6ykxJuoauwnqiiTlm51OryoKty4IHdM\nd6O/v6tiwn0SbCb+4cEF/Pj5Qxxt6MftC2G3joYN17W7+M1fqwDIy5i8pE4NBAnGjTqFTs4U8gmn\nkEAAgEEv85k7y/nXJ/fx1JvVFGXFj4iuLk8QnU7GZond3VEguNhUNWrfibqdvvOaZ8ATJCHOOMb9\nbNDrSIgz4nAJUUggEAgEgmuNKdl/dNgp1B9wjnlcZ7eTuGYtKbd/EH3CxF2ojvbUEIyGuD53JQZZ\njy/iJ9eWzR0ltwCQZE7EmaSVih3Lgv1r8pGtcXT/4Ul8NdVkeTTBSJHAHFK52TT7tF22TrctFiaj\njtLcBG5cmMujGyq4dVkBJTnxfP3++cTHnXuXoeWzMynKsnOoro9Xtzfxw2cO8P7hjnOaa1dVF5//\nyfvUnuRoOpnZqTNHgreXZy3iyws+y/LMJUSjsLNzLz2+PpKUVBy//jWLHt9IWn8YOambkuyxItW8\n0lRm5CdyuN7B8YZemh/7Nm0/+RGRaJS/7WpGr5O4ZamWn1TVowl1dQON53ROk+Heu5tgSzOGodBx\nX83xmOMkSWJWYTLeQIT2Xu+k8/Y4NSdObnrcWQd4FmdpOVaNnVrZXFRReGVbI//+9AH6BgLcuqyA\n1RWndytF3INIoTAho4RFb8YgG8Y4hXr9DgCSTJevs5tAMFXISonjwZum4Q9G+cnzh9l8qJ1gKMpj\nT+7lu0/tIxiOXu4lCq5Boooy0oWzu99/zvMoisqgN0yifXweXkqCGcdgAEVVY+wpEAgEAoHgamVK\ni0IOv3PCMbs79/N87V/Y1LoN9ZQvMG0urYNZWWIxFanlmHRGHip/YKQMSpZk/GW5/G1lPLtnWZgx\nbSnZn/sCAB2//DnJ9d0ANORqX5oyG2MLIxcCWZLYcH0J/9/HFp1XydfwXA99YCbzSlO5a00xZqOO\nJ9+o5pl3a4kqyhnPEwxHeW5THf5ghP977Rj+YGTcGFmS2VB2B0XxBazIXsLuAwG2v5lG/65VBI8t\nJdxWypJ3B/Hs34cUiVJeH0Bvd5GTNrZETJIk7rlBy5La9Mo2Iv0OAnUnOPzmdnqcflbNySLJbkJV\nVap6agHo8ffhCg6exzM1HlVRcPz1FdDpyH7k8yBJ+KqrJxyfP+TOae3xTDgmrERwhzwjv+qmJ519\neVbRkCjU0OEiHInyn88c5JVtjSTajXzjgflsuL5k0u5tzrfeBKA+14RZZ8ao0/KvhunwaO+XU7vr\nCQTXKivnZHLjgly6+338/s0a/uv5Q7g8IXoG/Ly+s+lyL09wDdLY4cYf1ATJbqdv3PeeM8XtC6Go\nKokxfoBKiTcTVdSRzCGBQCAQCATXBlNSFLLoLZh15nFOoWECkSB/OP48W9p28OcTr1LjrBuzvX1Q\nayGfGZfOR2few3eW/yOZcWM7lWXEpXOiwExELzE3rRzrjJlkPPhxFK8X9h4GYPdsK0GTDv97m4i6\nL3/A8ZmQm27j0Q0V3La8kG9/YhHZqXG8u6+Nnz5/GG8gPPkEwLv7Wvn/2bvv8Cbvc//j70d7eMry\n3htjszcECCHMELJHszpy0qbtr2na0560p+05PSftadOdNm3Tpjt7kUESCIEQSNjb4IH3nrIsyba2\n9Pz+kG0wnhAII9/XdeW6iPTo0Ve2wY8+ur/3be/1Ehulw9maIp4AACAASURBVGJ3s2FHzeB9vu5u\nar/3HSwbXiEvOodvzf4qDU0+Nu1twKBVsXpuOo/cuJwi7WzSLc0ozLEowiPIrfeg1NtAGh5OZSRE\nsKAwgaimk4O3uba+gwJY07+drt3Zic3tGKzKqrLVDDvPx2H/aCe+tjYiFi5Cm5qGNj0Dd20NQY9n\nxOMHRlaPFQptqNzIf+35Kc22UCVOfPTZb8/KTDpVKfTR8TYqmuxMy47hf74wl/y08RtMB3p6sG3f\nhj9Mz4kcPboRKoWaeltRSUriDbFnvT5BuBJJksTdK/N47MEFGHUqqprsKBUSkWEaNu1toKrJfrGX\nKHzKlNaFto7pNErc3gAO58R+n5/J1h/4jDQ5c+CDqdO3kG3e1zBqxfBEybLMxt117CttP+cwSxAE\nQRCEC+eSDIUkSSJGH02X2zriBURTbwsy8uAI+7drtww5rtnRhkpSEqMzoVGqB/uonC7eEAqJorVR\npISFmhZHLllK1IpVAPhUEpYoFe0L8wm6XFg3v33eX+eFFhdt4Hv3zmJqdgw11a18+F8/peoH36f+\nf34w5L+2v/0Fvy100efzB9hyoBGDVsX37ptNXLSeD44202Z1EvR5afn9b/G1t2H/cAdyIICzsZHN\nb+5Ficw3bpvG7dfkkJsSxXSDE43spy85B2/+NAzeIGltbpr7q1IGBOUgvoCP9YvSye1rxKfS4s/I\nI66njVUJ3sF+HsWWUP+cuQkzATjQfpSW3jaC8sQroEYTcDrpeu1VJK0W8w03AWCYVACBAK6qyhEf\nM14oJMsyxy1leANemrxVqFWKEcv1xxNh0GCO1FHTYmfT3npUSgWfWzMJ4wjNyEfSc/AAssdD19w8\nAkoJvUqLRqke7CkUlIO09rWRYIxHqRAjuAXhdOYoPbdcHfo9M3tSHA+sm4wsw+82FGOxnfsWHkE4\nWw39v2tm5oXC+3brufUVOjWOfuRKIQCLI/Sz3WFz8dL2Kp7YcByH81T10OGKTk42jF7JfaZmSx+v\n7azhT2+W8MSG49h7R/6wRRAEQRCEi+OSbDQNEKc309zbyps1m1mXuXLIG9bGnmYAFiTNQafShUbY\nWyspiMlDlmWaetqIM8SO+SZ3oHJoamzhkD4vsbfdgezxcKivEiQfEUuvRtpVRV9JCbHnv7fxBafX\nqvjSfBNVH/0NbZ8dj0INp00ckYNBPI2N9B45hPmW2zgRnU+P08eaeWlEGDTcujSbP7x+ghe2VnBH\nzwE8dbVIGg2Bnh5KfvEbNJXHuRFoy5lNStzywfOm9LXiB2q08Xi0RmYCWc0eah0NpEekAuAJePnN\n4Sdx+918w7yeSH8f5bGxlKbCzXWQXPcuTx3vxB/0caKrHK1Ky5qMaznWWcJxSynHLaVolRrSwlMw\nqo3kRWezNGXhWX+NrG9vJNDTQ8yNN6OKClXfGAuL6N78Dr2HDmIsLBrx62qO1NHY0Yssy8N6BXW5\nu+n2hIK2Xm09cdHZKM6yn9CAmXmxbDnQSJ/bz7KZySN+wjuavuOhqrfOnDjobQ1tH1NocARClW+d\nTgu+oF9sHROEUSyZlkSYTk1+WhThBg13rcjlmS0VPP5qMf95zyz02kv216hwBWmzOtFrleSlRrH7\nRBvtVid5qWffB87eNzCOfvjvEfMZlUJN/UFUr8vH81sr+dL6QtxeP398/QSRYRp+/uWFE+qT19YV\nCrCMOhVHKi1UNNq4e2Ue8wriz7rPniAIgiAI598lezV7fdYqmnpb2FK/nTpHI18ovGuw4qfe0QRA\nengKicYEii0lvF27hUmmXLo9Njx+DwmmuLFOz1RzIavSrxkWIkgKBfH3fQ5V/Xb09R+QH1uAPTkF\nd0M9st8/4ojyS1nv0SO0PvUntB43tXkLeDGQwzfvmE5RVgwQCoXsOz7AsuFlOp75F4TFExOzkGtm\npgAwKz+WgvRo1Id30WM5gDYjk+jl19L216fQVB7Ho9LiQk1C9SE8TY1oU0KBj7qxGh+wtzeMXoeS\nqZKS+C4/u1oP0dc/7arWXk9DTxPIMjUv/B0DUDLVR1O8jYZ4NWlN3bSWHabdrCZGZ+KRJQ9i9Efx\nyOyHKO+uoM7eSF1PI1W2WmRkii0lzIybOmJl2Gi87e10b92CKiaG6JWrB2/X509CFW2iZ/9eYu/4\nDArt8AvotPhwDle28/DvPsLnD6JRKVCrlGjUCgKRjWAGCQnZ2E109LlXNN22LJvs5EjK6qzcsChz\nyH2Bnh4UYWEjXlgHvV6c5WVokpJxGJXQC7rBSqHQ1oMm0U9IEMakkCRmTzr1++SamSm0WpxsO9zE\nn94s4aFbpqJQiDe2woUTDMp0dDtJjQsb3Ibc3n1ulWqnKoXG3z42EAoZtCr2lbYzryAehQICQRmr\nw0OHzUX8BHrltfZXNd2/bjJddjcvf1DFn98sxecPsnhq0jm9DkEQBEEQzp9LNuGIN8bxH7Mf4umy\nlyi2lPDTA4/zb0X3kBmZTmNPEzqllliDGYWkYFpsEcc6T1BqrWDg0jzBMHYopFGqWZ+9etT7V6Rd\nzfLUJSgVSlzJKbhra/C2t6FNTjmPr/LC8NtsKCMj6d78DpYNryCp1SQ++BW0KQXwjwO8s7eewkwT\nkiQhKRREXr2MUm0y3a++QK69ls8od2GKWA+EtvJ9bpKSrm0H6VXqcK29G218FEEkFMjsj5iEJy6F\na8vfpv6HP4CBcEKW8ZgS6PQqABmPOQmzpZGW7sZQENQvIyINQ0Ujhvp2bBkJNMUH8TXmkbJmMvzj\nz3yuLQXTjQ+iV+lIiI6is7OHWEMMsYYFLE5eAIQqjnY07eKN6k0cbD/KstSrJvy1srzyEgQCxN56\nBwrNqXJ6SaEgYtEirG9tpPfwISIWDK9ACotxoIt4D1dzDvH+qQQCMl5/gB6nD09ECypgcsQ0ShxH\nUUZ1nf03sp9SoWDOpDjmnPbGNOjx0Pnyi9g/eJ/4z32ByKuWDHucs7wM2evFOHUa7kDoonygUigo\nBwkEA6LJtCCcgzuvzaG920lxdRcvba/izuW5F3tJwhXM4nDjD8jEmwwkmEIhzI6jzdS3OYiO0GGO\n0LF4WhLR42xRlmWZuv5JlpFjbB+ranZQUmulqTMUCj14QyG/fbWYf71bPviBEkBZffeEQqGBSqHE\nGAPTc8wUpEfzg7/sY1dxqwiFBEEQBOEScMmGQgAGtZ4HptzL1vodvFmzmV8ffpLrMlfQ7uwkJypz\nsOnwdZkrONZ5grdrtzA7fjrAsMbSZ0uSJJRSaPuZNiUUBHmami75UKh723t0Pv8sqpgY/F1dqKKj\nSfp/X0eXnkE4UJhpoqTWyvYjzaFPvLv6+Nfmk5xstKFNupovxhgIqynBVVmBIS+foNuF4+m/oFAq\n2Ji8jI4dzSwoDGA2JJLg7uJg5CTWLpiEKUvCVVlxaiGSRMI113KbP459pe3EJOfj3tHIV2NWQ1ao\nebSERIYxmcpXvkNQgjcLvYCKWXHTyF80m8YP38ddfBx1uxVlauqor1mr1LAgcQ4ba95lX9uhCYdC\nzvIyeo8cQp+bR9jsOcPuj1i4GOtbG+nesonwufOQlEO3Iyoiu5CsMurUSnTGHgJyAEXAhzLoI+B1\nIvtVBLuSQH0USTf+6PqJcjfU0/bnJ/G2hQIdd33diKHQwNYx49RpuHu3A6FKIbUy9NfeG/TS3CdC\nIUE4W0qFggdvKOLHTx9ky4FGEmMMLJ2efLGXJVyhBkKVBJOBCKOG2ZPiKKuzUlJ3qq/PgZMd/OC+\n2WjUo2+b37CzhqNVFtLiwkiMGR7m6LUqosO1NHX28ssXj6LXKtFrVRRmmli3MIPXP6zlo+JWJAlk\nGcrqurl6jJ/78vpuKppstFn7UCqkwe1pSWYjOSmRVDbZsfd6zmpLtCAIgiAI598lHQpBaPT5yoxl\npEWk8PeS53izJjReOy38VDiTHJbI9NgpHO08TqfTAkBq+Pm7QB/YEuVtbhrnyIvLb7fT9foGJJUK\nf1cXuqwskr7yEKqoU30HPr9mEv/zjwM8v7WSklorx2u68AdkpueYuXtFHoaOJBofK8H23hYMeflY\nN28i4HBguv4GVuYs4KmNpbx/uBlNwlI+tyKbjAoHi6clERWWOeKa1hCaINZzEFp3bMXc6cE0K2fw\nftv7W1FZbFQVmOiOVJESlsSXrgkFNDHr1tP8+K+wvrORxC99ZczXHq4JY7IpnxNdZXQ4O4kbZ5KW\nHAzS+eJzIEnE3nHXiNuvNHFxRCy8Csfuj7Bt30b0tSuH3O9WhnoGGVUGutxW1Ao1aoUao9qIOmik\nozGaI10udDNAoTu3pqBn6t7yLp2vhqqbIpcuw75jO77OzuGvT5bpKz6GwmBAn52D+/BmNEoNCkmB\nRhH6hNgb8NHU00qEJvysttwJggAGnYqv3zqVH/3rEM9sqSA7KZKUOPH3SDj/2qynQiFJkvjKjaE+\ndx5fgO4eD+/sreej4lb+vqmcu1fkEaYfPoigtM7K23vqiY/W8807pqNUjDxn5NufmcGBsnZe+7AW\nlydAbkokkiSxdn46B8s7aOrsoyA9mtYuJ+UN3QRledR+eT97/sjgnxNjDEOec1Z+HJVNdg5XWlg2\nQwSqgiAIgnAxXfKh0IBJply+O/dhDrYfpdfbN6wX0NrMaznaeRyn38XNk9eMGwqcjYHqIE9T48c6\nj89qxW/twmftwlNfj6ehHnd9PUFnqIpE0mhI/vo3MeRPOqfzW9/eSNDlIu6uewibOQtleMSw6hZT\nhI6v3jSFpzaWcKTSQnS4lruuzWNmnhlJkpAjctBmZNJ79DC9xcfo3rIZZWQUptVrWaDV0tDew7v7\nG4mLj2L+nBzmDy+wGZEuMzTBp2f/PtRmM7qsHGSvB8uGV1DodGTceg9S9YuDlV4AhqIpaNPS6Tl4\ngJgbWiE2fMznyDflcKKrjAZH07jff2dZKZ7GRsLnzUeXkTHqcebbbqf36BG6Xt9A+Ow5g42oAZp7\nWjCqDTx21X8PC5U6bS4e2bMHkJGCKuy+iU9qGXXN5WV0vvQ8yshIEr7wAMbCInoPHRwxFPK2tIQq\nxWZM4/t7f4rNYydSE/r6qZWhNwx2j4Nuj40CU97HXpsgfBrFRRv4/JpJ/G7DcTbtq+eB6wsn9LjK\nJhsHyjuIDtNiitBhitCSEhsmmlYLIzo9FDqdVq0kwWTgnhV51Lf1sK+0ncMVnWQmRvDZ1fkkxhgB\ncHn8/P2dchSSxBfXFxJhHL51bECCycB1CzPYdbyNDpuLlNhQ0KlSKvj82gJ++cJR5k2Op7y+mz0l\n7bRa+kiOHT8MPXPts/JieWFbJa98UI2918ONi7PO6msiCIIgCML5c1ldgUZpI7k2bemI9yWHJXJD\n9hr6fE5uL1pHl+X8bddRhoejjIzCXV9PoLcXZdjZfxrss3RS91/fQ/Z6h9yujo8PbU+TZVyVFXRt\nfOOcQyFnWSkKnY7IpcuGhUGny0uN4rEvL6Slsw9zlA6d5tSPgSRJxKy/gZbf/oaW3/0GZJnYez47\n2Gj51quzUauU5KVEntXaVCYTmuQUPA31tD75h9BzqVTIfj8JX3yQiPSZ/G98JlHaU+eVJAnTddfT\n+scn6Nr4BklF3xrzOVLCQr0Jmnpbmc2MMY+1bd8GQNTylWMepwqPwHzrbXT86x90vPA8SQ+GKpZc\nfjcWt5X86JwRq4zMkaE3elaHhwhVNJ0uC0E5OLjl8VxY33kLgKSvPoQ+KxSyqWNj8TQ2IAeDSKd9\nCttXHNo61pisw+YJbRFTK0NvBAYqheocoZBTbB0ThHM3LddMstnIvtIOuns8GHRqVs9NIy5aT5hB\nPayKIijL/PWtMjrOGGmfYDLw4wfmiWlMwjAD4+dH69+jUSv53r2zeP9wM7uOt1LRaOOj463cdnWo\nKveVHdV0OdxctyCdzMSIcZ9PIUksn5XC89sqyUg49WFMZmIEv3t4MZIkEQjK7Clpp7LZPmoopNUo\n8XgDACScsV0tJlLH3SvyeP3DGt7cVce1s1NHrHASBEEQBOHCu6xCofGsTF8G8LHeeI/GWFiEY/dH\n1H7325iuu56o5dfi2PURfcXHSPzyV1GoR//kDaDrrTeRvV4iFixCm5qGNj0dbWoaSsOpC6WmX/4c\nZ1kJ7vo6dOkZZ7W+oMeDt60VfU7umIHQAIUkjbrVwThlGvrcPFyVFRinzyD8tCbLSoWCm5ec/Sd6\nkiSR9v3/wl1bi7u6Cld1FZ76OiIWLCJi7nwATLroYY8LmzETTUoqPfv2UvW7PxB5xz1Dwo/TpfSH\nG029LWOuxWfppO/YUbQZmeizxn8tkVctwbHrI3oP7qevZAnGwiJaetuA0QMVSZKYlm1m94k2UiPj\nOdHdid3jIFp39iOEgz4f9p0f4CwtwVAweTAQglAo5K6twW+zoTaZBm/vO34MJImD0X0QuibH4go1\nu9b0VwrVOurHfA2CIIxPIUmsnpfGX98uo7whtKX0cEXn4H3hBjXTcsx8bk0o7D9e3UWHzcXs/FgW\nFiXS5XDzUXEr9e09E57mJHx6uL1+6tocmCN1aDWj/27XqJWsnpfG0ulJ/L9f76Sm2QGE+vpsP9xM\nktnI+kUjb/MeyfJZKcSb9EzOMA25fSC0zE0OfYBT2Wgfsa+Qy+MfDIQAks3GEZ+jpauP7Yeb6e7x\nfKxQ6MPiFlotTm6/Jmf8g8/g8QbYWdyCo8/L3IJ4UsU2UEEQBOFT5ooKhS6kuHs/izYlla633sTy\nykt0v7uZQE/oostZUkLY9JErU+w7d2DdsglfRweapCTiP3//qKFG9Oo1OMtK6HzxeVK++W0k1cjf\nHk9jA6qYGJSGUxdZnqZGkGW06ekf85WGLvri7v0ctq3vEnPTLeftk2uFWoMhLx9DXv7E16JQkPz1\nb9Lyu9/Qse19NNNnYyiYPOKxBrWBaG3UmKGQLMt0PPcMyDLR166Y8Bri77mP+kd/SNsz/yDiO9+i\n0lYNnKpOGsmdy3NZf1UmO9tcnOiGTpflnEKhzheew75jO5JKRcwNNw25T20ObZPzWToHQ6FAXx+u\nqkpUGenUBDpIMibQ0tdGdmRG6DGK0IV3nb0BEKGQIHxcC4oSAEiPD6e718PRKguOXi92p5eWzj52\nHmthQWE8v3/tBL5AEIB1CzNIiw9VYUgS1G/poXyC05yET489Je24PAFWzkmb0PF6rYqkWCO1bQ6c\nbj9/e6cMSYIvrC1ArZr4B2YKhcTUbPOo9yeajRh1KqqabSPeb+3xALCgMIHc1EjmFsSPeFxUf5Pp\n7h7POYcxDqeXZ9+rwOsLsmpeGpFjbI87UyAY5I9vnKC4OvShSWNHLw/fNu2c1iEIgiAIl6vzX1Jz\nhVKo1USvXEXmT35G9Ko1BF1OJG1oksbAlKczycEgXW+/ia+tDSQJ8623jxoIARgmFxI2cxauipO0\nP/1P5GBw2DGepkbq//e/6XzxhaG3N4SqPrSpHz8UAtAmJRF/3+dRhY9fan6hqaOjibkxFIY4T5aN\neWxKeBI93l7snp5h9zl6utj9t5/RV3wMbX4+4fMWTHgN2tQ0lIvnE+i08Na/HmVjzbvA2IGKWqUg\n0qjBbAhdWHc6z20sfd/xYyjDwsn86c/R5wwdfT0YCnV2DN7mLDkBwSD2zNAEvoVJc/nBvG/xwJT7\ngFOVQh0uC0pJSfx57L8lCJ9GCkli0ZREUuLCmJIVw70r8/nqzVP4z3tmcf2iDAD+9k4ZvS4fHm+A\nKVkxg4EQQH5aqEryZMPIb7CFT6egLPP+oSaUComl0yc+uj07KQKvL8jvXzuOxe5m9dw0spLO7+9y\nhSSRkxxJp81Nd38AdLruHjcACSY9V09PRqUc5cOw/lDI1jv8HBMhyzLv7m/A6wtdLzV19p7V41/Y\nWkVxdRdFmSbCDWpazmPrAUEQBEG4XIhQ6CwpjUZib7uDzJ/9iqyf/hyF0UhfcTGyLA871l1dhb+r\ni4iFi8j941OETZ0+whlPkSSJhPu/iDYtHceuD2l54nGC7qF9J2zbt4Es4ywrGfpc/aGQLu38hEKX\nGn1OHigUuE6eHPO4gcqdnc276XINbe58/Jf/Q+yeMpxaib1XJZxVBVRQDvJ6Vi9elcTsmgDz42ax\nNnPFhKps4vShUKi5r3XEn5Ox+G02/FYruqysIU2uB6jMoXMfKf9w8Lbe/pCyNC50kVwYM4kEY9zg\nhLGBnkIACcY4VApRMCgIF0pRZqiCr9PmRqNW8PhDV/G1W6YMOSYpJjRqvKyh+6z/jRCuTB5vgD+8\ndoJmSx9zCuIGK2omIjsptLWrrL6bxBgDNy6e+Laxs5HT31vw1R3VQ7aKAXQ7QiFPdLhuzHNEhYd+\nH9lGCJbGU9Vk59t/3M2mvQ2DtzV3TDwUeu9gI9sON5ESa+TLNxaRGGOky+7G5w+M/2BBEARBuIKI\nUOgcqSIiUIaHYyyagr/bineEyWSOvXsACJ+3YMwKodMptFpSvvUIhsIi+oqP0fCTH+PrsgAQ6O0d\nPKffah28HcDT0ICkUqFJvDK3AikNBoyZmbhrawie0az7dGnhod4Gm+u28cSxpwgEQxd3Xo+LmGYH\n9mgd2z4zlZ2uMv5r90/YXPf+hJ5/Z/MeajxtWCenoO1xc1Mgn+syV0woWIo3xCIhsaNpNz898Dg7\nm/bg8rvGfRyAu64WAN1pfYRO540OBT2O1noqu2uQg0Gcx4+jjIrioKqVOIOZOMPQLQADlUIAScYr\n8+dFEC4VSWYj0eGhN/TTc8yEGzTDqiYkSWJSWhT2Xi9v7a4jGBTB0KeZ1eHmJ88e4nBFJwXp0dy9\n4uwmRGb19/s5tW1s/D6D52Lx1CRSYsPYfaKNJ147TvC0QHOgeig6Yuww61wrhdqtTn77ajG2Hi8z\nck/17GrqnFilz9EqCy9sqyTSqOHrt05Dr1WRYDIgA+3dE/v9LAiCIAhXChEKfUxh00K9hKyb3hly\nu7O8DMeeXSgjIjBMKjircyoNBpIf+gaRy5bjbW6i4Uf/S8uTv6fmkX9H9ngGq0NclRUE+vpof/qf\neOrr0Kalj9qH6EoQWTQZ2e/HXVM96jFF5gLumnQLRTEFdDgt7G49AEB740kUMniTY7lrxl2khafQ\n4+3l7dottPa1j/m8do+DjdXvolfpKbrurtBtOz+Y8LrDNEYemvFFpscW0dLXxosVr/Hdjx7le7t+\nzJvVm8d8rLu2BgBd5sgNsbu1QXxKSGn38XrZGzirKwn09uDLzcAb9FEUM/xnT6M4FQqlhItQSBAu\nJEmSBquF5kyKG/W4NfPSiTRqeO3DWnYcG7tZvnDlqm118Oi/DtLQ3svS6Ul84/ZpGHVn14A5McbA\n7Elx3HZ1DtnJZzcp9GxEGDX84LOzmJIVQ0mtlXf3n6rYsfZvHzOFjx0KRYWf6ik0UQ6nl1+/dIxe\nl4/7VufztVumsmhKAiqlgsaOXg6Ud2Cxjx7sdPd4+NMbJaiVCh66dSoxkaFqpgRTqJ9XW5dzwmsR\nBEEQhCuBCIU+prDZc9BlZtGzfy+9R48A0HeimObHfwXBIPH3fX5C08DOJCmVxN99L3F33UOgr5fe\ngwdQRUZhWnc9ifd/EQDb+1up+/53se/YjiYpibh77juvr+1SE1lUCIDzZPmoxygkBYuS5nHXpFvR\nKNRsqn0PT8CLtb4KAHVCAonGeB6Z8xBfKLqboBxkQ+VbYz7vq5UbcQfc3JC9GlN2AZqUVJwlJwh6\nJn4Rm1Tv4C7FTB6d/x3WZ60mzhBLn8/J9sYP8QV8oz5uMBTKGLn83+q1cTTfQJgrSNyBKk7seB2A\nujQ9ENo6dib1aZVCyaJSSBAuuBsXZ3Hvqnxm5I3evys9IZxvfyb0IUNVk+gt9Gnk8wf59UvHcPR5\nuXN5Lvetyh+1F89YFJLEV24sYvW8iTWn/jjUKiX3rysgMkzDhh011LSEBnAMNJqOHicUMmhVaFQK\nbL2jVwCfzusL8LtXiumwuVi3MJ0l00JbxpUKBUkxBurbe/jj6yd49J8HqW11jHiOI5WdeHwBbl6S\nRWbiqV5LA6FQe7cIhQRBEIRPlyu3rOQTIikUxH/ufhoe/W/an/knQaeT9n/9HSSJpP/3EMaiqR/r\n/FHXXIs+Nx+Q0aSkIkkSst+PpNHgrqlB0mgw33wr0StXX9FVQgARkyeDJOEaIxQaEKkNZ3naEjbV\nbWN740dENjegB8JTMwaPKYopYFJ0LqXWk5R0lY8YoJRZKzjUcYyMiDQWJc0DwDhlKt1NjThPlo3b\nJwpCfYFanngcCPUAmn3VEpbkr6Ns858pN/ZRMaWGwpihE9m62+qp/vXPiOjqw2+KRGkcPs4XwOK2\ncqDQwKwGmFPqxK05CRoN+8Ks6IJacqKGh0mn9xRKEpPHBOGCiw7XsmzG8LHdZ0owGVCrFDRf4s1u\nA8Eg//f0YWKjdDx4Q9HFXs4Vo7bVQa/Lx7KZyayck3qxlzNhEQYND6ybzC9fOMqf3jzBDz8/l+4e\nDwatCp1m7OsSSZKICtPSPYHtY0FZ5qm3SqlucTB/cjw3LR5aQRt+2tSxXpePnz13hK/eVERRVsyQ\n4473TxqbeUZImxAjKoUEQRCETydRKXQeaJOTMa1bT8Bmo+1vT4FSSfJD3/jYgdDg+VNT0aamDfav\nkVQqYtatJ3zeAjL+58eY1q674gMhAFWYEW1qGu6aaoK+8T9VXJ62lDC1kffqP8DTGtqOYU4/Fb5I\nksQtudcjIfFq5VuD/YcGBIIBXqp4HQmJO/NvQiGF/roYp4S+r33Hj09o3d62ViBUpRTo6aHr9Q00\nPvZjwho6mX7SSWnL8POUP/8XIrr6aIpT8+H0kQMhgC6XFZ9agfHWm1AFZMJcQTrSIunwdTPJlDti\nE+mBSqFwdRiR2vBh9wuCcHEoFBJJMUZaLM5Luq/Q3pJ2alsdHCjrOOepUReDPxC8pBt5n2wIDUeY\nnD58qMClbnKGibUL0um0ufnTmyV02lzjVgkNiArXBtX18QAAIABJREFU0tPnxR8YPnH1dPtK2jl0\nspO81Cg+v7ZgWE+/Wf0hz70r8/jKjVMIBGUef6WYPSfaBo/x+QOU1XeTZDZijtIPebw5UodSIdFm\nHTsU+vBYC5v21U/otQmCIAjC5UCEQueJafVatBmZKPR6Uh7+FoaCyRf2+dauI/GBL6GO/XSNE9fn\n5ff3FaoZ/1iVjtUZy3EH3KgsNnxKiZjEoZUzSWEJXJU8n3ZnBzub9wy5r8xaQYfTwoLEOaSGn/qU\nX5+VjUKvx3ligqFQR6hnkWnNOrJ+8Rvi7v0shsIidLl5qILQu38ve//8E3ptocbh7bVlRJc20m3S\nUnb7PIoT/Ng9I5fBW1xWJCQSFlxN2MJFAByMD71JKxyhnxCcajQ9kclpgiB8spJjjfgDQTpsl2az\n20AwyMbddQDIwMHyjou6noly9Hn5jz/u5uktFRd7KaMqbwhtG8xLjbrIKzk3N1yVSXZSBMXVXXh9\nQZbNHL86DiAqTINM6Hs0Go83wCs7qlEpFfzbugLUquGXr0umJfHTL81n2cwUZuXH8q07p6NVK3nq\nrVKK+6uDTjbY8PqDTD2jeghApVQQG6WnzeocNTz0B4K88H4lr35QM26IJQiCIAiXCxEKnSeSSkXq\nI/9J1s9/hT4392Iv54plyA9t8XKWl03o+KuS52PWRGFy+OmL1qEYob/TdZkr0Kt0vFP7Hs29rXgD\noQvTPa0HAVicPH/I8ZJKhWHSZHydHfgsneOuwdceCoU0cfEo9Xqili4j5RvfIvbm2wBYsNeCaf9J\nyja/BEDNS/9EIYNmzUqy+7d/1Tsa6fH28ui+X/L3kudw+kKfZFrcViK1EaiVahLv+zzFN82gMi30\n6exI2+EAYnQmdErdsC1rgiBcfMmxocrA5nGmKG3cVTtqIHMhq2H2lbbT0e1iRq4ZCdh/mYRCL2+v\nwtbrZW9J2yX5Zt7nD1LdbCcl1ki4QTP+Ay5BKqWCL60vpCjTxAPXT+aamSkTetxARdFYW8g27aun\nu8fD6nmpmCP1Ix6jUEjERRsG/z8vNYqHb5sGwEfFoWrhE7VWAIqyTCOeI8FkoM/tp9c1cq+/8oZu\nXJ4AQVke7JskCIIgCJe7K3/P0SdIoVaD+uymhAhnR5+bB5KEdeMb9BzYhy4zC11mFpELr8JVXYVt\n+zYSH3gQhTZ0kalWqFhvmIkqUIHfPHJJfrgmjDUZ17Kh6i3+b/+v0av0zEuYyXFLKclhiUOqhAbo\nMjPpPXIIT2MDavPY1Vq+jtCbJnV8/NBzZGejMBqhL/Tmz19VTfPJI8RUttEVq2fu0hupsIUmrdU5\nGql3NNLW105bXztVtlrumXQb3W4bWZEZQCisSpu1GEpfIC08edStYUa1gZ8v+SES0oj3C4Jw8SSb\nQ6HQ0++Ws+t4KylxYWQmhpOZGEFU//ju1q4+XvuwljC9mmk5MUNGjr/yQTUfFbfwowfmE6Y/v7+P\ngkGZjbvrUSokPnNtLi6Pn/IGG/Y+L5HGSzfIqGi0sat/C5HbG+Bkg43CzJFDgU+SLMv8c/NJWrv6\niInQ4fUHyU+9/LaOnc4cpeebd4zfa+90pojQ9K+tB5tIWxs+rArI6nCzeV8DkUYNa+enn9W5s5Mj\niI/WU1zThdcXoLLJhlIhjTqVbXACmdU5Yjh3pNIy+GeLzUVc1MgBlSAIgiBcTkSlkHBZUYaFEXfP\nZzEUFBKw2+nZs5vO556h7e9/wfLaq/QdPYKr8uSQx8TvCjWmzli6dtTzLk1ZyMr0ZcxLmIVCkvig\naRcBOcCipHnD+hYAaFNDU13cDQ3D7juTt6MdhU6HMnxoSCMpFITPnY/CYMCpVxLebKXxpWcACFu/\nDqVCSXp46JPW8u5KdjTvJkxtZG3mChzeHp449hdkZMz6U29upponkxuVxTWpS8Zck0JSjPi6BEG4\nuJLNYQA4nD6OVll4a3cdv3v1ON98YhePPXsYl8c/+Ma01+XjQH+ljr3Py9aDjbyztx6H00dF4/mf\nYLavrJ12q5NFUxIxR+oHg5XKC/Bc50sgGOSZ/i1jNy8JNSY+etob+4vpSKWFncdaqGyys7e0ncQY\nAyvmTKy65kqysCiBzMQI9pW284sXjtDjHLqN7NUd1Xj9QW5emjVu4+ozSZLEzPxYvL4ghyo6qW/r\nJSMhHK165KmwYzWbDsrykJ+dzkt0i6cgCIIgnC1RKSRcdqKWXk3U0quRg0F87W20/vlJeg8fgv4t\nE56GhsEm386KkziPHUWfl0/i7EWjnlOlUHFD9hoA3H4PTb0tKCUl6REjX6BrU0OTYTxNjWOuVQ4G\n8XV2oElIHDGEifvM3cTeeju7/vgo8SeaMNV3YUkMY/680FoMagNxBjP1jtDzrM9azaqMa5hsyucf\npc9jcXURZzAPnk+n0vHwzAfHXJMgCJcuU4SWJLMRg1bF59ZMwupwU9vWQ0mtlZONNh5/+RhuXwBJ\nAmR46f0qXt1RQ/cZW1lqWx3Dpit9HIGgzMZddSgVEusWhKo1clNCvW8qm+zMnhR33p7rfNp2qJmm\nzl6umprI6nlpbN7XwNGqTu5akXtRg/E2q5PntlagVEh87ZapyLJMYabpnEbQX+6MOjWP3DWDv75d\nxoHyDh7950G+fts0ks1Gqlvs7ClpJy0+jEVTzq0P3qy8ODbtbWDDjhqCskzuGD2bTq8UOtO2g010\n93hIMBloszrptLnPaT2CIAiCcKkRoZBw2ZIUCjSJSURdu5L2v/9l8PaBoEaWZSyvvgyA+ZbbJvwG\nQKcaeZT76ZSRUSjDwvE2jh0K+W02ZK8XdVz8iPdLCgWSVhvqlXSiCYDoG29CoTj1xmB2/Ax2t+xn\nQeJslqeFKoAyI9P47pyHOdh+hOlxUyb0ugRBuPRJksT/fmEukhT6c5LZSFFWDNfNT+fJN0sG+whN\nSosiTK/m4MlOosI0TM8xk5EQzqT0aH767GHqWkduTg+hfxsDQfmsAoiPjjbTZnWyeGri4NSmzMRw\nVEqJiqZLs1Kou8fD6x/WYNSpuPXqbFRKBflpURyptODo8xIZNrHpWI7+ypWI89DrR5ZlPjjawovv\nV+L1BblxcSZTs4c3Pf600aiVfOmGQhJjDLy5q47/e/og967K5919od+xn1mei+IcQ7zMxHCykyKo\nbgn9nchLObtQSJZljlRaePmDasINar64fjL/+4+DolJIEARBuGKIUEi47IXPnYvllReR/X4gVCkE\n0Hf0CO7qKsJmzkKfnXNen1OSJLSpqTjLSml/5l8EenuQlCoklQpJpRz8c6CnBwBN/Mih0IC0GYuw\nvLYVS2oUV828Zsh912Wu4LrMFcMeo1NpueqMJtiCIFz+FIrhb34VCokvrZ+MOVLH5n0NLJqSyLzJ\n8dzj8Q8LK+Ki9dS19SDL8pAwvKPbye4Tbew+0YbT7eeHn58zbCz3SHz+AM++W45CkrhuYcbg7WqV\nkozECKqb7bg8fvTaS+uS4uXtVbi9Ae5blT/4NUqJDeNIpYWaFgcHyjuIidSxfFbKYL+mMwWDMo/+\n4wC2Xi8LihJYMy+NxBjjOa2nu8fDPzaVc7ymC6NOxRfWFjC3YOzfDZ8mCknixsVZxJsM/P2dcv78\nZikACwoTyE87915LkiRx/7rJ/PBv+/H5g+SkjNxPCCDcoEavVQ2GQh3dTp7bWklxdRdKhcQX1haQ\nFh8KQy12EQoJgiAIV4ZL6wpOEM6BQq0h5d//AzkYpPP5Z3FVVRJwubBseAUUCsw333pBnleTEgqF\n7B+8P+6x2rS0Me+PTcjE8e2vMS0hU/T6EQRhREqFgtuX5bB+UcZgb5WRqlcG+rN02FxEGjXsL+tg\n1/FWKpvs/eeRCARl3thVy/3XTR7xuaqb7ZTWWbluQQbv7G2g1dLHitmpwxrr5qZEUtVkp7rFTlHm\npVPxUlbfzd7SdjITw1kyLWnw9pS4UM+mTfsbqOr/euwrbedH/zYPzQh9Zqpb7HQ5PKiUEh8Vt7Kr\nuJXpuWY+szx3QoEahPo9/frFozR29CIDhZkmvrC2YHDqljDUgsIEYqP0vPpBNXMK4rh6xsRG248l\nwWTga7dMxdbrGbMBuyRJJJgM1LY6eG1nDZv2NeAPBClIj+aelXmDgaA5Ui+2jwmCIAhXDBEKCVcE\nbUqox482NQ1XZQWWl1/A29pC5JKr0SScWx+C8egzs7ABmqRkUr75LWQZZL8P/H5kfwA54Ef2+5FU\nKrRp409Myc6ddUHWKQjClWW8ZruZCeHsK22nuKqL3SVt1Lf1IAEF6dEsLEpgRm4sP3nmELtPtJGR\nEMG8yfFD3ij7/AGefKOELocbWYa399RjitBx4+Lh22qnZsWwaW8D2w83XzKhkD8Q5JktJ5GAe1bm\nD6m8GpjuNhAI5aVGUdFoY/O+BtZfNfz1HavqAuDLNxYRDMI7e+s5UmlBluGhW6dOaD2ltVYaOnpJ\njQtj2cxklk5LEuH/OHKSI3nk7pnn9ZwTnTgXb9JT2+pg4+46osI03Lk8lzmT4oZ8z8xROtqszkuy\nQk4QBEEQzpb4TSZcUQYqcuw7dyBpNMSsv+GCPVfYrNkkPfQNDAUFKNSX7jhmQRA+XablmNmws4bn\nt1UCMHtSHHcsyyEmUjd4zG3LcvjtK8U8+14Fm/bV82/XTUarUeL1BThaZaHLEaqCeP2jWgAeuLFo\nxDe/ealRZCdHcKTSwtPvnsQcqWPNWY4NP99qWhy0djlZNCU01ep08SY9KqUCfyCIWqXgqzcV8V9/\n3c87e+tZNCVx8GvU2tXH9iPNFFd1oVYpmJxhQqtWMjPPzH/8cQ9VzfZh2/NG09jZC8Bd1+Z+rG1Q\nwidjcrqJvSXtLJmWyB3X5I74cx/bXyVWUmu9ZJusC4IgCMJEffrGXAhXNOOUaegnFaDPzSPurntR\nRV24C3BJqSRs6jQRCAmCcEmJNxn4yk1TUCpCjarvX1swJBACmJodw/99aT7XLUjH6vDws+eP8Og/\nD/LYc0d4d38jRp2KOf1vdosyTSyamjTSUyFJEtcvDFXYbD/SzMsfVFPf1nNhX+A4avubbE/JGl65\npFQoSOofO56dFEG4QcOtV2fj9Qd5aXsVEKo0+sPrJ9h6sIkOm4uC9OjBEeaSJJGdHEGvyzfhRsNN\nHaFQKLV/65pwaVs0JYE//vtSPremYNQqoIWFCahVCv68sYTi6q5PeIWCIAiCcH6JSiHhiqKKjCT1\nW49c7GUIgiBcVFOzY/i/L84nTK9GqxneKwcgLkrPLUuzyUqK4HiNFY1KgVatRKNWUJBuwhypIzJM\nw+q5aWNWxEzJMnHvqny6e9y8tbued/bW8+Ubiy7USxvXQCiUcUaV0IDk2DAaOnrJ6x9NvqAogQ+O\nNHOgvINrGro52WijubNvcCrY2jMqn7ISI9hf1kFNi4O4aMO462ns6CUmQotBN3ovG+HSIUnSYAg4\nmuzkSB6+dSq/eaWYP75xgu/ePZO0+PBPaIXCpaaq2c4TG44zNSuGm5ZkiX5hgiBcdkQoJAiCIAhX\noNgJNkKekRvLjNzYEe+769q8cR8vSRLLZiQjyzLF1V0cPNlBu9VJvGn8wORCqG11EKZXE3tGddSA\ngvRo9pa2MT3XDISmXt21Io9H/3mQv28qx+pwExWm4YvXF2LQDb9MykoKTa+qbnEwvzBhzLU4+rzY\n+7xME2PnrzgFGSYeWDeZP7x+gsdfKeb7980WYcCn1KGTHTj6vHx0vJWWrj7+895ZKETfMEEQLiNi\n+5ggCIIgCB+bJEmsnZ+OLMOmfQ0XZQ2hbV1uMhLCR61uWjQlgccfWkxGwqlKoszECK6amkhHtwt/\nQObeVfkjBkIAafFhKBUSNS2OcdfT1N9PKDVebB27Es2eFMetV2fT3ePht68U4/EGLvaShItgoJn/\n1OwYaloc7Ctpv9hLEgRBOCsiFBIEQRAE4byYnR9HXLSe3Sda6e7xTPhxHl+A3Sda8fmDH+v568bZ\nOgah8GqkseS3LM0mwqhhQWHCqJVTABq1koyEcOrberD3jv0aG/v7CaXEilDoSrVmXhqLpyZS397D\nnzeWEAzKF3tJwicoKMvUtfWQEGPgnpV5qFUK3uhv0C8IgnC5EKGQIAiCIAjnhUIhsWZeGv6AzHsH\nGif8uK0HG/nLW2W8+H7lOT+30+3ntQ9rgNBI87MVadTwi68s5P51BeMeO78wgaAss/tE25jHnai1\nAqE+RMKVSZIk7l2VT0F6NEcqLYMNyz8pHl8Ar09UKF0sHd0u3N4AGQnhmCP15KdF0WFz4XT7LvbS\nBEEQJkyEQoIgCIIgnDcLixKJDNOw/Wgzm/bV88/N5QSCY1cAHam0APD+4WZK66zn9LxPbzlJbWsP\ni4oSKMoyndM5VErFhHqBzC+MR61SsPNYC7Is09TZS8cZ08h6nF7K6rrJTAzHPMH+TsLlSaVU8JWb\nikiMMbDlQCMvf1A1uJXM6fZR39bDkYpOth1q4r0Djbg8/gmdt73bSbvVOer9QVnmR/86yE+ePYws\niwqli2GgOjG9fztqstkIQLOl76KtSRAE4WyJRtOCIAiCIJw3apWCVXPSeGl7FS9vrwZgVl4sRSOM\niAew93qoaXEQG6Wj0+Zm26EmJmcMDXX8gSCldVbarC6WTksaNlHN5w9wpLKT+Gg9n19bcMGbvBp1\nambnx7KnpJ1XdlTz3oEmFBJ8ds0kFvQ3nz5U0UlQlpkzKf6CrkW4NBh1ar5+2zQee/Ywm/Y2cKCs\ng/mFCWzaW0/gjC1lx6otPHzbNFTK0T+bDcoyv3j+CG5vgMceXDBkep0sy8gylNV309wZCh+qmu3E\nRelp7XJypMZKTWM3Hd0u7L0ebrgqk4KMcwtKhZEFZZnG9l72lob6B2UkhKbPJQ2EQp195KZEXbT1\nCYIgnA0RCgmCIAiCcF4tnZ7Euwca8PuD9Ln97ClpHzEU8vkD7Olvyrp8ZgpbDjZS1WxHluUhjaL/\n+nYZ+/rffFU02vjKTUVDgp+KRjteX5BpOWYUik9m6s8NV2VytKqLTXtDTbW1GiVPbSyltsXBbcuy\n+fBYKwBzJsV9IusRLr64KD0/+rd5vLW7jk37Gnhrdx0RBjVzCuIxRWgxhevYV9rO0SoLb3xUyy1L\ns0c9V31bD12OUM+qZ9+rIChDl92Nvc+Dvc+LhIQp4tS0s7+9XUaHzcVIBUO7S9pEKPQx9bp8lNZZ\nae1y0mZ1UlpnpccZ2iJm1KlIjw+FQsnmUP8wUSkkCMLlRIRCgiAIgiCcV3qtikfvn4dGpeD7f9nH\n4cpO7L0eIsNCb2Kdbj9bDzay5UAjzv6tNNNyzdS0Othf1kGHzUV8dGikvcXuYn9pO8mxRow6NYcr\nOtmyv5HV89IGn6+4uguAKZ/g6Pe4aANfvH4yv3/tBKvmpnLVlER+t+E4Ww81ceBkB/ZeL7PyY4mJ\n1H1iaxIuPr1WxW3LcshJjuTgyQ5uWpKFOfLU9sHpOWb+/fe72H2ijZuWZI1a1XasKrSlUpIYDE4V\nkkSEUU2iyUinzUVrl5NksxGvP0B7t4vIMA1XTUkkOzUavUrCFKHjO0/uodPmvvAv/ArmdPt45Mnd\nuDynejdFhWlYNCWBwkwTRZkxg9WLSebQv1stIhQSBOEyIkIhQRAEQRDOu4EJX/Mmx/P2nnq+8cQu\npmTFkJUUwXv9YVCYXs3iqYnkpEQSH20gOzmS/WUdVDfbB0OhncdakYFVc9KYnmvmkSd38+6BBq6d\nnYJKqcBid3GkshOtWkneJ7xdY1qOmd89vBitOvSG8Pv3zeJv75RzsLyDuCg9n18z6RNdj3DpmJEX\ny4y84VPstBolM/LM7DreRk2zg5yUkZuiH6vuQqmQuP+6AvaUtLN6bir56dGDIVKb1ckL2ypZPisF\njzfA9iPNfHZ1PnHRBmJjw+ns7AEgOkKLxe4a8TmEialqtuPyBJiRa+aamSmYIrQkmAxDqhkH6DQq\nzJE6USkkCMJlRYRCgiAIgiBcMKvmpuHzB6lusXO8povjNV2E6dXcsjSL5bNS0GlOXYoMTA07XmNF\nlkPTu45UdqLXqphTEIdWrWRRUSJbDzXxzt56Wix9HCwP9e5ZNCUBteqTn58xEAhB6A3hl28o5FhR\nAunx4UP6wAjCgDmT4tl1vI395e3kpEQiyzItlj5Kaq2cqLNS3R9CTM6IZn5hAvP7+1SdLsFk4OHb\npg3+/+xRtimaI/VUNtrw+YMX5e/HlaCq2Q7A1TOSKcwcfxtektlIcXUXz26pQKmUUCgkYqP0LJ6a\nOGYfKUEQhItFhEKCIAiCIFwwYXo1dy7PRe4f4d7r8rFkWhJ67fBLkNS4MDQqBftK2wd7CEWHa7l1\nafZg+LJsZjJbDzXx+oe1AKTEGlk1N415ky+Nhs6SJDE9x3yxlyFcwiZnRGPUqdh9vI1Ek4F39tYP\n9g+CUOCTnaxj3YKMj/1csVE6KhrB6nATbzJ87PNdTvyB4HkJYaqa7EhAdlLEhI7PTYmkuLqLbYeb\nhty+82gL/37n9MEqSkEQhEuFCIUEQRAEQbjgJEli0ZTEMY9RKRWsXZBOVZOdyRkmpmSZSDIbh2zT\nSIwxsmpuKu1WF8tnpTA5I3rEbRyCcKlSKRXcuTyXv75dxtNbKlApJeZNjqcww0RhponocO34J5mg\n2P5+Rp0216cqFCqrs/Krl45x/3UFI1ZaTVQgGKSm1UFSrHHClX9r5qczPTcWvz9IUJYJBmXeO9jI\n/rIOdhxt5rrzEPYJgiCcTyIUEgRBEAThkrF+Uea4x9xxTe4nsBJBuHAWTUkkEJTZdbyVO5fnkpk4\nsSqUsxUb1R8K2T9dzaY37q4jEJR5flslFrsbnz/IDYszR23sPZqmjj68vuDg1taJUEgSyf2j6QfE\nmwwcOtnJoZOdIhQSBOGSI0IhQRAEQRAEQfiELZmWxJJpSRf0OcxRoel3nbYL32y6ob2HJLPxovfN\nqW/robzBhlatpMfpY8POGgCCsswtS7PP6ly1rQ4Asj5maBemVzMpLYqSum667O4hUwlrWhy8tbsO\nfzDIw7dOQ6EQlY+CIHyyRCgkCIIgCIIgCFeggUohywUOhQ5XdPLEhuPctDiT6ydQ7TcRPn8Arz+I\n8SwatgeCQV58vxKAL14/mX1l7USFaTlaZeHtPfWE69WsnJs24fN1dIe+bolnVP6ci1n5cZTUdfPy\nB1Usm5GMSqVg4646iqu7Bo+paraTl/rJTlEUBEEQoZAgCIIgCIIgXIEijRrUKgXNlj58/iCHKzop\nzDSd12bHQVnmtQ9D1TgHyjs+VijU2NHLr186yp3Lc9lxtIXGjl4ee3DBiI3pR/LitirKG2xMzzEz\nPdfMjLxYAJbNSOax5w7zwvtVBIIya+anT+h87d1OAOKi9ef2gk4zMz+Wjbvr2F/Wwf6yjsHb81Ii\nmZQezZu76jhWZRGhkCAInzgRCgmCIAiCIAjCFUiSJIoyTRyptPDYc4epaXEQpldz+7IcFk1JOC9N\n2g+UddDc2QdAU2cfx6osON1+ZuXHoumfGjhRm/bVY+v18tTGUgJBGYCD5R0snsA2u53HWth6qIlk\ns5EHrp885LXFmww8cvdMfv78EV7+oBp/IDgkvDp0soP2bher5qaiVJza/tZhc6HXKgk/DyFahEHD\njx+Yx8kGGydqrdj7vCybnsSk9Gh8/iCb9zdwtMrCbctyPvZzCYIgnA0RCgmCIAiCIAjCFermJVkc\nrbRQ0+IgJkJLr8vP394p46PiFu5dlU9ybNg5n9vp9vHC+5WolApWzElh094GHn+lGIDntqrITo4k\nNS6M1LgwMhMjBrezjcTe6+FAWQdqlQKfPzRO3h8Isut467ihUGWTjaffPYlRp+Jrt0wZsbIoPtrA\nI3fN5GfPHeG1D2sJBGVuuCqTQFDmH5vK6XP7qW118KX1haiUCoKyTEe3i6QY43mbcKjTqJiWY2Za\njnnI7Rq1ksKMUHjXbOkj2WykuLqLd/bUEQS0GhU+X4DkWCO3L8tBe5ZhmyAIwlgubic4QRAEQRAE\nQRAumOTYMJZMT0KjVvC1W6by4wfmMTMvloomOz/8+wGOVlrO+dwvba/C3utl/aIMVs5ORQIkYFFR\nAhq1kuLqLt7eU8+Tb5TwnSf38NzWCnz+4IjnGpgYdvuyHFbMTuWzq/MpSI+moslOi6Vv1DW0WZ38\nfsNxZBm+cmMRcdGGUY+NjdLzyN0ziI3S8eauOjbsrKG0zkqf249GpeDQyU7+8lYpwaCMrceDzx88\nL1vHJmJ+YQIAf3u7DI8vwHPvVVDRZKe62U5pbRcVjTa2H27mFy8cwen2fyJrEgTh00H5wx/+8IcX\nexEATqf3vJ3LaNSe1/MJlwbxfb0yie/rlUd8T6884nt6ZRLf1yvTSN/XqVkxLJ+VgjlKj16rYm5B\nPGlxYRyu6OTAyQ6KMk1Eh2vP6nlK66w8v62K1Lgw7r+uAINOjSlCy+JpSaydn86quWlcMzOZKZkm\n0uLCsNjdFFd3oVJK5KdFDznX5n0NvLWnnsQYA/euzGd6rpm0+HB0GhUHyjuobnGwqCgR5RmTufaW\ntPGr/9/efcdXWd/9H3+dc3Ky9w5kEAgQCBD2FC1gUXCAA61UK/ZW61213raW2tr+9L5ba+fdcdfW\nXW2tiKOCqwoKWJG9RwhJyA7Ze49zrt8fgSiyQnLISa68n/805JxzXZ+L9yN8zaff8dp+mtsc3HL5\nSGaMjT5v3b7edqaMimB/VgV7Mys4kldNa7uDB29KpaKmmQPZnUu7gv292HKohEkjwxk7LPSC/m56\nYkiYL+U1LRzMrmRfZgUlVU1cmjqER1dM487rJjAvNYbymmYOZldRVN7A9LFRLpvBJH1H/+6a00DI\n1c/v7P/Ga6aQiIiIiIiJWa0WfL90itekURHcdU0K7R1ONu4tuqDrtbY7eOmDdCwWuGNxctcx9HMn\nDGHiF5ZGBfh6MmZYKAunx/OT26fi6WFl66EUI+oGAAAgAElEQVQSDMPoes9nB4t5bWMWIQFefPem\niXh5fr40asroCC4ZH0NeSX3XqWJf9O7WPCwWuGdJCpdPjet2/aGB3qxcPpmYMF9qG9sIDfQiOSGE\n79yYSnyUP5/sO84rH2UArtlkujssFgvfODE7qqiiEQtw5YzPT0rzsFn5j6vHkDIshP3HKnn3s9w+\nqUtEzE9NIRERERGRQWjSyHC87DZyjtdd0OfWfJpNeU0LV0yPZ1h0YLc+4+PVuZ9OaXUzeaX1AOzP\nquCv76fj5+3Bd29KJSzI+7TPfX3hKIZG+LFhTxE7jpR2fb+t3UFxZSPDogOYPibqguoHCAnwYuXy\nyUwZFcENl43AarHg6+3Bd2+aSFSob9fm2VHnWI7mal52Gw/elMo1s4dx8/wkokNPvbfNauVbS8YR\nGujF25/lkn2BufWUw+k8pZEnIuaippCIiIiIyCBktVoYFh3A8YpGmlu7t0/N7qPlrNtZQGSID0su\nubDj52eM7WzevLc1j3U78vnLmkN42Cw8cGPqWTe89rLb+PbScXjZbbz4r3RKqzqPiS+qaMQwID4y\n4IJq+KIgP0/uvX48s1I+X3YW6OfJQzdPJDTQC4uF0xozF5uHzcp1lw5n4fT4M77u72PnzqvGYhgG\nf15zkOLKM++31NzawaqPMtmwp7BX9TgNg8de2MnTbx/u1XVEpP9SU0hEREREZJAaPiQQA8gtqT/t\ntfqmNo7mV9PhcPL024d56M+f8eRbB/GwWfnm4jEXfArW+OFhhAd5s/toOa9uyALgP5eOIyk26Jyf\niwnz4/YrR9PS5uDPaw7R3uEk/8Rso7ionp+edjZhQd78+BtTWXnLJAL9PF1+/d5KTghh2bwkqupa\neeLlPWw7fOqSvLqmNv7nxZ2s31XAy+sycDp7Pssnt7ieoopGDmZXabaQiEnpSHoRERERkUFq+JDO\n5V/Zx2sZk9C5AXReST0f7y5kW1opHQ4nMWG+FFc2EeBrZ2RsELctHE1s5IU3Y+weVh69Yxp7Mypo\nbutg5tgoAny713SZmRJNen41/95fzLqd+VTVtwIQfxGaQgDB/l4E+1/Y5tt96coZ8fh6e/DK+gye\neSeNzw4Wc9sVo4kM8eWNTccorW7uem9RRSNxPcgL4MCxztPpmls7qKxrITyob/ZYEpG+o6aQiIiI\niMgglRhzsilUR01DK0+tPUxGQQ3Qucmyh83K8YpGIkN8eHTFNHy8evfrg5+3nUsmxPToszfNS2Jv\nZgXvbuncYNpqsTA03K9X9Qxkl6YOITkhhJfXHeVQdhU/eX4Hk0aGs+NIGbER/sybPJS/f3iUrKLa\nXjSFKru+LihrUFNIxITUFBIRERERGaRCA72JCPbmcE4VL6/LIKOghjEJIVwxPY5xw8Noae1g3c4C\nZqZE97oh1Fu+3naWfSWJF94/AkBUiA92jwtbwmY2kcE+PLgslZ3pZaz6KJMdR8rw9rTxjStG4+Pd\nmVdWYQ3zJg3t9jVLq5s4nFNFSVUTuSX1eNqttLU7KShrYNLIiIv1KCLiJmoKiYiIiIgMYgunxfOP\n9RnsySgnKsSH7908EavVAnQ2YpbOHe7mCj93yYQY/Lw9+Gh3IVOTI91dTr9gsViYPiaKiUnhVNW3\nEuTniY+XB07DwM/bg6yi2m5dx2kYrN9ZwJufZNPhcHZ9/+pZw/jnv7MpKGu4WI8gIm6kppCIiIiI\nyCA2d0IM73yWQ11TO4tmJnQ1hPqrSaMimDRKM1a+zNNuO+W0NKvFwoihQRw4VklhWcM594GqrG3h\n+ffSSM+vIdDXztK5I4mPCiAq1AdfLw8+3JGvppCISakpJCIiIiIyiHnabXx94Wj2ZZafcjy7DHzz\nJg3lwLFK/rbuKA9/fTJWy6kNP8Mw2HKohFc+yqC51cGkkeHcfmXyaaeuxUX6czS/hsaWdvy87X35\nCCJykelIehERERGRQW5aciR3XZOC3UO/HphJalI4U0ZFkFVYy+YDxae9/ty7aTz/3hGcBtyxKJn7\nrh9/WkMIICUxFAPYc7S8D6oWkb6kf/VFRERERERM6pbLR+LlaeP1jVnUNbbR4XCSfbyOsuomth4u\nJTbCn59+czpzU4dgsZx56eCMMVEAbEsr7cvSRaQPaPmYiIiIiIiISYUGenP93OGs+jiT1zZm4Wm3\nsWlvEaPiggG4fGos4cHnPmo+PNiHpNgg0vOqqa5vJSTAqy9KF5E+oJlCIiIiIiIiJrZgSiwJUQFs\nOVTCJ/uKAMgoqMFqsTBpZHi3rjFzbBQGsOOIZguJmImaQiIiIiIiIiZmtVr4xpWjsQCGAbERnSeR\njY4PJsD39D2EzmRaciQ2q0VLyERMRk0hERERERERk0uMCeRrC0aycFocD96Uyui4YK6cEd/tzwf4\nepKSGEpeST3FlY0XsVIR6UvaU0hERERERGQQ+Oq0uK6vf/D1yRf8+ZljozhwrJJth0u57tLhrixN\nRNxEM4VERERERETkvCaODMfTbmV7WimGYbi7HBFxATWFRERERERE5Ly8PT2YPDKCsppmsovr3F2O\niLiAmkIiIiIiIiLSLTPGRgGwYXcRx47Xkl9aj/MMs4achoHD6ezr8kTkAmlPIREREREREemWlMRQ\n/H3sbD1cwtbDJQDcPD+JK6Z3blptGAY708t4fWMW3p4e/PDWKfh669dOkf5KM4VERERERESkWzxs\nVv5z6TgWzYznyhnxpxxTn1dSzy/+sYen1h6msq6VoopGnn8vTfsPifRjatmKiIiIiIhIt41JCGFM\nQggAhWUNHMqp4u8fHmXT3iIMYNLIcJbNS+JvH6SzN7OCA8cqSU0Kd2/RInJGmikkIiIiIiIiPTJl\ndAQAG/cWERrozfe+NpH7b5hAdKgvX1swEoB1OwvcWaKInEOPZgq1t7fz8MMPc/z4cWw2G0888QRx\ncXGnvOf999/nhRdewGq1MmvWLB588EGXFCwiIiIiIiL9w6SREfztw6N4eth4YNkEYiP8u16Ljwpg\nTEIIR/KqeeTZbdisVpITgrl2TiL+PnY3Vi0iJ/VoptC7775LYGAgq1at4p577uG3v/3tKa83Nzfz\nm9/8hhdffJHVq1ezZcsWsrKyXFKwiIiIiIiI9A+Bfp7cd914vve1iac0hE5aPDMBgKq6Vspqmvho\nVyE/fm47u9LL+rpUETmDHs0U2rp1K0uXLgVg9uzZ/OhHPzrldR8fH95++238/Tv/UQgODqampqaX\npYqIiIiIiEh/M2lUxFlfS0kM5df/OZsgf0+gcynZmk9z+POaQ0weFcGw6ACq61v52oKR2D20u4lI\nX+tRU6iiooLQ0FAArFYrFouFtrY2PD09u95zsiF09OhRioqKSE1NdUG5IiIiIiIiMpCEBXl3fb14\nZgKTR0Xw4r/S2ZNRzp6McgDGDQ9l0sizN5dE5OI4b1Po9ddf5/XXXz/le/v37z/lz2c7YjA3N5eH\nHnqI3/72t9jt514zGhLii4eH7XzldFtERIDLriX9h3I1J+VqPsrUfJSpOSlXc1Ku5mO2TCMiAvj1\nyEg27y8iu6iWNzdmkVFUx8LZw91d2kVntiyl00DO9bxNoWXLlrFs2bJTvvfwww9TXl5OcnIy7e3t\nGIZxyiwhgJKSEu69915+9atfMWbMmPMWUl3ddIGln11ERADl5fUuu570D8rVnJSr+ShT81Gm5qRc\nzUm5mo+ZMx0TG8ToIYF8sDWXHYdLKLusDovF4u6yLhozZzmYDYRcz9W06tGizTlz5vDBBx8AsHHj\nRmbMmHHaex555BEee+wxUlJSenILERERERERMTmr1cL44WFU17dSUNbg7nJEBp0e7Sm0ePFitmzZ\nwi233IKnpye/+MUvAHjmmWeYNm0awcHB7Nq1iz/+8Y9dn1mxYgULFixwTdUiIiIiIiJiCqlJ4WxL\nK+XVjzN5YFkqXnbXbSsiIufWo6aQzWbjiSeeOO37d999d9fXX953SEREREREROTLpiZHsCs9gt0Z\n5fztg3TuukarTUT6is78ExEREREREbexWa18a0kKsRH+bEsrpaK22d0liQwaagqJiIiIiIiIW3nY\nrFwxPQ7DgPe35pFTXEd6XjX7MivYllbC0fxqd5coYko9Wj4mIiIiIiIi4krTx0TxxqZjbNp3nE37\njp/2+u1XjuayiUPdUNm5Nbd2YBgGvt52d5cicsHUFBIRERERERG3s3tYufPqsezOKMfLbsXLbsPb\n0wO7h5U1n2bz8roM4qMCSIwJPO+1nE6DDocTzz7YtPp3r+2nqKKBWxeOZlZK9EW/n4grqSkkIiIi\nIiIi/UJKYigpiaGnfT861Jffrt7HK+sz+OFtU7AAFoul63WH00ltQxvBAV5YLRaefOsg2cV1/OjW\nKUQE+1y0ep1Og9ySejocTp59J40Dxyq5beEozRqSAUNNIREREREREenXUhJDmTo6gl1Hy/mvP27G\n4XQyf3IsC6fFEeDryXPvHmF7Wim+Xh7cszSFg9mVdDgMfv/6fh65bcpFa9JU1rXQ4XCSHB9Me4eT\n7WmlZBXW8O3rxndrRpOIu2mjaREREREREen3bvzKCOweVtraHXjYrLy3NY+Vf9nKX9YcYntaKSEB\nXjS1dvDSv9LpcBgE+nlSXNnEk28dosPhdHk9NQ2tlFY1AZAcH8LDt05mySWJVNW18o/1GS6/n8jF\noJlCIiIiIiIi0u9Fhvjy+F0zuvYZ+vf+43ywPZ+d6WXYPaw89LWJ/OH1A5TVdB5p/83FyXyy7zh7\nMyv424dHuWNR8ilLznpj6+ESnn0njQkjwgCICvXFZrWy5JJE9mVVkF/auaTMw6Z5GNK/qSkkIiIi\nIiIiA0J40Of7A311ahxfmTiUHUdKCQ7wIibMj+ljI3l3Sx42q4XRcSGMjgvhF6/sYfOBYsKDvLl2\nTqJL6ti4twiAA8cqgc49j05KiAogr6Se4xWNxEcFuOR+IheL2pYiIiIiIiIyINk9rMwZH0PKsM7N\nqaePiQJgxJBAvDxteHnaeODGCYQFerHm0xx+9coevvOHT0nLrerxPcuqm8gqrD3le1GhnzerEqI7\nG0F5pfU9vodIX1FTSEREREREREwhNsKfu68Zy60LR3d9L9jfi5XLJxMW6E16fg0Nze2s3pCF0zB6\ndI/NB0sA8PPuXHgTEuCFt+fni3ASTswOyi9p6OljiPQZNYVERERERETENGamRBMb6X/K9yKCffjJ\niql8/5ZJzBwbRUFZA8+/e4T1uwpoaG7v9rWzj9fxwfY8/H3s3HDZCODUpWMAsRF+WC0WzRSSAUF7\nComIiIiIiIjpBfp6EpjgSWigF7uOlrH1cAlbD5fw5qZjzBgbxfzJsV1Lv86krrGNJ986iMNhcPe1\nYxkWHci7W3MZNzz0lPd52m0MCfclt6SOv32QTkJ0ADPGRl3kpxPpGTWFREREREREZNCICvHlZ3fN\npL6xjczCWjbsKeTTA8V8emIz6mExgSycFsf+rAomJoUzYmgQDqeTp9Yeorq+lesuHc64xM5Tx37z\n7TlnvMe8ybG8sSmLTfuOA7DjSBm/vH9unz2jSHepKSQiIiIiIiKDSmSwD5HBPowYGsTCaXEczK5k\n094ickrq2ZVexq70MgAOZVfx6B3TePOTbNLza5iYFM5VsxLOe/15k4ZyaWoM+aUN/POTYxzOrWbj\n7kLGJwRf7EcTuSDaU0hEREREREQGLavVQmpSOA8sS+V3983hrmvGMjI2iKHhfuSV1rN2cw4fbM8n\nKsSHO68ei9Vi6dZ1bVYriTGB3L4oGU+7laffOkBeifYZkv5FTSERERERERERwGKxMCslmh/eOoUl\nlyQCsHZzDl52G/ddPx5f7wtfbBMe5MM3F4+hubWD367eR3Flo6vLFukxNYVEREREREREviQ1KQxf\nr84m0B2Lkxka4X+eT5zd9DFRfPuGVBqa2/nNq/uoqG12VZkivaKmkIiIiIiIiMiX2D1s3H1tCnde\nPYbpY3p/etiVs4axbN4Iqutb+c2r+6htbHNBlSK9o6aQiIiIiIiIyBlMGBHG7HExLrveohkJXDUr\ngbLqZh7/2y6yj9e57Nr9SWubg2PHa91dhnSDmkIiIiIiIiIifeT6S4dz7ZxhVNa28MtX9nCsyHzN\nk5c+TOfxv+0mv1Qba/d3agqJiIiIiIiI9BGLxcLSucO574bxdDic/N+bB6iub3V3WS5TVtPM9rRS\nAPZlVri5GjkfNYVERERERERE+tikkRHcPH8kdU3tvLc1193l9EhpdRN1TafujfThjnwMo/PrA9mV\nbqhKLoSaQiIiIiIiIiJuMH/yUMKDvPn3/mJqGno+W6imoZV1O/Jp73D0+BrPvpPGz/++G6fT6Nb7\nnYbBz17axS//sYcOhxMAwzDYnV5GkL8nSbFB5ByvO61pJP2LmkIiIiIiIiIibuBhs3LVrAQ6HE5e\n/TgT48QUG8MwaGxpP2uDprK2hXe25PLYX3fw4Y583t6cw6sbsnht4zHqGttoaum4oDpa2x3sOFJK\nVlEtezLK2ZdVQVv7uRtM9Y1tNLZ0UFzZxEe7CgGobWyjrqmdEUOCmJQUjgFsP1x6QbVI3/JwdwEi\nIiIiIiIig9Wc8TFsPlDMjiNlVNW30tLaQXltC61tDiaMCOM7N07AarEAcCS3ine35pGeV83JdlF5\nTS4ets7XP95dyIY9hQT42PnWtSmMGRbarRqyCmtxnGhAPfPOYTocBgsmx/L1haPO+pmqL+yDtPaz\nHGamRHVtLB0f6c/0MVG8tzWP1zZmERPuy7jEsAv9q5E+oJlCIiIiIiIiIm7iYbNy/w0TiA71Jauw\nlvLaFiKCvIkK8eHAscquWTgNze384Y0DHMmrZmRsECsWJXP51FiaWzuob2pndFwwPl4exEX409jS\nwW9e3cfbm3O6tRwsPb8agEBfOx2OzvdvSys553K0mhNNoaHhfrS2OXhz0zHySxsAiIvyJyzIm/tv\nGI/FYuEvaw5RXNnYq78nuTg0U0hERERERETEjQL9PPnZnTNobGnH38eOxWKhtrGN//f8dt7YdIzJ\no8LZmV5GW4eTZfNGsGhGAtB50tfJptHVc4YxNiEEi8XCsaJanlp7iDWbc8gorGHepFiiw3wJDfDi\n5XUZ2KwWIkJ8iAj2JjLYl0M5VdisFlYun8yBY5VU1DazYU8RezMrmD4m6ow1n5wptHhWAh9uz+ez\nQyXEhPkCEB8ZAMDo+BDuWJTMs++m8X9vHuTH35iKr7faEP2J0hARERERERFxM6vVQoCvZ9efg/w8\nuXl+Es+9e4TVH2eRW1KPp93KpalDut4TGezDlFERFJQ3kBwfjOXEMrMRQ4N49I7pPP9uGvuPVZKW\nW423p42rZw9j6+GSM94/aWgQQ8L9GBLux/GKRjbsKeLVjzMprWpizvgYQgO9T3n/yY2xwwK9Wf7V\nUfziH3sormzCz9uD0ECvrvfNGhdNXmk963YW8Ow7h7n/C8vhxP3UFBIRERERERHph2amRLNuRwG7\nM8oBWDA5Fj9v+ynv+c+l43AaBjbrqbvD+PvYuf/GCexKL2PX0XJ2pZex5tNsLMAPb51CU2sH5TXN\nlFU3U1XXwqUTP282DQn34+rZw1i3M5+3Ps1hzeYcUhJD+frlo4gK7ZwNVFXX2RQKCfAiItiHmWOj\n2JZWSlykf1dz6qRl80ZQVN7A/mOVrPk0m+svHeHqvyrpITWFRERERERERPohq8XCisXJrP00hwlJ\n4cydEHP6e6wWrJx55o3VYmH6mChGxQWzN6OcDofBmIQQkmKDznvv6y8dzqIZ8ew4UsrmA8Ucyq7i\n1Y8zeWBZKvD5TKFg/85ZQcvmJZFTXMeU0ZGnXctmtfKtJeP46Us7eXdLHvGRAUxNPv190ve00bSI\niIiIiIhIPzUsOpAHlqUyb9JQPGw9+xU+2N+LiUnhAMxKie7253y8PLhs4lAe+cZUIoN9yCis7dq4\nuqq+lQBfO3aPzppCArx44luzWDAl9ozX8vexc/8NE/Cy23juvTQKyhp69CziWmoKiYiIiIiIiJjc\nTfOTWHJJIjPGnnnj6PMZFRdMc2sHheUNGIZBTX0rIf5e5//gF8RG+HPn1WNpa3fy1NpD3ToZTS4u\nNYVERERERERETC4i2IcllyR2zey5UCPjOpecZRTU0NzaQWu7g5CAC2sKAUwZHcHcCTEUVzbx1qfZ\n/HrVXnKK605735/fOsiz7xzuUa3SfWoKiYiIiIiIiMg5jY4LBjqbQtX1n28y3ROLZyZgscB7W/M4\nklfNaxuyTnm9qq6FXUfL2Xq4tOteg1WHw8kr6zM4drz2olxfTSEREREREREROaeIYB+C/T05klfN\n/mOVAIQFeZ/nU2cWFerLzBPL2EIDvThaUHNK0+NQTlXX13szy3tR9cB3OKeKj3YX8n9vHqS2sc3l\n11dTSERERERERETOyWKxsGBKLI0tHbyx6RieHlZmjzv9NLTuumPxGB6/awb/cdVYAP6y5hAfbM+n\nsaWdQ9mVXe/bfXRwN4UyCmsAqGts47l303Aart2HSU0hERERERERETmvhdPiCD8xO2jh9LgeLx8D\n8LBZiQnzIzk+mEUz42loaue1jVl870+fsS+rkvAgbxJjAjiaX0NDc7urHmHAySyoxWKBscNCOJxT\nxYfb8116fTWFREREREREROS87B427rx6LHPGR7NoRoJLrmmxWFj2lSR+c+8cbpqXRKCfJx0OJ5NG\nRjB+eBhOw+BY0cXZT6c/q2loJae4jpziOuKjArj72hSC/T3557+zXfr34eGyK4mIiIiIiIiIqY2K\nC2bUiU2nXcnfx86VM+JZOC2O3JJ6hkb4kVHQuXTq2PE6UpPCXX7P/uz5945w+MTeSqNigwn09eTu\na1L49aq9PP32YR67Yxq+3vZe30czhURERERERESkX7BaLQwfEoiX3UZiTCAA2Rfp5K3+LC338822\nkxOCT/xvCFfNHkZFbQsb9xa55D5qComIiIiIiIhIv+PvYyc61Jfs43U4na7dYLk/cxoGVouFiGBv\n7r1uPBO/MEtq4bQ4rBYLezIqXHIvNYVEREREREREpF8aMSSQljYHxZWN7i7lvA4cqyC/tL7X16lv\nbMPhNEiICmDK6AgsFkvXa/4+dkbHB5NTXEd1fWuv76WmkIiIiIiIiIj0SyOGBgHw+qZj/foUsubW\nDv74xkGeWnsY4xzHxtc0tNLS1nHOa1WdaPaEBHif8fXJoyIA2JtZ3sNqP6emkIiIiIiIiIj0SzNT\nohgVF8yBY5X8z4s7KaronzOG8krqcRoGJVVNFJWfucamlg5+/Ox2XvxX+jmvVd3VFPI64+uTRnYu\nJ9uToaaQiIiIiIiIiJiUt6cHK2+ZxDUnNlj++d93c/gLmzD3F7klny8b25ledsb37Mkop6m1g2NF\ndee81smmUGjgmZtCoYHeJMYEcDS/hsaW3s2eUlNIRERERERERPotq9XCdZcO565rxtLe4eB3q/ez\naZ9rTt9yldySzkaPxQK7jp65KbTjSCkAlXUttLY7znqtqvoWAELPsnwMOpeQOZwG+7N6t+G0mkIi\nIiIiIiIi0u/NSonmoa9Nwtfbg799cJSNe4swDIO2czRY+kpeST2+Xh5MGB5GcWXTaZtA1zW1kZZb\n3fXn0qqms17rfMvHACaN7NxXaPWGLL735GcUlDX0qG41hURERERERERkQBgVF8yPvzEFHy8baz/N\n5pl30vjunz6jtqH3J3H1VFNLB6XVzSREB5A4JBCAvC+cQuY0DF76VzpOwyA61BeAknM0harqWrEA\nQf6eZ33PkHA/hoT7Ud/UTnV9K/9YdxTDMGhp66CovIFN+4p49IUdrN9VcM7aPS7gOUVERERERERE\n3CoyxJfLp8TxzpZctqd1Lsn69EAxV88e1ue1VNY289IHnRtHD4sJID4qAID80nomJnVuCL3m02z2\nZlYwJiGEhdPi+MMbBzhaUENVXSsLpsRi9zh1vk51fQuB/p542M49j+e+68dTUdvMxj1F7M2s4L7f\nf0pz66knm328q5Dli8ae9RpqComIiIiIiIjIgLJwehwf7S7EZrXQ3uHkk31FNLd1MDYhlJTE0It2\n37Z2BzvTy0gaGsSOI6W8vz2f1jYHCVEBXD4lrut9eSc2nt6WVsK7W/KIDPbhP5eOo7Wtc6nbxj2d\neyJ52q3Mnxzb9TnDMKiubyUu0v+8tUSH+hId6ktUiC/FlU1YLDBiSCBhQd5EBPtwOKeKI3nV57yG\nmkIiIiIiIiIiMqD4edv58TemYLNZeX9rLv/eX8y/tuWzYU8R/33HNCJDfC/KfTfsKeK1jVldfw4O\n8GL5gpHMmRCD1WLBMAwC/TzJL60np7iOv76fjrenjftvnIC/jx1f71PbMFsPlZzSFKqobaHDYRAR\n7NPtmiKCffj53TNP+76vl8d5m0LaU0hEREREREREBpyYMD8ig31YPDOBcYmhXDZxCK1tDp55J40O\nh/Oi3DOjoAaAMQkhLJoZz9MPL2Bu6hCsFgsAFouF+Ch/KutaeeLl3XR0OLlnSQpDw/0Aut4HEOTn\nybHjdadsOp19vPMUs8SYwF7XOq4bM6bUFBIRERERERGRASsyxJfv3jyR269MZmZKFNnH63j7sxyX\n38dpGGQV1RIe5M33b5nEsq8k4ettP+19CSf2FepwGNxy+UgmjAg/5fV7rxvHVbMSWDZvBAD//Hd2\nVxPLlU2h8GAfhkb4nfM9Wj4mIiIiIiIiIqZw61dHk1VYy3tb8kgZFsro+JBuf/ZQdiUf7MjntitG\nE/WF5WcNze2s/TQHu4eVhuZ2xg8PO+d15k0aSl1jG5dNHMrwIac3d6aMjmTK6Eja2h18tKuQnell\nNLa08+2l48kursVqsZAQHdD9hz6H+68ff87XNVNIREREREREREzB19uDu69NwWKx8Mw7aTQ0t1NR\n08y+zIqzfqbD4WTjnkL+8MYB0nKrWfQhifIAABIPSURBVLfz82Pc92dV8JPntvPxnkI+2JEPwMi4\noHPWEBrozR2Lx5yxIfRFnnYbP1g+mYlJ4aTlVvPEy7vJK2kgNtIPL7vtAp767M63t5KaQiIiIiIi\nIiJiGklDg1hyyTCq61t569Ns/rL2MH988wCl1U2nvdcwDH69ai9/X5eBh4cVXy8Pdh4po6G5nRfe\nP8If3jhAY0s7l0+NxWbt3A9oZGywy2r18rRx3/XjuXxqLEUVjXQ4nAwfcu6mkytp+ZiIiIiIiIiI\nmMriWQlsPljMpr1FGEbn947m15yyLAygvLaFzMJaRgwN5NtLx/P+tjw+3l3Iw09tpam1g7hIf+66\neiyxkf4kxgSSXVTHkDDXnmxmtVpYfvkoIoJ9+Ocn2UxMOvfyNJfeu8/uJCIiIiIiIiLSB2xWK4tn\nJnQ1hACO5p9+PHvmidPEpo+JIiTAi1kp0QA0t3Vw9ewEfnL7VGIj/QGYlRLN1xeOwvKFE8Rc6atT\n43jyu5eetjH1xaSZQiIiIiIiIiJiOrPHxfDhjgJCArwoKGsgPb8GwzBOaepkFnY2hUadWBI2fEgg\n9yxJISrE12WbPV8I60VqOJ2NmkIiIiIiIiIiYjp2Dyv/8x/TsVjgqTWH2Z1RzqZ9xxka7kdCdABe\ndhuZhbV4edqIjfz86PbpY6LcWHXfUlNIREREREREREzJw9a5a05yQgi7M8r5+4dHgc4ZOUPC/Siu\nbCIlMRSbdXDurqOmkIiIiIiIiIiY2twJMV1fl9c0k11cR15JPQDjh/fdxs79jZpCIiIiIiIiImJq\nnnYbC6bEnvK9DoeTqvpWwgO93VSV+6kpJCIiIiIiIiKDjofNSmSwj7vLcKvBuWhORERERERERGSQ\nU1NIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERE\nRGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERE\nRERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNI\nRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQUlNIRERERERERGQQ\nshiGYbi7CBERERERERER6VuaKSQiIiIiIiIiMgipKSQiIiIiIiIiMgipKSQiIiIiIiIiMgipKSQi\nIiIiIiIiMgipKSQiIiIiIiIiMgipKSQiIiIiIiIiMggN2KZQQ0ODu0uQi0C5mlNra6u7SxAX0s+p\n+ShTc1Ku5qQx1Xz0s2oeytKczJ6r7bHHHnvM3UVciPb2dp555hmef/55HA4HISEh+Pv7u7ss6aWO\njg6efPJJnnnmGXx9fYmOjsbT09PdZUkvtbe389RTT/Gvf/0LLy8vQkJC8PT0xDAMLBaLu8uTC9TR\n0cHTTz/N008/TUdHBz4+PoSEhCjPAUxjqjlpTDUnjanmo3HVPDSemtNgGU8H1EyhtrY2Hn/8cRoa\nGrjzzjv56KOPKCwsdHdZ4gKrV68mMzOThx9+mJCQELy9vd1dkrjA//7v/1JRUcGiRYvYvn07q1ev\npr29Xf+hM0A9+eSTVFRU8L3vfY/CwkLWr1+P0+lUngOUxlTz0phqThpTzUfjqjloPDWvwTKeDoim\nUHl5OQDNzc1kZ2fz/e9/n9mzZ+Pv74/VOiAeQc7gZK4AjY2NLFmyhJSUFCIiIigqKnJjZdIbX/x5\nzczM5Lvf/S6zZs1i7ty57Nu3jw0bNgBgGIY7y5RucjqdANTX17N//37uvvtuUlNTiY6OprKyEqvV\nqiwHGI2p5qQx1Zw0ppqPxlXz0HhqToNxPO3Xy8eqq6v5xS9+wapVqygoKCA5OZl58+bh7+/Piy++\nyPr166mqqqK4uJi4uDh8fHzcXbJ0wxdzLSwsJDY2ltzcXHbv3k1jYyN/+MMf2Lt3L7m5uSQmJuLr\n6+vukqUbvvzzOnbsWI4cOcLmzZuZP38+TU1NpKWlUVRUxIQJE5RrP9fa2sqjjz6KxWLp+vc1JiaG\n0aNHA1BbW0tOTg6XXnqp/h/NAUJjqjlpTDUnjanmo3HVPDSemtNgHk/7dQvzpZdewsfHh7/+9a8E\nBQVx//33Exsbi9VqZfbs2axfv56bb76ZiooKPvjgA3eXK930xVwDAwP58Y9/zPLly9mwYQMZGRms\nWrWKe+65h5aWFtavX+/ucqWbvpzrQw89xMMPP0xmZiaPPPIIjzzyCBMnTiQsLIyysjJ3lyvnUVFR\nwbZt20hLS6OgoACA6dOnd73+8ccfM3LkSHeVJz2gMdWcNKaak8ZU89G4ah4aT81pMI+n/bIpdHJa\nZUBAAElJSdjtdlasWIHNZuPll18GID4+HoDJkyfj7++Pl5cXoOmz/dnZcm1oaGDz5s3cd999bN68\nGYBx48YRHBzctZGXcu2/zpTrHXfcQUNDA+vWrePVV19lxYoV/OUvf+Gmm27i0KFDtLe3u7lqOZ+8\nvDyuuOIKjh8/zsGDB2lpaQE6N1JsaWkhPz+fyy+/HICdO3eadjqtGWhMNSeNqeakMdW8NK4OfBpP\nzUnjaT9tCp1cg9nW1kZDQwNNTU0APPTQQ7zwwgs0NTXx9ttv88knn1BSUsL27du7pllqumX/dbZc\nf/CDH/CrX/2K5cuXExISwjvvvMOBAwfYsmWLch0AzpbrypUr+f3vf09bWxvBwcEUFhaSnp6Oh4cH\ndrvdnSVLN0yYMIGVK1cyf/58du/eTU5ODgB2ux273c6oUaM4dOgQ3//+93njjTe0dr4f05hqThpT\nzUljqnlpXB34NJ6ak8bTfrCnUF1dHcXFxQQHB3d97+QxjL6+vqxevZrU1FRCQkIYOnQoe/bsobGx\nkYSEBP75z3/y2muvsWzZMpYuXerGp5Avu9Bct2/fTlNTE9/85jcpLCzk73//O9dee61y7Wd68vNa\nXV1NTEwMa9eu5aWXXmLJkiVcdtllbnwKOelMeZ5ks9mwWq0kJiayefNmGhoaSEpKwtPTk5ycHH76\n059SUlLClVdeyb333ktAQIAbnkC+rLa2lqeeeorGxkaCgoLw9fWlra0Nm82mMXUAu9BcNaYODD35\nedWY2r+dKVOHw4HVatW4OsDU1dXxzDPP0NLSQmBgID4+PrS3t2s8HeAuNNfBMp66tSnkcDi47bbb\nyM3NJTk5mcDAQODzaViRkZFkZmZy7Ngxhg4dSlBQEK2trQBcccUVLFiwgGXLlpGcnOyuR5Az6Emu\nbW1tOJ1OZsyYQXJyMtdcc41y7Wd6+vNqtVqZM2cOs2bN4qabbiIlJcWdjyEnnC3Pk0fhnjz9xGq1\n4uPjw65duwgLC+OTTz5h9uzZxMTE8OCDDzJmzBg3P4mctGfPHh599FGio6MpLCxk1apVLFmypOv1\nqKgojakDUE9y1Zja//X051Vjav91tkxPjqUaVweODRs28PjjjxMaGkpBQQEbN25kwYIFQOfMEP2O\nOjD1JNfBMp66dV7i8ePH8fT0xGazkZaWRltbW2dRVisWi4UjR44wdepUWlpaeP7553nrrbd45ZVX\nCA8PB8DDw0NTK/uh3uYK5pmKZya9zfXkfxBJ/3CuPAF27drFgQMHgM6NMMvKyli5ciXbtm2joaGB\na6+9VksW+pnCwkJGjhzJgw8+yMqVK2lpaeHo0aNd/++0xtSBqbe5gsbU/qi3uWpM7X/OlSloXB1I\niouLWbp0KStXruTyyy9n+PDhQOcsav2OOnD1Nlcw73japzOF8vPz2bRpU1d3zeFwMHfuXCwWC3v2\n7CEhIYHQ0FCcTic/+9nPePPNN7n11luZPn06drud7du3c9tttzF37ty+Klm6Qbmak3I1lwvN8/33\n32fhwoUEBgaydu1a9u/fzw9+8APuuusuvL293fw0AqdnWlJSwpQpU4iMjKS0tJRDhw5xzTXX4OHh\noZ/RAUS5mpNyNZ8LzVTjav/15SxzcnKYPXs2DoeDBx54ALvdTklJCRMmTOCJJ55g9erV3Hbbbfr5\n7OeU6wUwLjKn09n19f3332/ccMMNxpYtWwzDMIyWlhbDMAyjqanJeOSRR4zXXnvNqKqqMgzDMI4e\nPXqxS5NeUK7mpFzNxVV5FhUV9VHFcj5nyvSzzz4zDMMwHA5H12tZWVnGHXfcYdTW1nb9Wfov5WpO\nytV8XJWpxlX3O1OWmzdvPuW1/Px84/XXXzeys7ON5cuXG3/961+NpqYmt9Qr3aNce+aiz2s7eUxm\nTk4OHh4eLF26lDVr1mAYBl5eXjgcDnx8fJg/fz579+6ltrYWoGs6l47Z7J+UqzkpV3PpbZ4nl5QN\nGTLEPQ8gpzlTpmvXru3ap6KjowOA/fv3Ex8f37VXVGxs7Cmfl/5FuZqTcjWf3maqcbX/OFOWb7/9\ndtcBKgBxcXHceOONJCYm8t///d+8+eabXcvCHA6H22qXs1OuPXPRlo9t27aNX/7yl+zbtw8/Pz9S\nUlIYPXo0SUlJ7N27l6qqKsaOHYvT6ezaiT8tLY1PPvmEZ599Fi8vL5KTk7vW4Ur/oFzNSbmai/I0\nn+5mCp3r3Tds2MBXv/pV6uvr+c53voPVamXs2LHKtJ9RruakXM1HmZpHd7Ps6OggJyeH6upqQkND\nOXjwIIZhMG/ePADtGdTPKNfeuShNobKyMh599FFuv/12QkND+eijj6iurmb27NldG2+tW7eOyZMn\nd3XQOzo6eO655ygsLOTee+/lyiuvdHVZ0kvK1ZyUq7koT/O5kExPHmH84Ycf8tRTT5GZmcmKFStY\nvHixm59Cvky5mpNyNR9lah4X+t9I27Zt4+233+aVV15h7969XHfddcTHx7v7MeRLlGvvuawp5HA4\nePLJJ8nMzCQ7O5v4+Hiuv/56EhISCA4O5oUXXmD+/PkEBgbi5eVFQUEBpaWlpKamkpubS3p6OkOG\nDOGxxx5j2LBhrihJXEC5mpNyNRflaT49ybSkpITU1FSys7MpKipi3rx5rFy5Upn2I8rVnJSr+ShT\n8+hJlsXFxUycOBGbzcaiRYuIioriO9/5zqBvHPQnytW1XNIUKi0t5Uc/+hGenp5ERkby2GOPUVFR\nwdKlS/H29iY6OprMzEz279/PnDlzCAwMJCgoiN/97nf84x//YMSIEVx66aWkpKS44JHEVZSrOSlX\nc1Ge5tPTTH//+9/z8ssvM2bMGK6//nrGjx/v7keRL1Cu5qRczUeZmkdvsnzllVeIj49n/PjxJCQk\nuPtR5AuUq+u5pClUWFjI+vXr+d3vfkdKSgp5eXns2rWLyspK5s2bh2EYhIWFsXXrViZMmEBTUxM/\n+clPiI6O5oc//CFz5swZtOv3+jPlak7K1VyUp/n0NtNZs2Zp34p+SLmak3I1H2VqHr3NclAcRT4A\nKVfXc8lvAmFhYXz729/G6XTS0dFBfHw8zz77LJs2beLQoUPYbDb8/f3x9vYmLCwMu93O7bffzpNP\nPqkuej+mXM1JuZqL8jQfZWpOytWclKv5KFPz6E2W48aNc3f5chbK1fVcMlPIz8+PuLg4LBYLTqeT\nP/3pT6xYsQJ/f39WrVpFZGQku3bt4tixY11r+0aMGOGC8uViUq7mpFzNRXmajzI1J+VqTsrVfJSp\neShLc1Kurufh6gtmZGQAEBQUxK233oqPjw/btm2jvLycxx57DD8/P1ffUvqAcjUn5WouytN8lKk5\nKVdzUq7mo0zNQ1mak3J1DZc3hUpLS7nqqqu6joabMGEC//Vf/4XFYnH1raQPKVdzUq7mojzNR5ma\nk3I1J+VqPsrUPJSlOSlX13B5U6impoaf//znfPTRR1x33XVcc801rr6FuIFyNSflai7K03yUqTkp\nV3NSruajTM1DWZqTcnUNi2EYhisvuGPHDtLS0li+fDmenp6uvLS4kXI1J+VqLsrTfJSpOSlXc1Ku\n5qNMzUNZmpNydQ2XN4UMw9B0LRNSruakXM1FeZqPMjUn5WpOytV8lKl5KEtzUq6u4fKmkIiIiIiI\niIiI9H9WdxcgIiIiIiIiIiJ9T00hEREREREREZFBSE0hEREREREREZFBSE0hEREREREREZFBSE0h\nEREREREREZFBSE0hEREREREREZFBSE0hEREREREREZFB6P8DhLT6+jPCo1EAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f86a1e07e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Res = (Result[-test_size:][['return','nn_ret','rnn_ret']]).cumsum()\n",
    "Res[0] =0\n",
    "Res.plot(figsize=(20,10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f86945dd898>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIUAAAIpCAYAAAA8QogCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAPghmkIAAAAAAAB+iKYQAAAAAACAH/r/f/XJlPZNG/oAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f86a1c6c358>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Res.columns =['Market Baseline','Simple Neural Newtwork','Attention RNN','Do Nothing(0)']\n",
    "Res.plot(figsize=(20,10),title=\"Performance of MarketVectors algo over 27 months compared with baselines\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f86945372b0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86944af588>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f86944ca518>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86921b1c18>]], dtype=object)"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAFhCAYAAACcQbj0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2BAFDwCADVHwAADY0P8DUzL2SfMOBEQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f86a1922160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Labeled.hist(column='return',by='class')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1, -1,  0])"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Result['class'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACUoAAAJ8CAYAAAAiIQltAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f869204ef28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "g = sns.FacetGrid(Result, row=\"class\", col=\"rnn_pred\", margin_titles=True)\n",
    "g.map(sns.distplot, \"return\",);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f8691e5e710>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3580fd0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3d0ae48>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3dba1d0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a337dba8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a337d438>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3fa7f98>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a366b978>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a326ecf8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a4172630>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a41f8550>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a41c4cc0>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3da80f0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3b3cc18>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a430a748>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a433b470>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3b179b0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a37fc2b0>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f86a44e9240>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3443400>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a37468d0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3875d30>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f874ba7f780>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3c98710>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f874bb15da0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3150080>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3488ba8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f874bb9d320>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f8763bea748>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f8769e507b8>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f86a41a3e10>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a424d3c8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a4133780>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a335e390>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a3def128>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f86a45329e8>]], dtype=object)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfAAAAFmCAYAAACSk8i4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f86a35b4e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Result.hist(by=['class','nn_pred'],column='return',sharex=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Result.index.dayofweek"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}