examples/Xgboost_XGBClassifier.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# A sample example to tune xgboost classifier showing a complex search space definition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mango.tuner import Tuner"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import uniform\n",
    "\n",
    "param_dict = {\"learning_rate\": uniform(0, 1),\n",
    "              \"gamma\": uniform(0, 5),\n",
    "              \"max_depth\": range(1,10),\n",
    "              \"n_estimators\": range(1,300),\n",
    "               \"booster\":['gbtree','gblinear','dart']\n",
    "             }"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# userObjective"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.datasets import load_wine\n",
    "\n",
    "X, Y = load_wine(return_X_y=True)\n",
    "\n",
    "count_called = 1\n",
    "\n",
    "def objective_Xgboost(args_list):\n",
    "    global X, Y, count_called\n",
    "    \n",
    "    print('count_called:',count_called)\n",
    "    \n",
    "    count_called = count_called + 1\n",
    "    results = []\n",
    "    for hyper_par in args_list:\n",
    "        clf = XGBClassifier(**hyper_par)\n",
    "        result = cross_val_score(clf, X, Y, scoring='accuracy').mean()\n",
    "        results.append(result)\n",
    "    return results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "conf_Dict = dict()\n",
    "conf_Dict['batch_size'] = 1\n",
    "conf_Dict['num_iteration'] = 40\n",
    "conf_Dict['domain_size'] = 10000"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Defining Tuner"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "tuner_user = Tuner(param_dict, objective_Xgboost,conf_Dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'param_dict': {'learning_rate': <scipy.stats._distn_infrastructure.rv_frozen at 0x1a163d1898>,\n",
       "  'gamma': <scipy.stats._distn_infrastructure.rv_frozen at 0x1a163d1dd8>,\n",
       "  'max_depth': range(1, 10),\n",
       "  'n_estimators': range(1, 300),\n",
       "  'booster': ['gbtree', 'gblinear', 'dart']},\n",
       " 'userObjective': <function __main__.objective_Xgboost(args_list)>,\n",
       " 'domain_size': 10000,\n",
       " 'initial_random': 1,\n",
       " 'num_iteration': 40,\n",
       " 'objective': 'maximize',\n",
       " 'batch_size': 1}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tuner_user.Config()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count_called: 1\n",
      "count_called: 2\n",
      "count_called: 3\n",
      "count_called: 4\n",
      "count_called: 5\n",
      "count_called: 6\n",
      "count_called: 7\n",
      "count_called: 8\n",
      "count_called: 9\n",
      "count_called: 10\n",
      "count_called: 11\n",
      "count_called: 12\n",
      "count_called: 13\n",
      "count_called: 14\n",
      "count_called: 15\n",
      "count_called: 16\n",
      "count_called: 17\n",
      "count_called: 18\n",
      "count_called: 19\n",
      "count_called: 20\n",
      "count_called: 21\n",
      "count_called: 22\n",
      "count_called: 23\n",
      "count_called: 24\n",
      "count_called: 25\n",
      "count_called: 26\n",
      "count_called: 27\n",
      "count_called: 28\n",
      "count_called: 29\n",
      "count_called: 30\n",
      "count_called: 31\n",
      "count_called: 32\n",
      "count_called: 33\n",
      "count_called: 34\n",
      "count_called: 35\n",
      "count_called: 36\n",
      "count_called: 37\n",
      "count_called: 38\n",
      "count_called: 39\n",
      "count_called: 40\n",
      "count_called: 41\n",
      "53.937219\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "start_time = time.clock()\n",
    "\n",
    "results = tuner_user.maximize()\n",
    "\n",
    "end_time = time.clock()\n",
    "\n",
    "\n",
    "print(end_time - start_time)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Inspect the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best hyper parameters: {'booster': 'gbtree', 'gamma': 0.011667904935745432, 'learning_rate': 0.6859901826452589, 'max_depth': 1, 'n_estimators': 201}\n",
      "best objective: 0.9220306513409962\n"
     ]
    }
   ],
   "source": [
    "print('best hyper parameters:',results['best_params'])\n",
    "print('best objective:',results['best_objective'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sample hyper parameters tried: 41\n",
      "[{'booster': 'dart', 'gamma': 0.3873414168569206, 'learning_rate': 0.19560021429563113, 'max_depth': 6, 'n_estimators': 188}, {'booster': 'gblinear', 'gamma': 0.7199956896001947, 'learning_rate': 0.7567770987013972, 'max_depth': 7, 'n_estimators': 22}]\n"
     ]
    }
   ],
   "source": [
    "print('Sample hyper parameters tried:',len(results['params_tried']))\n",
    "print(results['params_tried'][:2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sample objective values 41\n",
      "[0.9164750957854406, 0.8942528735632185, 0.8549808429118775, 0.899808429118774, 0.9053639846743295]\n"
     ]
    }
   ],
   "source": [
    "print('Sample objective values',len(results['objective_values']))\n",
    "print(results['objective_values'][:5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plotting the actual variation in objective values of the tried results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABugAAAFpCAYAAAB+h67yAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdd3iUVdrH8e9JQugBQodAQgIiHUkI0kFRsVfs0hHb2taylrWs+9pF17KKgjQrsvZekF5DlyIlhNBb6CH9vH88E5yaBEjP73NdXmPOc+Z5ziTDZDL3ue/bWGsRERERERERERERERERkeIRVNILEBEREREREREREREREalIFKATERERERERERERERERKUYK0ImIiIiIiIiIiIiIiIgUIwXoRERERERERERERERERIqRAnQiIiIiIiIiIiIiIiIixUgBOhEREREREREREREREZFipACdiIiIiIjIKTLGzDHGZBXxNUKMMdYY82tRXqc4GGMGGmPmG2MOuh7TtCK+3geu60ScxH22GWM2FuW6ToYxpqXrMYwr6bWIiIiIiEjhUYBORERERETKDGPMR65gxe0FmPuLa+4VxbG201HagkJFwRgTA3wJRALjgaeBqQW8rzHGDDLGfG2M2WmMSTfG7DPGzDbG3GuMqVqESy9S5SkAKyIiIiIiBRdS0gsQERERERE5Ce8CNwCjgLcDTTLGRAHnAjuBb4twPTcCRRocstZmGWPaAMeK8jrF4DygMnCvtbZAgTkAY0wd4DOcn+dB4DsgGQgHBgKvAn8zxlxirV1bCOvsC9hCOE9h2QK0wXnsIiIiIiJSTihAJyIiIiIiZYa1doYxZj1wljGmi7V2aYCpIwADTLDWFlkJSmttclGd2+s664rjOkWsiet2R0HvYIwJBv4H9Ae+B2621h5wO14J+DfwEPCz6zmx93QWaa3ddDr3L2zW2kygPPz8RURERETEjUpcioiIiIhIWfOe63aUv4OuoM4wnCyocW7jTY0xTxpj5hljdhljMowx240xHxpjzvRznhO9v4wxrY0xnxlj9hpjcowxvVxzfHrQGWMqG2P+Zoz5wRizxVWOMcVVcvMCr7kDjDEWaArEuK6X+98415yAJRCNMbWNMS8YY9YbY9Jc1/nRGHOOn7kDXOd53Bhzlmt9h4wxx4wxM4wx3fL+tvv9Xl/vKjN52Bhz3Biz0hjzsDEm1M9j/KdraLbbY+yVzyVuwQnObQCucQ/OgRO8stY+DEwDIoB/BThPkDHmQWPMn67v01ZjzCvGmJp+HlPAcqPGmJtc36uDrvOsMcY86v54vea3NcZMcHse7DHGzDLG3Oo6PhLIdE0/1+vn/7hrjk8POmPMb66xdgGue7Pr+HNe43Vdz5d1rp/XQdfzckCA75uIiIiIiBQRBehERERERKSsmQRkADcaY6r5OX4hTsDrV2vtZrfx/jiZVik4WVmvAYuAa4FFxpj2Aa53hmteBPABToDwSB7rq+86dw3gF2AM8DUQC/xgjBnqNjcRpxfbEeCA6/9z//s6j2tgjAkH5rse0wHXNb8AegK/uoI//sQD83AqqryHk5nWB5hujGmV1zW9rv8i8DHQGuf78qbrnM+7HmduxZbcxzjb9fUEt8eYXwZibhD2JWvt8TzmPeO6HRIgWPYG8AjwO/AfnO/X/Tjfp8r5rAEAY8wknMfZAicg+BZwCPg/4HtXYNh9/mXAEmAwsArnefA5UAl40DVtqdvaN+P585+Vx3Imum4HBzieOz7ZbT0tXOt5CNiNUyL2M6A9TvbhsDyuJyIiIiIihcxYW5pK64uIiIiIiOTPGPMpTmBtmLV2otexr4DLgEHW2mlu4w2BY9bao17zzwLmANOttZe6jbfEydwCeMZa+4SfdcwBzrbWhriNVQHqWmu3e82tjRNQqwdEWGvT3Y5tA9KstS39XCMEJ8vqN2vtALfx8cBw4G1r7R1u42cCi3ECQa2stVtd4wNwAoYAt1hrP3C7z504AbY3rLV3e6/Bz5p64wSQtgDx1to9bmv9GidI+rC19kW3+/wbeAzoba2dU4BrhOL03QsBor2Crf7m7wYaAN2ttQtcYx8ANwF7gVi370Vu6czLgUettc+5ncfnZ+EKdr6HE9AabK1Nczv2DPA4cJe19i3XWENgIxAKnOv9eI0xEdbabW7fM5+fr9vc3OfheGvtSNdYNWAXTmC3ubU2221+U5zAZ4K1tpvb+BygB3CdtfYzt/E6OD/LaCDqdEuEioiIiIhIwSiDTkREREREyqJ3XbceWWLGmMbARTgZQl+5H7PW7vYOzrnGlwEzcUoMBnsfx+mZ9u+CLsxam+YdnHONH8TJHquHk013ylxZXzcCh4FHva6zDifYVhmnRKS3me7BOZdxQA5Odl1BDHfd/is3OOe6dhbwd5zyooEy+AqqHn/1Td9agPm5c5r4OfZqbnAOwBXQehBnncP9zPd2D07W5kj34JzL08BBnEBgrqE4GZRv+gtG5gbnTpW1NhUni68J4B3UuwXnb/1JuQPGmFiczMpP3YNzrnMdAJ4CqgFXns66RERERESk4ELynyIiIiIiIlLqTAc2AT2NMW2stWtd48Nw/s6ZaK3N9L6Tq+zgaJwAWV18/yYKx8m2crfcWptxMoszxnTACQD1wgmieJdRbHoy5/OjLVAFWOgK/HmbDvwDOMvPsQTvAWttujFmL1CngNfv4nYd73OtNcbsBFoZY2r4C4oWkDnF+f7KxMz0HrDWbjDG7ABaGmNqWmv9li119alrjxP0vd8Yv8tKA9q4fX226/aHAq79VEzEeb4PAX5yGx8MpAOfuI11d93WMcY85edcDV23bfwcExERERGRIqAAnYiIiIiIlDnWWmuMGQc8h5Op9XfjRE6G4wRoxnnfxxhzP/AKTg+6X3HKMx53zb8K6IBvIA2cUoIFZozp6Tp/EPAbTibfEZwMtS7ApQGuczJquW53BjieO17bzzF/AT2ALMBfBuGpXr+Ja96pBuj2utYUAjTD6dGWl4g81rQ7wH124QRLwwjcVzDcddsQeDKP62e5/X/u990nk7IQzcbp73eFMSbMWnvYGBOPE2SbZq1NcZtb13V7geu/QGoUzVJFRERERMSbSlyKiIiIiEhZNQGnd9dgV7+yc4AY4Hdr7Ub3icaYSjhl/HYAba2111lrH7LWPmmtfQrfrDl3J9u4+5842W3nWmsvstbeZ619wnWdxSd5rkAOuW4bBTje2GteYSvy67uyFnO/Xz692dy5MhYb4ARcl/mZ0tDPGPy1/sN5nD73MSy21po8/qvkdp/cIOjpZkoGZJ2G8pOBqjj9GMHJpgO38pYuuY/hznwew6iiWq+IiIiIiHhSgE5ERERERMoka+1u4GucXmVX8FfPs3f9TG8I1ATmuO53gjEmDP+lIE9VS2CPv95jQN8A98mm4NlrAGtwyiqe5Vq/t/6u26Uncc6TkRsE6+d9wBjTGidAt+E0ylvmys2E/Lsxpkoe8x533U6y1qb7Oe7zfTfGtMLJ8tsYqLwlnOgd+CfQwRjjLyPRnwWu2wsLMDfHdXsyP/9ck3ACyENcQerrcbIFfwywnt6ncA0RERERESkCCtCJiIiIiEhZ9p7r9u/AlcA+4As/83biBLS6GmOq5w66ghpvUPDeawWRBNQ3xrRzHzTGjAbODXCf/UADY0yBSl+6glAf45SQ/JfXdVoBdwEZwAcntfKCe991+09jTG75RIwxIThlRA0wvhCuMxmYBbQGpnoHyIwxIcaYZ3EyyLYTuATlfcaYZm73CwZecq1zQgHWMQYnK3K8MaaW90FjTLgxxj3IOwGntOddrpKn3vNzy3Firc0BDgDNC7AOD9baJJz+er2Ae3DKcX5orc3ymrcAmA9ca4wZ4n0e15o6GWPqnewaRERERETk1KgHnYiIiIiIlGU/4/Qmi3d9/aarNKIHa222MeZN4AFglTHma5w+cOfgBLlmEji77WS9ihOIm2eMmYpTPjEe6A78D7jaz31+w8ni+9EYMxsnuLbMWvtdHtd5EFdgxtV7bCZQHydYVQO43VqbXDgPyZO1dpYxZgxwP7DaGDMNSAUuBtq61jKmEK6TZYy5Cuf7dimQaIz5DkjGCUYNBKJwerFdaq3dE+BU84EVrp/HIZzMtg44JTRfKcA63jXGxAK3An2NMT+7rSEaJzPtPZzAKNbaPcaYm4CpwCxjzPfAKpznWiecjM5Wbpf4DbjGGPMVTnZiFjAjQBamt0k4mYz/5/a1P9e7rjPRGHMvsAinFGcE0Bnn59YVJ8gtIiIiIiJFTAE6EREREREps6y11hgzHvi3a+i9PKY/AuwBhgOjcYITvwCPAc8V4pq+M8Zc7jrv9TjBlkU4QZQz8R+gexoIAy7BCfYE42SgBQzQWWv3G2O6AY/iZA/ejxMkmw+8ZK39tZAeUqDr/90YsxS4A6f3WSVgo2s9Y6y1mYV0nf3GmHNwAo83A+fhBMaO4pT6fB0Ya61NzeM0fwOuwymDGoXTc/BV4MkAJTH9rWO0Kzg42rWG2jiZj1uAF/HKVrTWfm2MiQP+gRMIvgAnU24t8IKf9WXhBHYvwal280+gIAG6acCbQHVgubV2ZYD1J7uCjHcDVwE3ua6zC+f7+JrrVkREREREioFx+kqLiIiIiIiIiDFmH7DbWtsu38kiIiIiIiKnSD3oRERERERERABXD7Y6wLaSXouIiIiIiJRvKnEpIiIiIiIiFZoxpg5OidALcTayTivZFYmIiIiISHmnEpciIiIiIiJSoRljWuL0htuM08fwFWttTsmuSkREREREyjMF6ERERERERERERERERESKkXrQiYiIiIiIiIiIiIiIiBQj9aArYvXq1bNRUVElvQwREREREREREREREREpZkuWLNlnra3vPa4AXRGLiooiISGhpJchIiIiIiIiIiIiIiIixcwYs8XfuEpcioiIiIiIiIiIiIiIiBQjBehEREREREREREREREREipECdCIiIiIiIiIiIiIiIiLFSAE6ERERERERERERERERkWKkAJ2IiIiIiIiIiIiIiIhIMVKATkRERERERERERERERKQYKUAnIiIiIiIiIiIiIiIiUoxKfYDOGBNhjHnfGLPDGJNujEkyxrxmjKlzEucwxpjhxpgFxpgjxphUY8wyY8zdxphgP/NbGWMeNsZMN8ZsNcZkGGN2G2O+Msb0L9xHKCIiIiIiIiIiIiIiIhVJSEkvIC/GmBhgHtAA+ApYB8QD9wADjTE9rbX7C3CqScAtwB7gU+AYMAD4D9DHGDPIWmvd5j8DXAesAb4HUoDWwGXAZcaYe6y1rxfCQxQRKXX2HU3ngwVbSMvM4ZKOjWnftFZJL0lERERERERERESkXDGecanSxRjzE3A+cLe19g238THAfcBYa+1t+ZzjCuALYDMQb63d5xqvBEwFrgCGWWsnut1nKLDCWrvM61x9gV8AC0RZa3fm9xji4uJsQkJC/g9WRKQU+Hn1Lh75fBX7j2WcGOvdqh6j+8TQs2VdjDEluDoRERERERERERGRssUYs8RaG+czXloDdMaYaGATkATEWGtz3I7VBHYCBmhgrT2Wx3km42TP3WWtfcvrWHtgFbDUWhtbwHX9DJwHXGOt/V9+8xWgE5Gy4EhaJk9/s4ZpS7YFnNOuSRi39onm4g6NCQku9RWSRUREREREREREREpcoABdaf6E9RzX7c/uwTkAa+0RYC5QDTg7n/M0ct0m+jmWO9bFGFO7gOvKdN1mFXC+iEipNn/Tfga+NjvP4BzA6h2HueeT5fR7eQYT524mNUMvgyIiIiIiIiIiIiKnojQH6Fq7btcHOL7BdXtGPufZ57pt4edYtNv/n5nfgowxkcC5QCowK7/5IiKlWVpmNv/+dg03jlvA9oPHPY6FBBnaNw3ze79tB47z1Ddr6PH8dMb8/Cf7jqYXx3JFREREREREREREyo3SHKCr5bo9FOB47nh+mW/fum7vN8aE5w4aY0KAp93m1cnrJMaYysCHQGXgKWvtgXyuKyJSav2x/RCXvjGHcXM2413puFWDGnx5Z0++/VtvPr+jBwPbNcJf67mDqZm8Pn0jPZ+fzmNfrCJpX8BqwyIiIiIiIiIiIiLiJqSkF3Aacj8uzq+J3ifAzcCFwBpjzNc4GXADgBicTLxWQHbACxkTDEwBegKfAi/nuTBjbgVuBWjevHl+j0NEpNhkZefw9oxN/Oe3DWTleL58GgMjerbggQtaU6VSMABdmtfhnVtiSdx7lPdmb+Z/S7eRkeVRdZj0rBw+XJjMR4uSubB9I0b3iaFTs4JWDRYRERERERERERGpeEpzBl1uhlytAMfDvOb55epfdxnwALALuAUYDmwDegH7XVP3+Lu/Kzj3ATAImArcbK13vonPNd+11sZZa+Pq16+f11QRkWKzed8xBo2dzyu/rPcJzjWtXZWPRp7N45e0PRGccxddvwbPXdWBuQ+fw139WxJWxXd/h7Xw/apdXP7WXK5/dz6/r9tDPi+XIiIiIiIiIiIiIhVSac6g+9N1G6jHXCvXbaAedSdYa7OAV1z/nWCMqQp0Bo4Dq73v5yqD+RFOcO4jYLC1NmCmnZyeCXM3c3GHxjQIq1LSSxEpV6y1fLBgC89+v47jmb4vYdfERvDEpW0Jq1Ip33PVr1mZBy5ozW39Yvh08Vben7PZp38dwILEFBYkptC6YU1u7RPNpZ2aEBpSmveEiIiIiIiIiIiIiBQfU1qzG4wxMcBGIAmIcWXC5R6rCezEyQCsb609pcZHrlKUY4FJ1tqhXsdCcTLmLgcmA8Pc11BQcXFxNiEh4VSWV6GMnbmJ535YR7PwqnwwohuRdauX9JJEyoVdh9J46H8rmbV+r8+x8OqhPHtlBwa2b3TK58/MzuG7lTt5Z+Ym1u06EnBeo7AqjOjVguvjm1GzAIFAEREREREREZHS6lBqJs98t4YVWw9yWacm3NG/JcFBJv87ikiFZIxZYq2N8xkvrQE6AGPMT8D5wN3W2jfcxscA9wFjrbW3ucYq4fSUy7TWbvI6T5i19rDXWFfgJ6AS0Mlam+h2rDLwOXARMB649VSCc6AAXUFMTdjKQ9NWnvi6Xo3KTB4eT9smYXncS0Ty8/WKHfzzyz84dDzT59iANg147qqO1K9ZuVCuZa1l1oZ9jJ25iXmb9gecV7NKCDefHcmwHlHKlhURERERERGRMmfP4TQGv7/IY6PyJR0bM+bazqoeJCJ+ldUAXQwwD2gAfAWsBboB/XFKW/aw1u53zY0CNgNbrLVRXudZiFPG8g/gCNAOJ/iWDlxlrf3Ja/4EYCiwD/gv4O+bNMNaOyO/x6AAXd6stYyesoSf1+z2GK9ZJYTxQ7oS3yK8hFYmUnYdTM3gn1+t5psVO3yOVQ8N5slL2zEoLgJjimZn16pthxg7axPfr9pJToBfMaHBQVx5VlNG9YmmZYMaRbIOEREREREREZHClLw/lZvHLyQ5JdXnWO9W9Rh7SyzVQktzVykRKQllMkAHYIxpBvwLGAjUxSlt+SXwtLU2xW1eFIEDdA8C1+Nk2FUFduBkzz1vrU3yc80ZQN98lva0tfap/NavAF3+MrNzeHjaSj5ftt1jvHJIEG/d2IUBbRuW0MpEyp6Z6/fy0LQV7D6c7nMsvkU4rwzqRLPwasWyluT9qYybk8jUhK2kZQZOQj6vbUNG94kmLkoBeREREREREREpnf7cdYRbxi9kzxHfz1xydW5WmwlDu1KnemgxrkxESrsyG6Ar6xSgK5icHMu/v1vL+3M3e4wHBxlevLojV8dGlNDKRMqG1Iwsnvt+HVMWbPE5FhocxIMXtGZErxYElUA99JRjGUyen8SkeUkcSPUtt5krNrIOo/tEM6BNwxJZp4iIiIiIiIiIP0uTDzBswmKfNiKhIUFkZHluSm7ZoAZTRsTTuFbV4lyiiJRiCtCVEAXoCs5ay39nbOKln/70Ofb4xW0Y2Tu6BFYlUvotTT7A36euYPO+Yz7H2jYO49XrOtO6Uc0SWJmn4xnZfLZkK+/NTmRryvGA86LrV+fW3tFccVZTqlQKLsYVioiIiIiIiIh4mr1hL6OnLCE1I9tj/PLOTXjkwjaMnLyYP7Yf9jjWtHZVpoyIJ7q+2nqIiAJ0JUYBupP30cJkHvtyFd5PzTv6xfDgBa2LrG+WSFmTkZXDG9M38NbvG316vQUZuL1fDPece0apa1CclZ3Dj6t3MXZmIqu2Hwo4r37NygzrGcVN3SKpVbVSMa5QRERERERERAS+X7WTez5ZRma25wcvQ7pH8uSl7QgKMhxJy2TU5AQWJKZ4zAmvHsqkYfF0iKhVnEsWkVJIAboSogDdqfl+1U7u/WQ5GdmeKeI3xDfj31d0IFjl76SCW7/7CPdPXe6zQwsgsm41xlzbidjI0t3TzVrL/MT9jJ2ZyMz1ewPOqx4azA3xzRneqwVNaqs8hIiIiIiIiIgUvU8WJfPoF6t8NkXffW4r7hvQyiOJIC0zm7s/XsbPa3Z7zK0eGsx7Q+LoEVOvOJYsIqWUAnQlRAG6Uzdnwz5unZLgkz5+YftGvHZ9ZyqHqPSdVDw5OZb3527mxZ/+9KlxDnBTt+Y8elEbqlcOKYHVnbo1Ow7z3uxEvlmxgyzvd74uIUGGyzo1YVSfaNo0DivmFYqIiIhIaXMoNZPKlYJUFl2KRFZ2DulZOWXubyspGzKzcziWnkXtaqElvRQJ4J2Zm3j+h3U+409c0pbhvVr4vU9Wdg6PffEHnyZs9RgPDQ7i9Rs6M7B94yJZq4iUfgrQlRAF6E7Piq0HGTphEQdSPRuw9oipy7uD46ihN8pSgWw7kMrfp65g4eYUn2MNalbmhWs60r91gxJYWeHZfvA478/ZzMeLkn2C8+6a1KpCbFQ4cZF1iI2sw5mNahISXLpKeYqIiIhI4bPWMn/TfsbOcqowhIYEcXWXCEb3iSaqXvWSXp6UA5nZObw5fSPvz91Mdo5lVO9o7vXKlBE5Hau2HWLIhEWkHMugS/PajO4bw3ltGhKkalGlgrWWF378k3dmbvIYDw4yvHh1R66OjTil+wcZePbKDlwf37zQ1ywipZ8CdCVEAbrTt3HPUW4Zv5Cdh9I8xjtG1GLC0K7UrVG5hFYmUjystUxbso2nv1nD0fQsn+MXd2zMvy9vT53q5Wfn3aHUTD5YuIUJczez72hGvvOrhwbTuXltYiOdoN1ZzWtTs4r61omIiIiUF/n1MQ4ycGH7xtzWN0a9fuSUbdxzlPunLmflNs/n2DWxETx/VQdtCpTTlp1jGfjaLDbsOeoxHl2vOqP6RHPlWU2VFVyCsnMsj3+5io8XeWXAhQTx1o1dOK9twwKf691Zm3j2e98MvIcHnsltfaMV9BepYBSgKyEK0BWOHQePc8v4hWzae8xjPLp+daaM6EZT9aWScmrf0XQe/XyVTw1zgLAqITxzRXsu69Sk3L6xS8vM5otl23lvViKJ+47lfweXIAOtG4WdyLCLjaxDRJ2q5fb7JCIiIlJeHc/I5rMlWxk3ezPJKakFuk/vVvW4rW8MPWLq6v2fFEhOjmXivCRe+HEd6X5aCQAMaNOQN288S8ETOS1TE7by0LSVAY/Xq1GZYT2juLlbJLWqadNpcUrPyub+T1fw3aqdHuM1Kofw3uA4usfUPelzTk3YyiOfryLbq5XHqN4tePSiNvodJVKBKEBXQhSgKzwpxzIYNmERK7x2sjWuVYUpI+Jp2aBmCa1MpGj8smY3j3y+0m8GWe9W9Xjxmo40rlUxgtPZOZZf1uxm8vwkErYc8Nt/Lz8NwyoTFxlObGQd4qLq0KZxGJW0A1ZERESkVEo5lsHk+UlMmpfk0/KgoDpG1OL2vjGc364RwSodJwHsOHicB6etYO7G/fnO7dYinPeGxBGmah1yCtIys+n/8gyfClH+VA8N5vr45gzv1UKb0ovBsfQsbvtgCbM37PMYD68eyqRh8aeVmf3z6l3c9fEyn88xru4SwQtXKzNXpKJQgK6EKEBXuI6mZ3HblCXM2ej5C7N2tUpMGNqVs5rXKaGViRSeI2mZPPPtGqYmbPM5VqVSEI9e1Iabu0VW2Pr06VnZ/LH9MEu2pJCQdIAlWw6w/1j+ZTC9Va0UTOdmtZ0Mu6g6dGleh1pV9Ye2iIiISElK3p/KuDmJTE3YSlpm4E1Z57VtyKje0ew8dJy3Z2xi3a4jAedG16vO6L7RXHFWUyqHKPtJHNZavli2nSe/Xs2RNN9WAue3bcianYfZduC4x3jbxmFMGh5P/ZpqtyEnZ+zMTTz3w18lD0ODgxjYvhHfr9pJVo7/z2dDggyXdWrCrX2jObNRWHEttUI5mJrBsImLWZZ80GO8Sa0qTB7RjZYNapz2NRYk7mfkpASftiXKzBWpOBSgKyEK0BW+9Kxs7vt0Od+v2uUxXi00mLG3xNK7Vf0SWpnI6VuYuJ+/f7bC549AgE7NajPm2k7E1D/9N4flibWWpP2pJCSlsGTLARK2HGCjVz3/gjAGzmhQk9ioOsRF1iEuMpxm4SqLKSIiIlIcVm07xNhZm/h+1U4CfE5NaHAQV57VlFF9oj0+MLXWMmP9Xt6esYlFm1MCXqNhWGVG9GrBjd0iqVE5pLAfgpQhKccyeOyLVfzwxy6fYzWrhPCvy9txReem7DmSzuDxi/hzt2cAOKpuNaaM6Eaz8GrFtWQp4w6lZtL7xekcdgsGj+zVgscvacuOg8eZMHczHy1M5lhGdsBz9D2jPqP7RtM9WuV7C8vuw2l+/41H16vOlJGF21Lnj+2HGDphkU+VpPgW4YxTZq5IuacAXQlRgK5oZOdY/vnVH3y0MNljvFKw4dXrOnNJxyYltDKRU5OWmc0rP//JuDmb8X5ZDgky3H1uK+7oF6PSBwV0MDWDpckHSEhyAnYrth4M2EsiL/VrVia2uVMSMzayDu2a1CI0RD8DERERkcJgrWXWhn2MnbmJeZsClxesWSWEm8+OZFiPKBqEVcnznEu2pPD2jER+XevbwzlXWJUQBnePYmjPKOrVUBZURTN93W4emraKfUfTfY71iKnLy4M60cTtQ/mDqRkMn7iYpV7ZNQ1qVmbKiG60bqR2G5K/51IrX94AACAASURBVH5Yy9iZiSe+rlk5hFkP9adO9dATY4eOZ/Lhwi1MmJvE3iO+z89cHSNqMbpPDAPbq3zv6diy/xg3j1/I1hTPDdLtm4YxcVh8kfx+2LzvGLeMX6jMXJEKqNgDdMaYOkC2tfZwkVygjFCAruhYaxnzy3remL7RY9wY+Nfl7bnl7MgSWpnIyVm94xD3f7rCZ8cWQEz96rx6XWc6RtQugZWVHxlZOazeccjJsHMF7fz9QZ6fyiFBdGpW28mwc5XFrF0tNP87ioiIiMgJmdk5fLtyB2NnJuZZmrJRWBVG9GrB9fHNqHmSmQUbdh/hnZmJfLV8e8DScZVDgrg2rhm39olWJlQFcCw9i39/t4aPF231OVY5JIh/XHgmQ7pH+W0lkJqRxR0fLmXGn3s9xsOqhDBhWFdiI8OLbN1S9u04eJx+L8/w6EH24AWtubN/S7/z07Oy+XLZdsbOSiRx77GA520eXo1RvVtwTWwzqoaqROLJWLvzMIPfX+QTCO3mymY72d85JyNQ1l5k3Wp8oMxckXKrUAN0xpgmwABgj7X2R69j7YBJwFmuoXnACGvt+pO+UDmgAF3RGz9nM898u8Zn/O/nncFd57RU2r+w/eBxlicf5NDxU2suX5S2Hkhl3OxEMrN9X4uH92zBQwNbqxZ5EbDWkpyS6vSwSz7AkqQDrN9zxCd7sSBaNahBbGQdusfU5cL2jZVhJ4Vu456jrN99hB4xdRUQLsOstWTlWCopE1oqkEPHM5m3cR8xDWpwRkNlmIgTIPlk8VbGz05kx6G0gPPOaFiD0X1iuLRTk9N+b7X94HHGzU7kk0VbOZ7pv3RccJDhko6Nua1vDG0aq8dTebQ4KYW/T11Bckqqz7EOTWvx6nWdaNkg79epzOwcHvhsBV8t3+ExXqVSEG/fFEv/MxsU6prLEmsty7ceJOVYBj1b1tPfsF4emrbCo8d8g5qVmflg/3yDajk5ll/X7mbsrESWbDkQcF549VCGdI9icPdIj4w88W/JlhSGTVjsUW4UYECbBrx5Y5dief7mlZk7eUS8+g0K4LyHWZZ8gM7NahNRR4Hbsq6wA3SPA08DL1lr/+E2XhVYB0QA7lGRbUD7iphNpwBd8fh86TYenLaSbK+dkUN7RPHEJW397oCT8ikrO4d1u46QkJRCwpYDLNlygJ15/PFfGjWpVYWXr+1Ej5h6Jb2UCuXQ8UyWuoJ1CVtSWL71IGmZJ1cWs3t0XT4c2U2vOVJoZq7fy/CJi8nOsUTWrcbnt/egrkpxlTnH0rMYOSmB+Yn76d+6Pm/e2IXq6n0k5VzKsQyufnsem/c5O/9vPrs5j17Uhmqheu5XRHuPpDNx3mY+WJCc56a5s6PDGd0nhn6t6xf6RssDxzKYND+JifOSOJgaeA39Wtfn9r4xxLcI12bPciA9K5sxv6zn3VmJPpvxgoMMd/VvyV3ntCzwBpqcHMvT36xm0vwtHuMhQYZXru3E5Z2bFtbSy4x9R9N55PNV/LLGKSvbPbouU0bEqz2Dy/rdRxj42iyP3prPXdWBG+Kbn9R5EpJSGDsr8cT32Z+qlYK5Ni6Ckb2VFRzIzPV7GT0lwedv/avOasoL13Qs1s10eWXmvj+0K3FRysytqLz78oYEGf52Tivu6B+jDZ9lWGEH6GYAvYE4a+0yt/HbgP8C+4F/AMeB54GmwOPW2udOafVlmAJ0xee3tbu548OlPn2mrujchJcGddILWDl1OC2TZckHWZKUwpLkAyxLPkhqHk2VS7urujTlqcvaqTlwKZCZncOaHYdZ4gr0JmxJYffh/MtifjiyGz1bKrgqpy8jK4cBY2Z67PS+JjaClwd1KsFVyanwzvYf0j2Spy9vX4IrEila1lpGTV7i0wMsqm41Xrm2M7GRdUpoZVLcEvce5b3Zm/nf0m0epd3cGQMXtm/ErX1i6Nys6Mu6p2Zk8cmirYzLJ4uvS/Pa3NY3hgFtGmrzVRm1dudh7vt0ud8yqtH1qjPmus6n9Jyz1vL6bxt59VffQlFPXdqWoT1bnNJ6y6KfV+/ikc9Xsf9Yhsf44xe3YWTv6BJaVekyclKCx+/D6PrV+fnePqccwNy45yjvzUrki2Xbycj2/7oaZODijk0Y3Sea9k1rndJ1yqNvV+7gvk+X+1QwKsnN/crMlVzWWmZv2MfYWZuYu9F/X95OEbUYc11nYurXKObVSWEo7ABdEtAMqGGtPe42/jNwLnCbtfY919gFwA/AQmtt91NbftmlAF3xWpyUwvCJiznilabev3V9/ntTrGpyl3HWWrYdOH4iWJKQdIA/d59aWcLSJrx6KM9e2Z6B7RuX9FIkgII+/y7t1IQ3bjjL/0lETsKkeUk8+fVqn/FPbz2bbtF1S2BFcqru/HAp363aeeJrY+CrO3uqv6iUWx8s2MLjX/7h91iQgdv7xXDPuWeoLHQ5tjT5AO/OTOSnNbsCvlevHBLEoLgIRvaKJqpe9eJdIM5GmK9X7GDszE1s2HM04LyWDWpwW98YLiuEcptSPLJzLO/OSmTML3/6bSUwtEcUDw8887Q/H5gyP4knvl7t8xy/+9xW3DegVbnOwDySlsnT36xh2pJtfo9XDw3m17/3pXGtqsW8stJlcVIKg96Z7zH2zs2xDGzf6LTPvedwGhPmJfHBgi0+n4G569myLqP7xNC7Vb1y/ZzMz0cLk3nsy1U+/17vG3AGd59bsu1x8srMfXlQJ644q+Jl5lYkmdk5fLdyJ2NnJbJ2Z/7FB6tUCuKRC9twy9mR2kBUxhR2gO4okGmtreM2FgQcBqoA9a21B9zGM4DD1toKl5urAF3xC9ToNS6yDuOHdKVWNWUmlRW5GUxOqcoUlmw5UKAMJm+hIUF0iqhFdL0aBJXCv6lb1KvOlWdFUL+mytaVNYfTMvlu5U4e+XzVibHQ4CAWPnquav/LaTmankXfF3/32Y0MTt/D7+7urQ8Jy5ABY2ay0evD3/ZNw/jyjp4q/yTlzobdR7jkjTk+VS28tWsSxqvXdVZvunIkJ8fy+597GDszkUVJKQHn1a5WicFnRzK4RxT1SkHZ5pwcy2/r9vD2jI0+vYDcNalVhRG9o7m+azOVKS7Fkvencv/U5ST46dfVKKwKLw3qSO9W9Qvtet+s2MH9U30zcm4+uzlPX9ae4HL44en8Tft54LMVbD94PM95F7ZvxNs3xxbTqkofay2D3pnv8Vw8q3ltPr+9R6EGg46kZfLJoq2Mn7OZXYcDZwW3aRzGbX2juahD4wpVYcpay9szN/Hij3/6HHv6snYM6RFV/IvyI6/M3CcvbcuwCpSZW1EcS8/i08XOv928Xk9DggxZOb7xm14t6/HSoI4VfiNEWVLYAbrjgLXWVnMb6wQsA1ZZazt5zd8LhFlrT/rdtzEmAvgXMBCoC+wEvgSezg0CFuAcBhgG3Aq0A4KBP4EJwFvWWr/18IwxPYDHgbNxAo8bgfeBNwLdx5sCdCUjeX8qN49f6NMA+sxGNZk0PJ6GYVVKaGWSl0OpmSzd+lcPsBVbDwVs5J6XutVDiY2sQ1xUHWIjw2nfNIzKIcqelKKRk2Pp/8oMtuz/6/XmiUvaMryX3kDLqXv1l/X857cNAY8/PPBMbu8XU4wrklOVnpVN2yd+8umTC/pjW8qf9Kxsrnhrnsfu3+qhwVzYobHfLIvQkCAePL81I3q10A7gMiw9K5uvlu/gvVmJeWaiNa1dlVG9W3Bt12alshehtZbFSQd4e8ZGfvfqB+SudrVKDOkexZAeUYRrQ1apYa3lk8VbeebbNX5bHlzRuQlPX9a+SDbszly/l9umLPH52/WSjo0Zc23ncrOpKi0zm5d/+pPxczf7ZCGFBBl6t6rn829nwtCuFbZE3y9rdjNqsufngUVZCSMjK4dvVuxg7KxNrN+d92vxiF4tuK4CbDaw1vLcD+t4d1aix3hwkOGVUpqZNmXBFp746g/fzNxzWnLfeWdU6CzI8mLvkXQmzUtiyoItefbl7dYinNv6xtAhohaPfbGKn1b79p+sWSWEZy5vz+Wdm+i5UQYUdoAuEYgEWlprN7vGHsLpNzfWWnu71/zjwEFr7UnVbjPGxADzgAbAV8A6IB7ojxNg62mt9V+U1fM8k4FbgD3AN8AxYADQFvgfMMh6fSOMMZe7jqUBnwIpwKVAa2CatXZQQR6DAnQlZ8+RNIa8v9gnPbhZeFWmDO9WImVU5C/WWpJTUklIOnAiQ27DnqOnVK6yVYMaJ4JxcZF1iKxbTb+YpFi99ftGXvrprx15ZzSswU/39tHzUE7J3iPp9H3pd48Pl+rVCGXf0b+y6apUCuKX+/qq+XsZsHbnYS78z2y/x2pUDuHX+/vSqJY2Dkn58My3axg/Z7PH2JhrO3FVlwhmrt/Lg5+tYM8R32oI3VqE8/KgTnpNK2MOp2Xy8cJk3p+7Oc8qF+2ahHFrn2gu7tC4zGQNr9lxmLGzNvHNih342V8BQNVKwVwf34yRvaNpWlu710vSniNp/ON/q5i+bo/PsdrVKvF/V3Tg4o5F20pgafIBhk9czMFUzw9be7eqxzs3x5b5QMgf2w9x36fL/QbhWzWowavXdaZN4zCu/O9cVm47dOJYs/Cq/Hxv3wrXbiQrO4cL/zPb4/t17pkNGD+0a5Ff21rLjD/3MnbWJhYkBs5mrlW1EoO7RzKklGQzF7as7Bwe/WIVUxM8NwhVDgnivzd14dw2DUtoZfkLlJl7U7fm/Ovy8pmZWxFs3neM92YnMm3Jyffltdby+dLtPPX1ao6k+5a0vbhDY/59RXtVcirlCjtA9yFwPTAFGI6T2bYIaA5cba390m1uC2ATsNxa2+Ukr/MTcD5wt7X2DbfxMcB9OMHA2/I5xxXAF8BmIN5au881XgmYClwBDLPWTnS7TxhOtlwtnCBggmu8CjAd6A7cYK39JL/HoABdyTp0PJORkxazOMkz2bJejcpMGt6Vdk3ULLe4ZGTlsHrHIad/lysot+/oyZerrFIpiE4RtU9kyHVpXofa1fQLSErWnsNpdH9+ukeGzOd39KBL8zp53EvEvye++oPJbv0H6lYP5cs7e3Lpm3M8PvQ598wGjBsSp0BwKfflsu3c++nygMcv7tCYt246qbfIIqXSjD/3MHTCYo+xyzo14T/Xdz7xOnUwNYPHv/yDb1fu9Ll/jcohPHFpWwbFRuh1rZTbdSiNCXM38+HCZI76+ZAoV+9W9RjdJ4aeLeuW2Z/p1pRU3pudyKeLtwYs2xoSZLiscxNu6xujkq0l4IdVO3n0i1UcSPXNQujXuj4vXt2RBsVUQWfD7iPcMn6RT5nBzs1qM2Fo1zL5wWlWdg7vzNzEa79u8CmxZgyM6NmCBy5oTZVKTgBu1bZDXP7WHI/A9l39W/LABa2Lc9klburirTz0v5UnvjYGfrynD60bFe9rxPKtB3l31iZ++CNwP9DQkCCuiY1gVO9oWpSTjezpWdnc8/Fyfly9y2O8ZuUQxg2JKxP9vGet38toP5m5F3dszJhrO6lKVBmyLPkAY/PpyxsaEsSg2AhG5vPvcNuBVB78bCXzE33zlerXrMyL13Skf+uKmbVcFhR2gC4eyO1yegwIdf2XCJxprc1ym3sr8A7wvrV25ElcIxonsJcExFhrc9yO1cQpdWmABtbaY3mcJzd77i5r7Vtex9oDq4Cl1tpYt/HhwHhgsrV2iNd9zgF+A2ZZa/vm9zgUoCt5xzOyueujpfzmtZuuLP1iLosOpmY4wbgtTsnKFdsO5tuLxJ/6NSsTF1nHFZALp23jsHJTIkTKl1GTE/hlzV8lB66La8YL13QswRVJWZS07xgDxsz0+ADkqUvbMrRnCz5dnMzD/1vlMX/sLbFc0O70m8xL0Xnhx3W8PWPTia+b1q7q02NgwrCu+kNKyrR9R9MZ+Npsj81XTWtX5Yd7exNWxbec3NcrdvD4F6s4nOYb3DmvbUOeu6pDudzNX9Zt2H2Ed2cl8uXy7T67+nMFBxku6diYUb2jad+0/GyG3Hc0nYlzk5g8P8nv8zbXgDYNuL1fDLGR4cW3uArq0PFMnv56NZ8v2+5zrGqlYB6/pA03xjcv9uDwtgOpDB6/iMR9nh9TtWxQgykj4stUr6DN+45x/9TlLPPTm7Fp7aq8PKgT3WN8P0956uvVTJyXdOLrSsGGH+7pQ8sGNYpyuaVGWmY2/V6a4RGovSY2gpcHdcrjXkUrad8xxs1J5LOEbQE/lzEGBrZrxK19ojmrDG80PZqexegpCczd6BnAqFs9lEnD48vU76bynplbnuXkWGas38M7MxNZtDnvTNYh3U+uL29OjmXCvCRe+HGd30y8G7s157GL2uj5UQoVaoDOdcIhwOtA7vaPdcD11tqVXvPmAD2AW6y1H57E+UcC7wHvWmtH+zmem103wFr7Wx7n+Rk4D7jIWvuD17FqOAFGgDrW2oOu8Q+Am4AbrbUfe90nBDiEE5CsYa3NMwVIAbrSITM7h4f/t5LPl3q+ea8cEsSbN3bhvLalN7W9LNlx8DjvzNzEvE372ZhH/4lAjIHWDWvSJbIOcZF1iIsMp1l41TK741Yqlt/W7mbEpL9e76uFBrPosQHU0JsiD1nZOXywYAtbUlK5vmvzYt9FWtrd9dFSj8ySZuFV+e3+foSGBJGTYxk0dj5L3BrNN6lVhV/u76s336XYiImLPTYJvXRNR6Ys2KLyT1JuWGsZPnGxR9+hIANTR3cnLipwkGLXoTQenLaC2Rv2+RyrWz2U567qwPnagFAqbNp7lGe/W+uz4dFd1UrBXNe1GSN6tSjXpUqPpGXy8aJkxs/Ju6xnfFQ4d/SPoZ82XxSJuRv38eBnK9hxKM3nWGxkHV4Z1KlEW1rsO5rO0AmL+GO7Z7uNprWrMnlEPDH1S3egylrLBwuTefa7tX57wl8TG8ETl7b1uwEDnPK3A16Z6VHS+OzocD4edXaF+Nv+nZmbeP6HdSe+Dg0J4vcH+pWKUrj7jqYzeV4Sk+bn3fsqvkU4l3RsTNeocFo3rFlm+sQeOJbB0ImLWbHVM6jctHZVpoyIJ7qU/9vzJ1BmbqdmtZlYRjNzy7OMrBy+Wr6d92Yn5tsLcmTvFlwbd+q9IDfsPsJ9U5f7/K4BiKxbjTHXdtKGoVKm0AN0rpNWBdoDB4FN7lluruOhOKUwDfBVbgCsgOd+CXgAeMBa+4qf428CdwJ3WGvfzuM8HwE3AHdaa//rdSw3gw6gu7V2gWt8MRAHxFlrl/g55x9AO6CttXZtXo9DAbrSIyfH8n/fr/XpixEcZHjh6o5cExtRQisrHzKycrjwP7PYtDdgQquPqpWC6dystqt/XB3Oal6HWlULv2m3SHHIys6h5wvTPT6see6qDtwQ37wEV1X6jPn5T16fvhGAsCoh/HBvn1Lxx2ppsHLbQS57c67H2H+u78zlnf9qXr5u12Eufn2ORznV0X2ieeSiNsW2Tjk5vV6YzrYDf2XMffu3XliLT/mnO/vH8OAFZ5bACkVOz6R5STz59WqPsXvObcV9552R732ttXywYAv/9/1a0jJ9dwAPcn0IXDPAh8BS9I5nZNPv5d8DBqPqVg9laI8obj47skJ9SJielc2Xy7YzdmaiT6aUu1G9W/DoRW0qRFCiOKRlZvP8D+s8srNyVQo23HfeGYzuE1Mq+jMdScvk1slLfMqQhVcPZdKweDpElM4snt2H03hw2kpmrd/rc6xu9VCevapDgao3fLNiB3/7eJnH2KvXdeLKs8r35y4HUzPo8+LvHpm2o3q34LGL25bgqnwdS89iasJWxs3e7FPZwVtYlRDiosLpGhVO16g6dIioVSrLK+46lMYt4xf69EmMqV+dD0Z2K1PZq97yysydPDyeJvp7usTlbuB5f06STzDVXdvGYYzuW3h9eTOycnhz+gbemrHJ4zMCcDbM3dY3hnsHnKFKZKVEkQToipIx5l1gFDDKWjvOz/H/Ax4FHrXWPpfHeW4EPsQplxlvrU1xjYcAnwJXuaaeyLAzxqwHWgGtrLUb/ZxzLk5WYA9r7Xw/x28FbgVo3rx57JYtW7ynSAmx1vLfGZt46ac/fY49dlEbRvWJLoFVlQ8T527mqW/W5DmnUVgVYqP+yo47s3FNKpWRRvEiBfHKz3/yxvS/fm10alabr+7sWYIrKl2OpGXS/bnpHv1qzm/bkHcH+7w/qXCstdw0biHzNv31IU67JmF8c1cvnx2rz36/lndnJZ74OjjI8N3dvTizUVixrVcK5mh6Fu2f/OnE10EG1vxrIFUqBfst//T93b1ppf5FUoas23WYy96c61FeJzayDp/eevZJfeiQuPco909dwfKt/suovXJtJ85WWfoSMW3JNh74bIXPeFTdaozsHc01sREnek9VRNk5lp9X7+LtmZs8MqPd/fOStozo1aKYV1b+rNh6kPunLve7IbR1w5qMua5Tqesxn5aZzT2fLOOn1bs9xquHBvPekDh6xNQroZX5982KHTz+5R9+M6sGtGnI81cXvPywtZbB7y/yyJKuWz2U6X/vR61q5XfTxXPfr2Ws2/v0mlVCmPVg/1K7gSErO4fvVu1k7MxE1uz0zcLxp3JIEJ2a1SY+KpyuLcKJjaxT4lVjNu87xs3jFvoEGztG1GLisHjCS+n3/2TsP5rO0AmLWbXd83dNk1pVmDyiW4UpIVva7D6cxvtzN/PRgmSO5NGXt1fLeozuG02vlvWKZNPOsuQD3D91BZv9bBpq0ziM167rrOpFpUB5DNA9CzwCPGKtfT6P8wQB3wIXAruBr4FUYAAQA2zFCcZdYK392XWf/AJ084DuuGXdBaIMutLpo4XJPP7lKrw2F3B7vxgeuqC1djiepCNpmfR9aQYpxzJOjAUZaN0ozAnGuTLkmtZWuUop37ampNL7xd89xn64pzdtGitwAoED+eMGxzGggpcanrV+L4PfX+QxNmVEPL1b1feZeyw9i/PGzPQo6xQXWYepo7uXmfIzFcWy5ANc+d95J76Orled6Q/0A5zfned6lX+KbxHOp7dWjPJPUvalZWZz+Ztz+XP3kRNjNSuH8P09vU+pxGFWdg5vz9jEf37b4NGHE5wy6CN6tuCBC1pX6GBQSbjm7XkkuJVWbtM4jLvPacn57RqViiyl0sJay7xN+3ln5iafsq3GwDs3q2fsqcrMzuGt3zfyxvSNPtkBxsCtvaO5//wzSmVGDzivbY998QefJmz1GA8NDuL1GzozsH3jElrZXw6mZvDEV6v5esUOn2M1KofwxKVtGRQbcdLvT5L2HeP812Z5bOK4qVtz/u/KDqe95tJox8Hj9Ht5hsfjfWhga+7o17IEV1Uw1lrmbNzHu7MS/ZaezkuQgbZNwugaFU58VDhxUeHUr1l8fWRX7zjEkPcXse9ohsd4j5i6vDs4rsSDh4Upr8zcicO60jGidgmtrOLZuMfpy/vFssB9eYMMXNKxCbf2KZ6+vKkZWTz/wzomz/dNFAoNDuKBC85gRK9ovX8rQYECdKeVumKMqWaMudsY850x5g9jzCav47WMMTcaY244hdPnbgkI9AwO85rnl6vs5mU45TJ3AbcAw4FtQC8g91XNvaB+oVxbSq8buzXnrRu7EOq1s/ftGZt45PNVPm/8JW/vzd7sEZyrFhrM3H+cww/39OaZK9pzeeemRNSppg8cpdxrFl6NXi09d8J+unhrgNkVS06OZZKfN4oAT369mtSMwLvNyrucHOvRpwKcHXb+gnMA1SuH8ORl7TzGErYc4LMleq6VNn/uOuLxtfuuxZpVKvHkpZ4/x0WbU5i2ZFuxrE3kdD3/wzqP4BzAv69sf8r9x0KCg/jbua344o6ePrvArYVxczZz6Rtz+GO7/gQrLhv3HPEIzgH896YuXNihsT7c8WKMoWfLekwZ0Y1pt3X3+EDYWrjnk2U+PZEkfxv3HOWat+fx2q8bfP5Gj6hTlU9v7c4jF7UptcE5cF7bnr+6A7f1jfEYz8jO4Y4Pl/LJouQSWplj1vq9XPDaLL/BufgW4fxwT2+ujWt2Sn/LR9Wrzp1ewamPFiWzLPlAgHuUba/+st4jONcwrDLDepSN7FljDL1b1WfKiG78dG8fHrnwTAa0aVCgFiQ5Fv7YfpgJc5O4/cOldP2/Xznn5Rk8NG0FnyVsZcv+YxRVcsjipBSuf3eBT3Du/LYNeX9o13IVnAPn74cJw7pyQTvPza0pxzK44d0FzN14csFVOTnWWhYnpTBy0mIGjJnF1IRtfoNzVSsFM7RHFDMf7M/rN5xVLME5gGqhIfzr8vZMHh5PwzDPIHlGdg7Pfr+OG95bwNaU1GJZjxTcKQfojDGdgbXAqzjZaW2BKK9ph4HHgA+MMd1O8hK5NQgDNS9o5bpdn9+JrLVZ1tpXrLWdrbVVrbVh1tqBwBqgM3AccG+cEPDartKYLYAsINH7uJQdF3ZozIRhXake6vlm/pPFW7nzw6Wk+WmGLL72HElj3GzPfwqjekeX6freIqfjuq7NPL7+fOk2vZ4AMzfs9VtuAWD7weMepUErmm9W7vApKfPwwLx7kZ3ftiED2jTwGHvuh3UemyWk5HkHL87wKl95UYdG9D3DMxD77PdrOaCfo5Ry09ft9ukBddVZTT16Zp6qDhG1+PZvvfyWBNyw5yhXvDWXN6dvICvbt2edFK5PFnlu/Dg7OpwW9aqX0GrKjriocN66qYtHEDMtM4cRkxL0oVgB5eRYJs7dzMWvz2aFn9Kh18U148d7+xDfIrwEVnfyjDH848IzecyrZ3COhX98voq3Z2wqsgBGIKkZWfzzyz8Y/P4inx6TocFBPHZRGz4ZdfYpb7rIdVu/aI/XDWvhsS/+KHev4X/uOsL/lnpusrpvwBlUDS29weNAWjeqyei+MYwb0pVl/zyPn+7twzNXtOeyTk1oXKtKgc6RuO8YUxO28eC0lfR9aQbdnv2NOz9ayqR5SazZcbhQM38qFAAAIABJREFUNsX/vm4Pt4xfyJE0z42e18RG8N+bupTbjPsqlYJ568YuXBfn+bnDsYxshk1YzA+rdpbQysqvnBzLT6t3cfXb8xj0znx+XbvH77zw6qHcf94ZzPvHOTx1WbvTfv08VX3OqM/P9/blsk5NfI4t2pzCwNdmMXXx1mL/vSOBnVKAzhhTF/gOaAYsxclO8ylWbJ2f9HjAAFec5GVya4Sd7ypT6X79mkBPnMBaniUm83ELUAWYaq11L7I93XU70M99+gDVgHnWWv+dsqXM6NmyHh+NOps6XjXQf1y9i2ETFnMkzbf2unh647eNpGb8FXyoWz1UvfykQju/XUOP15TDaVn8tHpXCa6odJgwN8nja+/NEe/NSmS9VzCjIkjPyvbpi3pppyZ0iMh7l50xhicvbUeVSn+9RTqYmslz368tknXKqfF+TnvX/TfG8Mzl7ans1rT7QGqmT0alSGmy50gaD3620mOseXg1nr68XYB7nLwqlYL55yVt+WhUN5rW9tz0lZVjefnn9QwaOz/gxg85felZ2Xy+bLvH2A3xzUtoNWVP3zPq88zl7T3G9h1NZ/jExX77e8lfdhw8zuD3F/HUN2tIz/IM4tSrEcq4wXG8cE3HMpkZM6pPNC9d09EnA/WFH9fx7PdrySmmSj5Lkw9w8etzmLLAt7pF28ZhfPO3XozqE10opdMrhwT7/FtYs/NwwMoaZdVLP63zaKESU78618RGlNyCCklQkKF1o5rccnYkr99wFvMfOZc5D/fn1es6cUN88wL3PdtzJJ3vVu7kya9Xc9Hrs+n8r58ZNmER/52xkYSkFNKzTm5D61fLtzNqcgJpmZ6vESN6teDFqzueVB/csig3M/f2fr6ZuXd+tJSPSzgzt7xIy8zm40XJDBjz/+ydeVxU1RvGnzMMi+yyCIossqsoqLih4pbivpSaWaa2qGWpmZllWmn2q8xSK3NJraxcc0tzy31H3AEFBQQEBQRkX2fO7w9U5txhQIZh5s5wvp/PfMZ55565B2e59573fZ/nBKZsvITLSVV3wrvZmWPRiACc+bAPpvf1EYXnpI25MVa81A4/vNROqRO2oFSGOX9fx5u/X0JGHk9tiAF1f7FmAmgK4AiAzpTS71CRLKuKfY/ve9dmB5TSOACHUNGVN03w9OcALAD8TiktAABCiDEhxJ8Q4iXYFoQQJfMfQkhHAF8ByAewUPD0dgAPAYwlhAQrjDED8MXjhz/X5u/hiJdAV1tsmxqCZoJKoHPxmRi39gIy8/mPlSoSHhYoHfin9/XRy4slDkdTmEqN8Hx79mJMWIHe0LiTno+TsRlMbO2EYDhbV/7ulsspPtkZ2eCquP66kIR72ZWnUFIJwfv9VIkHsLjamWN6Xx8mtu3SPYQnZGl0jhz1iXmQzzwWdtABgJu98vu4JSKZv48cUSKXU8zedh2ZCl2eRhKCZWODYGVWswxWbQnxcsD+mT3wQnvlRc4rSY8waPkpbDyf2OCOHdrgcHQa05Vt08iYe6jVknGd3TClJ1u4eDs9H2//eYmRweNUQCnFrispCFt2EqerkGkLa+2EgzND9d63eHSwK1a90gEmUnY5bu2pBHyw/Xq9dpaVlsux9FAMRv18VqnAQUKAab29sGtaN6WCorrS3cdBqZPju0MxeKDgp6zPhCdkKXXUzBngb7BJouaNzTGyXXP87/k2+G9WT1ye3w+rx3fAG91bINDV9pkkkPOKy3EsJgPfHIjBqFXn0OazQxiz6hyWHLyF4zHp1RbLbzyfiJlbrir51c7u74tPBrdsMJ7chBB8OKDqztyPdtzAT8fu8PMjNckpLMNPx+6g+9fH8NGOG4hXURAW2NwGK19uj2Oze2F8F3dRdswODWyGQ++FKqm2AMB/N9MQtuwkDkTygnJdQ9T5shJCrgJoAyCYUnrlcew+gCaUUqVPIyGkBEAepdRB+FwN+/ECcBZAEwC7USGp2RkVyb5YACGU0szH23oASACQSCn1ELzOBVQkECMB5AFoDWAQgBIAz1NKD1ax7xGoSNQVA9gMIAsVXnZ+j+Nj6DP85wUHB9OIiIja/NkcHZH6qAjj111AXAb7w+vpYIGNbyhX73KAaX9dxr7rle3zbnbm+G9WT6WLDQ6noRGblof+359kYsdn94JHA5WFWrA7kjEqbu9mix1vd8P+G/fx1p+XmW2XjGqL0QK5DkMlr7gMPZccZxZAX+3qjoWCKuPqKC2XY/CKU7idXpkI8nWyxL7pPWBsoIsC+kJmfgk6fPHf08cmRhJELwyrcrGmtFyOQStO4Y7gfdz7bg9+TOWIinWnE7BobzQTe7+fL94VJJnrgwORD/DxzhtVSvmG+jpiyai2cLJ+NuktTs288ssFJkkyMcQDnw3TXJdkQ0Eup3h30xXsE0iOje7QHN+Mass9uh+TXVCKebtu4N8byouEVqZSfDasNZ5v72JQ/1/n4zPx5m8RyCth5fmea+mEH8e107g83+20PLy39SoiU5TEr+Bub47vxgSig3v9SYam5xWj79ITjBzhoDbOWPlyh3rbpzaglOKFn88ynTXt3Wzx91shBvV5rQ2FpeW4kvQI4QlZuHg3C1eSHqGolpYPEgK0bGqNjh526NTCDsEejeFoaYqfjt3Bt4dYpyNCgIXDAzC+i7sm/wy9YltEMubuuKEkHfp69xaYN6jhJC3rSsqjIqw/nYDN4UkoKFX9me3t54jJoV7o4mmnN99zSin+vJCExftuVvl9fKF9c3w6rBWs66HgjlMJIeQSpTRYKa5mgi4PgBSA+ZMkVQ0JugwANpTSWvd4EkJcUdHhNgCAPYD7AHYB+JxSmqWwnQdUJ+g+ADAWgBeARgBSARwE8BWl9G41++6GCg+9rqiQwrwDYD2AFZTSZzq68ASdfpFVUIpJG8KVdO6drc2waXIX7rmgwLXkRxj+0xkmtuKldlVqHHM4DZHnV55hLtTe6uVVo6+YIZJbXIYuXx5hpHCf/FZQSjHp14s4HlPZXWdnYYIjs3qKQhaivvnuUAxWKHjvmZsY4cQHveFoZVrNKGUuxGfixTWs4vfcgf6Y2lNJVICjRc7FZeKltZXvS8um1tg/o4fK7c/HZ2Ks4H38cIC/knQNh6MrolNzMeKnMyhV6O7o5GGHTZO7PFO1vCbIyCvBRzuuV+n9YdPIGF+MCMBQfi5aZ5KzCtHjm2NM7MDMHvB3VhKm4TwDxWUyjFt7Xkkaa3Z/X7zTp/6T22LnfHwm3t10pUqZra6e9lgyui2aN9aNj099E5mSg4kbwvEwny086NTCDr9MCNbIQqlcTrH+TAK+ORhTZefmy53d8PGglrDQggrO7+fuYsHuKCa2YVJH9PZrUvUAPeBg1ANM2XiJiW2d0lVv/BG1QZlMjsiUHFy8m4XwhGxEJGbhUWHtpX6b2pjhvqDrUiohWDomUCMeuPrO4eg0TPvrstL33MHSBMZGEkgIASGAhBBIHt9XPq74t5GEPH2eKGzHbCt5Mlbx+Wq2J1VsLwHkckBOKeS0Inn05N9ySkHpk+cUn69ie3ktt6/q9eWV2z/ML1HqzHyCVEIwLKgZJod66vX5UMLDAszaehVXqpDrdLFthCWj2yLEq1b9VZxaoOkEXQEAOaXUSiFWZYLusX9cIYAiSmnjWu9Mz+EJOv0jv6QcUzdeUpLVCHCxxu5p3bW2ACFmKKUYt/YCzsVnPo0FuFhjz7TuvDKHw3nM1ohkzNle6dHjaGWKs3P7NLiuJmG3hZO1KU5/WPn/kJRZiH7fn2A8RsZ2dMVXL7TV+ly1SXpeMXp+c5ypXpvR1wfvPaO8pZDZ265h+6VKY/pGxkY4PCvUYBe09IFfzyTgs38qP/sj27ng+xeDqh0jfB/NjCU4/F5PnRmMczhPKCqVYeiPp5kuT2szKfbPDNW6ygSlFNsi7uHzf6KqrG4eFtgMC4e3hq254Rd61BffHozBj8cqC0iCXG2xa1o3Hc5I/8nML8HIlWeRlFXIxJePDWrQC8uqFpRNpBJ8OMAfk0I8DP76MuFhAcavu8BIngMVhT2/vdYRTazU7wy+l12I2duu4Xy8smx2EytTfD2qrVaTYzI5xciVZ3BdoSDazc4ch94L1XjHoDYol8kRtuwko8L0XMsm+GVCRx3OSvzI5RR3MvJx8W4WLiZk4eLdbKQ8UuWapBozYwl+frkDevvrb4JX06jqzOWoj4WJEcZ1dsOkbi3QzECU1cplcqw+GY/vD8dWmZB8vXsLfBDmp5e/y2JHVYJO3VXC+wDMCSHP8ivYBYAJgLtq7ovD0SqWplKsmxiMwW2aMvHIlFxsPHdXJ3MSGydvP2SScwAwdwBvm+dwFBncpinjx5iRV4Jjt5Qr/g0ZmZzit7N3mdj4Lu5MktLN3hzv9vFmttl8MRmXEg3bf2vFkdtMcs7ewgRvhnpWM6J6Phroz5g/F5XJ8Nme6GpGcOqbmLSa/eeEfDTQH7bmle9jcZkcn+6J4v4RHJ2z+N9oJjkHAF8+30YnEvCEEIzp6IoDM0PRyUO5Q2HPtVSELTup5H3KeTbKZXJsu8R6547t2DCkp+sTe0tTbJjUkTlWA8AH2643WM/RnVfuYeofyn58AS7W2Pdud7zevUWDuL5s4WCBv98KgZ/gPOHm/VyMXnUOyYKk7rNAKcX2S/cwcNmpKpNzg9s2xcGZoVrvXDOSECwe0QaKb2tSViF+UigI0Ce2X7rHJOckBPggrOEpptQWiYTA18kKL3d2x7Kx7XBmbh+cmdsHy14MwrjObvBpYlnja1iZSbHx9c48OSegi6c9Nk3uAgdLXqRUVxytTDFngB/OftQX8wa3MpjkHABIjSSY1tsbu6Z1g6+T8vdt3ekEDP3hNCJTcqoYzakP1E3QPTHWmVDdRo+75xYBoACOqLkvDkfrmEqNsOKldghrzRpQf3soFmm5hmFkrC5yOcVX+28xsR4+Dujuw1ugORxFLEylSjJbmy8mq9jaMDl2K52pFDeRSvBSJzel7d4M9YSXIyshPG9nJMpkyjI8hkB8Rj42hbOfhel9fZiEbm2xtzTFRwPZBYH/bqbhUBQ3fNYVsWl5zGM/55oXG6p6H4/eSsdB/j5ydMihqAf443wSExvdoTmGtNWtlKSrnTk2Te6Cjwf5w0TQnZ6WW4JX14dj/q5IFJbyKvLacDwmA2m5lVKDFiZGXDZUQ3g5WmLN+A7M57VUJsfkjRGIz8ivZqTh8fu5u3hvyzUlv6Rpvb2w461u8HmGohZDwsnaDFundEUHd1Z0KjGzEC/8fBa3Hij7xqkiM78EU/+4hNnbril10VibSbF8bBB+fKmdzuTk2zS3UfIKW3UiTqkIROwUlcrw/X+sF9oL7ZvDz7lhfXY1hYttI4xo54IvR7bB4Vk9cXl+P6wZ3wFv9miBQFdbSBWyuo5WptgyuSs6VlGkwwECXGywfWoI/PlnUS28m1jiq+fb4NSc3ni7l7dSYY0hEeBigz3vdMebPVpAaKV3Oz0fI346gx+O3Ea5ga7LiAl1JS6DAFwCkAdgBKX0uFDi8rF33E8AhgAoB+BPKY3X2Mz1BC5xqd+k5RbjuaUnmBPbIW2b4sdx7XU4K92y60oKZm65ysT2vtsdAS42OpoRhyNehF6NEgKcmdsHTW0Mp/qqOl755QIjFzyqQ3N8Ozqwym3Pxj3EuLUXmNi8QS3r1FUmVqb9eRn7btx/+tjNzhz/zeoJE2nd5E/lcopRq1iT+mY2Zjg8q6dWfEU4lVBK0fazQ8z5w+kPez+T5KhcTjFm9TlEJGY/jTlbm+G/93vWKYnL4ahDWm4xBiw7iWwFrxgPe3Psm95DVL8rMQ/yMHPLVdy8r7yI3cLBAkvHBKK9W4NzW1CLN36LwH83054+bgiy09qmquspd3tz7HgrBPaWtfOh1TcopVh5PA5LDsYwcUKAL0e2qbKQqyFRWFqOt/+8zPgzAxWJtfUTOyK4hoTE4eg0fLTjupKnHVBRVPvNqLaiuA7JLS5D36UnGN/Brp72+OvNziDCVWKRsvL4HXxzoPJzbCKV4PjsXgbVZSMmCkvLcTXpEbIKSxHq66gRf0ZDh1KKzIJSlMnkFd5r8pq82hSff+LVRp+OrdX2KrziFLc1qpW3XdW+eYz3naSW2yvu//FYEyOJwR+HVXE+PhPvb71WpdxskKstvhsTCE/HmgtOOdWjUQ+6xy/4OYD5AOQAIgC0AWAGYNXjf3cBYASAAPiAUrpUvanrNzxBp/8IPWQA4PfXOiHU11FHM9IdJeUy9F16gtHHHxbYDCteaqfDWXE44oVSikErTjMLhu/388W7fX10OCvtcDstD/2+P8nEakrmz9pyFTuupDx9bG5ihP9m9TSoC92ryY8wQiFpCwArXmqHYRrqTrh5PxdDfjjNVKRPCfXER4NaauT1Oc9G6qMihHx19OljCxMjRH4e9syLTjEP8jB4xSnGE+D17i0wf0grjc+Vw1GFXE4xfv0FnLlTKWsulRD8/VYIAl1tdTizqiktl2PZf7FYdSIOQjsNCQGm9fbGu3186lwMYcik5RYj5KujzDFk17RuCBLh+63vrDhyG98dZrtvOrg3xp9vdDZYzxdKKf63/xbWnGTrto2NCL5/MUjnXblioUwmx+xt17D7aioTr85vK6+4DIv2RmNrxD2l58yMJfh4UEu80tldVJKhe66lYvqmK0zs+xcDMbJdcx3N6NnJLihF6JJjyCuuLMTi59scDkffySsuw8J/orHtUtXHknmDWuKVLu56U0ghRjTtQQdK6acAZgEoBtAJQCNUJOOmAugOQPr4uRkNNTnHMQzGd/VAG8GC8vzdkSguUzalN3T+PJ/EJOeMjQhm9/fT4Yw4HHFDCFHybdkSkQx5FUa8hsavAu+5jh6Na+y0/XhwS1ibVXZkFJbK8Pk/UfUxPZ1AKcXXAongABdrDBF4ntaFlk2t8Vo3Dya27nRCraSROHUnRiBv6etsVasLGT9nK7zRg+0e3XAmgfsAcLTKL6fjmeQcAMzq7yvK5BxQ0b0wZ4A/tk3tCnd7tltVToEfjt7B8z+fwW3B95NTybaIZCY55+9shcDmXCWjPni3jzdGdWATEZcSs/H+tmsGeZ4ok1N8tOOGUnLOzFiCNa8G8+ScAsZGEnw/JggTQzyYeHGZHG/+HoFdCsVsAHAhPhMDl5+qMjkX6GqLfdN74NWuHqJKzgHA0LZN0UNgk7F4303kKHRsi5WVx+8wyTlrMyne6uWlwxlxOBxO3bEyM8aS0YFYM74D7AUyyMVlcszfHYVX14fjQU7Dtn6qD+pUPkgpXQbAHcBMANsBnAUQDmA3gDkAPCmlP9R1khyOLjGSECweGcDo8SZmFmLl8TjdTUoH5BaX4Yejt5nYy53d4WZfs1wXh9OQGRHkwlTr38suwpm4h9WM0H9yCsuw4zK7eDAxpEWN4xwsTfGhwH/rYFQajihIbekzJ28/xLl4drF77oCWGl8wmfmcL5ramD19XC6n+GRnpEEu+ImV2AdsAkAdD4jpfb3hotA9KqfAvJ03lPx6OJz6IDIlR0mCrounHaaEin8BsoO7Hf6d3gPjOitL5UWm5GLwD6fxy6l4/psoQC6n2BLB+qOO7ejKq6TrCUIIvhzZBiFe9kx83/X7WHIoRsUo/aS0XI7pm68oeTFbmUrx+2ud0dtPuSOsoSOREHw6tBVm9fNl4uVyiplbrmLDmQQUl8nw5b83MXbteaaIFqjodp7Vzxd/T+0KL5FKkhFCsHB4AHOd9DC/FN8cvFXNKN1zL7sQv51NZGJv9/aGrbluPP04HA5H0/Rv7YyD74WiXysnpedO3X6IsGUnsedaahUjOepSZ30PSmkmpXQFpXQMpbQ7pbQrpXQkpfRbSqlhrKhxGjxtm9sqGxkfj2tQZt5rT8Yz/iMWJkZ4p4+3DmfE4egHNubGGBTgzMSECxSGxpaIJBQpdBk3tTFDWGvlk7uqeKmjm5KU1oLdUSgq1e+uZbmc4itB91wPHwd0F1QOawILUyk+HdqaiUUkZmN7FVIVnPohRpCg83WqfYLO3ESKhcPZ9/HavRz8FZ5Up7lxODVRWFqO6ZuuoExWmcCyaWSM718MgpHIOjBUYWEqxZcj22DDpI5wtGK9RErL5fhi302M++U87mUX6miG4uNsXCaSsyoX+U2kEr2QmtNnTKQS/PxKB3g3YRMoPx+PwyYD+a0vKpXhzd8jsO/6fSZuZ2GCTZO7oFOL6j3VGjKEEEzv64NFI9hiYQD4/J9o9FpyHGtOxkPoWuPlaIEdb4dgel8fSI3ELenbwsECbws6z/4KT8KVpGwVI3TP94dvo1Qmf/rY2dpMqduRw+Fw9B0HS1OsGd8BS0a1VfJBzykqw/RNV/DOX5fxqFDZ85RTe8R9tOZwRMT7/f3goGAWWiqTY/7uSKjr46hPpOcW45dTCUzszVBP5v+Dw+GoZqzA8P5Q1ANkFRjmiYxMTpWqSsd3dX/mBQLJ465lxTXglEdFWCHo4NU39lxLZbwIAeDDAf4qtq47Ya2d0FfgUfLl/psG+7kTG0KJSz81EnQA0LelEwa0ZhP83xy4hfQ8LiuiCyil2Hg+EWNWncMXe6ORbaDfp0V7oxH/sICJff1CGzS10T8/0N5+TXBoZigGVyElfD4+CwOXnUJUKpeOBYDNF9mE0KAAZ9iYG+toNg0Hm0bG2DCxIxws2e6bT3ZF4mRsho5mpRlyi8vw6voLOCH4O5ramGHrlK41Sp9zKhjfxR0rxraDsRGbpXuQq3wu8Fq3Ftg3vQfaNhenFHFVTO3pBQ8FVR5KgXk7I1GukAQTC7ce5GLHFbbgbVY/X4P1jeRwOA0bQghGB7ti/4we6FxFQc3e6/fR//uTOB6TroPZGRY8QcfhPCM2jYwxfwhr+nvmTmaDaOtdcfQ20w3jYGmi5I3D4XBU07mFHXPhWSaj2HHZMLuZ/ruZhpRHlRX4plIJxnZUlhmrjtbNbJQkMdeejEesnvoGlZTL8K1ArmpYYLN6XZgihOCzYa1hZlx5qveosAxf7b9Zb/vkVCCTU9xOZzvsfdWQuHzCp8NawcKkcuEnr7gci/fx91HblMvkmLP9OubvikT43Sz8cjoBYctOKi086zv7b9zHpnC2y/ulTq4YEKA5r0xt09jCBD+Oa4flY4MYn1MAyCspx9y/bzSIgrvqyCooxaEoVvxGWFzEqT9c7cyxbkJH5pgtk1O8/edlpeIefeFhfgleWnMeF++ynVAtHCywbWpXpa5BTvUMDWyGdRM6opGKRFAzGzP89UZnLBjaSu+SRWbGRlg0IoCJRd/Pxe/nElWM0B1LDsQwHYs+TSzxfHsX3U2Iw+FwtICrnTk2vdkFnwxuycgSA0B6XgkmbrjIPZ7riFoJOkLIAnVump48h6NthgU2Qzdv1idg0d6byCkSv5GxusRn5Cst1Ezv66PU4szhcFRDCMGLgiTV5ovJBrkg+OuZu8zjEUEusLOovSfDrP6+cLYW+Kjt0s+u5T/PJzHeIMZGBLP7+9X7fl3tzDG9rw8T2xpxDxfvZtX7vhsyiZkFKC2vrPq2tzCpU8d5U5tGeE/gQbP7aipO3TasxJCYKS6T4e0/L2ObQCY2Pa8EE9aH49PdkXovwwsA93OKMHfHDSbm6WiB+UNa6WhGmoMQguFBLjj4Xii6e7PSwjdScnDytmF7w9bEjsv3GMm2Fg4WVVZKc+qPQFdbLB/bjpEyzC8px2u/XkRaFZ1SYib1URHGrDqHqFQ2udiyqTW2TumK5o25h7k6hPo64s83O8NW0Nn6fHsX7J8ZihBvzcuma4sePo4YGtiMiS09FIMHOeL57F+Iz8SRW2yXyJwB/qKXEeVwOBxNIJEQvNHDE3vf7Y7WzayZ517u7AYfNRVjOBWoeyT5DMCntbg92Z7D0WsIIVg0PAAmRopGxiVYamBG3oosPRQLmbxyQdzd3rzW3TAcDgd4oYMLpAq6jXfS83FZxP4K6nDrQS7OxWcysYndPNR6LUtTKRYMZReFwxOy8PflFHWnpxNyi8vwg0Ce8+XO7nCz187i1BvdPeEjqFKft/MGykQoG2QoCDs91fGfEzIxxAOtmrIXQvN3RaK4TP+TQmLnyQL5oWjV1tq/nUvEkB9O4cY9/ZVKlMkp3ttylSk6MzYiWDG2HcxNDKcoq6lNI/z+WiclCeCfjt7R0Yx0D6VUyRv3xY6uIELTK069E9baGfMGsYot93OK8dqvF1FQUq6jWdWO+Ix8jF51Tkkmt4N7Y2ye3EXJE5JTO9q7NcaOt0IwPKgZ+vo3wZrxHfDdmCDYNNJ/Odr5g1vCSqEIuKBUhkV7o3U4o0oopfjqAOslHezeGM+1bKJiBIfD4Rgmvk5W2Pl2N7zT2xsSUrFG/LHg3IVTe9RN0J2s4XYNQCkAAuARgBOP4xyO3uPpaImpAiPjjecTcS35kY5mVH9cTX6EfTdYQ+/Z/f2UWpo5HE7NNLEyQ1/BRdxmQXeqvvPb2bvM4y6edmgpSCrUhoEBzujl58jEvvz3pl75Pq09GY/swsoFbwsTI7zTx1tr+zeRSvCFQDYoNi0f604nqBjBqSsxD1h5S786yFs+QWokweKRAUxnxd3MQvx8PK7Or81RTVZBKV5eex5n49jCA3sLE8YnEwDiMgowcuUZ/Hj0tih9c2pi9ck4nI9nu2s/CPMzSI8oiYQodReH381CeELD7C6+lJiNOwqyvFIJwQvtm+twRg2b17u3wKtd3ZlYVGoupm+6whRNipGo1ByMWX2OkToHgB4+Dtj4eieDSCKJAU9HSywf2w7rJnZEf4FPrT7TxNoMs8NYhYl9N+6LwtvoYFQariSx6z1zB/rzQgYOh9MgMZFKMDvMD9vfCsHyse1gwRXW6oxaq+yU0l6U0t7V3NoDcASwCIANgH8opb01OXEOR5e83csL7kIj4103RH/RVBsopUpeRW294hQUAAAgAElEQVRcbDC4jf56kHA4ukbYfbr3+n3kFRuGRG52QSl2CLrbhD5ytYUQgoXDAmCqUBSQVVCKbw7eqmaUeEjPLcYvp9hE2ORQrzrJHapDZ097pcXW5f/dxr3sQq3Oo6EQk8ZKemkiQQcA7dwa4+XO7G/Iz8fjEJ+Rr2IEpy7czynCmNXncE3QFefdxBJ7p3fHtqld4WbHdsKWyym+PRSLF9ecR2Im2z0iZq4lP8J3h2KZWHdvB7zR3XD9hgNdbdHDh5WD++lYw+yiE3bPPdfSiXc56RBCCBYMaYU+gi7PI7fSsfCfKNFKfUfczcLYNefxMJ8tohoY4IxfJgQbVCcup/54pYs72ggKQxbsjtKpYkC5TK507dGvlROCPbgMMIfDadi0d2uMIFdbXU/DIKi3NhhKaT6l9FMA3wD4hhDSq772xeFoGzNjIywaznYkRKbkYuO5uzqZT31wIjZDqZJ67kB/SIQl4xwO55kJ9XVEU5tKX7WiMhn2XEvV4Yw0x+aLyShR8N1ysW2kEdkXN3tzvCvoONsUnoxLieLvdFh+5DaKFBYUHCxN8EaPuiUt1eXjQf5M5XpRmQyf7RGHbJChEfNA8xKXT/ggzB8OlpWejqUyud56M4qZ+Ix8jPr5HNNVBACBzW2wdUpXNLVphA7udvh3Rg+M7eiqNP5SYjYGLT+FLReTRP/eFJSUY8bmKyhXKDJrbG6MpWMCDf6cb1pv9thyIjZDr2VK1SG3uAx7r7PnIWM7KX+mOdpFaiTBDy+1U/J4+e1cItYLvH7FwInYDLyy7gLyilkZzjHBzfHDS+1gKjXS0cw4+oaRhCgpBiRlFWKlDgsotl26h/iMyqIbCQHmhNW/lzSHw+FwGg7a0Klbigqpyw+0sC8OR2uE+jpicFu2m+zbQ7F6Z+JdFXI5xVf72SqxHj4O6KbHxtMcjhgwkhCMDmYXvrZc1H+Zy3KZXKlA4dWu7hozTX8z1BOejhZMbN7OSFHLyMVn5Ct1Jczo66Mz+Qd7S1PMHejPxP67mYZDUQ90Mh9DpbhMhruZbGeir5Oliq1rj00jY8wfwnozno3LxO6rhpHoFwOqJNpCvOzx55tdYGdRmSC1NJXiqxfaYs34DrBXiAMV3jkf/n0DkzdewsP8Eq3MXR0+2xOl9Jn9ZlQgnKzNVIwwHDq3sEOwe2Mm1tC66PZcTUVxGVtc08PHsZoRHG1hYSrF+okdmcIuAPhiXzQOiujY/e+N+3jjt4vM5wiokOr8+oW2GjsX5DQc2ja3xatdWJnXn0/EIU4HigFFpTJ8f5jtMB/dwRU+Giy+4nA4HA6n3s+WKKWZqPCh61Tf++JwtM2CIa1gqbDYml9SLhoj47qw+1oKbgk6AD4c4K9iaw6HUxtGd2jOVIVev5eDqFT9rtg/HJ2G1JzK4gQzYwlerKKrRF1MpUZKPmq3HuThV4HnnZj49lAMI3vsYW+OsZ3cqhlR/7wY7Ir2bqwExef/RKOwtFzFCE5tic8oYN53F9tGsDLTrOfOsMBm6C4omPliXzRyCg1DLleXhCdkYexqZYm2sNZOWD+xI3POp0j/1s44MDMUff2Vu4YPR6dhwLKTOHIzrV7mXBf2Xk/Ftkv3mNjLnd3Qr5WTjmakXQghSl10B6Ie4HZanooRhsfmi0nM49HBzWFk4J2T+oSTtRk2TGJ/eygFZmy+Igr/860Xk/HOX5dRJmM7hWf188Ung1tyfy6O2rwf5sdI7ZbJKObrQDFg/ZkEpOdVFtmYSiWY2c+nmhEcDofD4dSeek/QEUKsANgCsKhpWxXjmxNC1hNCUgkhJYSQu4SQZYSQxjWPZl5nMCHkECHkHiGkiBASTwjZRgjpqmJ7U0LINEJIOCHkISEknxBykxCyghDiXtUYTsPDydoM7/f3ZWJ7r9/HqdsZOppR3Skpl+Hbg2yV2PCgZggQaMFzOBz1cLUzV1pc1/cuug2CRNnIds1ha25S9cZqEuLlgJHtXJjYd4djkSrochEDV5Ky8e8Ntrp9dpgfjHVcRS6RECwe2YZZfE15VITlR27rcFaGRWyaUN5Sc91zTyCEYNGIAJgoeDM+zNcfb0axcvRWGsavu4C8EjZhPbpDc/w0rj3MjKuXaHO0MsUvE4Lx5cg2aCTY9mF+KV7/LQIf7bghmoT4vexCfLTjBhPzbmKJTwa3UjHCMOnl56gkI/jz8TgdzUa7RKbkIDKl0jOTECh1+XN0j7+zNVa+3J45dheXyfH6bxFIztKdl+wvp+Ix5+/rEFqwfza0Fab39eHJOU6dsDbTvWJAdkEpVgmOB5O6tUBTm0ZamwOHw+FwGgbaWCl6HxUSlwm1HUgI8QJwCcAkAOEAvgcQD2AGgHOEEPtnfJ2vAewF0B7AAQDLAVwGMBzAGULIK4LtpQCOAPgRgBWATQBWAUgH8C6Aa4SQhnX1ylHJ+C7uCHBhL+zn74rUqZFxXfjjfBIj62RsRPB+P66xzuFokrEd2U6qnVdS9PY3Iyo1B+EJrB/cpG4e9bKvjwe1hLVZZRV5YakMC/8RV9cypcoSwW2b22BQQFMVI7RLy6bWeE3w/qw7laDkm8ZRjxhBgs7P2VrFlnWjhYMF3u7lxcT+Ck/C5aTsetmfobP7agom/36J8dEEgDd7tMA3o55doo0QgnGd3fDvjB5VGqZvCk/CoOWncEXH75NMTjFryzXGL8rESIIVY9uhkUnD8oqqqotu97VUJGXqLvGhLYTdcz19HeFiyxeexUior6OSksDD/BJM+vUicoq02z1NKcV3h2Lwxb6bTNxIQrB0dCAmdtON1y7H8BjatqlOFQN+OnaHKdqxNpPirZ5e1YzgcDgcDkc91ErQEUJCa7j1J4S8Tgj5F8B8ABQVSa7ashJAEwDTKaUjKKVzKaV9UJGo8wOw+Bnm6gxgNoA0AK0opW88fp1RAMJQkTxcKBg2EkA3VCTpWlNK36WUzqaU9ny8rc3j1+RwIDWSYPGINoxk3d3MQr2svs0tLsOPR9lOipc7u8PN3lxHM+JwDJN+rZwYH6O84nLsj7yvwxmpz69n7jKPu3nbw7eefBkcrUwxRyC3eyDqAY7eEo903PHYDFwQJCznDvCHRESSYTOf82U8bcrlFJ/sugG5sAyeU2uEiU4/Z8130D1hak8vtHCoFKigFPh4xw1RezOKkd/P3cXMLVdRLvj8fxDmh48HqSfR1sLBAtundsV7z/kqyQXezSzEqFXn8P3hWJTp6L1aeewOwu+yv1MfDvRHq2b1k1AWOwNaO8NLwedUJqdYdVL/zuNrQ2FpOXZfYTtRhMVDHHHxUic3TBUkB+6k5+OtPy6htFw7vyVyOcXn/0RjxVHWq9HESIKVL7fHCx2aa2UenIYBIQQLh7eGiRGrGLDkUP0rBtzLLsTv5xKZ2LTe3rAx16xsOYfD4XA4gPoddMcBHKvmth/AGgADUJEAOw5gSW12QAjxBNAfwF0APwme/hRAAYDxhJCapDPdUfF3XqCUpis+QSk9BiAPgNAJ2/Px/T5KqfBsd/fje+6ezXlKoKstXuksMDI+HoeEhwU6mpF6rDkRj2yFijRLUyne7eNdzQgOh6MOJlIJXmjPyjVuDtc/mcvM/BLsvsYu8E0Mqd/K6XGd3BAo6ExZsDsKRaW670CUySm+FnTP9fBxQIig+lfXWJhK8enQ1kzs4t1sbBd4UXFqjzBBV1/JagAwMzbCouH65c0oJiilWHHkNhbsjoKipQ0hwBcjAjCtt3edJNqkRhLMeM4Hf78VwiRSgYrfiuVHbmPUz2cRn5Gv9j7U4XJSNpYJZG1DfR0xKcRDq/MQExIJwVu92PPd7RH3kJZbrGKE/vPvjQdMZ4iDpSn6tlT2UOSIizlhfhjclu3IPxuXiY933qh3b65ymRyzt19TOsaYmxhhw6SOCGvtXK/75zRMPB0t8ZZAMeDPC0m4Ws8ejN8djkWpQhFNUxszTGjAx0kOh8Ph1C91kbgk1dzkADJR0YH2BoB+lNISFa+jij6P7w8Jk2SU0jwAZwCYA+hSw+vcBlAKoBMhhFkhI4SEokLC8j/BmKjH9wMJIcL/oyGP74VjOA2c2WF+cLCsNDIulcl1YmSsLum5xfjldDwTmxzqCXuFv4nD4WiOFzuyPi8XErK0vlBbVzZfTGaqtl3tGqGPf/0u8EkkBItHBECxKeVedhF+OKp7H7XdV1NwS5Cg+VDQ8ScWwlo7Kb1X/9t/E9kFpTqakf6TV1zGSERLCODlWH8ddADQ3ccBI4KaMTGxejOKCbmcYuHeaHx3mPXclUoIlo9th1e6aM5uOsjVFvumd8crXZS7k67dy8GgFaew8XyiVs4X84rLMGPzFcgUugXtLUzw7ei2oury1QXDg5qheeNKecdSmRxrT8ZXM0K/2RzOyluO6tBc5z6pnJqRPJaRbO/GFiptv3QPPwq62jRJcZkMb/95GTsupzBxm0bG+OONzugmskIkjmHxVi8veCgo+lAKzNtZf4oBN+/nYucV9rP+Xj/fGr1oORwOh8NRF7XOwimlkhpuxpTSJpTSfpTS9VV0oT0LT0yvYlU8/2QlzreGuWYB+BCAE4BoQsgaQsj/CCFbARwCcBjAFMGwfQB2AOgH4AYhZDkhZAkh5CiATwD8gAp/Og7nKTaNjDF/SEsmdvrOQ/xzXT9k65YduY3issqvqoOlKV7vzj0EOJz6wruJFYLdGzOxLRH600VXJpNjo0D6ZUJXDyU5t/ogwMVGqYp17al43E7TnY9acZkMSw+xpyzDg5ohwMVGRzOqHkIIPh/WGmbGlaeC2YVlSv55nGfndjqbYPdwsNDKYs68wa1gJfBm/GxPVDUjGjZPukA2COR5zYwlWDshGMMCm1U9sA6Ym0jxxYg22DCpIxyt2MKn4rKKgq5Jv15Eel79dmx9ujsKyVls8nbJ6LZoYmWmYkTDwdhIgik9lbs0sgywaOFOeh4iElkfRGHREEe8mBkbYe2rwXAXWBAsPRyL3VdTVIxSn4KScrz260UcimblxB2tTLFlShe0d2usYiSHoxnMjI2wUKAYEJWai43nE1WMqBvfHLjFdNb7OlnihfZcvpXD4XA49YeYy+SerGjlqHj+SVzZgV0ApXQZgOcBSAG8CWAugNEAkgH8WoX0JQUwCsBnqEgUTkeF51xvACcB/EUp1b2WFkd0DAtshm7e9kxs0d5o5BZr17y7tsRl5GPLRTYxMOM5H1iYSlWM4HA4mmBsJ7aj4u9L93TmSVRbDkY9wAMF+S9zEyOMDtbeAt+sfr5wsq5c6C6TUXyiw67lP84nMt1TxkYE7/fzq2aE7nG1M8e7fXyY2JaIZEQIvKk4z0as0H+uHuUtFXG0MlXq1DwUnYbD0eLxZhQLxWUyTP1DuQvEykyKP17vjN5+9dsB3NuvCQ7ODEVYayel547HZCDs+5M4EPmgXva9+2oKdgg6AiZ0dUcff+W5NFRGd2jOJFCLymTYcCZBhzOqH4SS2l087ZRkWDnixt7SFBsmdoStwA/rg23XEZ6guWP4o8JSvPzLBZyNy2TizRs3wvapXeHv3DB9KznaJ9TXEUME8q5LD8VqXIr4XFwmjsVkMLE5Yf5aKUDkcDgcTsNFzAm6mnhyhKxxJY4QMgfAdgC/AvACYAGgA4B4AH8SQr4RbG8GYAsqknLTADRFRcJwECo87U4SQoZXs7/JhJAIQkhERkaGqs04BkiFkXEAY2SckVeCpQdjdDirmvn2YAwjd+Rhb46xvJKWw6l3BrVxhpVCIvxhfimO3EyvZoR4EHafvNC+OWwaac843crMGAuGsD5qFxKylBbetUFucRl+PMZKS73c2R1ugup2MfJmD094N2FlGOftjNSbRLGYiBF0cPo5aydBB1R4MwYJvBk/2xOFwtJyFSMaHnnFZZi4IRz/3WQTlw6WptgyuSuCPey0Mg87CxOseqUDloxqC0tBIVR2YRmm/nEJH2y7hvwSzb13yVmF+GRnJBPzc7LCR4NaqhjRMDEzNsKbPVj1iF/P3kWeyAvtakNJuUwpUTu2o7L8Kkf8eDpaYs34YOa6s1Qmx+SNERqRTE/PLcaLq88reX15N7HE9qkhcLfnSV2Odpk/pBVz3ZRfUo6Fe6M19vqUUnx1gFWS6OjRmPtzcjgcDqfeEXOC7kmHnCptKGvBdlVCCOkF4GsAeyilsyil8ZTSQkrpZQAjAaQAeJ8Q4qkw7EmH3TxK6WpK6QNKaS6ldD8qOuuMASxXtU9K6RpKaTClNNjR0bGGP5NjaHg5WmJqT08m9vv5RFy/V79GxupyOSkb+wXV2h+E+XMfCg5HC5ibSDFM4B+1+WKSiq3Fw/V7j3BJII81IURznk3PyqA2zujpyx5nF/97E48KtStJtvpEHB4VVi7gWppK8W4fb63OQV1MpBJ8MYKVDYpJy8P604bXNVLfxOiogw6o8CX6cmQbpsI75VERlh/RvTejGMjML8G4tRdwPp7tLHnSBdKqmXa7QAghGB3siv0zeihJHQPAtkv3MHD5SVzUQDdruUyOGZuvIE8h4WcilWDFS+24n04VvNzZnelKyisurzcZNV1wODqNke20aWSMAQHOOpwRpy50amGHJaPbMrFHhWWY9OtFZOaXqP26yVmFGL36nFLhSdvmNtg6pSucbbgsLkf7OFmb4f3+rMPNvuv3cSJWM0XxByIf4JogIT13oD8I4d1zHA6Hw6lfalyBJ4Qs0NStlnN70nKkymPuiSaTKo+6Jwx5fH9M+ASltBBAOCr+H9o945hrALIAuBNC7IXPczgA8HZvb8YXoMLIOJLpUhMDlFIlv6HA5jYY1IZfqHM42kJYuX4iNgOpj4pUbC0Ofj17l3ncw8cB3k20l4x4QkXXcmuYSitPZ7IKSvH1Ae11LaflFmOdIJk1OdQT9pamKkaIjy6e9kreGsv+u81IdnJqJlawkOmrxQ46AGjVzBqTBN6M604l4NaDXK3OQ2ykPirC6NXncCOFrenzdaroAvHQobSfq505tkzpijkD/GBsxC4AJmcV4cXV5/DNgVsoLVe/o/WHo3dwOYldcPx4oL9WOzz1CQtTKSaFsF10604loKjUMNwNhPKWI9u58EStnjM8yAXv92OXTBIzCzF54yUUl9X+c3s7LQ+jVp1FYmYhE+/cwg5/vtEZdhYmdZovh1MXxnf1QBuBv/OC3ZFqfdYVKZPJsUSgetS/lRM6uGunu57D4XA4DZtnaZH5DMCnGrrVhifJsf6EEGaehBArAN0AFAE4X8PrPFkhU9XK9iSuWG6vcgwhxBSV3XuG5xrO0QhVGRnfSMnBHyKrwD0ek6HkU/AhrxLjcLRKm+Y2aK3QvUEpsC3ing5nVD0ZeSXYe+0+E5vUzUM3kwHgbm+Bd3qz3WqbwpOUOvzqi2X/3UZxWeXiuYOlKV7v3qKaEeLk40H+jERpUZkMn+2J0uGM9IuH+SV4mF95WmgilcDdTvsSpzP7+aKpQmdDuZxi3s5IyEVWIKQt4jLyMerns4jPKGDiQa62oukCMZIQvN3LGzvf7qYkNyunwMrjcRi58gxuCxLAz0LE3Sz8cJTtouzt54gJgkQuh2ViiAcsTCqTVpkFpXrR3V4TyVmFOH3nIRMb24lL2hsC7/TxxqgObKHNpcRsvL/tWq1+/6/fe4Qxq88hLZftvuvr3wS/vdYJVmbakzLncKrCSEKweGQAFJcrEjMLsVIgNV9btkYkI/5h5bmChABzBojbS5rD4XA4hsOzJOhOavD2zFBK4wAcAuCBCh84RT5HhY/c75TSAgAghBgTQvwJIV6CbU89vp9MCHFRfIIQMhAVib5iAGerGPPx44ScIp8BkAK4SCmt/ZUyp8HQ09cRgwVGxt8ejEG6ho2M1UUmp/haoLHe09cRIV4OOpoRh9NwEXo+bo1IFl3H7RM2hSehVMGfzMPeHL18devNMLmnJzwd2S6YeTtvoLyefdTiMvKxNYLtRpjxnA8sBL5S+oC9pSnmDvRnYoej03A4Ok3FCI4iwu45b0dLSHUgFW1pKsWnQ1lvxkuJ2Uqf04ZAZEoORq86h9Qc9ryru7cD/nyjM2zNxdUFEuBig73vdq+y4CEqNRdDfjiNDWcSnnmxPaeoDDM2X4Xi5g6WplgyOpAXYtWAjbkxXunKyjavORlfp05GMbDlIvs7EORqC39n7cq7cuoHQipkjkO8WIGffdfv45tn9EI/F5eJcWsvILuQ9VwcHtQMq8Z34J2WHNHQtrktxndhf6NXnYhHnJrei4Wl5Vj2H1vMMibYVSfqIBwOh8NpmNS4ckAp7UUp7a2JmxrzextAOoAVhJBdhJD/EUKOAngPFdKW8xS2dQFwE8ARwWtsB/AfACcANwkhvxFCviaE7AGwDwABMJdSmqkwZjGAewD6ArhFCPmZEPIdIeQCKvzpigDMUOPv4TQwFgxpBUuFhdq8knIs2ndThzOqZNeVFNxS8MshBPhwgH81IzgcTn0xLMiFkWlMeVSkVOUuBkrL5UpePBNCPCCR6Hax11RqhC8EXcu3HuQpSXFqmm8PxjCJVA97c6Vkqz7xYrAr2rvZMrHP9kShsLRcxQjOE2KF/nM6lA8Ma+2E51qySfP/7b9VJz8ifeN8fCbGrjnPeG0BwMAAZ6ybGCzaJLqZsRE+HdoaG1/vBCdrtkawpFyOz/+JxoQN4XiQU32xF6UUn+yKVJKp/XZ0WzjokfyuLnmjuydzXL6fU4ydV8Tb3V4T5TI5tl1iE3T6fLziKGMileDnVzrAR9CJu+pEHDaFV98BeuRmGiZsCEd+CXu8f6WLG74fE8S9yTmi4/3+fszxrFQmx/xdkaC09gWO608nICOv8hzJVCrBzOdUOe1wOBwOh6N5RH2m9biLLhjArwA6A3gfgBeAFQC6CpJqql5DDmAQKpJ60QBGPn6dLgD+BRBGKV0uGJMCoD2ApajorpsE4B0Azo/n0p5Seq7OfyDH4KnKyPifa6k4dVszRsbqUlwmw3eHWfvGEUEuaNWMV9FyOLrAppExBrdhO263iFBOa3/kfeYC1sLESElSSVeEeDtgRFAzJvbd4dh68/O7nJSN/ZEPmNgHYf56vYglkRB8MaINjBQSrimPirDiSN1kgxoCMWls1bYuE3SEEHw2rDUaKXQ75BSVYfG/4igQqm8OR6fh1fXKC80vBrvix3HtYSoVfxdIDx9HHJwZiiECJQYAOHX7IcKWncQ/11JVjt9xOUXp+de6tUAvP912O+sTjlamSgmsn4/H1Xtndn1xPCaDkS20MDHC0MBm1Yzg6CM2jYyxYVJHpUT8J7sicSK26uvP3VdTMGXjJaUO0bd7eWHR8ACdF2FxOFVh08gY84e0ZGJn4zKxp5pjY1VkFZRi1Yl4JvZa9xaikMDmcDgcTsNB9KtIlNJkSukkSmlTSqkJpdSdUjqDUpol2O4upZRQSj2qeI0ySukySmkXSqk1pVRKKW1CKR1CKT2kYr8ZlNLZlNKWlFIzhX1PopTeqmoMh1MV47u4M/5SADB/V92NjOvCH+cTmapqEyMJZvXjVWIcji4Z28mNeXw4Og0PRdbxsuHMXebx6GBXUfmRzBvcClZmlZ0xhaUyLPwnWuP7oZTiq/3sqUBgcxsMauOs8X1pm1bNrDFJ4E/1y6l4xDzgqt7VIZS49HPSrSxS88bmmPGcDxPbcTkFZ+PE15mrSXZcvoepfygvNE8J9cRXL7DJZ7Fja26CH15qh+Vjg5jfNaAi4frupiuYufkKcopYObrEzAIs2B3JxPydrbiXjhpM7ukFqcJn5m5mIf4VFGboC5sF8pZDA5uJtpOUUzeaNzbHugnBMDOuXOqRySmm/XkZN+/nMttuPJ+ImVuuolwgnTt3oD/mDOC+5BxxMyywGbp5s7Kui/ZGKx0Xq+OnY3eYgh6bRsaY2lPomsPhcDgcTv0i+gQdh6PvSI0kWDyyDWNkfDezEKtOxOlkPjlFZfhRYKL8Shd3uNqZ62Q+HA6ngo4ejeHpUOmjViaj2HFZPHJaV5KycTX5ERN7VeDRo2scrUwxRyDVeyDqAY7e0qyP2vGYDIQnMHVC+HCg4Sxkzezni6YKlcPlcopPdt14Zu+rhgalVEni0leHHXRPeL17C6VE4Se7IlFSrrsCofpk/ekEzNp6Tcm/88MB/vhoUEu9/H4SQjA8yAUHZoaii6ed0vO7rqZi4LKTTxOvZTI5pm++ioLSyvfYVCrBDy+14/5RauBi2wgj2zEW5lh57I7e/Ram5RbjWEw6ExMWBXEMi0BXW6wY2465/swvKcdrv15E2mM/9JXH7zyWBKzchhDgy5FteIKCoxcQQrBoeABMFNQrHuaX4ttn9F1MzirExnOsdP87vb1h00g8xYccDofDaRjUOUFHCAkhhMwhhKwghKwjhKxXcVuniQlzOPpIkKstXu7MXgivPBaHhIcFWp/L6hNxeKRg/m1pKsU7fby1Pg8Oh8NCCMGLAjmtzReT1fJSqA9+E/i59fJzhKejZdUb65BxndwQ6Mr6qC3YHYWiUs0kJWRyiq8PsN1zPX0dEeLloJHXFwOWplJ8OrQVE7t4NxvbRZQwFhOpOcXIU6i+tjSVopkIpJGMjSRYPJL1ZozPKMDak/EqRugnlFJ8dzgWC/ey3bJPFprf6qX/C80uto3w1xtdMG9QS2YhEqj4/L38ywUs3heNbw/G4JqgkOKTIa3go+OOTn3mrV5eUGy8vPUgD0dupaseIEK2RSQziWt/ZysENrfR4Yw42qB/a2d8Mpg9lt/PKcZrv17EF3uj8c0BNokhlRCsGNsO4zrz5C1Hf/B0tMRUwXH+jwuJSkWFVfH94ViUKsgWN7Mxw3iRFR9yOBwOp2GgdoKOEOJDCLkI4BSA/wGYBmDi49sEwe1JnMNpsHwQ5g8HS5Onj0tlcizYrZ6Rsbo8yCnG+jMJTGxqT0/YWZioGMHhcLTJCx2aM3Ja8RkFiICwNx4AACAASURBVEjM1uGMKkjPLca+G/eZ2KRuLXQ0m+oxkhAsHhHALKjeyy7Cj8dua+T1d11JwS2FbilCKjp0DI2w1s7o7efIxP73701kF5TqaEbiRal7zslSNN1awR52Sj5aPxy9g8RM7RcI1QdyOcVne6Kw4gj7/TY2IvjhJcNaaJZICN4M9cSed7vBX9ChSSmw9lQCVguSr8+1dMIrBvR/oAs8HS0xSOAR++OxO6IpnqkJuZxiSwQrbzm2o6tofqM49ctr3TwwQZBwiErNxS+n2etBU6kEa18N5r6EHL3k7V5ecLevVAOiFJi380a1nqHRqbnYeTWFic3q78e7zTkcDoejE9RK0BFC7AEcBdABQDqAbQAIgGIAfwA4AiD/cSwTwG8AftfAfDkcvcWmkbFSFeOp2w/xz/X7KkZonuVHYlFcVnmi6mhlite6i3ORncNpiDhYmqJfKycmtjk8WcXW2uOPC0kok1UuRno6WqCHt3g7xgJcbDBB4KO25mQ87qTXzUetuEyG7w7HMrERQS5oJfAZNQQIIVg4PACm0spTxezCMiXvPQ4QI/SfE4G8pSIfDvBnCnFKyuWYvztKbxIMqiiTyTFr61X8JpCnamRshF8mdMSQtoa50OzvbI3d73TD5FBPVJdjaWJlim9GteWJGA0wrTerNHEt+RHOxmXqaDa142xcJpKzFHynpRKMbNdchzPiaBNCCBYMbY2+/k1UbmNpKsXvr3VC72q24XDEjJmxERYNZxUDolJzsfF8oooRwDcHbzHyrn5OVkqSxhwOh8PhaAt1O+hmAnABcAGAF6V07ON4DqX0VUppfwDNACwB4ACgiFI6qc6z5XD0nOFBzRDipWxknFv87EbG6nInPR9bBAbxM/r6wNyEG8RzOGJCKHO570ZqrczONU1JuQx/XWAvcCeGeEAiEfei76x+vnCyNn36uExGMW9n3bqW/zifiJRHCgudRhLM6udbp3mKGVc7c0zv68PEtkQkI+JulooRDRPlDjpxJegaW5jg40EtmdjJ2Aylrlh9orhMhikbL2HX1VQmbm0mxR9vdEZPX0cVIw0DU6kRPh7UEn+90QUuto2q3GbpmECukKAhWja1Vkpw/Hj0joqtxcXmi0nM40EBzrAx5/5KDQkjCcGKl9ohwEW5mMjOwgSb3uyCzp72VYzkcPSHUF9HDGnLdjsvPRT71HNRkbNxD3E8JoOJfTjQD0Yiv7bhcDgcjuGiboJuMAAK4GNKaWFVG1BKCyilHwJYDmAKIWS0mvvicAwGQggWjWCNjDPySrD0GY2M68KSg7eg6GnfwsFCKRHA4XB0Tw8fR2bBtbhMjj3XUqsZUb/su34fD/MrZQ2tTKV4vr34q++tzIyxYEhrJnYhIQs7LqeoGFE9OUVl+PEYuyD7Shd3uNqZqxhhGLzZwxPeTVivwXk7I1FWjWxQQ0PsHXQA8EJ7F3RuYcfEFv6jnQIhTZNbXIZX14XjqMAHzNHKFFundkUH98Y6mpn26eplj/0ze+B5QdX/5FBP9PAx7CSltpkm8Gs+F5+JSyKQoK6OrIJSHIpKY2JjO3HJ04aIhakU6yd0ZPxRm9qYYeuUrmjD/Qg5BsL8Ia1gaVpZfJxfUo5FAn9aSim+FqhBdGphh95+vIOUw+FwOLpD3QSdFyoSdKcE8arKNL96fD9ZzX1xOAaFl6Mlpvb0ZGIbzyfi+r2ajYzV5VJiNg4KLtA/CPODsZHaNpQcDqeeMJIQjA5mE2BbBBXw2oJSig1n7jKx0cGuzMWvmBnUxhmhgk6aL/+9iUeFtfdRW30iDo8KK5MZlqZSvCNYsDVETKQSJdmgmLQ8bBD4mTZUymVy3E7PZ2J+IuugAyoKhBaPDICxUWV1eHpeCb47FFvNKPGRkVeCsavPI1zQxelmZ46/p4bA39nw5GZrwtrMGN+9GIT1E4PxfHsXLBjSCnMN0BdT17R3a6ykgvHTMXF30e24fA+lCsUULRwslBL1nIZDE2sz7JzWDRNDPPBatxbYNa2bUgEOh6PPOFmb4f3+rLLF3uv3cTK2sltuf+QDXLuXw2wzd6A/l4PmcDgcjk5Rd3XeGBVyluUKsUIASisSlNI0ADkA2qq5Lw7H4Hi7tzfcFLou5LSiI0Em17wfTFVVYoGuthgY4KzxfXE4HM0wOtiV8RaKTMlFZEqO6gH1xOWkR7ihsF9CgAkh7lqfh7oQQrBoeGuYKPioZRaU4ptadi0/yCnGekFCampPzwYjH9fVyx7Pt2c7dL4/fJuR+2yoJGYVorS8cgHcwdIE9pam1YzQHd5NrDAl1IuJ/Xbubr0WCGmSe9mFGLP6HKLv5zJxPycrbJ/aFW72ht3NWhN9/J3w3ZggvNa9hegliPWVdwRedEdvpSMqVfvH5meBUorNAmn7Fzu68kXoBo6TtRk+G9YaC4a2gpO1Wc0DOBw9Y3wXdyU51/m7I1FcJkOZTI4lgmuAAa2d0d6t4XTeczgcDkecqJugSwVgQwhRLKFPAyAlhDCtQYQQYwDWALh2AofzGDNjIywczkqv3UjJwZ8XVBsZq8vRW+lKleZzB/AqMQ5HzLjYNkKoQJ5M6COjDYRdUn39m8Dd3kLr86gL7vYWSouqf11IwuWkZ5cmW34kFsVllUkYRytTvNa9hcbmqA98PKglbBpV+hYVlcnw+Z4oHc5IHIjdf07IO33YAiFKgY933qiXAiFNcjstD6N+PoeEhwVMvL2bLbZO6YomfKGZowW6etkjyNWWia08Hqej2VTPpcRs3FHo7pVKCF7QA3lqDofDqQtSIwkWj2jDFDomZhZi5fE4bLmYzJxHGEkIPhjgp4NZcjgcDofDom6CLgEAAaBoYHXx8f0rgm0nPt6PeqYvHI6B0suvCQa3YY2MlxyIQXoVRsbqIpNTfH2A7Z7r5eeIrl7cCJzDETtjBR6Ru6+koqhUprX9388pwv7IB0xsYoh+JqWm9PSEpwObWJy3MxLlz+Cjdic9H1sEXQgzn/OBuYl+yHxqCgdLU3wokM07FJ2G/6LTVIxoGAj958SeoKuqQCgyJRcbz93VyXyehWvJjzBm9Tk8EJwf9fBxwB9vdIaNubGKkRyOZiGEKBV8/HvjPuIy8lWM0B3C7rnnWjrB0Uqc3b0cDoejSQJdbfFKZ1bxY9XxOHx3mJX1HhPsCi9HLvPK4XA4HN2jboLu0OP7AQqxjahI2n1CCPmJEPImIeRHAD+iwq9ul/rT5HAME6GRcV5JOb7Yd1Njr7/j8j3EplUuGhACzAnjviQcjj7Qt6UTHCwrJRTzSsrx7437Wtv/n+eTmK4a7yaW6Oatn8l9U6kRvhjB+qjdvJ+LX8/erXHskoO3oNhc5OlggTHBrqoHGDBjO7qinRvbPfLpnigUlparGGH4xAoSdH7O4k7QAY8LhNqyBULfHopFmgYLhDTF2TsPMW7teWQr+D8CwOA2TbFuQscGlyjn6J6+LZvAX+F7Tinws8i66HKLy7D3eioTe7FTwzxucTichsnsMD84KEiOl8rkyCqo9KA2M5Zg5nM+upgah8PhcDhKqHtVuxvANAA9APwMAJTSfYSQzQDGApiqsC0BcBPAwjrMk8MxSJxtzDCrny8W7o1+GttzLRVjgl3R3cehTq9dXCbD94IqsZFBLmjVzFrFCA6HIyZMpBK80L45Vp+MfxrbcjEZL3Sof4mq4jIZ/gpnJTUnhnjotTRuiLcDhgc1w+6rlYuW3x+OxeC2TdHUplGVYy4lZuNgFNshNjvMD8ZG6tY36TcSCcHiEW0w9MfTT5O3KY+KsOLIHcwd2DCLP2Ie6F+CDgAWDGmFEzEZyC+pSK7ml5TjrT8uicqHpVQmx+bwZJQKOl1f6uSKL0a0gRH3WePoAEIIpvX2xrubrjyN7bqSgpnP+aB5Y3H4IO65msrIMlclm83hcDiGjE0jY8wf0hIzNl+t8vnXu7fgPowcDofDEQ1qJegopbEAPKp46mUAxwC8iAr5yxwABwAspZSK00Gbw9Exr3Z1x9+X7yEqNfdpbP7uSOyf0QNmxkZqv+7Gc4lIzamshjcxkuC9fr51miuHw9EuYzq6Mgm68LtZuJOeD+8m9SvH8s+1VKbK1NpMiufbu9TrPrXBvMEtcfRWOvKKK5ISBaUyLPwnGj+/0kFpW0opvt7PSgQHutpiYICzVuYqVlo1s8bEEA+sO13pT/jLqXiM6+QGN3txLE5ri+IyGe5mFjIxn3r+bmoKJ2szzO7vi8/+qSwQupz0CJeTHulwVjXzVi8vzAnz0+tiAY7+M6hNU3x3OPapl1G5nGLNyXgsHB5Qw0jtIPSsHR3cnCe0ORxOg2NYYDNsjUjGmTuZTNzW3BhTenrpaFYcDofD4Sij0RJwWsFaSulzlFI/SmknSukCnpzjcFQjNZJg8UjWyDjhYQFWn4hXPagGcorK8OOxO0xsfFd3uNo1rMVTDkff8XK0RCcPOya2NSJZxdaagVKqJP04tpObQUjJNbEyw5ww1gx+f+QDHLuVrrTt0VvpCL+bxcTmDvDniQEA7/XzhbNC1XG5nOJg1INqRhgmcRn5jAysi20jWJnpjx/a+K4eaONio+tpPDMfDfTHh/w7yBEBRhKCtwSLu5svJiM9T/cysZEpOYhMqSz6IwQY3UBlmTkcTsOGEIKFwwNgIlC+eKe3N6z16HyNw+FwOIZPw9Ro4nBERpCrLV7u/H/27jy+rrLa//hnZW7TpOkcmg5pC22BMrdhukIZZBCUQRTwMtWriOJFAS+iXuGi8kP0XkUUB7xiRZkEmS4olFnmthRlTkvblM5T2mZo5rN+f5yTNHtnPjnJSU6+79crr9397Gfv5zm8SpOcddZaUwJjtz3/IWWxT+b21K9fWMmumj39WvKyM7g81NReRAaH80J9Y/7yxjrqGyMdzO69JWU7Ahm9aQYXHjG1kzsGl88dPpWDJgWDEtc9+g419U0t500R5+Yngtlz82eN48gZg7MHX6KNyM7g0mOmB8ZeX13ewezUNRj7z7WWnmb88NMHkJsVf7Z+f0gz+OHZB+jT7jKgnHlIERNH7vmgQn1jhN+9uLqTO/pHOHvu2JnjKCpov4yziEiqmzFuBP/R6sN5+0/M54IU+r1GRERSQ1wfhzezRcCdwIPuvrur+SLStf84eTZPvLOJbVXRsnL1jRG++8g73Pn5kh59WnzTrlrueCn4BsFl82cwOjcrofsVkf5x6py9uP7Rd1vKMm6vrueZ9zdz6gF79cl6C18J/vtx4r4TUir7Nj3NuPGsA/jUL16iOflpbXkNtz33Id+I/QL/4LJ1LN9c1XKPGVxz8tDssdaRw6cHMzuXriknEnHShlAZtdJNVYHzmRMGV4AOYP+JI3n8io/x4oqtgZ5VA0V6mnHkjDHsu5f658rAkpWRxpeOncH1j77bMvan19bw5fkzKBienJ+5d9c38sibGwJj581T9pyIDG1fPGY6c4pGsqmihhP2ndCrNiIiIiJ9Id56VScCJwC/NLMHgT+5+9OJ25bI0DNyWCb/edp+fP2+PY2MX1yxjcfe2sgnD5rY7efc8vRy6lpl14zPy2bB0cWJ3KqI9KNhWemceXARf3xtTcvYPUvW9kmAbv3OGp58d3Ng7JIU/PdjTtFILjqyOFDK8zd/X8mZh0xk0qjh/OSp5YH5Zx1cxH4TFSBobXZhPnnZGVTWRQPHO3c38OHWqkEZpIpX2wy6wdF/Lqx4bC7FY3OTvQ2RQefceZP5+bMrWj5cV13fxMJXyvj6icnp+fzXtze1/JsMMHZEFifsOyEpexERGUhUBUNERAayeEtc3gXsBkYAFwJPmtlaM7vJzPZL2O4AM5tkZneY2QYzqzOzMjO7xcxG9fA5p5nZIjNbZ2Y1ZrbKzO43syM7ucfM7GIze97MymP3rTazP5tZcn7zkpR2xsETOSr0w+P3H3uPitqGDu4I+nBLZZv+VF8/cWZK9I4SGcrODX0C/sUVW1m3I/EJ7H96bU2gp9asCXkcOT01f6G9+qSZjM/LbjlvaHL+8+F3+MMrZWzctaePUFZ6Gld+XN/yw9LTjLnFwR/FFg+xMpelm0IBugkK4ooMJTmZ6fzbvwTL/f7+5TKqWgXJ+tO9i4PlLT992CQy09XRQkRERERkIIvrJ3Z3vxCYQDQ49zQQAYqAa4C3zewNM7vCzMb1ZnNmNgN4A1gALAZ+CqwCvga8ambdetfQzG4GHgMOBZ4AfgYsA84AXjazC9q5Jwd4FFgIFAJ3A7cAfwfmAnq3ThLOzPj+mcFGxlsq6/jJouWd3LXHj54opdV760wfm8tn505K9DZFpJ/NKRrJAUV7+qa5w/1L1yV0jdqGJu4Jvbl3ydHFPSqxO5jk5WRy3SeDnyl6bVU5P36yNDB24ZFTU6rEZyLNmxYsczmUAnSVtQ2s31nTcp6eZkwfpyw0kaHmgiOmkJ+z54Nwu2oauKtVxnt/+XBLJUvX7AiMnTdvSgezRURERERkoIj7I3Xuvtvd73L3k4HJwH8AbwEGHEI0mLbezP7PzD5rZtmdPK4jvwTGA1e4+5nufq27Hx979izgxq4eYGaFwDeAzcB+7v6F2HPOAU6O7fd77dz6P8DpwE2x+77q7t9y94vdfTrwZByvR6RLM8aN4EvHBj+Ne+erZby9blen972xppxF7wVL011zyiwy9MlZkZQQzqK7f+naQLZbbz385np27t6TrVswPJMzDy5K2PMHotMO2IuP7TM2MNbY6r9pXnYGlx+3d39va9AoKQ4G6JaUleOeuL+TA1nrHoUAxWOGq6eJyBCUl5PJJUdPC4z99sXV1DY09es+7l0crKBxxPTRTFPpWhERERGRAS8h79y7+yZ3/x93PwQ4EPhvYAPRHnenAfcAG3vyTDObDpwElAG3hS5fD1QDF5pZV795TCX6Ol939y2hfT8HVAKBTL9Y5t5lwBLgO+4eIcTdu1dzUCQOlx+3N1NaZWxEHL7z8Nsdvhnv7vzwbx8Exg6eXMDJ+xf26T5FpP986uCJ5GTu+ba9YVctf1+xNSHPdvdAPzaIfvJ+WFZqBxzMjO+fMYesjPZ/HLps/gxG52b1864GjwMmjQz8t9u4q5Z1O2o6uSN1tO0/N3R674lI0IKjihne6vvltqo67g+VnO9LdY1N/GVZMKte2XMiIiIiIoNDwlNr3P0dd78GmAJ8HFhKNEttZKc3tnV87LgoHCBz90rgZWA4cEQXz1kB1AMlZhb4mLyZHQPkES3T2dr5RP/b/AHIN7MLzOxbZnapmemj9NLncjLT+d4Z+wfG3lq3i7tfb79kzjPvb2FJWbCszbWnzk7Z0nQiQ1F+TianHTAxMHbf4sS8AfjaqnI+aNVPK82ipR2HguKxuVw+v+239vF52Sw4urj/NzSIZGekc8jkgsDYkrKhUeYy3H9u5gQF6ESGqlG5Wfzr4cGA2K9fWEVDU5vPePaJp97bzI5WGfAjh2Vyyhx9SE9EREREZDDok9p3ZrYXcBXRMpGHxfmYWbFjR823VsSOnfaCc/dy4JtEe+a9Z2a3m9lNZvZnYBHwFPCl0G3zYseRwErgj8D/A34DLDez28wstdMKJOnmzxrPaQfsFRj70ROlbKmsDYw1RZybnwhmzx0/ezxHTO9Wi0YRGUTOLwmWuXz6/c1srazr9XMXvrI6cH7y/oUUFQzr9XMHi8vmT2d6qBTY10+cyfCsjA7ukGYlQ7QPXZsMOgXoRIa0L35seiCjeP3OGh5+c32/rB0ub3nWIUUquSsiIiIiMkgkLEBnZsPM7F/N7EngI+BHRMtdGrAMuLKHj2zOuOuo8VbzeEEH11u4+y3A2URLbn4RuBb4DLAWWBgufUm07x1Ee9MtBQ4gmml3AtGA3VeA73a0XizTbqmZLd26NTHlx2Ro+u7p+5HbqmROZV0jNz7+fmDOX5atY8WWPb1wzKK950Qk9Rw2dRQzxu0JJDVGvE1Zq55aW76bp0L9Ky85qrhXzxxssjPSueW8gxk5LBOAj+83gc/OnZTkXQ0O80J96BYPkQw6lbgUkdbG5+e0+b7xqxdWJrRXbHvWlu/mpQ+3BcbCPWtFRERERGTg6nWAzsxONLM/AJuBO4mWtUwH1gM3A/u7+1x3v7W3a4WXjh27/K3HzK4BHgAWAjOAXKKZfauAu8zsR6FbmiMiG4GzYmU7q9z9WeAcIAJcZWbtNqZx99tjr3nuuHHj2psi0i2FI3O4+qRgsO2Rf2zgpRXRX8RrG5r46VPBJNOzD5nE7ML8ftujiPQfM2vTV+a+JWtxj/8NwD+9tobW7x/ut1d+m6yooeDASQW8/u0TeOmbx/Hbi+aSkd4nRQZSzqFTR5HWqpryqq3VbKvqfVbnQLatqo5tVfUt51kZaUwd01VLZBFJdV86Zgbprf5BXLW1mife2dSna963JJg9d9DkAvbdS78HiIiIiIgMFnG9+2Rmc8zsZjNbCzwJXACMAKqJloP8ODDV3b/l7u938qjONGfIddS7Lj80r6O9zicaKHzU3a9y91XuvtvdlwFnEQ0kXm1m01vd1tzM6wl3r2n9PHf/J7CaaEbdvt19MSLxuujIqewX+kX7u4+8Q21DE394pYyNu/aUvMzKSOOqkzqt+ioig9zZhxaRmb7nDcDV26rjLiu4u76RexZ/FBi75OjiIdu/MicznUmjhid7G4PKiOwM5hQFf1RbkuJlLpeH+s/tM35E4E15ERmaJo8ezhkHB3vF/uK5D3v1IZrONDZFuP+NYIDufGXPiYiIiIgMKvF+PPwt4BtAEdEMtmeAi4BCd7/Y3Z/x3v8mUho7dhRt2Cd27KhHXbPTY8fnwhfcfTewmOh/h0PaWXtnB89sDuANnQY9kjQZ6WnceNYcWr9fvnpbNT9+spTbnvswMPfiI6cOqb5RIkPRmBHZnLRfYWAs/An67nrozfVU1Da2nI/OzeJTB03s5A6RtoZamctS9Z8TkQ58Zf7egZ/Z399YwfOlfdPy4PnSrWyu2JOxnJuVzif1PVxEREREZFDpTf2md4FvApPd/SR3/1Ms4JUozQG1k8wssE8zywOOBmqA17p4Tnbs2FGtyebx+lZjz8SOc8KTzSybPcHBsi7WFkmIQ6aM4nMlwbJ2v3tpdeCN9bycDL4yf+/+3pqIJEG4v8zjb29k1+6GHj3D3Vn4cllg7PySyeRkprd/g0gHwgG6JSkeoAv3n5up/nMiErP3+BGcOif4IZq+yqK7d0kwA/6TB00kNzsj4euIiIiIiEjfiTdAd5i7H+juP3b3jfEubmZFZjalvWvuvhJYBBQDl4cu30C0j9yd7l4de1ammc02sxmhuS/GjpeaWVFo/VOJBvpqgVdaXfob0f50J5vZx0PP+y7RspsvuHvfNhUQaeWak2czdkS7bQ8B+PL8GYzK7fi6iKSOf9l7bCBbtq4xwiP/XN+jZ7yycjsrtlS1nKenGRccMTVhe5ShY17xqMD5exsqqKztWcB4MCndpAw6EelY+ANzb6zZwWurEvvBhU27ann2gy2BsfNK2v21WkREREREBrC4AnTu/maC1l9KNBDWka8AW4BbzexhM7vJzJ4FriRa2vI7reYWAe+zJ/ut2QPA08AE4H0z+0Osf96jwOOAAde6+/bmG9y9HriYaODub2Z2v5n9t5m9EFtzK3Bp3K9aJA4jh2fyndPab3s4IT+bBUdN6+cdiUiypKVZmyy6exav7dEn9H8fyp47ZU4he41UiVzpuTEjstl7/IiW84hH35BORe7O8s1VgbFZyqATkVbmFI1k/qxg8ZZwWfreeuCNtURafcufXZjHQZM6at0uIiIiIiIDVW9KXCaKdXQhlkU3F1gIHA5cDcwAbgWObB1U6+QZEeATRIN67wFnxZ5zBPBX4GR3/1k7970UW/svwLHAFcB04HbgUHfvqvedSMKdeXARR04f02b8yhNnMixLZelEhpJzDptEWqjPzTvrK7p175rt1TzzwebA2OePLk7g7mSoGSplLjfsqqWqrlV56ewM9hqZk8QdichA9NXjgll0L324jX+s7ai9ec9EIs59S4O9Z8+bNxmzDn+tFhERERGRAWogBOg65e5r3X2Bu+/l7lnuPtXdv+bu5aF5Ze5u7l7czjMa3P0Wdz/C3fPdPcPdx7v76e6+qJO133P3c2Nzs9x9srt/yd3X9cFLFemSmfH9M+eQmb7nF/AZ43I557BJSdyViCTDxIJhHDsz+An9e0L9aDpy56traJ1sd0DRSA6dMqrjG0S6UDIt+PdnyerUzKAr3RQMgs8szNOb4iLSxtzi0Rw+LfjBhURl0b2ycjtry2tazrMy0jjzkKJO7hARERERkYFqwAfoRCRo7/Ej+Pn5hzJ9XC4HTy7gNxfOJSNd/yuLDEXnzgv2m3n0HxvYXd/Yweyo6rpG/rwk+Mn7S44qVpBBeiWcQfePdTupbWhK0m76TummYHnLmeo/JyIduDyURffUe5v5YFP3Mt07c2/owzifmFNIwXD1oRYRERERGYz0rr7IIHTKnEKevXo+D19+dKDvj4gMLSfsO56xI7JbzqvqGnn8rY2d3vPgsnVUtirRN3ZEFqcftFef7VGGhkmjhlNUsKeHYX1jhLfW7UrijvrG8s2VgfNZE/Q9WETa97F9xnJgqC/cL59b2atnllfXs+jdYInq8Id1RERERERk8FCATkREZJDKTE9rU+L2vlB2XGuRiLPwlbLA2OdKppCdoR6W0nvzikNlLlOwD13ppmCAbmahMuhEpH1m1iaL7rG3NlC2rTruZz64bB31TZGW82ljczli+uhO7hARERERkYFMAToREZFB7Nx5kwPnS9fsYEUoy6fZSx9uY+XWPW8MZqQZFxwxtU/3J0PHvFC/pcWrUytA19gU4cOtwRKXs1TiUkQ68fF9JzCzVaZtxOHXL8SXRefu3Bv6EM658yarRLWIiIiIyCCmAJ2IiMggNm1sLoeHZ83VtAAAIABJREFUAiMdZdH9/uXVgfPTDtyL8fk5fbY3GVpKQn3o3lizg6aIJ2k3ibemfDf1jXsyV8aOyGZMqxKzIiJhaWnGV+YHs+j+smwdG3bW9PhZb6zZwYdb9nxIICPN+PShkzq5Q0REREREBjoF6ERERAa580uC/WcefHM9dY1NgbHV26p5rnRrYOySo4r7emsyhOw9fgSjc7NazqvqGnl/Y0USd5RY4fKWswrVf05Eunb6gXsxZfTwlvOGJuf2v6/q8XPC2XMn7juBcXn6kICIiIiIyGCmAJ2IiMggd8qcQvJzMlrOy6vrefq9LYE5fwj1njtocgGHTAn2DBPpDTNj7tTg36lUKnPZpv+cyluKSDdkpKfx5fkzAmP3LvmIbVV13X5GRW0Dj721ITB2bsnkDmaLiIiIiMhgoQCdiIjIIJeTmc5ZhxQFxu5d8lHLnytrG3jgjXWB6wuUPSd9oCRUbnVJWeoE6JaHejuq/5yIdNfZhxZR2KqkdG1DhDteWt3JHUGP/mMDtQ17SuxOHJnDMfuMS+geRURERESk/ylAJyIikgLOnRcsc/niim2sLd8NwF/eWEdVXWPLtXF52XzigL36dX8yNMwrbhugc0+NPnSloQDdzEIF6ESke7Iz0vniMdMDY398dQ27ahq6dX/rD90AfGbuZNLTLGH7ExERERGR5Eh2gG4d8FGXs0RERKRT+03M56BJIwNj9y9dSyTi/OHVNYHxCw6fSlZGsn8EkFS0/8R8hmelt5xvq6pn1bbqJO4oMWobmigLvQ6VuBSRnji/ZHKgT2dlXSN3hspPt+ed9bt4Z/2efp5m8Nl5Km8pIiIiIpIKkvrunLvPc/dpydyDiIhIqghn0f156Tqe+WALq1sFFjLTjc8dPiV8q0hCZKSncVioD92SFOhDt3JrFZFWiYCTRg1jRHZGxzeIiIQMz8rg3/4l+KvvHS+vZnd9Ywd3RIWz546dOY6igmEJ35+IiIiIiPS/XgXozGySmf3EzN41syozawxdH2Vm3zazb5mZPqovIiLShz550F4My9yTvbSpopZvPfh2cM6BExmXl93fW5MhJFzmcnEK9KEr3aT+cyLSexceOZW8VsH9HbsbuPv1jgvK7K5v5JE3NwTGzlP2nIiIiIhIyog7aGZmJwJvA18D9gWGA4FC+O6+AzgD+AFwbPzbFBERka7k5WRy+oHB3nLbquoC5xcfVdyPO5KhqL0+dIOd+s+JSCLk52Ry0VFTA2O/fXEVdY1N7c7/69ubqGzVQ3bsiCxO2HdCn+5RRERERET6T1wBOjObDDwAjAT+DzgH2NHB9DuIBu7OjGctERER6b7zSjouX3nolAIOmlzQj7uRoeiQKQVkpu/5zNba8ho27qpJ4o56b7ky6EQkQT5/9LRAtvvmijoeeGNdu3PvXRzMrvv0YZPITFdhGhERERGRVBHvT/dXAfnAn939THd/EKjvYO6TseO/xLmWiIiIdNOhUwrYZ/yIdq8tOFptX6Xv5WSmc+CkYCB48SDvQ7d8c1XgfKYCdCISpzEjsjk/9GGaX7+wksamSGDswy2VLF0T/AzsefPUQ1ZEREREJJXEG6A7GXDgu11NdPcyoA7Qu4IiIiJ9zMw4t53+NBPyszllTmESdiRDUSqVuaysbWD9zj0ZgOlpxozxuUnckYgMdpceM71NpvH/vRXsNXfv4rWB88OnjWbaWP3bIyIiIiKSSuIN0E0Fatx9RTfnVwHtf5y/C2Y2yczuMLMNZlZnZmVmdouZjerhc04zs0Vmts7MasxslZndb2ZHdvP+35mZx772jue1iIiI9IezD51EVqgE1oVHTFVZLOk3h08LBehWd1QJfeALZ89NG5tLdkZ6B7NFRLpWODKHcw6bFBj75XMriUQcgLrGJv6yLFj2Mpx1JyIiIiIig1+879RFunuvmWUS7VVX2dXcdu6dAbwBLAAWAz8FVgFfA141szHdfM7NwGPAocATwM+AZcAZwMtmdkEX938S+DzRQKOIiMiANjo3i08fVtRyXjA8U2/sSb86dOoobE9yCKWbK9lR3VE19IFt+Wb1nxORxLvs2Bmktfp3csWWKha9twmAp97bzI7dDS3X8nMylAUvIiIiIpKC4g3QrQWyzaxtDa22jgMygA/jWOeXwHjgilivu2vd/XiigbpZwI1dPcDMCoFvAJuB/dz9C7HnnEO0VKcB3+vk/nHAb4H7iAYLRUREBrzvnr4fXzp2Op86aCILF5QwZkR2srckQ8jIYZnMLswPjIV7KQ0WpZuCATr1nxORRJg6JpdPHTQxMHbbcytx9zblLc8+dBI5mcrcFRERERFJNfEG6J6JHb/U2SQzGwb8kGi/uid6soCZTQdOAsqA20KXrweqgQvNrKtC/FOJvs7X3X1L6wvu/hzRzL5xndx/e+x4efd2LiIiknzDszL41qn7cuv5h3Dw5IJkb0eGoJLiYDXywdqHLhygm1UYV9V2EZE2vnJcsHPC2+t3cdfrH/HSh9sC4+31lhURERERkcEv3gDdT4FG4BtmdnF7E8zsKODvwMHAbqLZcD1xfOy4yN0jrS+4eyXwMjAcOKKL56wA6oESMxsb2uMxQB7wdAev4RLgTOAyd9/ew/2LiIiIDFkl04KVyBevHpwBunCJS2XQiUiizJyQx0n7TQiMXf/ou4HzgyYXsO9ewYxkERERERFJDXEF6Nx9FfBVIAu4w8w2AKMAzOxxM/sIeBE4jGj23KXuvrmHy8yKHZd3cH1F7Dizi72WA98EJgDvmdntZnaTmf0ZWAQ8RTuZgGY2lWivuj+5+8M93LuIiIjIkDZvWjCD7p31u9hd35ik3cRnW1Ud21v1zsvOSGPqmK6KN4iIdN9Xjw9m0TVFPHB+vrLnRERERERSVrwZdLj7b4GzgY1AIdFgnQGnApNif94InO3u98SxxMjYcVcH15vHu6zb5e63xPaaAXwRuBb4DNFeegvDpS/NLA34A1AFXNHjnYuIiIgMcePzcigeM7zlvDHivPnRziTuqOeWh8pb7jNhBOlplqTdiEgqOnBSAR/bZ2y714ZnpXN6qE+diIiIiIikjrgDdACxzLJpRMtA/jdwN3Af0cyzzwLT3f2R3m6yA83vjninswAzuwZ4AFgIzAByiWb3rQLuMrMfhW65EjgW+KK77+jxxswuNbOlZrZ069atPb1dREREJCXMKx4dOB9sZS5LVd5SRPrBV0O96Jp96qCJjMjO6OfdiIiIiIhIf+n1T/vu3gA8GvtKpOYMuZEdXM8PzWuXmc0HbgYecverWl1aZmZnES2hebWZ/drdV5nZPsCNwO/d/a/xbNzdbwduB5g7d26XAUQRERGRVFQybTT3v7Gu5XxJ2SAL0IUy6GYpQCcifaBk2mjmTh3F0jXBz4aeq/KWIiIiIiIpLa4MOjPLSfRG2lEaO3bUY26f2LGjHnXNTo8dnwtfcPfdwGKi/x0OiQ3vD2QDC8zMW38RzaoDWBEbO7Mbr0NERERkSCqZFsygW/bRDuobI0naTc+1yaArVIBORBLPzLg81ItudmEeB0/uspuDiIiIiIgMYvFm0G0ysweAP7r7C4ncUCvNAbWTzCzN3VvezTGzPOBooAZ4rYvnZMeO4zq43jxeHzuWAb/rYO5pRPvt3Q9UxOaKiIiISDumjB7O+LxstlTWAVDbEOGdDbs4dMqoJO+sa+7epgedMuhEpK/MnzmO80smc8/itYzIzuDGs+Zgpp6XIiIiIiKpLN4AXT6wgGiW2Vrgj0SDdV1ls3Wbu680s0XAScDlwM9bXb6BaB+537h7NYCZZRLtL9fg7itbzX0R+CpwqZn9xt3XN18ws1OJBvpqgVdi6/4D+EJ7ezKz54kG6L7t7h8m4nWKiIiIpCozY9600Tz+1saWsSWrywdFgG79zhqq65tazvNyMthrZH8UkRCRocjMuOnsA7nyxJnkD8skJzM92VsSEREREZE+FleJS+CLQHPm3BTg28D7ZvaamX3ZzEZ3fGuPfAXYAtxqZg+b2U1m9ixwJdHSlt9pNbcIeB94JvSMB4CngQmxPf7BzG42s0eBxwEDrnX37Qnas4iIiIjEHB4qczlY+tAt39w2e07ZLCLS18bn5yg4JyIiIiIyRMQVoHP337n78UAx0SDZB0QDXSXAL4ANZvagmZ0Vy2yLSywTbi6wEDgcuJpoltytwJHdCarFSmN+gmhQ7z3grNhzjgD+Cpzs7j+Ld48iIiIi0rF5xeEA3Q4iEU/SbrqvdFNV4Fz950REREREREQkkeItcQmAu68FbgJuMrNDgYuA84DxwJnAGcAOM7uPaAnMrvrFdbTGgm7MKyMaJGzvWgNwS+wrbu4+vzf3i4iIiAw1sybkkZ+TQUVtIwC7ahpYvqWS2YX5Sd5Z50o3VQTO1X9ORERERERERBIp3hKXbbj7Mnf/OtFSk6cB9xHt7TYa+DLwUqLWEhEREZHBIS3NmBvOols98Mtclm4OZdApQCciIiIiIiIiCZSwAF0zd29y97+5+/nAQcDS2CU17RAREREZgkpCfegWl+1I0k66p7Epwsot4QDdiCTtRkRERERERERSUa9KXLbHzLKJlra8EDipL9YQERERkcEj3Idu8ertuDtmA/PzW2Xbd1PfFGk5HzsimzEjspO4IxERERERERFJNQkLnpnZMUSDcucA+ezJmNsE3A3cmai1RERERGTwOKBoJDmZadQ2RINemyvqWFtew5Qxw5O8s/Yt31wZOJ9dqPKWIiIiIiIiIpJYvQrQmdlM4CLgX4EpzcNADfAw0aDcU+4eaf8JIiIiIpLqsjLSOHhyAa+t2tN7bnFZ+YAN0JVuCgbo1H9ORERERERERBItrh50ZvZVM3sdeB/4FjA1dukF4N+ACe7+r+7+pIJzIiIiIlIybUzgfMnq8g5mJl84QDerUP3nRERERERERCSx4s2gu7XVn0uBPwJ/dPe1vd+SiIiIiKSaknAfurKBG6ALl7hUBp2IiIiIiIiIJFq8AbrtwL3Ane6+JIH7EREREZEUdMiUAtLTjKaIA7B6WzVbKmsZn5eT5J0F1TY0Uba9OjC2jwJ0IiIiIiIiIpJgcZW4BPZy939XcE5EREREuiM3O4M5E/MDY0vLdiRpNx37cEsVsRgiAJNGDWNEdq/aNouIiIiIiIiItBFXgM7dGxO9ERERERFJbSXTQmUuB2AfunB5y9mFyp4TERERERERkcSLN4NORERERKRH5oX70A3AAF2p+s+JiIiIiIiISD/osl6PmT0b++Mad18QGusJd/cT4rhPRERERFJAOED3/qYKKmobyM/JTNKO2lq+KRigm6UMOhERERERERHpA91pqDE/dvygnbGe8K6niIiIiEiqGpWbxT7jR7BiSxUA7vDGmh0cN2t8kne2R+kmZdCJiIiIiIiISN/rToDuhthxWztjIiIiIiLdVjJtdEuADqJlLgdKgK6itoENu2pbztPTjOnjcpO4IxERERERERFJVV0G6Ny9TTCuvTERERERka6UTBvNXa9/1HK+ZAD1oVsR6j83bWwu2RnpSdqNiIiIiIiIiKSytGRvQERERESGjnAfurfW7aK2oSlJuwkq3VQVOFf/ORERERERERHpK3EF6MzsOjO7qgfzrzCz6+JZS0RERERSx8SCYRQVDGs5r2+K8M+1O5O4oz2WhzLoZqn/nIiIiIiIiIj0kXgz6P4L+EYP5l8JXB/PQmY2yczuMLMNZlZnZmVmdouZjerhc04zs0Vmts7MasxslZndb2ZHtjN3HzP7ppk9a2ZrzazezDab2SNmdlw8r0NEREREog6fFsyiWzxAylyWbgoG6GYqQCciIiIiIiIifWRAl7g0sxnAG8ACYDHwU2AV8DXgVTMb083n3Aw8BhwKPAH8DFgGnAG8bGYXhG75PvBDYALwV+B/gJeB04BnzeyK3r0yERERkaFrXjhAV5b8AJ27UxrOoFOJSxERERERERHpIxn9tM5ooDaO+34JjAeucPefNw+a2U+IZuXdCFzW2QPMrJBott9m4EB339Lq2nHAs8D3gD+1uu0J4GZ3fzP0rGOBp4Afm9n97r4xjtckIiIiMqSF+9AtW7ODxqYIGenJ++zYtqp6yqvrW86zM9KYMnp40vYjIiIiIiIiIqmtz98FMbPPAHnARz28bzpwElAG3Ba6fD1QDVxoZrldPGoq0df5euvgHIC7PwdUAuNC4wvDwbnY+AvA80AWcFR3X4uIiIiI7DFjXC5jcrNazqvrm3h/Y2Und/S9cP+5fSaMID3NkrQbEREREREREUl13QrQmdnXYj3bVpnZqtjwuNZj7XytNrMdwL2AA4/3cG/Hx46L3D3S+oK7VxItOTkcOKKL56wA6oESMxsbel3HEA0ePt2DfTXEjo09uEdEREREYsysTRbd66u3J2k3UeH+c7Mm5CdpJyIiIiIiIiIyFHQ3g64AKG71BZAeGgt/TQVGAsaeMpI9MSt2XN7B9RWx48zOHuLu5cA3ifaTe8/Mbjezm8zsz8AioiUrv9SdDZnZVOAEYDfw9+7cIyIiIiJthfvQLUlyH7pwBt2swhFJ2omIiIiIiIiIDAXd7UH3MNFSkxANuN0B7AK+3sk9EaACeMfdV8axt5Gx464OrjePF3T1IHe/xczKiO77i60ufQgsDJe+bI+ZZQN3AdnANe6+o5O5lwKXAkyZMqWrR4uIiIgMOSWhDLqlZTtwd8ySU1ayNBSgmzkhLyn7EBEREREREZGhoVsBOnf/J/DP5nMzuwOocfc/9NXGuqH53RvvcqLZNcD/A24FfgFsAmYDNwF3mdnB7n5NJ/enA38EjgbuA/67s/Xc/XbgdoC5c+d2uT8RERGRoWbfvfLIzUqnur4JgO3V9azcWs3e4/s/cy0ScZaHS1wWKkAnIiIiIiIiIn2nuyUuA9w9zd0nJnozIc0ZciM7uJ4fmtcuM5sP3Aw86u5Xufsqd9/t7suAs4D1wNVmNr2D+9OBPwGfAf4MXODuCrqJiIiI9EJGehqHhbLoFq9OTpnL9TtrWgKFAHk5GRTm5yRlLyIiIiIiIiIyNMQVoOsnpbFjRz3m9okdO+pR1+z02PG58AV33w0sJvrf4ZDwdTPLAO4BzgPuBj7n7o1drCciIiIi3VBSPCpwnqw+dG36z03IS1qpTREREREREREZGuIK0JnZEWa2zMxu68bc/43NndvDZZoDaieZWWCfZpZHtNxkDfBaF8/Jjh3HdXC9ebw+tEYW8ADRzLk7gQvdvQkRERERSYh5AySDrk3/OZW3FBEREREREZE+Fm8G3eeAg4AXuzH3NeDg2D3d5u4rgUVAMXB56PINQC5wp7tXA5hZppnNNrMZobnNe7zUzIpaXzCzU4kG+mqBV1qNZwMPAWcAvwMWuHukJ/sXERERkc4dNLmArPQ9P46u31nD+p01/b6PcP+52QrQiYiIiIiIiEgfy4jzvmNjxxe6Mffx2PG4ONb5CtHA2a1mdgLwPnB47FnLge+0mlsUu76GaFCv2QPA08CJwPtm9hCwCdiXaPlLA6519+2t7vk18AlgG9Eedde1U+boeXd/Po7XJCIiIiJATmY6B00eyZKyHS1jS1aXU3RIUSd3JV7p5qrA+cwJCtCJiIiIiIiISN+KN0A3Cahz941dTXT3jWZWRzSA1iPuvjJWGvN7wClEg2YbgVuBG9y9yzpI7h4xs08QzcI7DzgLGA6UA38FbnX3RaHbpsWOY4HrOnn8891/NSIiIiISNq94dCBAt7isnDP7MUDX0BRh5RYF6ERERERERESkf8UboBtGqGdbF+qAuN7pcPe1wIJuzCsjmg3X3rUG4JbYV3fWnN/9HYqIiIhIvOZNGw3Pr2w5X9LPfejWbK+mvmlPJfNxedmMzs3q1z2IiIiIiIiIyNATbw+6LUCemU3samKs71s+0XKRIiIiIiItDps6itaVxFdsqaK8uiefA+ud0k3B7LlZyp4TERERERERkX4Qb4Dutdjx8m7MbZ7zepxriYiIiEiKys/JZL+98gNjS8r6L4uudHNl4FzlLUVERERERESkP8QboPsd0XKS15jZpR1NMrMvAdcAHrtHRERERCRgXvHowHl/lrlcvikYoJtdqACdiIiIiIiIiPS9uHrQuftTZvYAcA7wKzP7KvB/wBqiwbhi4JPA/kQDeX9x978lZMciIiIiklJKpo1m4StlLef9mUG3PJxBpwCdiIiIiIiIiPSDuAJ0MRcTDcZ9BphDNBjXWnM3kXuBf+vFOiIiIiKSwsIZdO9sqKC6rpHc7N78qNq12oYmyrZXB8b2GT+iT9cUEREREREREYH4S1zi7jXufi5wInA30ey5OqAWKAPuAo5398+5e00C9ioiIiIiKWhcXjbTx+a2nDdFnGUf7ejzdT/cUkXE95xPHj2sz4OCIiIiIiIiIiLQuww6ANz9WeDZBOxFRERERIaoecWjWbVtTzbbktXlfGyfcX26Zmmo/9ysCSpvKSIiIiIiIiL9I+4MOhERERGRRJk3LVjmcnE/9KFr039OAToRERERERER6ScJqeFjZuOAqcBwd/97Ip4pIiIiIkNHSagP3Zsf7aS+MUJWRt99nqw0FKCbVagAnYiIiIiIiIj0j16942FmnzKzZcAm4HVCpS7NbJSZPRH7ym33ISIiIiIy5E0ePYzC/JyW87rGCG+v39mnay4Pl7hUgE5ERERERERE+kncATozuxZ4CDgYsFZfLdx9B7Ab+Djwifi3KSIiIiKpzMzalrlcvaPP1quobWDDrtqW84w0Y/rYEX22noiIiIiIiIhIa3EF6MzscOBGoBG4EhgLbO5g+p+IBu4+Fc9aIiIiIjI0lBSPCpwv6cM+dOHsuWljc/u0nKaIiIiIiIiISGvx9qD7Wux4k7v/DKKfeu7AC7HjvDjXEhEREZEhIJxBt6SsnKaIk57W4c+ZcQv3n5up8pYiIiIiIiIi0o/i/Zjwv8SOv+hqortvB6qAojjXEhEREZEhYOb4PEYOy2w5r6xtpDSU6ZYobfrPTVCATkRERERERET6T7wBuvFApbtv6+b8BiArzrVEREREZAhISzPm9VOZy3AG3Sxl0ImIiIiIiIhIP4o3QLcbGG5mXd5vZvlAAbAjzrVEREREZIiYVxwsc7m4DwJ07t4mM08ZdCIiIiIiIiLSn+IN0C0H0oEDuzH304AB/4xnITObZGZ3mNkGM6szszIzu8XMRnV9d+A5p5nZIjNbZ2Y1ZrbKzO43syM7uecoM/urmZWb2W4ze8vMvm5m6fG8FhERERHpXEm4D93qctw9oWtsq6pnx+6GlvOczDQmjx6e0DVERERERERERDoTb4Du/4gG3a7tbJKZ7Q38EHDg4Z4uYmYzgDeABcBi4KfAKuBrwKtmNqabz7kZeAw4FHgC+BmwDDgDeNnMLmjnnjOAvwPHAA8BtxEt0/lT4N6evhYRERER6dqcopEMy9zzWagtlXWs2b47oWuEs+f2GZ9HepoldA0RERERERERkc7EG6D7ObAF+IyZ/d7MZre+aGbTzezbwBJgHFAG3BHHOr8k2u/uCnc/092vdffjiQbJZgE3dvUAMysEvgFsBvZz9y/EnnMOcDLRQOP3QvfkA78FmoD57v5v7v4fwMHAq8A5ZnZeHK9HRERERDqRmZ7GIVMKAmOJLnMZ7j83U+UtRURERERERKSfxRWgc/cKotlnFcBFwLtEA2mYWRWwAvg+MBLYDpzt7nU9WcPMpgMnEQ3u3Ra6fD1QDVxoZrldPGoq0df5urtvCb2O54BKokHE1s6Jjd3r7ktbza8F/jN2+uVuvxgRERER6bZwH7olqxMboFse7j9XOCKhzxcRERERERER6Uq8GXS4++tEM8oeJFrC0mJfw2NHiJa1LHH3ePrPHR87LnL3SGjtSuDl2FpHdPGcFUA9UGJmY1tfMLNjgDzg6Q7WfqKd5/0d2A0cZWbZXb0IEREREemZw8N96Po4g25WYX5Cny8iIiIiIiIi0pWM3tzs7muIlrkcBRwJTATSgU3AK+6+tRePnxU7Lu/g+gqiGXYzgWc62WO5mX0T+Anwnpk9TDSrbwbwKeAp4EvdXdvdG81sNbA/MB14v1uvRkRERES65ZApo8hIMxojDkDZ9t1sqahlfH5Or58diTgrwgE6lbgUERERERERkX7WqwBdM3ffAfw1Ec9qZWTsuKuD683jBR1cb+Hut5hZGdE+eF9sdelDYGG49GUi106EpqYmKioqqKyspKamhkgk0vVNMuSkpaUxbNgw8vLyyM/PJz09PdlbEhERicuwrHTmFI3kH2t3towtLivn9AMn9vrZ63fWUF3f1HKen5PBhHwVRRARERERERGR/pWQAF2SNJfR9C4nml0D/D/gVuAXRDP8ZgM3AXeZ2cHufk2i1jazS4FLAaZMmdKDx7ZVX1/PmjVrGD58OAUFBRQVFZGWloaZdX2zDBnuTiQSobq6msrKSrZt28bUqVPJyspK9tZERETiUjJtdCBAt2R1YgJ0pW36z+Xp5yoRERERERER6Xdx96DrB81ZaiM7uJ4fmtcuM5sP3Aw86u5Xufsqd9/t7suAs4D1wNVmNj1Ra7v77e4+193njhs3rrPtdaqpqYk1a9YwduxYioqKWrKi9CaShJkZ6enp5OfnU1RUxNixY1mzZg1NTU1d3ywiIjIAlRQH+9AtLtuRkOeG+8/NVHlLEREREREREUmCLjPozOzZ2B/XuPuC0FhP1QGbgVeBu929spO5pbHjzA6u7xM7dtSjrtnpseNz4QvuvtvMFhMN1B0CrGq19tzY2m+0vsfMMoBpQGOr+X2ioqKC4cOHM2rUqL5cRlLQqFGj2L17NxUVFfr7IyIig9Lc4uD3rw82VbCrpoGRwzJ79dzl4f5zhQrQiYiIiIiIiEj/606Jy/mx4wftjMXrQuDbZnasu5d1MKc5oHaSmaW5e0vW9swxAAAgAElEQVTjNTPLA44GaoDXulirualIR6lszeP1rcaeBf4VOAW4JzT/GGA48Hd3r+ti7V6prKykoKBf2txJCsrLy2Pnzp0K0ImIyKBUMDyLWRPyWjLe3OGNNeUcP3tCr57bpsSlMuhEREREREREJAm6E6C7IXbc1s5YPOsVEc1qm0S0L9zn2pvo7ivNbBFwEnA58PPQ+rnAb9y9GsDMMoEZQIO7r2w190Xgq8ClZvYbd1/ffMHMTiUa6KsFXml1zwNEy2KeZ2Y/d/elsfk5wA9ic37V85ffMzU1NRQVFfX1MpKicnNz2bhxY7K3ISIiErd500YFSlIuXr2jVwG6hqYIq7ZWB8ZU4lJEREREREREkqHLAJ27twnGtTfWE2Z2APBP4IQupn6FaODsVjM7AXgfOBw4jmhpy++0mlsUu74GKG41/gDwNHAi8L6ZPQRsAvYlGig04Fp33958g7tXmNkXY/c+b2b3AuXAp4BZsfH7evzCeygSiZCWNpDbBMpAlpaWRiQS6XqiiIjIAFUybQx/eu2jlvMlZeW9et6a7dXUN+353jg+L5tRuVm9eqaIiIiIiIiISDySEv1x97eBSmBMF/NWEu0Ft5BoYO5qollytwJHtg6qdfKMCPAJ4ErgPaL95q4GjgD+Cpzs7j9r576HgWOBvwOfBv4daACuAs5zd+/GS+01M+uPZSQF6e+OiIgMdiXFowPnb63bSW1DU9zPK91UFThX/zkRERERERERSZbulLjsK58Fcrqa5O5rgQXdmFdGNBuuvWsNwC2xr25z95eJBvdEREREpJ8Vjsxh8uhhrC2vAaChyXnzo50cOaPTz3h1qHRTReBc5S1FREREREREJFl6HaAzs0KiGWZzgfGx4S3AUuBBd2+3CZa7P9nbtUVEREQktc0rHs3a8pYWwiwpK48/QNeqnx3ALAXoRERERERERCRJ4g7QmVkmcBPR0o/Nz2nOYHPgIuAnZvYL4FvuXt+bjYqIiIjI0HP4tNE8uGxPgG7x6vj70C3frBKXIiIiIiIiIjIwxBWgM7M04BHgZKJBuRrgDaD53ZMi4DBgGPB1YH8zO7W/+raJiIiISGqYF+pDt+yjHTQ2RchI71kr5dqGJsq2VwfG9pkwotf7ExERERERERGJR8/e2djjy8ApsT//ACh092Pc/fzY1zHABOB7RLPpPg58pde7FREREZEhZdrYXMaOyGo5313fxLsbKjq5o30fbqmi9UfFpowezvCsZLZjFhEREREREZGhLN4A3QKigbfvuvt17l4ZnuDuVe7+X8B1RLPsPh/3LkWSzMw6/EpLSyM/P5/Zs2dzwQUX8Nhjj3X5vIULF7bcf8kll/T9CxARERmkzKxNFt2Ssp6XuSzdFPxxdab6z4mIiIiIiIhIEsUboJsNRIBbuzH3VqAJmBXnWiIDmrtTWVlJaWkpd911F5/85CeZP38+27dvT/bWREREUkLJtGCA7vU4+tCVbg4G6GYVqryliIiIiIiIiCRPvHV96oBad6/qaqK7V5nZLqJZdCKD3kMPPRQ4j0QibNu2jddee4177rmH2tpaXnjhBc444wxefPFFzPRXX0REpDfCGXRLy8qJRJy0tO5/j1UGnYiIiIiIiIgMJPEG6N4BjjazMe7eaZqQmY0BCoCX4lxLZEA588wz2x2/9NJLueqqqzjqqKOorKzk5Zdf5oknnuDUU0/t5x2KiIikln33yicvO4PKukYAduxuYOXWKvbpQZBteZsMOgXoRERERERERCR54i1xeVvs3u92Y+53Y3Nvi3MtkUFjzpw5fOELX2g5f+GFF5K4GxERkdSQnmYcOnVUYGxxD/rQ7appYOOu2pbzjDRj+liVuBQRERERERGR5IkrQOfufwZ+BPy7mf3ezKaH55jZNDO7A/h34GZ3v793WxUZHGbPnt3y5127diVxJyIiIqkj3IducQ/60K0IZc9NH5dLVka8n1MTEREREREREem9LktcmtmznVyuAC4CLjKztcB6wIFJwOTYnF3A4Wb2jLuf0Mv9igx427Zta/nzlClTkrgTERGR1BEO0C3pQYCudLP6z4mIiIiIiIjIwNKdHnTzu/msKbGvsILYM7ybzxEZtGpra7nrrrtazk888cQk7kZERCR1HDhpJFkZadQ3RgDYsKuWdTt2M2nU8C7vLd0U6j+nAJ2IiIiIiIiIJFl3AnQ39PkuRAaxSCRCeXk5r732Gj/4wQ947733ALjkkkuYN29ekncnIiKSGrIz0jl4UkGg99ySsvK4AnQzCxWgExEREREREZHk6jJA5+4K0A1wxdc+nuwtJFXZD0/r1/XMrNPr+++/P5///Oe58sor+2lHIiIiQ0PJtNGBAN3i1eWcdcikTu9xd5ZvVgadiIiIiIiIiAwsacnegEiqycrKIjc3F3dVdRUREUmkeaE+dIu70Ydua1UdO3Y3tJznZKYxZXTXWXciIiIiIiIiIn2pOyUuO2VmhwLnA3OB8UR7zW0FlgD3uPubvV1DZCB56KGH2oxVVVXxwQcfcO+99/Lmm29y2WWXcf/99/Poo48yfLjeBBQREUmEQ6cUkGYQiX0GZuXWarZX1TFmRHaH9yzfVBU4nzkhj7S0zrPhRURERERERET6WtwBOjPLBX4LnNs8FJpyDHC1md0LXOru1fGuJTKQnHnmmR1eu+6661iwYAF33303zzzzDFdccQX/+7//24+7ExERSV15OZnsNzGfd9ZXtIwtKdvBKXMKO7ynNFTecqbKW4qIiIiIiIjIABBXgM7M0oBHgOOIBuY2As8C62JTJsWuTQTOA8ab2UkeR80/M5sEfA84BRgTW+th4AZ339GN+y8Bft/FtIi7p4fuywa+AFwMTAdygLXAU8D/uPuanr2SvtPfPdikY1lZWfzqV7/iscceo6KigoULF3L99dczefLkZG9NREQkJZQUjwkE6BavLu88QLepInCu/nMiIiIiIiIiMhDEm0F3EXA80ABcDfzS3SOtJ8SCeJcBP43NvRC4syeLmNkM4BWipTMfAT4ASoCvAaeY2dHuvr2Lx/wDuKGDax+L7e1voXUzgGeAo2Nr3gPUAfOAfwcuMrOj3P29nrweGRry8/M58sgjefLJJ2lqauLZZ5/l4osvTva2REREUkLJtFHc8fLqlvMlZZ33oSvdHCpxWagAnYiIiIiIiIgkX7wBuguI9pr7D3f/RXsTYgG7X5pZOvAzokG9HgXogF8SDc5d4e4/bx40s58AVwI3Eg0Cdsjd/0E0SNeGmb0a++PtoUtnEQ3OPQOc1Dr4aGY3ANcB3wA+35MXI0PHmDFjWv68YcOGJO5EREQktcwtHh04f3fDLqrqGhmR3fbH2kjEWREqcakMOhEREREREREZCNLivO8goIloD7qu/C/QCBzckwXMbDpwElAG3Ba6fD1QDVwY64XXY2Y2BzgCWA88Hro8PXZ8PJwZSDSTD2BcPOvK0LB9+57EztzcuP6KioiISDvGjshm+rg931sjDsvWtF/1fP3OGnbXN7WcjxyWyYT87D7fo4iIiIiIiIhIV+IN0OUBle5e09XE2JxKYEQP1zg+dlwUDpK5eyXwMjCcaJAtHl+KHX/n7k2ha+/GjqfGSnW2dnrs+HSc60qKq6ys5NVXX20533fffZO4GxERkdRz+LRgFt3i1e2XuSzd1DZ7zsz6bF8iIiIiIiIiIt0Vb4BuGzDSzMZ3NTE2pwDoqldc2KzYcXkH11fEjjN7+FzMbBjRMp0Rohl+YY8DDwIfB942s5+Z2Y/N7FngP4GfA+2W9pShraGhgS9/+ctUVFQAMHHiRI499tgk70pERCS1zAuVuVzcQR+60lB5y5mFPf28mIiIiIiIiIhI34i3B92rwNnAfwFf6WLuDYARzXjriZGx464OrjePF/TwuQCfjd33uLv///buPFquqk70+PeXkJCBJCQMiQgkYVRB2yHMNMggYrcTCji0A7QKTgtRWG0/hQa0XY9+3QgKNkKDgtjd6HOAVoHGxTwpQjPIY45JDBAg80gCSX7vj3MKKkVV5U6pqlv3+1mr1q6z9z7n7JO1s6vu/d2999zawszMiDiaYq+504E3VBXfAPxHnVl3L4uIE4ATAHbcccc+NE+d7KqrrnpV3sqVK3n00Ue58sorefLJJwEYNmwYF1xwASNHjmx6vfvuu4/TTjutR/f+1Kc+xfTp03vfaEmSukhtgO7+uUtYs3Ydm282fIP8x91/TpIkSZIkdai+Bui+B3wQODEixgFnZeaT1RUiYheKAN5HgeTV+8j1V2V9ouzDuSeU6UV1LxwxCvgR8C7gCxT7zq0CDgC+C9waEcdk5tX1zs/Mi4GLAWbMmNGX9qmDHXXUURutM2nSJC688MIe1X3wwQd58MEHe3Tvww8/3ACdJGnI237iaF4zYRTzlq4G4MW16/njU0uZURO4q13icjcDdJIkSZIkqUP0aYnLzLwZOI8iSPZR4LGImB0Rd0TE7RExB3gM+Eh5yrmZeUsvb1OZITehQfn4mno9EhFvAPYHngKuaVDt74FjgK9n5kWZ+WxmLsvMa4GjgRHAd3pzX3W30aNH89rXvpYjjzyS8847jyeeeIJjjz223c2SJKkrRQR71+xD9/uafeheWreemfNXbJC3+xQDdJIkSZIkqTP0dQYdmfmViPgTxSy5ScCO5avaQuDMzOzL7LnHyrTRHnO7lmmjPeoaObFML22yTOW7y/Sm2oLMfCAiFgFTI2KrzOzt3noahDIHdiLkcccdx3HHHTeg15QkaSjZa9okrr7/mZeP/1CzD93sBSt5ad0rn9+Tx2/OlmOaLzstSZIkSZLUKn0O0AFk5gURcQnwDmAGsG1Z9DxwD/DbzFzdx8tXgmNHRMSwzFxfKSiX1TwAeAH4XU8vWC5d+XFgPXBpk6qbl+k2da6xOa/M3nuxp/eWJEnSwKmdQXfv7MWsW58MH1asgv7Ycy5vKUmSJEmSOle/AnQAZQDuV+VrwGTmzIi4HjiCYh+486uKzwLGAhdl5kqAiBgB7Ay8lJkzG1z2GGAi8OvMnNvk9rcBewJfi4g7MnNNVdmZFP9uf8jM5fVOliRJ0qa1yzZbsOWYESxZ9RIAy9es5dFnl7HHdsXq6I/X7D+3uwE6SZIkSZLUQfodoNvEPg/cCXw3Ig4DHgH2AQ6hWNry61V1X1uWzwGmNbjeCWV68Ubu+y3gPcBhwKMRcR3FbL0DgL3L91/q5bNIkiRpgAwbFuw1bRK/ffi5l/PunrXo5QDdq2bQuf+cJEmSJEnqIMPa3YBmyplwM4DLKAJzp1DMkvsusF9v9n+LiNcDBwJPAdds5L5PA28FzgFWA8cDXwSmlG15a2be1bunkSRJ0kDae9qGy1xW70P3mDPoJEmSJElSB+v0GXSUS1Ee34N6s4FoUv5Is/I69ecDp5YvSZIkdZi9avahu3vWYjKT1S+tZ86iVS/nR8Cuk7dodfMkSZIkSZIa6vgAnSRJklTPHtuNZ/SI4bzw0joAFqxYw+yFq1ixei2Zr9TbcdIYxoz0a68kSZIkSeocHb3EpSRJktTIiOHDeNvUiRvk3T1r4av3n3N5S0mSJEmS1GEM0EmSJGnQ2mvaq5e5fPw595+TJEmSJEmdzbV+JEmSNGjtNX3DGXR/mL2I6VuP3SBvtykG6CRJkiRJUmcxQCdJkqRB6y07TGTE8OCldcWmc39etIpFK1/coI4z6CRJkiRJUqdxiUtJkiQNWqNHDueNr52wQd6KNWtffr/ZsHjVjDpJkiRJkqR2M0DX4TKz3U3QIGXfkSQNFXtNn9SwbOdttmDkZn7llSRJkiRJncXfVnSwYcOGsX79+nY3Q4PU+vXrGTbM/+KSpO6397TGATr3n5MkSZIkSZ3I3953sNGjR7Ny5cp2N0OD1MqVKxk9enS7myFJ0iY3Y+okIuqX7T55i9Y2RpIkSZIkqQcM0HWwcePGsXz58nY3Q4PU8uXLGTfOWQOSpO43YcwIdp9c/zNvtwb5kiRJkiRJ7WSAroONHz+eVatWsXjx4nY3RYPM4sWLWbVqFePHj293UyRJaom9G+xDt7tLXEqSJEmSpA5kgK6DDR8+nKlTp7JgwQKefvppli1bxrp168jMdjdNHSYzWbduHcuWLePpp59mwYIFTJ06leHDh7e7aZIktcRedfahGzViGDtMHNOG1kiSJEmSJDW3WbsboOZGjhzJTjvtxLJly1iyZAnz5s1j/fr17W6WOtCwYcMYPXo048aNY8qUKQbnJElDSr0ZdLtNHsewYQ02p5MkSZIkSWojA3SDwPDhw5k4cSITJ05sd1MkSZI60uTxo5i61RjmLFz1cl6jfekkSZIkSZLazSUuJUmS1BX2rlnm8nWvcS9WSZIkSZLUmQzQSZIkqSuccNBOjBxefL3dcswIjn7r9m1ukSRJkiRJUn0dH6CLiO0j4gcR8UxErImI2RFxXkT0aL3HiDguInIjr3UNzo2I+GRE3BwRiyLihYiYFRE/jYjdBvZJJUmS1B+7Th7HDacczHkfejM3nvJ2JowZ0e4mSZIkSZIk1dXRe9BFxM7AncC2wNXAo8DewJeAIyPigMxcuJHL3A+c1aDsL4FDgWvr3HsU8H+BdwOPAf8BLAe2K8/bDXi8l48kSZKkTWiHSWPYYdKYdjdDkiRJkiSpqY4O0AH/ShGcOykzz69kRsS3gS8D3wI+2+wCmXk/RZDuVSLirvLtxXWKz6EIzv1v4LTMXF9zrn+SLUmSJEmSJEmSpF6LzGx3G+qKiJ2AmcBsYOfqAFlEjAPmAQFsm5kr+3D9PYE/Ak8DUzNzXVXZzhSz4+4F9sl+/CPNmDEj77nnnr6eLkmSJEmSJEmSpEEqIu7NzBm1+Z28B92hZXp97ey1zFwO3AGMAfbt4/VPLNNLq4NzpY9Q/NtcDoyPiI9FxP+KiBMiYpc+3k+SJEmSJEmSJEnq6CUudy/TRvu8PQEcQbEX3A29uXBEjAY+BqwHLqlTZa8ynUAxi2+rqrKMiAsplt2sDexJkiRJkiRJkiRJTXXyDLoJZbq0QXklf8s+XPvY8rxrM3NunfJty/QbwD3AG4FxwGEUAbvPA6f34b6SJEmSJEmSJEka4jo5QLcxUaZ92R/uhDK9qEH58DKdBxyVmQ9l5orMvBE4mmLm3VciYmTdhhVLYd4TEffMnz+/D82TJEmSJEmSJElSt+rkAF1lhtyEBuXja+r1SES8AdgfeAq4pkG1xWV6XWa+UF2QmQ8Asyhm1L2+3smZeXFmzsjMGdtss01vmidJkiRJkiRJkqQu18kBusfKdLcG5buWaaM96ho5sUwvbbKHXOXeSxqUVwJ4o3t5b0mSJEmSJEmSJA1xnRygu6lMj4iIDdoZEeOAA4AXgN/19IIRMQr4OMUSlZc2qXpDme5Z5xqb80pwcHZP7y1JkiRJkiRJkiQBbNbuBjSSmTMj4nrgCOALwPlVxWcBY4GLMnMlQESMAHYGXsrMmQ0uewwwEfh1Zs5tcvtrgT8B74yId2Tmb6vKTqdYdvOWzHx2Y89x7733LoiIORurp5dtDSxodyOkFrPfayiy32sost9rqLLvayiy32sost9rKLLfa6iy7/fO1HqZkZmtbkiPRcTOwJ3AtsDVwCPAPsAhFEtb7p+ZC8u60yj2hpuTmdMaXO824EDgvZn5q43c+0DgemAk8EtgDrAXcBAwHzgwM3u7vKY2IiLuycwZ7W6H1Er2ew1F9nsNRfZ7DVX2fQ1F9nsNRfZ7DUX2ew1V9v2B0clLXFLOhJsBXEYRmDuFYpbcd4H9KsG5noiI11ME554CrunBvW8v7/1z4GDgJGAn4GLgrQbnJEmSJEmSJEmS1Bcdu8RlRbkU5fE9qDcbiCbljzQrb3DOw8CHenOOJEmSJEmSJEmS1ExHz6DTkHRxuxsgtYH9XkOR/V5Dkf1eQ5V9X0OR/V5Dkf1eQ5H9XkOVfX8AdPQedJIkSZIkSZIkSVK3cQadJEmSJEmSJEmS1EIG6CRJkiRJkiRJkqQWMkCntouI7SPiBxHxTESsiYjZEXFeRExsd9ukTaHs49ng9Wy72yf1R0QcHRHnR8RtEbGs7Nc/3sg5+0fENRGxKCJWRcSDEXFyRAxvVbul/uhNv4+IaU0+AzIirmx1+6XeioitIuLTEfHLiHgyIl6IiKURcXtEfCoi6v6c6Xivwa63fd8xX90iIv4pIm6IiLllv18UEfdFxBkRsVWDcxzzNaj1pt873qubRcTHq/rypxvUcczvo83a3QANbRGxM3AnsC1wNfAosDfwJeDIiDggMxe2sYnSprIUOK9O/opWN0QaYKcBf0HRl58CXtesckS8D/g5sBr4CbAIeA9wLnAAcMymbKw0QHrV70sPAFfVyX9oANslbSrHABcC84CbgD8Dk4EPAJcA74qIY7Jqw3PHe3WJXvf9kmO+BrsvA/8D/BZ4HhgL7AucCZwQEftm5txKZcd8dYle9fuS4726SkTsAJxP8bPuFg3qOOb3Q7z6e6PUOhHx38ARwEmZeX5V/rcpPggvyszPtqt90qYQEbMBMnNae1siDbyIOIQiQPEkcDDFL6/+PTM/Vqfu+LLeBOCAzLynzB8F3AjsB3wkM/1rQ3W0Xvb7acAs4PLMPK51rZQGTkQcSvFLqt9k5vqq/CnA3cAOwNGZ+fMy3/FeXaEPfX8ajvnqAhExKjNX18n/FvA14MLM/HyZ55ivrtDLfj8Nx3t1mYgIigD1dOAXwKnAZzLzkqo6jvn95BKXapuI2IkiODcb+F5N8RnASuDjETG2xU2TJPVRZt6UmU/U+cvxeo4GtgGurHyJK6+xmmJGEsDnNkEzpQHVy34vDXqZeWNm/qo6QFHmPwt8vzx8e1WR4726Qh/6vtQV6gUpSj8t012r8hzz1RV62e+lbnQScChwPMXv6etxzO8nl7hUOx1aptfX+QFneUTcQRHA2xe4odWNkzaxzSPiY8COFB9yDwK3Zua69jZLaqnK58B1dcpuBVYB+0fE5pm5pnXNklpiu4g4EdgKWAjclZkPtrlN0kB4qUzXVuU53msoqNf3Kxzz1a3eU6bV/dkxX92uXr+vcLxXV4iI1wNnA9/JzFvLVQTqcczvJwN0aqfdy/TxBuVPUATodsMAnbrPFOCKmrxZEXF8Zt7SjgZJbdDwcyAz10bELGAPYCfgkVY2TGqBd5Svl0XEzcAnM/PPbWmR1E8RsRnwifKw+od0x3t1tSZ9v8IxX10hIk6l2INoAjADOJAiSHF2VTXHfHWVHvb7Csd7DXrl95orKPba/dpGqjvm95MBOrXThDJd2qC8kr9lC9oitdIPgduA/wcsp/iQ+iJwAnBtROyXmQ+0sX1Sq/g5oKFoFfBNis3j/1TmvYlis/lDgBsi4s2Z2WgJEamTnQ3sCVyTmf9dle94r27XqO875qvbnApMrjq+DjguM+dX5Tnmq9v0pN873qub/APwFuDAzHxhI3Ud8/vJPejUyaJM3c9FXSUzzyr3r3guM1dl5kOZ+Vng28Boii9wkvwcUBfKzOcz8x8y838yc0n5upVi1YDfA7sAn25vK6Xei4iTgFOAR4GP9/b0MnW816DTrO875qvbZOaUzAyKFWE+QPHHpvdFxFt7cRnHfA0qPen3jvfqFhGxN8WsuXMy866BuGSZOuY3YIBO7VSJoE9oUD6+pp7U7Sobyx/U1lZIrePngFTKzLXAJeWhnwMaVCLiC8B3gIeBQzJzUU0Vx3t1pR70/boc8zXYlX9s+kuK4MNWwI+qih3z1ZU20u8bneN4r0GjamnLx4HTe3iaY34/GaBTOz1Wprs1KN+1TBvtUSd1m+fLdGxbWyG1TsPPgfKL4XRgLa8sESJ1u8oyOX4OaNCIiJOBC4CHKAIUz9ap5nivrtPDvt+MY74GvcycQxGg3iMiti6zHfPV1Rr0+2Yc7zVYbEExdr8eWB0RWXkBZ5R1/q3MO688dszvJwN0aqebyvSIiNigL0bEOOAA4AXgd61umNQm+5WpH1oaKm4s0yPrlB0EjAHuzMw1rWuS1Fb7lqmfAxoUIuKrwLnA/RQBiucbVHW8V1fpRd9vxjFf3WK7Ml1Xpo75Ggpq+30zjvcaLNYAlzZ43VfWub08rix/6ZjfTwbo1DaZORO4HpgGfKGm+CyKvyz5kRuoqptExB4RMalO/lSKv8AF+HFrWyW1zc+ABcCHI2JGJTMiRgH/WB5e2I6GSZtKROwTESPr5B8KfLk89HNAHS8iTgfOBu4FDsvMBU2qO96ra/Sm7zvmqxtExOsiYkqd/GER8S1gW4pfvi4uixzzNej1tt873qsbZOYLmfnpei/gv8pql5d5PymPHfP7KTLdn0/tExE7A3dSfLBdDTwC7AMcQrG05f6ZubB9LZQGVkScCfw9xQzSWcByYGfgr4FRwDXAUZn5YrvaKPVHRLwfeH95OAV4J8VfCt5W5i3IzFNr6v8MWA1cCSwC3gvsXuYfm35ZUYfrTb+PiJuBPYCbgafK8jcBh5bvT8/Myg8yUkeKiE8Cl1H81fj51N9TYnZmXlZ1juO9Br3e9n3HfHWDcjnXfwZuBWYCC4HJwMHATsCzFMHqh6vOcczXoNbbfu94r25X/j7zDOAzmXlJTZljfj8YoFPbRcQOwDcopsJuBcwDrgLO6ukm29JgEREHA58F3kLxS9yxwBKK5XGuAK7wQ0uDWdWXtkbmZOa0mnMOAL5OsczrKOBJ4AfAdzOzJ0uGSG3Vm34fEZ8CjgL2BLYGRgDPUSwRckFm3tboIlKn6EGfB7glM99ec57jvQa13vZ9x3x1g4jYE/gcxTYk2wNbAisp/qj6NxRj+Kt+d+OYr8Gst/3e8V7drlmArix3zO8jA3SSJEmSJEmSJElSC7kHnSRJkiRJkiRJktRCBugkSZIkSZIkSZKkFjJAJ0mSJEmSJEmSJLWQATpJkiRJkiRJkuBayjgAAAg+SURBVCSphQzQSZIkSZIkSZIkSS1kgE6SJEmSJEmSJElqIQN0kiRJkiRJkiRJUgsZoJMkSZIktV1EnBkRGRE3t7stkiRJkrSpGaCTJEmSpDapCkplTf60suzMNjVtwETEm8tnObndbZEkSZKkTmGATpIkSZI6zzTgjPI12L2Z4jk2FqBbADwG/HmTt0iSJEmS2myzdjdAkiRJkqTMvAC4oN3tkCRJkqRWcAadJEmSJEmSJEmS1EIG6CRJkiSpg0TEbOCmquOseV1W55xREXFSRNwSEQsi4sWIeDYiroqII5vcq3LNt0fEthHx7Yh4PCJWVe+LFxGjI+K9EfFvEXF/RMyPiDUR8Ux5j3c1uj7ww/Jwap1nObOqbmU/vpubtPctEfGjiJgTEasjYnFE3BkRJ0fE5g3OOa687uzy+G0R8dOImFc+w5/K557Y5L77RMS/R8Ss8r4ryzbcEhGnR8T2jc6VJEmSpHpc4lKSJEmSOst8YDxQCRg9V1O+tPogInYFfgPsWmYlsAyYDLwPeF9EXJiZn29yz12AK8tzVgMv1ZR/iFcCbQAvAGuB11Td45zMPLXmvOeA0eXzrC+frdqKJm3aQEScDHwbiDJrKTAW2K98HR8RR2bmvCbX+ChwGTCiPH8zYDrwZeCIiNg3M1fUnPNJimev3HcNxbPvWL4OAuaW15UkSZKkHnEGnSRJkiR1kMzcC/hA1fGUmteXKmURsSVwPUVw7kaKYNHozNwS2BL4CkUQ7HMR8SUaOxdYAhwGjM3M8cDuVeVLgIuBQ4CtM3NMZo4FtgPOoAjonRIR7615lilA5b5z6zzLv/Tk3yQi3l22MYCrgZ3KZ9wC+ASwHHgT8LOIGN7gMtsAPwAuB3Yszx8HfLFs/x7A39XcdwxwfnnfHwO7ZOaozJxQ3nsG8M/A8z15DkmSJEmqMEAnSZIkSYPX14FpFMG5d2bmbZm5BiAzl2bmuRQBLIDTIqLRKirrgcMz88bMXF+e/3ilMDOvyswTM/PmzFxYlT8vM78BfK3MOmkgH67KP5Xp7cAHM3NWef8XM/MK4G/K8v2BoxpcYwxwZWZ+JjPnluevyszvUQThAD5Sc86eFEG8lcDxmTmzUpCZKzPz3sz8u8y8pp/PJ0mSJGmIMUAnSZIkSYNQRATwt+XhOZm5tkHVqyiWvNwaeFuDOldk5lP9aM5vynS/JjPY+iQi3gS8oTz8Zmauq62Tmb8C7i4Pa4Ns1f6xQf7VZbpLOWuuYkmZjgS26lmLJUmSJGnjDNBJkiRJ0uD0BmBS+f6yiHi23guYR7EcI8DUBte6Y2M3i4jJEXFWRNwVEQsjYm1EZEQk8HBZbQyv7J03UGaU6Vrglib1fltTv9aizHyyQdkzVe+r2z8TeJRiz7rfR8RXI+LNAx2ElCRJkjT0NFreRJIkSZLU2barer9ND88Z0yC/6R5qEbEfcA3FvnYVK4BVQALDKWboAYwFFvSwPT2xbZkuqCzf2UBlBuC2DcqXNzm3evbhiMqbzFwXER8GfglMB84uX6si4k7gF8DlmbmqybUlSZIk6VWcQSdJkiRJg1P1LK4pmRk9eF3W4FqvWjayoty37j8pgnP3A38FjM/McZk5OTOnAPtWn9K/x2ooB7hezy6W+QDwOuCDwMXAQ8Bo4HDgX4FHI+KNA3lPSZIkSd3PAJ0kSZIkDU7PVr3flAGi/SiWxlwHvDszr83M2tloUzbh/Suz+7aJiM2b1Nu+TOcPdAMy88XM/EVmnpiZb6SYsfhZYBGwA3D5QN9TkiRJUnczQCdJkiRJnWd95U1ENJqR9hCwrHz/4U3Ylh3KdH5mPt2gzuFNzq88S19n1t1TppsBBzepV2nDH/p4nx7LzIWZeRHw1TLrLRGx1aa+ryRJkqTuYYBOkiRJkjrPsqr3W9arkJlrgR+Uh5+MiAObXTAiJvWxLUvLdHJETK5z3e2Bk5qcX3mWus+xMZn5IPBweXhaRAyvrRMRfwXsUx7+Z1/uU89GZuwBvFD1vuEyoZIkSZJUywCdJEmSJHWex4EXy/efbjKL7pvATIrZZddFxFciYptKYURMiIgjI+Jy4LY+tuV2YCXFDLifRsRu5bWHR8Q7gZtpvu/bQ2U6PiKO7WMbKjPV/hL4WURML9swIiL+hleCcncCV/XxHvV8OCLuiIgTI2KnSmbVs59dZt2VmUsG8L6SJEmSupwBOkmSJEnqMJm5CriiPPw/wIqImBMRsyPiX6rqLQLeATwAjAXOAZ6PiMURsRRYAlwLfAIY2ce2LAVOLQ8PAh6LiOXACuA6YAJwfJPznwRuKA9/EhHLyueYHREn97ANvwa+QhEIfD/wp4hYXLbhx8B44I/AMZk5kDPZAtgf+D4wMyJWR8QCiuDpdRT73j0D/O0A3lOSJEnSEGCATpIkSZI60xeAM3llBtqOwFRg6+pKmTkLmEERhPs1MI8iWDcSmAX8kiKAtF9fG5KZ3wf+mmK23AqKGXtPA+cDf0ERHGvmaOBcipmBI8rnmEovlr3MzHMpnvPHwFxgDMUSk7+jCN7tnZnP9PR6PfRfFP+uP6QIgi6lCEguB+4GTgf2yMxHB/i+kiRJkrpcZDZbiUSSJEmSJEmSJEnSQHIGnSRJkiRJkiRJktRCBugkSZIkSZIkSZKkFjJAJ0mSJEmSJEmSJLWQATpJkiRJkiRJkiSphQzQSZIkSZIkSZIkSS1kgE6SJEmSJEmSJElqIQN0kiRJkiRJkiRJUgsZoJMkSZIkSZIkSZJayACdJEmSJEmSJEmS1EIG6CRJkiRJkiRJkqQWMkAnSZIkSZIkSZIktdD/B31ISN5q44XDAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 2160x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Size = 201 \n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "fig = plt.figure(figsize=(30,5))\n",
    "plt.title('Variation of Objective',fontsize=20)\n",
    "plt.plot(results['objective_values'][:Size],lw=4,label='BL')\n",
    "plt.xlabel('Iterations', fontsize=25)\n",
    "plt.ylabel('objective_values',fontsize=25)\n",
    "plt.xticks(fontsize=20)\n",
    "plt.yticks(fontsize=20)\n",
    "plt.legend(prop={'size': 30})\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plotting the variation of Max objective values of the tried results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Size = 201 \n",
    "\n",
    "import numpy as np\n",
    "\n",
    "results_obj = np.array(results['objective_values'])\n",
    "\n",
    "y_max=[]\n",
    "for i in range(results_obj.shape[0]):\n",
    "    y_max.append(np.max(results_obj[:i+1]))\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "fig = plt.figure(figsize=(30,5))\n",
    "plt.title('Max variation of Objective',fontsize=20)\n",
    "plt.plot(y_max[:Size],lw=4,label='BL')\n",
    "plt.xlabel('Iterations', fontsize=25)\n",
    "plt.ylabel('objective_values',fontsize=25)\n",
    "plt.xticks(fontsize=20)\n",
    "plt.yticks(fontsize=20)\n",
    "plt.legend(prop={'size': 30})\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# See the Result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{   'best_objective': 0.9220306513409962,\n",
      "    'best_params': {   'booster': 'gbtree',\n",
      "                       'gamma': 0.011667904935745432,\n",
      "                       'learning_rate': 0.6859901826452589,\n",
      "                       'max_depth': 1,\n",
      "                       'n_estimators': 201},\n",
      "    'objective_values': [   0.9164750957854406,\n",
      "                            0.8942528735632185,\n",
      "                            0.8549808429118775,\n",
      "                            0.899808429118774,\n",
      "                            0.9053639846743295,\n",
      "                            0.9053639846743295,\n",
      "                            0.9220306513409962,\n",
      "                            0.9109195402298851,\n",
      "                            0.9053639846743295,\n",
      "                            0.8436781609195402,\n",
      "                            0.899808429118774,\n",
      "                            0.860536398467433,\n",
      "                            0.9053639846743295,\n",
      "                            0.8660919540229886,\n",
      "                            0.7651340996168582,\n",
      "                            0.9051724137931035,\n",
      "                            0.8660919540229886,\n",
      "                            0.8660919540229886,\n",
      "                            0.8885057471264369,\n",
      "                            0.8716475095785441,\n",
      "                            0.9053639846743295,\n",
      "                            0.8886973180076628,\n",
      "                            0.8383141762452108,\n",
      "                            0.899808429118774,\n",
      "                            0.9109195402298851,\n",
      "                            0.899616858237548,\n",
      "                            0.8716475095785441,\n",
      "                            0.8942528735632184,\n",
      "                            0.9109195402298851,\n",
      "                            0.8886973180076628,\n",
      "                            0.9053639846743295,\n",
      "                            0.8660919540229886,\n",
      "                            0.9109195402298851,\n",
      "                            0.899808429118774,\n",
      "                            0.8940613026819925,\n",
      "                            0.9164750957854406,\n",
      "                            0.8942528735632184,\n",
      "                            0.8940613026819925,\n",
      "                            0.8942528735632185,\n",
      "                            0.9053639846743295,\n",
      "                            0.8886973180076628],\n",
      "    'params_tried': [   {   'booster': 'dart',\n",
      "                            'gamma': 0.3873414168569206,\n",
      "                            'learning_rate': 0.19560021429563113,\n",
      "                            'max_depth': 6,\n",
      "                            'n_estimators': 188},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.7199956896001947,\n",
      "                            'learning_rate': 0.7567770987013972,\n",
      "                            'max_depth': 7,\n",
      "                            'n_estimators': 22},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 4.271932433395348,\n",
      "                            'learning_rate': 0.295517664199457,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 299},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 4.900025909071044,\n",
      "                            'learning_rate': 0.5337794023290698,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 108},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 0.3536485846213011,\n",
      "                            'learning_rate': 0.8515465846252355,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 107},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 4.86618939254914,\n",
      "                            'learning_rate': 0.9767311773382229,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 179},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 0.011667904935745432,\n",
      "                            'learning_rate': 0.6859901826452589,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 201},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.020068184927398147,\n",
      "                            'learning_rate': 0.1382847274708281,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 154},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.08649552117724557,\n",
      "                            'learning_rate': 0.957769685814364,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 241},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 3.8542502616743537,\n",
      "                            'learning_rate': 0.9191520656363247,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 1},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.15933625384121186,\n",
      "                            'learning_rate': 0.9080778428227618,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 59},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 4.880774001907471,\n",
      "                            'learning_rate': 0.14379577591879678,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 57},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 0.5644719578468316,\n",
      "                            'learning_rate': 0.8215171058514628,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 299},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 4.920671461587199,\n",
      "                            'learning_rate': 0.2376497831284553,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 274},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.6100114690621694,\n",
      "                            'learning_rate': 0.9667126364880959,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 2},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 0.26559898071896626,\n",
      "                            'learning_rate': 0.9852919915344887,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 269},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 4.622001963100803,\n",
      "                            'learning_rate': 0.2112186678350978,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 223},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 3.7811271724864888,\n",
      "                            'learning_rate': 0.25048357957896383,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 133},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 4.724139189387041,\n",
      "                            'learning_rate': 0.9990783864572119,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 38},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 4.867979323312713,\n",
      "                            'learning_rate': 0.07076576404706736,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 179},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 0.7841003789654138,\n",
      "                            'learning_rate': 0.10741092583425316,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 84},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 4.942487948626461,\n",
      "                            'learning_rate': 0.947672559329843,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 83},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 0.08511755539532206,\n",
      "                            'learning_rate': 0.0393473780948258,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 20},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.03801236357024229,\n",
      "                            'learning_rate': 0.9637202316555209,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 121},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.18045866879006167,\n",
      "                            'learning_rate': 0.06293528846422902,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 286},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 0.10215666648900446,\n",
      "                            'learning_rate': 0.9053859471721212,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 168},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 4.993496300431588,\n",
      "                            'learning_rate': 0.21747842533560724,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 12},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 4.984312647518595,\n",
      "                            'learning_rate': 0.2529128502717738,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 251},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 0.0045158389903648954,\n",
      "                            'learning_rate': 0.29487867285072833,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 44},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.14392505397580135,\n",
      "                            'learning_rate': 0.37201304661638146,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 213},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 4.927663287059414,\n",
      "                            'learning_rate': 0.006049399471735328,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 215},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 4.363304584766794,\n",
      "                            'learning_rate': 0.08152623069247222,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 200},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 4.87821194109707,\n",
      "                            'learning_rate': 0.04686812203839741,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 36},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 0.18425914869700022,\n",
      "                            'learning_rate': 0.03847255144747974,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 255},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 4.66839413601658,\n",
      "                            'learning_rate': 0.9508091635175131,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 144},\n",
      "                        {   'booster': 'gbtree',\n",
      "                            'gamma': 0.10921211173208156,\n",
      "                            'learning_rate': 0.12026105041094026,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 285},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.04969987647705454,\n",
      "                            'learning_rate': 0.8613852316047141,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 147},\n",
      "                        {   'booster': 'dart',\n",
      "                            'gamma': 4.914241169463478,\n",
      "                            'learning_rate': 0.933385718727159,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 21},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.047209459390731845,\n",
      "                            'learning_rate': 0.8338751785684253,\n",
      "                            'max_depth': 9,\n",
      "                            'n_estimators': 78},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 0.04275440542318487,\n",
      "                            'learning_rate': 0.9865381721117884,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 184},\n",
      "                        {   'booster': 'gblinear',\n",
      "                            'gamma': 4.435083542957001,\n",
      "                            'learning_rate': 0.3953986308156986,\n",
      "                            'max_depth': 1,\n",
      "                            'n_estimators': 297}],\n",
      "    'random_params': [   {   'booster': 'dart',\n",
      "                             'gamma': 0.3873414168569206,\n",
      "                             'learning_rate': 0.19560021429563113,\n",
      "                             'max_depth': 6,\n",
      "                             'n_estimators': 188}],\n",
      "    'random_params_objective': [0.9164750957854406]}\n"
     ]
    }
   ],
   "source": [
    "import pprint\n",
    "pp = pprint.PrettyPrinter(indent=4)\n",
    "pp.pprint(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}