CO_FHMM_Final_DRED_NILMtk.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Imports necessary library using NILMTK enviroment\n",
    "from pylab import rcParams\n",
    "import matplotlib.pyplot as plt\n",
    "from os.path import join\n",
    "import nilmtk as ntk\n",
    "import nilmtk.disaggregate as ntkd\n",
    "import nilmtk.metrics as ntkm\n",
    "from matplotlib.sankey import Sankey\n",
    "%matplotlib inline\n",
    "\n",
    "rcParams['figure.figsize'] = (14, 6)\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "loaded 1 buildings\n"
     ]
    }
   ],
   "source": [
    "#Uploading the dataset and define the location of data\n",
    "h5_path = '/Users/Dhruvin suthar/Desktop/Thesis work/test_of_database/DRED/DRED' \n",
    "dred = ntk.DataSet(join(h5_path,'DRED.h5'))\n",
    "print('loaded ' + str(len(dred.buildings)) + ' buildings')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MeterGroup(meters=\n",
       "  ElecMeter(instance=1, building=1, dataset='DRED', site_meter, appliances=[])\n",
       "  ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])\n",
       "  ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])\n",
       "  ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])\n",
       "  ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])\n",
       "  ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])\n",
       "  ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])\n",
       "  ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])\n",
       "  ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])\n",
       "  ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])\n",
       "  ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])\n",
       "  ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])\n",
       "  ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])\n",
       ")"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Load the data from H5 file for a 1 week (appearant dataset: 05.07.2015 to 10.07.2015)\n",
    "dred.set_window(start=None, end='2015-07-10 00:00:00')\n",
    "elec = dred.buildings[1].elec\n",
    "elec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dd27febb70>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the aggregated power of appliances \n",
    "mains = elec.mains()\n",
    "mains.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>power</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>apparent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:08+02:00</th>\n",
       "      <td>223.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:09+02:00</th>\n",
       "      <td>223.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:10+02:00</th>\n",
       "      <td>224.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:11+02:00</th>\n",
       "      <td>224.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:12+02:00</th>\n",
       "      <td>225.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             power\n",
       "                          apparent\n",
       "2015-07-05 00:00:08+02:00    223.0\n",
       "2015-07-05 00:00:09+02:00    223.6\n",
       "2015-07-05 00:00:10+02:00    224.2\n",
       "2015-07-05 00:00:11+02:00    224.8\n",
       "2015-07-05 00:00:12+02:00    225.4"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#opning the aggregated power dataset\n",
    "mains_df = next(mains.load())\n",
    "mains_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['apparent']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Findind the type of power\n",
    "mains.available_ac_types('power')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['apparent']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#available recodered power\n",
    "elec.submeters().available_ac_types('power')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "apparent    16.584037\n",
       "dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# type of power\n",
    "elec.mains().total_energy() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12/12 ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])))=1)])ance=1)])"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>(2, 1, DRED)</th>\n",
       "      <th>(3, 1, DRED)</th>\n",
       "      <th>(4, 1, DRED)</th>\n",
       "      <th>(5, 1, DRED)</th>\n",
       "      <th>(6, 1, DRED)</th>\n",
       "      <th>(7, 1, DRED)</th>\n",
       "      <th>(8, 1, DRED)</th>\n",
       "      <th>(9, 1, DRED)</th>\n",
       "      <th>(10, 1, DRED)</th>\n",
       "      <th>(11, 1, DRED)</th>\n",
       "      <th>(12, 1, DRED)</th>\n",
       "      <th>(13, 1, DRED)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>active</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>apparent</th>\n",
       "      <td>1.251359</td>\n",
       "      <td>0.67942</td>\n",
       "      <td>5.41584</td>\n",
       "      <td>1.579026</td>\n",
       "      <td>0.618707</td>\n",
       "      <td>0.23525</td>\n",
       "      <td>0.012125</td>\n",
       "      <td>0.081595</td>\n",
       "      <td>0.78638</td>\n",
       "      <td>0.038197</td>\n",
       "      <td>1.176727</td>\n",
       "      <td>1.065199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reactive</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          (2, 1, DRED)  (3, 1, DRED)  (4, 1, DRED)  (5, 1, DRED)  \\\n",
       "active             NaN           NaN           NaN           NaN   \n",
       "apparent      1.251359       0.67942       5.41584      1.579026   \n",
       "reactive           NaN           NaN           NaN           NaN   \n",
       "\n",
       "          (6, 1, DRED)  (7, 1, DRED)  (8, 1, DRED)  (9, 1, DRED)  \\\n",
       "active             NaN           NaN           NaN           NaN   \n",
       "apparent      0.618707       0.23525      0.012125      0.081595   \n",
       "reactive           NaN           NaN           NaN           NaN   \n",
       "\n",
       "          (10, 1, DRED)  (11, 1, DRED)  (12, 1, DRED)  (13, 1, DRED)  \n",
       "active              NaN            NaN            NaN            NaN  \n",
       "apparent        0.78638       0.038197       1.176727       1.065199  \n",
       "reactive            NaN            NaN            NaN            NaN  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#opening the dataset appliancesvise\n",
    "energy_per_meter = elec.submeters().energy_per_meter() # kWh, again\n",
    "energy_per_meter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plotting the Sankey diagram with availabe appliances\n",
    "#elec._plot_sankey()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Dhruvin suthar\\Anaconda3\\envs\\nilmtk-env\\lib\\site-packages\\networkx\\drawing\\nx_pylab.py:611: MatplotlibDeprecationWarning: isinstance(..., numbers.Number)\n",
      "  if cb.is_numlike(alpha):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(<networkx.classes.digraph.DiGraph at 0x1dd2a3e8470>,\n",
       " <matplotlib.axes._axes.Axes at 0x1dd2888d5c0>)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the wire diagram with availabe appliances\n",
    "elec.draw_wiring_graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "###INdividial Appliances proparties "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "##taking the of fridge\n",
    "fridge_meter = elec['fridge']\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_sample_period': 1,\n",
       " 'model': 'GT',\n",
       " 'measurements': [{'type': 'apparent', 'physical_quantity': 'power'}],\n",
       " 'sample_period': 1}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Proparties of Fridge\n",
    "fridge_meter.device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loading the fridge data from h5 file\n",
    "fridge_df = next(fridge_meter.load())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>power</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>apparent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:08+02:00</th>\n",
       "      <td>99.210000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:09+02:00</th>\n",
       "      <td>99.179070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:10+02:00</th>\n",
       "      <td>99.148140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:11+02:00</th>\n",
       "      <td>99.117209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:12+02:00</th>\n",
       "      <td>99.086279</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               power\n",
       "                            apparent\n",
       "2015-07-05 00:00:08+02:00  99.210000\n",
       "2015-07-05 00:00:09+02:00  99.179070\n",
       "2015-07-05 00:00:10+02:00  99.148140\n",
       "2015-07-05 00:00:11+02:00  99.117209\n",
       "2015-07-05 00:00:12+02:00  99.086279"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fridge_df.head() # Visualize the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "apparent    5.41584\n",
       "dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fridge_meter.total_energy() # kWh and type of data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dd2835c9b0>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fridge_meter.plot() # plotting the fridge dataset for a 1 week"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dd2a3d8be0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "##Good sections : To check the data is recorderd proparyly or not. \n",
    "#when we plot the raw power data then we see there is one large gap where, supposedly, the metering system was not working.\n",
    "# In our case, metering us working properly\n",
    "good_sections = fridge_meter.good_sections(full_results=True)\n",
    "good_sections.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[TimeFrame(start='2015-07-05 00:00:08+02:00', end='2015-07-10 00:00:00+02:00', empty=False)]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#selectin of data\n",
    "good_sections.combined()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#As well as large gaps appearing because the entire system is down, we also get frequent small gaps from wireless sensors dropping data. This is sometimes called 'dropout'. \n",
    "#The dropout rate is a number between 0 and 1 which specifies the proportion of missing samples.\n",
    "#A dropout rate of 0 means no samples are missing. A value of 1 would mean all samples are missing\n",
    "fridge_meter.dropout_rate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "##For television\n",
    "television_meter = elec['television']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_sample_period': 1,\n",
       " 'model': 'GT',\n",
       " 'measurements': [{'type': 'apparent', 'physical_quantity': 'power'}],\n",
       " 'sample_period': 1}"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Proparties of television\n",
    "television_meter.device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>power</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>apparent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:10+02:00</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:11+02:00</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:12+02:00</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:13+02:00</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-05 00:00:14+02:00</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             power\n",
       "                          apparent\n",
       "2015-07-05 00:00:10+02:00      0.0\n",
       "2015-07-05 00:00:11+02:00      0.0\n",
       "2015-07-05 00:00:12+02:00      0.0\n",
       "2015-07-05 00:00:13+02:00      0.0\n",
       "2015-07-05 00:00:14+02:00      0.0"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#visulize the data\n",
    "television_df = next(television_meter.load())\n",
    "television_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "apparent    1.251359\n",
       "dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "television_meter.total_energy() # kWh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dd28093f98>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "television_meter.plot() #plotting the television data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dd28105240>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "good_sections_t = television_meter.good_sections(full_results=True)\n",
    "good_sections_t.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[TimeFrame(start='2015-07-05 00:00:10+02:00', end='2015-07-10 00:00:00+02:00', empty=False)]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "good_sections_t.combined()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "television_meter.dropout_rate() #dropout rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training model for submeter 'ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])'\n",
      "Done training!\n"
     ]
    }
   ],
   "source": [
    "#Train the data using Combinatorial Optimisation \n",
    "co = ntk.disaggregate.CombinatorialOptimisation()\n",
    "co.train(elec)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimating power demand for 'ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])'\n"
     ]
    }
   ],
   "source": [
    "##Disaggreagtion via CO\n",
    "disag_filename = join(h5_path, 'Weekly_COMBINATORIAL_OPTIMIZATION.h5')\n",
    "\n",
    "mains = elec.mains()\n",
    "\n",
    "try:\n",
    "    output = ntk.HDFDataStore(disag_filename, 'w')\n",
    "    co.disaggregate(mains, output)\n",
    "except ValueError:\n",
    "    output.close()\n",
    "    output = ntk.HDFDataStore(disag_filename, 'w')\n",
    "    co.disaggregate(mains, output)\n",
    "\n",
    "for meter in range(1, 2):\n",
    "    df1 = output.store.get('/building1/elec/meter{}'.format(meter))\n",
    "    df2 = dred.store.store.get('/building1/elec/meter{}'.format(meter))\n",
    "\n",
    "    \n",
    "output.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Disaggregate via CO\n",
    "#output = ntk.HDFDataStore(h5_path + 'Weekly_COMBINATORIAL_OPTIMIZATION.h5', 'w')\n",
    "#co.disaggregate(mains, output)\n",
    "#output.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MeterGroup(meters=\n",
       "  ElecMeter(instance=1, building=1, dataset='CO', site_meter, appliances=[])\n",
       "  ElecMeter(instance=2, building=1, dataset='CO', appliances=[Appliance(type='television', instance=1)])\n",
       "  ElecMeter(instance=3, building=1, dataset='CO', appliances=[Appliance(type='fan', instance=1)])\n",
       "  ElecMeter(instance=4, building=1, dataset='CO', appliances=[Appliance(type='fridge', instance=1)])\n",
       "  ElecMeter(instance=5, building=1, dataset='CO', appliances=[Appliance(type='laptop computer', instance=1)])\n",
       "  ElecMeter(instance=6, building=1, dataset='CO', appliances=[Appliance(type='electric heating element', instance=1)])\n",
       "  ElecMeter(instance=7, building=1, dataset='CO', appliances=[Appliance(type='oven', instance=1)])\n",
       "  ElecMeter(instance=8, building=1, dataset='CO', appliances=[Appliance(type='unknown', instance=1)])\n",
       "  ElecMeter(instance=9, building=1, dataset='CO', appliances=[Appliance(type='washing machine', instance=1)])\n",
       "  ElecMeter(instance=10, building=1, dataset='CO', appliances=[Appliance(type='microwave', instance=1)])\n",
       "  ElecMeter(instance=11, building=1, dataset='CO', appliances=[Appliance(type='toaster', instance=1)])\n",
       "  ElecMeter(instance=12, building=1, dataset='CO', appliances=[Appliance(type='sockets', instance=1)])\n",
       "  ElecMeter(instance=13, building=1, dataset='CO', appliances=[Appliance(type='cooker', instance=1)])\n",
       ")"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Metrics==========================================================================\n",
    "disag = ntk.DataSet(h5_path + 'Weekly_outputDRED.h5')\n",
    "disag_elec = disag.buildings[1].elec\n",
    "disag_elec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Dhruvin suthar\\Anaconda3\\envs\\nilmtk-env\\lib\\site-packages\\sklearn\\metrics\\classification.py:1135: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2     0.213910\n",
      "3     0.237900\n",
      "4     0.564750\n",
      "5     0.477773\n",
      "6     0.047019\n",
      "7     0.122449\n",
      "8     0.000000\n",
      "9     0.000000\n",
      "10    0.082645\n",
      "11    0.031250\n",
      "12    0.164589\n",
      "13    0.215054\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Accuracy with F1-score of COMBINATORIAL_OPTIMIZATIONw\n",
    "f1 = ntk.metrics.f1_score(disag_elec, elec)\n",
    "print(f1)\n",
    "f1.index = disag_elec.get_labels(f1.index)\n",
    "f1.plot(kind='barh')\n",
    "plt.ylabel('appliance');\n",
    "plt.xlabel('f-score');\n",
    "plt.title(\"CO\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training model for submeter 'ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])'\n"
     ]
    }
   ],
   "source": [
    "#Train the data using Factorial Hidden Markov Model(FHMM)\n",
    "fhmm = ntkd.fhmm_exact.FHMM()\n",
    "fhmm.train(elec)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function nilmtk.datastore.hdfdatastore.DocInherit.get_with_inst.<locals>.f(self)>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# name of file to save disaggregated output to\n",
    "disag_filename_1 = 'outputDRED_FHMM.h5'\n",
    "output_f = ntk.HDFDataStore(disag_filename_1, 'w')\n",
    "\n",
    "output_f.close\n",
    "#Disaggregate via CO\n",
    "#output = ntk.HDFDataStore(h5_path + 'Weekly_COMBINATORIAL_OPTIMIZATION.h5', 'w')\n",
    "#co.disaggregate(mains, output)\n",
    "#output.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "##Disaggregation with factorial hidden markov model\n",
    "fhmm.disaggregate(mains, output_f)\n",
    "output_f.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dd2be79208>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAECCAYAAAA/0+q6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzsnXl4U8XawH9ZutGytWUHRQQUREARioAipWWXxQsHxCuoINd9AS+KCugVRL0CouCCioh+AqNy2RQQURFFAUFAlH2RUsraQqFL2iTn++MkIUmTNk2TNC3ze54+zZkzc+adk5N5zzsz7zs6VVWRSCQSicQf9OUtgEQikUgqLlKJSCQSicRvpBKRSCQSid9IJSKRSCQSv5FKRCKRSCR+I5WIRCKRSPxGKhGJRCKR+I1UIhKJRCLxG6lEJBKJROI3xvIWIARIl3yJRCLxD11JGS4HJcLx48f9LpuYmMiZM2cCKE34I9t8eXC5tflyay+Urc3169f3KZ8czpJIJBKJ30glIpFIJBK/kUpEIpFIJH5zWcyJSCQVBVVVyc/Px2q1otOVOKdZJk6ePInJZApqHeHE5dZeKLnNqqqi1+uJjo72+3mTSkQiCSPy8/OJiIjAaAz+T9NoNGIwGIJeT7hwubUXfGuz2WwmPz+fmJgYv+qQw1kSSRhhtVpDokAkEjtGoxGr1ep3ealEJJIwIthDWJLKi2o2o+bl+FW2LM+dVCISiURSGchIg5P++8T5i1QiEonEhVmzZtGtWzdSUlJITU1l27ZtADz11FPs27cPgDfffDOkMp0/f5758+eHtM4Kh8VcLtVKJSKRSBz89ttvfPvtt6xevZpvv/2WxYsXOzyXX3/9dZo3bw7AW2+9FVK5srOzWbBgQanKqKpaprF+iW9IJSKRSBycOnWK+Ph4oqKiAIiPj6du3boADB48mB07dvDyyy+Tn59PamoqjzzyCABffvklffv2JTU1lfHjx2OxWIpcOykpiWnTpnH77bfTu3dv/vjjD4YPH06nTp1cFMQ777xDnz59SElJ4fXXXwfg5Zdf5u+//yY1NZWXXnrJa760tDS6du3KhAkT6NmzZ5lCHkl8Qy4DkUjCFOui91HTDgf0mrpGV6Efdr/X8127dmXmzJl06dKFW265hf79+3PzzTe75Hn22Wf56KOPWLt2LQD79+9n+fLlLF26lIiICCZMmMCSJUsYMmRIkevXr1+fFStWMHnyZJ588kmWLl2KyWSiW7dujBgxgvXr13P48GG++uorVFXlnnvu4ddff+XZZ59l7969jjq95WvQoAEHDx5kxowZTJs2LYB3TuINqUQkEomD2NhYVq9ezaZNm9i4cSMPPvggEyZMYOjQoV7L/PTTT/zxxx/06dMH0HxdEhMTPebt0aMHAC1atCA3N5e4uDji4uKIiori/PnzrF+/nvXr1zvy5ebmcvjwYRo0aOByneLyNWzYkHbt2pX5Xkh8QyoRiSRMKc5iCCYGg4FOnTrRqVMnrr32Wj7//PNilYiqqgwZMoQJEyaUeG37MJlOpyMyMtKRrtfrsVgsqKrKI488wt133+1SLi0trUid3vJVqVKlRDkkgUPOiUgkEgcHDhzg0KFDjuM///yThg0bFskXERFBYWEhAF26dGHlypWOkONZWVkcO3bMr/pvu+02Fi9eTE6O5u+QkZHBmTNniI2N5eLFiyXmk4QeaYlIJBIHubm5PP/882RnZ2M0GmncuDGvvfZakXx33XUXKSkpXH/99cyePZvx48dz5513oqoqRqORqVOnelQ+JdG1a1f2799P//79AahSpQpvvfUWjRs3pn379iQnJ9OtWzcmTpzoMd/lFtYkHNCpaqXf+E+Vm1KVDtnm8iM3NzdkwzFGoxGzuXx8C8qDyt5e9ch+7cOVTR0e6L622dNzZ1vaXaIruxzOkkgkEonfhGQ4S1GUeUA/4JQQopUt7b/A7UABcBC4VwhxznZuAjAKsACPCSHW2NJ7AbMAA/CBEOKVUMgvkUgkEs+EyhKZD/RyS1sLtBJCtAb2ARMAFEVpCQwDrrOVeVtRFIOiKAZgDtAbaAncacsrkUgkknIiJEpECPEjkOmW9o0Qwj5Y9ytgn4UbACwSQpiEEIeBA0AH298BIcQhIUQBsMiWVyKRSCTlRLiszroPWGz73ABNqdg5ZksDSHNLT/J0MUVRxgBjAIQQXh2ffMFoNJapfEVEtrn8OHnyZEj3E7nc9i6pzO0ttP03Go0uod19aXNUVJTfz3+531FFUZ4DzMD/2ZI8rQZQ8Ww1eVxaJoSYC8y15ynLqptwWbUTSmSbyw+TyRSyZaqVfbWSO5dLe82Fhej0Wnfpa5tNJlOR598eeLMkynV1lqIoI9Em3O8SQtgVwjGgkVO2hsDxYtIlEkmAyMzMJDU1ldTUVNq2bUu7du0cxwUFBR7LtGvXjvPnz5e6LovFwqBBg4rNM3z4cBcnQ0n4UW6WiG2l1dNAVyFErtOp5cBniqLMAOoDzYDNaBZKM0VRrgLS0Sbfh4dWaomkchMfH+8Icjh9+nRiY2N54IEHglKXwWDgf//7X7F5Pvvss6DULQkcIbFEFEVZCPwCXKMoyjFFUUYBs4GqwFpFUbYrivIugBDiT0AAfwGrgYeFEBbbJPwjwBpgt5ZV/BkK+SUSiTa/aA/3PmHCBI97dXjKM2/ePJeIup999hmTJ0/GbDbTokULQAtbMnDgQFJTU0lOTmbLli2Aq5Xz9ttvk5ycTHJyMvPmzQPg8OHDJCcnM27cOLp168Y///lP8vPzg30rJE6ExBIRQtzpIfnDYvJPBaZ6SP8a+DqAokkkYcsHv53kcFZgO8SrakYz+qY6pS63Z88eVq9ezbJlyzAajYwfP55ly5a5DEd5y9OvXz/uuOMOR4DG5cuX8+9//9vl+kuWLCE1NZWHH34Yi8VSRBH8/vvvLFmyhK+++gqLxULfvn3p2LEjMTExHDx4kDlz5tCiRQtGjx7NmjVrGDBALtwMFeU+sS6RSMKfDRs2sGPHDnr37g1o4d7r1avnU55BgwZRr149duzYQYMGDUhLS+PGG2902biqTZs2PP3005hMJnr27Ml1113ncu1NmzbRt29fYmJiAOjVqxebN2+ma9euNG7c2GHRXH/99UUi/kqCi1QiEkmY4o/FECxUVWXo0KGMHz/erzz9+/dnxYoVNGzYkD59+rgsQQUtEvAXX3zBunXrePTRR3nkkUe44447fJLNOaS8wWC4LFZghRMydpZEIimRW265hRUrVpCZqfkMZ2Zmkp6e7nOevn37smrVKpYvX+6IvOvMsWPHqF27Nv/85z9RFIVdu3a5nE9KSmLVqlXk5eWRk5PDmjVrSEry6CYm8ez5EDSkJSKRSEqkRYsWjB07lqFDhzrCvb/yyisuOw4Wlyc+Pp7GjRtz5MgRrr/++iLX/+mnn5g7dy5Go5HY2Fjeeustl/M33HADAwcOpG/fvgCMGDGCFi1acPhwYLcPlpQeGQq+BMLFCS2UyDaXHzIUfPCo7O11hIJv1ASdzWFVhoKXSCQSSVgjlYhEIpFI/EYqEYlEIpH4jVQiEolEIvEbqUQkEolE4jdSiUgkEonEb6SfiEQicaFRo0Zce+21juN58+bRqFGjYkpIwgvpbCiRSMqR6OhoRzh4iaQkpBKRSCQlkpaWxmOPPUZurrb1z5QpU2jfvj0bN25kxowZ1KxZk71799K6dWveeuutIrGxJJUXqUQkkjBl17Zcss9ZSs5YCqrVMNDqxuI94vPz80lNTQXgiiuu4MMPPyQxMZGFCxcSHR3NoUOHePjhh1m1apUm565dfPfdd9StW5cBAwawZcsWOnToEFC5JaUgxEFIpBKRSCQueBrOKiws5LnnnuOvv/5Cr9dz6NAhx7m2bds69uO+7rrrSEtLk0rkMkIqEYkkTCnJYggl77//PrVq1WLt2rVYrVaaNGniOCdDsV/eyCW+EomkRLKzs6lduzZ6vZ4vv/zSZUMpyeVNSCwRRVHmAf2AU0KIVra0eGAx0Bg4AihCiCxFUXTALKAPkAvcI4TYZiszEnjedtkpQoiPQyG/RHK5M3LkSMaMGcPKlSvp3LlzyCINS8KfUFki84FebmnPAOuEEM2AdbZjgN5AM9vfGOAdcCidyUAS0AGYrChKzaBLLpFcZuzfv79IWpMmTfj2229ZuXIlEyZMcOTp1KkTCxYscOSbOnUqQ4cODZmsEk+EdmY9JEpECPEjkOmWPACwWxIfAwOd0hcIIVQhxK9ADUVR6gE9gbVCiEwhRBawlqKKSSKRSCQhpDznROoIITIAbP9r29IbAGlO+Y7Z0rylSyQSiaScCMfVWZ68lNRi0ougKMoYtKEwhBAkJib6LYzRaCxT+YqIbHP5cfLkSYzG0P0sQ1lXOFCZ21to+280GNA5tdOXNkdFRfn9/JfnHT2pKEo9IUSGbbjqlC39GOAcqKchcNyWfptb+g+eLiyEmAvMtR2qZdn2NFy2TQ0lss3lh8lkwmDb2jTYVPbtYt25XNprNlvQ6bR2+tpmk8lU5Pm3+/6URHkOZy0HRto+jwSWOaWPUBRFpyhKR+C8bbhrDdBDUZSatgn1HrY0iUQikZQToVriuxDNikhUFOUY2iqrVwChKMoo4CgwxJb9a7TlvQfQlvjeCyCEyFQU5SVgiy3ff4QQ7pP1EolEIgkhOlUNcaCV0KMeP37c78LhMswRSmSby4/c3NyQ+WB4G+rwJRT8iRMnmDhxIu+//36R8oMHD2bixIm0adMm8EKXgco+nKUesS3NbtAYXUQE4HubPT13tuGsEiNpVt5ZJolE4hclhYI3m83UrVvXowKRhANyPxGJRBJmLF68mHXr1mEymcjNzWXGjBmMHDmS7777jry8PMaOHcv+/ftp2rQp+fn5jnILFy5kzpw51K1bl6uuuorIyEimTp3K2bNneeaZZ0hPTwfgxRdfpH379uXVPEkZkEpEIglTfvzxR06fPh3Qa9aqVYtbb7212DyeQsEDbN26lW+//ZaaNWuSlnbJZWvBggXExMTw7bff8tdff9Grl+YDfOLECd544w1Wr15NXFwciqLQsmVLACZNmsT9999Phw4dSE9PZ/jw4axfvz6gbZWEBqlEJBKJC96Gs2699VZq1iwaaWjTpk3cd999ALRs2ZIWLVoAsH37djp27Ogo069fP0cI+Q0bNrBv3z7HNS5evMjFixeJi4sLeHskwUUqEYkkTCnJYgg1xU34e9rJsLhFO1arleXLlxMTExMQ2STlhwwFL5FIykRSUhL/+9//ANizZw+7d+8GtM2qfv31V86dO4fZbObrr792lOnatSvz5893HO/atSukMldqQrzgVioRiURSJkaMGEFOTg4pKSm8/fbbtG3bFoB69erx6KOP0q9fP4YNG0azZs2oWrUqAC+99BI7duwgJSWF2267jU8++aQ8myApA9JPpATCxX8glMg2lx/h4CcSSHJycoiNjcVsNjNq1CiGDRtG7969g1qnNy4bP5H6V6CLjAKkn4hEIqngTJ8+nQ0bNmAymejatatj5ZYkuKinT0BkFCTUCnpdUolIJJKgMWnSpPIW4fIk54L2FwIlIudEJBKJROI3UolIJBKJxG+kEpFIJBKJ35Q4J6IoSh20vTvaADWAc8AOtP3OTwRXPIlEIpGEM14tEUVRWiiK8gXwF3A3EAGcsP2/G/hTUZQvFEVpGRJJJRJJSGjWrFmZr5GWluZwQKwsVJg2OXltqBZL0KsrzhKZD7wO3CWEMLmfVBQlEhgAfAjcHBTpJBJJhcTe4Q4aNKi8RQkY/rTJYrGEbLvjS1zSIpZTGVCrblBrk86GJRAuTmihRLa5/AgHZ8NmzZqxf/9+l7RvvvmGN998k4KCAmrWrMns2bOpVasW06dP58iRI5w4cYLjx4/z0EMPcdddd9GvXz8OHDhAo0aNGDJkCCNGjGDChAns3LkTg8HA5MmT6dy5M4sXL2b16tUUFBRw9OhRBg0axNixY4vI9P333/PKK69gsViIj49HCEFWVhbjxo3j6NGjREdH89prr9GyZUumT5/O0aNHOXXqFIcOHWLy5Mls27aN77//nrp16zJ//nwiIiJISkqif//+bNy4EYDZs2dz1VVX8cQTT5CSkkK/fv1c7od7m0aNGsXLL7/ML7/8QkFBASNHjuTuu+9m48aNzJgxgzp16vDnn3/yww8/BP7L84DD2bBeI8jQoizroqK14xIImrOhoijjgfXAViFE5XX1lEjCkLjTKzCaMgJ6TXNUPS7Wur3U5Tp06MCKFSvQ6XR89tlnvP3220yePBmA3bt3s2LFCvLy8ujRowfdu3fn2Wef5d1332XBggUAvPvuuwCsW7eOAwcOcOedd7JhwwZAi/a7bt06YmJi6Nu3L927d3fZFfHs2bP8+9//ZsmSJVxxxRVkZWUBmiNjq1atmDdvHj/99BOPP/64I/rw33//zeeff86+ffvo378/77//Pi+88AIjR45k3bp1DqfHuLg4vvrqKz7//HMmT57skNcT7m369NNPqVq1Kl9//TUmk4mBAwfStWtXR5u+++47rrjiilLfa//RASqY8t3SgktJE+spwERApyjKr2gK5UfgV09DXP6gKMqTwGg0G+wPtD3V6wGLgHhgG3C3EKJAUZQoYAHQDjgLDBVCHAmEHBKJxDsZGRk8+OCDnDp1ioKCApfOsWfPnsTExBATE0OnTp3Yvn071apVcym/ZcsW7r33XgCaNm1Kw4YNHWHhb7nlFuLj4wHo3bs3mzdvdlEiW7dupWPHjo467aHlN2/e7NhdsUuXLmRlZZGdnQ1At27diIiIoEWLFlitVrp16wbAtdde67IXysCBAx3/X3jhhVLdk/Xr17N7926++uorAC5cuMDhw4eJiIigbdu2IVYgTmQGdg+akihWiQgheiiKYkDrtG+x/T0OxCqK8huwXgjxvL+VK4rSAHgMaCmEyFMURQDDgD7ATCHEIkVR3gVGAe/Y/mcJIZoqijIMeBUY6m/9Ekk444/FECwmTpzImDFj6NGjh2O4xo57GPjShoX3pbyv17Tni4rSYkfp9XqMRqMjXa/XY3GabHa+rv2z0WjEarU66igsLPQq+5QpU7jttttc0jZu3BiyIclwoEQ/ESGERQixWQgxXQgxEGgKTAaaARMCIIMRiFEUxQhUATKAZOAL2/mPgYG2zwNsx9jOd1cUJfj2mkRymZOdnU3dutoE7eeff+5ybs2aNeTn55OZmckvv/xCmzZtiIuLIycnx5HHOVz8wd27SU9P5+qrrwa0DaqysrLIy8tjzZo1RbbJbdeuHb/88gtHjx4FcAxndezYkSVLlgBaxx0fH++IEuwry5cvd/xv164dAA0bNuSPP/5wtM2uRNzb1LVrVxYsWOA4f/DgQXJzc0tVf2XAFz+RBOBWp796wC/Aa8CGslQuhEhXFOV14CiQB3wDbAXOOc3BHAMa2D43ANJsZc2KopwHEoDynxGVhC3qwT1wVXN0eulb6wt5eXmODhVgzJgxjBs3jn/961/UrVuXG2+80WVI6IYbbmDEiBGkp6fzxBNPULduXRISEjAYDKSkpKAoCiNHjuSZZ54hOTkZo9XCjInPOqyF9u3b89hjj3HkyBEGDRrkMpQFkJCQwGuvvcbo0aOxWq0kJiayaNEixo4dy9ixY0lJSSE6Opo33nij1G0tKCigX79+WK1W5syZA8Bdd93FvffeS9++fenSpYvDqmjRooVLm0aPHk1aWhq9evVCVVXi4+OZN29eqWWo6BS7OktRlD+BGLS5kJ+An4QQewNVuaIoNYEv0YakzgGf244nCyGa2vI0Ar4WQlxvk6enEOKY7dxBoIMQ4qzbdccAYwCEEO0KCgr8lrGyh4/2RGVqc8EfW8ma9ChxIx4idtA/veYLlzafPHnS0blWBP773/8SGxvLQw895FN+VVUxH9wDQETTFixatIgdO3Ywbdq0YIrpkZtuuok1a9aQkJAQ8rqDQeGBPbjvSKWLjsHYsHGJZU0mE3Xq1HFJi4yMhACEgj8OtAauBP4GGiqKkiaECJTNlgIcFkKcBlAUZQnQCaihKIrRZo00tMkBmlXSCDhmG/6qDmS6X1QIMReYaztUy7J0M1yWfoaSytRm6+EDAOTs301eMW0KlzabTKaQ+RUEQnFarVasVqvP13F+aTWbzVgsllKVLwvu7VVVFYvFEhYvD8HEl/aZTKYiz79tiW+JlDSxnmqbWL8RbSjrUeD/FEX5G20oa4MQYplPNXnmKNBRUZQqaMNZ3YHfgO+BwWgrtEYC9jqW245/sZ3/TghR6R1dJAFAPiVBYdy4cWUqP3ToUIYOLZ+1MZs2bSqXeisbJc6JCCEswBbb33RFUWqgDRWNBZ4E/H5tEkJssoVW2QaYgd/RLIivgEWKokyxpX1oK/Ih8ImiKAfQLJBh/tYtuVyQ6y7CCoclIr+XgGNzEwk1/kyst0YbXlqHNldSJoQQk9FWezlzCOjgIW8+MKSsdUokkvJGmoaVhZI81v8ErgUOozkZvgn8KIQ4HALZJBKJRBLmlGSJvITmUBjY2AsSieQyRw5nVRaKCwUfKYRYVJICsYUikUjCHDl84iuNGjUiNTXV8Td79mwABg8ezI4dO0p9vV27drFu3Tqv53fs2MHEiRN9vl5SUhKZmUUWZZYa99DupZXDHxYvXsxzzz0X1Dq8UdL34C/FWSI7FUX5EPg/IUSRMLiKotRD21fkHkDuKSIJTzyEy5AUT3R0tCOQYSD4888/2blzJ927d8ddmZvNZtq0aVPEwTAUuId2Ly85gsul59/1ewgcxSmRLsAzwA5FUbKAvcAFoCrQHG2Xw/lok+0SieQyYv369bz++usUFBRw5ZVXMnPmTGJjY9m+fTuTJk0iNzeXqKgoFi5cyOuvv05+fj6bN2/mkYceYv+2LZw8c5Zj57KJj4/nrrvuckTHzcnJ4fnnn2fnzp3odDqefPJJ+vbtW6T+efPmsXbtWsxmM++99x5NmzYlNzeX559/nj179mA2mxk3bhw9e/YkLS2Nxx57jLy8PFRVZcqUKbRv356XX36ZAwcOkJqaypAhQ2jVqpVDjunTp5Oens7Ro0dJT09n9OjRjBo1CoCZM2fyv//9j/r16xMfH0/r1q154IEHXOQ7e/YszzzzDOnp6QC8+OKLRcK5eMvjayj7nTt38uKLL5KTk0N8fDwzxj1OnYR4Bj/yODe0bMnGbb+TnZPD9DdmccMNN7h+D488woABAwLyLHhVIkKIM8BTiqI8CyQB16MpjizgFWCzEMJ7ZDKJRFImtmV8yrn8vwN6zRrRV3JjPe+e+wD5+fmkpqY6jt07nMzMTGbNmsXixYupUqUKc+bMYe7cuTz88MM8+OCDvPPOO7Rt25YLFy4QExPDU089xc6dO5k6dSqqxcL0bVv4Y+8+/vfV18TExDj28wB44403qFq1qmPY5dy5cx5ljI+PZ82aNcyfP593332X119/nVmzZtG5c2dmzJjB+fPn6du3L7fccguJiYksXLiQuLg49u3bx8MPP8yqVauKhHZ3lgPgwIEDfP755+Tk5HDLLbcwYsQI/vrrL77++mvWrFmDxWKhZ8+etG7duoh8kyZN4v7776dDhw6kp6czfPhw1q9f73MeT6Hsn3/+eUaNGsW6devo3r07zz//PB999BEJCQksW7aM1957n+nPPg1om2F99cG7fPfb78yYMYPFixe7fA+BxBc/kQJsjoUBrVkikYQlJQ1nbd26lX379jkUS2FhIe3atePgwYPUrl2btm3bAhQbDDG1S2diYmKKpG/YsIG3337bcVyjRg2P5Xv37g1A69atWbVqFQA//vgja9eudexdYjKZSE9Pp06dOjz33HP89ddf6PV6Rwj6kujevTtRUVFERUWRmJjI6dOn2bx5syP0PeCibN3bsW/fPsfxxYsXuXjxos95Sgplf/DgQfbu3cuwYZqrnNVqpXbVuEv3p+st2v1pcS3Hjh3zqb3+UqISkUgk5UNJFkN5oaoqt956q0tnD/DXX395DNnuiSrR0V6v7cs17PHFDAaDI7S7qqrMnTuXpk2buuSdPn06tWrV4vvvv6egoIAmTZr4JKNzDDN7Pb7uBGu1Wlm+fLlHRelLnpJC2auqSvPmzVmxYoWjjPr3AYczZ2REhCa33hD0sC4yrKnk8qDybwMdMtq1a8eWLVs4fFhzF8vLy+PgwYM0bdqUkydPsn37dkB7szabzcTFxTm9hRf/PXTt2pWPPvrIcextOKu4svaOfteuXYAWxr527dro9Xq+/PJLh9JxD+3uCx06dGDt2rXk5+eTk5PjdbVT165dmT9/vuPYLktp83jj6quvJjMzk99++w3QrMG9h4p333P9HgKHVCKSyo1cnFVq7HMi9r+XX37Z5XxCQgIzZ87k4YcfJiUlhdtvv52DBw8SGRnJO++8w/PPP09KSgrDhg3DZDLRqVMn9u/fT2pqKsuXr/BSq8bjjz/O+fPnSU5OJiUlpcg8RXE88cQTFBYWkpKSQnJyMq+99hoAI0eO5IsvvqB3794cOnTIY2j3uXPnFndpB23btqVHjx6kpqYyevRo2rRp43HY7qWXXmLHjh2kpKRw22238cknn/iVxxuRkZG89957vPzyy6SkpNCjRw9++8ODEnJ6/p2/h2XLyhLy0K2KkswzWwDGecCYQG2JG2LU48eLrFD2mXCJ7hpKKlObrb9+j/rhTHRJXdGP9h4sMFzanJubG7Jd8coj/L1qNsOxw6DTobuyackFAkig2puTk0NsbCx5eXnccccdvPbaa1x//fUBkLBsOA9n2dHFVIE6DbyUuISn584WxbfE1zCfdjYEegDWEiWRSHxEVVWsyz5DPXOyvEWRlAcVeHRx/PjxpKam0rNnT/r06RMWCqQ88XVifSbwoqIok+WyXklAOJ+FunIR6roVGN5cWN7SSCQ+Y98BUaLhqxJ5FKgLjFUU5TRO7xFCiCuCIZikkmOb3CSvdBObflOB33wrF7YvQs5VVRp8VSLhudZQUnGxhMqglb2VRBJMfFIiQogy7xsiCT5q9jms40agf/g5dG2Tyluc4rFbIhKJpAgrs0JAAAAgAElEQVSq1QonjkF8LXTR3n1NwgGflvgqihKlKMpURVEOKYpy3pbWQ1GUR4IrnqRUHNPWiVu/W1nOgvhACatk1F3bUI/65llcWbBMuB/rob3lLUZwkcOKvlFYAAUmyCzrisHgW+K++onMBFoBd3HpMfgTeDAYQkkuAyzFKxHrrBewvvREiIQJE86chJP+L0cPFMePH+fee++lc+fOdOrUiUmTJlFQUFDeYknCFF+VyCBguBDiF2xLfYUQ6UDJC5AlEk/I4SwXfA2nEWxUVeX++++nV69e/Pzzz2zYsIGcnBxeffXVQNUQoOtIilI+83++TqwXuOdVFKUWcLasAiiKUgP4AM3SUYH70MLOLwYaA0cARQiRpSiKDpgF9AFygXuEENvKKoOkHAixk1vYd16m/PKWAICffvqJqKgohg4dCmgxo1544QU6duzIL7/8wsyZM7nmmmsAbZOqSZMm0bRpU48h2BcvXszatWvJy8vjyJEj9O7dm+eeHl+ezZMEAV+VyOfAx4qiPAmODaneABYFQIZZwGohxGBFUSKBKsCzwDohxCuKojyDtq/J00BvoJntLwl4x/ZfAmHfT7pQwnBWwKgom1Ll5xVJqnbsOBF5gVUuhTHRZDes7/X8vn37ijjPVa1alQYNGpCSksKKFSu45pprOHnyJCdOnKB169ZMmzbNYwh20DZCWrNmDZGRkdx6663ce/c/8V67pCLi63DWs2gWwR9oe4rsB44DL5alckVRqqFtavUhaGHnhRDngAHAx7ZsHwMDbZ8HAAuEEKoQ4leghk2hSSoaoVIiFQUPSqQ88BZFV1VVbr75Zlau1BZtrFixgn79+gFaCPY5c+aQmprK4MGDHSHYAbp06UK1atWIjo6mefPmHLOlSyoPvi7xLQCeAJ6wDWOdEUIE4r23CXAa+EhRlDbAVuBxoI59b3chRIaiKLVt+RsAaU7lj9nSXPaBVxRlDDDGVp7ExES/BTQajWUqH0pM1atzDi04W80wb3N+lSqct312r0u1mDnl5VxpyatalWzAePE8zJhIjWemoa9WdI+K8v6eC8+fwb5ruMFgQKfTkds4OH68zj96o9G1C2jRogWrVq1ySb9w4QIZGRm0a9eO+Ph49u7dy4oVK/jvf//ryDdv3rwiIdh37NhBdHS0I4/BYHCZ+3GvOxSUR53+oJoNmNEMaV9l9uh55WN5+54p/uCTdIqizAB+ADYIIU77VZP3+m8EHhVCbFIUZRba0JU3PI1NFFFmQoi5gD0sp1qWwHrhEpjPF9TzWrdcUFBQJplD0WY1K9Px2b0u1emtvKxyWC9cAKDwrx3a9VZ+jr7HoCL5yvt7Vk9ceg8ym80+78tRFjwFJOzUqRNTpkxh4cKFDBkyBIvFwqRJkxgyZAiRkZH079+f2bNnk52dTfPmzTGbzdx66628//77TJkyBZ1Ox65du2jVqhUWiwWr1eqoQ1VVRxh2eztDSXkEnPQX1bFHStnvky/lTSZTkeffFoCxRHwdzsoBxgHpiqLsUBTlTUVR/mGzSsrCMeCYEGKT7fgLNKVy0j5MZft/yil/I6fyDdGG1SQVDLW41VmFQfRmD5NVUEUIVfiXEtDpdHzwwQesXLmSzp07c8sttxAVFcUzz2jvdn379mXZsmXcfvvtjjLeQrBLLg98Hc6aCJrTIdAR6IsWHj4OMPhbuRDihKIoaYqiXCOE2At0B/6y/Y1E28t9JGAPfr8ceERRlEVoE+rn7cNeEipWhI9i5kSsY4MYZccansGo1ezzJWcKEQ0aNODjjz/2eK5WrVocPXrUJS0mJsaj4hg6dKhjlRfAggULUAtMcPxokbySiouvw1lxQGegK3AbcAWwBghEOJRHgf+zrcw6BNyLZiEJRVFGAUeBIba8X6Mt7z2AtsT33gDUX3kI05dsj3gxsYPuLxGmSoQLzkpEpWK9EUguZ3ydZcpCW531JnCfEGJPoAQQQmwHbvJwqruHvCrwcKDqrrRUhGWt3oazLlclku20DWxl1iEV6UVH4hO+KpEXgVvQlvr2UxRlPZoVslnuLxKGhOu4vzNeo/gGWXY1TJXIxezyliBEVIBnU1IqfJ0TmQKOrXJvRAuD8jXafEhc0KSTVF68WSLWy9MSUaXfjKSs6CgXHe3rnEg82nxIV6AbcA2aT4cMER+OVKThLHdZ3awo1WJBZ/B77UaJ1w8bwlUuiaQEfB3OOgZsBn4ExgIbhRDh4WIrKUpF6JC8ymhLj4nVlr1azFAWJeJOmFoirhZYBfj+JBIbviqRmkIIU1AlkZSdCmCAOLDPTbgrE3tnqre5MAW6Pw1XJeI8V1OOOiQtLY2RI0fy3XffOdKmT59ObGwsDzzwgMcyixcvZufOnUydOjVUYkp8JgROq75kEkKYFEXpBtyNFmYkHfhUCPFd8SUlIaUivcB6nftwVyIB7vTDVolUpC+vDFwu7byM8HVnw9FoodlPAEvQYlV9pijK/UGUTeIvFWFOxKkzcfENsSsN+xBWoDudcF2dVQE618GDBzN16lT69u1Lly5d2LRpU5E83377LbfffjuZmZk88cQTTJw4kf79+7sEb1RVlZdmv0NycjLdu3dn2TLNl3jChAl88803AIwaNYqxY8cCsHDhQl599VXS0tLo2rUr//73v+nWrRt33nkneXlyVP0S4b2fyHggVQixw56gKMpi4Evg/WAIJikDFaBDcunMrdZLSqPIcNbla4lYF72PmnY4oNXoGl2Ffpj/735ms5mvvvqKdevWMWPGDBYvXuw4t2rVKubOncsnn3xCjRpakMuTJ0+ydOlSDhw4wL333kvflO58/cOP/LV/P2vXriUzM5M+ffrQsWNHOnbsyKZNm+jRowcnTpzg5MmTAGzevJkBAwYAcPjwYebMmcN///tf/vWvf/H111/zj3/8owx3pPKgi4gEVUUtcN4+IPh9ga+xsxLQQpE4sxeID6w4kssGT9aHdqD90wVoTsS9cw5XBRsmFlJJgR/79OkDQOvWrTl27JgjfePGjcyZM4cFCxY4FAhAr1690Ov1NG/enNOntditW3b+wYDU7hgMBmrVqkXHjh3ZsWMHHTp0YNOmTezbt49mzZpRq1YtTp48ydatW7npJs0fuVGjRrRq1cohQ1qac1Dvyxtj1RpE1EgIfb0+5vsJmKEoytNCiFxFUWKBacDG4IkmKTX2jqiCDWe5WAf2dMdw1uVoiWify2Ix+EvNmjU5f941jte5c+do1EiLexoZGQloYd2do8NeccUVHD16lEOHDtGmTRtHuj0/XBq29Bbapl69epw/f57vv/+ejh07cu7cOVasWEFsbCxxcXFkZWURFRXlyG8wGMjPD48dIQOOHy87Or0HmyAE70y+WiIPAK2B84qinATOAW2AfwVLMIkf2B+8cH3bdsbFEnFWKO7DWWVsS5HVX2G6t3uAvzPVakX1Y8vd2NhYateuzYYNGwDIysri+++/p0OHDsWWa9iwIe+//z6PP/44e/fuLTZvUts2rPj2eywWC2fPnmXTpk20bdsWgHbt2vHBBx+QlJREhw4dePfdd0usW+JGbNWQVufr6qwMoKuiKA2B+sBxIcSxEopJQk2wvb0DifucyKUT2j99oCwR9+GsMLVErAFe4ns6A/JyUa+42vMbajHMmjWLZ599lv/85z8AjB07lsaNG5dYrmnTpsyePZt//etfzJ8/33MmVaV311vY+udfpKamotPpeO6556hdW9t3LikpiR9//JGrrrqKhg0bcu7cOZKS5A7YpSIiIqTVFatEFEWpAjwPtAK2AdOk8ghn7PMJFWw4y8VHwvbZ3vGVVTG6Fw/X4axA47BCSn//mjdvzhdffFEk3TktPj7esTrLOeR7q1at+OGHHwB44403XMrv378fNS8XnU7HxEceZNKVrjshAtx5553ceeedAERERHDgwAHHuUaNGrn4r3jzW5E4U/4T67OB24E9wGDg9aBLJPGfcH3L9oTV25yI7X+whrPCdaivIn13kjDH6SUyDOZEegM9hBDjbZ/7BV8kid9U1OEsT1aJPlB+Iu5zImHaWV92YU8qgLUs8YmSlEisfedAIUQaUD34Ikn8pwJ1Pt5WZwV8Yt3tOFyVSLAskQr0SEicCJjFHPwHoKSJdaMt3InOyzEy9EkYEa5DNZ4op4l1NWyViLM1Vn5iBJ9K3biwQxcZFfQ7XpISOYW2l7qds27HKtAk0EJJ/EOtUMNZ3ibW3f1EZNiTSsVl0szyRBdXFTXnAtSpjyE6xsWfJxgUq0SEEI2DWrsN22ZXvwHpQoh+iqJcBSxC84jfBtwthChQFCUKWAC0Q1NoQ4UQR0IhY4UgXDtIT3ibWLe6rc4qpk1q5mkwm9HVrue9niJ+ImF6j4KmRGSvfdlhMKJrcGXIqivdAvLg8Tiw2+n4VWCmEKIZ2v7uo2zpo4AsIURTYKYtn8ROhXqb9eJsaEdX8hJf69OjsD5Xgr9rRbknHjzWy4PJkyfz/vuXwuENHz6cp556ynH84osv8t5775X6uklJSWRmZhZJ/+abb5g9e3aprqWqKqop36vneyCZPn067777rsdz/fv3D3r9FYFyVyI2B8a+wAe2Yx2QDNgXpX8MDLR9HmA7xna+uy2/BErVYarZWahbfw6iMCUJ4G2Jr1sU30B3qOGqVMJkYv2mm27it99+A8BqtZKZmenigf7bb7/Rvn37gInXo0cPHnnkkdIVys2BjDTIuRAwOfxh+fLlQby63efL/6KhotyVCPAGWpRg+68oATgnhLAP5B1D28ME2/80ANv587b8EihVB2md9SLWd19Fzc0JokDFCeBBcUDww56EqxIJk/ms9u3bO5TI3r17ueaaa4iLi+PcuXOYTCYOHDhAq1atyMnJQVEUevbsSffu3VmzZg0Aubm53H333aSkpJCcnOwI8w4wb948eg0YSPe77+XAkb8BbUOr5557DsBr6Hir1cqECRPo1q0bI0aM4O5Ro1j5/Q9QUOAi++DBg5k8eTJ33HEHXbt2Zfv27YwePZrOnTvz6quXBi3uu+8+evXqRbdu3fj0008d6d9//z09e/YkJSUFRVEc6fv27WPw4MHcfPPNfPjhh470Zs2aAVrwycGDB3P//fdz66238sgjjzispJ07d/KPf/yDXr16MXz4cEdk4hJR3f6XitA+S74GYAwKiqL0A04JIbYqinKbLdmT7lV9OOd83THAGAAhBImJiX7LaDQay1Q+lOTFxZENREZEULMEmU+dPY0KJNSsib5qNZdzwWizJess+mrV0Rm0R+58VCR2n+qa1atjtNVnNuVwFoiKjsHkds4d+8+xOFlz42Jxfl/1dm8C3WbVai1VuJGzBj32tyajwYDOaGTu5gwOZfq3X4Y13wyqim5/uktk3ibxMYzpcGkOyWh07QIaNmxIREQEJ06c4Pfff6dDhw5kZGSwfft2qlWrRsuWLalSpQpms5mPP/6YqlWrcvbsWfr06UOfPn348ccfqVevHgsXLgQgOzsbo9GITqcjMTGRtSuWM+/t2by7cBGzuvfAYDCg1+sxGo3o9XpOnz7NypUr2b9/PyNGjGDgwIGsWLGC9PR01q9fz5kzZ+jSuTNDe6ag1+swOMmv0+mIjo5m+fLlzJ07l/vuu4+1a9dSo0YNkpKSePDBB4mPj2fWrFnUrFmTvLw8evbsSf/+/bFarYwfP56lS5dy5ZVXkpWV5ZDp4MGDLFmyhIsXL9K5c2fuu+8+ImyhRYxGIwaDgV27dvHjjz9St25d+vXrx7Zt27jxxhuZOHEiH3/8MYmJiSxdupTXXnuNWbNmlfz9GQxYbG1y/45Kwmg0XnoJ8/AdeyIqKsrv579clQjQGeivKEofIBqohmaZ1FAUxWizNhoCx235jwGNgGOKohjR/FaKDLQKIeYCc22H6pkzZ/wWMDExkbKUDyXW7GwACgoLS5RZtb39n808i87k+kYX6DarhQVYHxqM7pYe6EdoQxdWp82EsjIz0cVoQeNU27i5yVx46VyVahRHcbJaL1x0OS4w5XvMH8g2q4f2Yp32b/RPTUV3zfU+lbE4raAxm83o9AasVmuZx/3dy1utVsdqHaPR6HHlzk033cSmTZvYtGkTY8aMoW7dumzatIlq1arRrl07zGYzhYWFTJkyhU2bNqHT6Thx4gQZGRk0a9aMF154gRdffJGUlBSSkpIwm82oqkrPnj2xWCxcf01zVq3fgNlsxmKxOGSyWq306NEDq9XK1VdfzenTpzGbzfzyyy/07dsXq9VKfHw8N9sCMlqtVlQn+VVVJSUlBbPZTPPmzWnevDkJCdpAxRVXXMHx48epVq0ac+fOZdWqVQAcP36c/fv3c/bsWZKSkmjQoAFms5mqVas6ZEpOTsZgMFC9enUSEhLIyMigfv36ju/KYrHQtm1bateujdVqpWXLlhw5coTY2Fj27NnDkCFDHPLWrl3bp9VSqkULFKqi+ry6yh4xy2w2O5SIt+/YHZPJVOT5t7exJMpViQghJgATAGyWyFNCiLsURfkcLczKImAkYLeJl9uOf7Gd/04IER7jAOGAY1gozKaJCjWFoG5cBzYlUlIoeJ3eoJmYAR/OKtvlfKpy7x/a/13bfFYinlaNjb6pjv8yHD2oXbNBA3SlDMhnnxfZs2cP1157LfXr1+e9996jatWqjhhZS5Ys4ezZs6xatYqIiAiSkpIwmUxcffXVrFq1iu+++45p06bRtWtXnnzySQBHGHeDXo/Z4jmasqfQ8d4bWfS8vbxer3e5ll6vx2w2s3HjRjZs2MCKFSuIiYlh8ODBmEwmwPteKu7h5y0eZHeuyx4mX1VVmjdvzooVK4pvRwUnHOZEPPE0MFZRlANocx72gcgPgQRb+ljgmXKSLzxR/RhIDcVYvF25Of/4vMbOCvCcSHlG8S2N7Kp6aUVaQOr2v+hNN93Et99+S40aNTAYDNSsWZPs7GyXzaEuXLhAYmIiERER/Pzzz44Nqk6cOEFMTAz/+Mc/eOCBB/jjjz/K3JT27dvz1VdfYbVaOX36NL9s2ez3tS5cuED16tWJiYnhwIEDbNu2DdBC0P/yyy8cPXoU0ELgl5Wrr76azMxMxxxTYWFhiWHyL1Fx3o3LezjLgRDiB+AH2+dDQJFNBIQQ+cCQkApWkShVh2t761JDsL+Gp7dOT5F7nT8HK+xJSCbW7W+0pVQipQzZ7uOFS12iRYsWZGZmMnDgQEfatddeS05ODvHx2mamd9xxByNHjqR3795cd911NG2qReTds2cPU6ZMQafTERERwbRp08osT9++ffnpp59ITk6mSZMm3NC6NdVi4/zqZ2+77TY++eQTUlJSaNKkCTfeeCMACQkJvPbaa4wePRqr1UpiYiKLFi0qfQVOREZG8t577zFp0iSys7OxWCyMHj2aa665pkzXDTfCRolcTqgH96CeOIa+c0qAL+zHryoUlognBz+vw1m2/0HbTyQE7fVDh6BawRAeAwMGg6HIG7N7WPf4+HiPwzSNGjXitttuK5JuDxuv5lygTYtr+WLOm4BrGHlPoeNBG4qaNGkSsbGxZGZm0q9PH669+ircb7BzqPpOnTrRqVMnl3P2+QHnFVnOJCcnk5yc7JI2btw4l2PnUPR2+dzrmjp1quNzq1atWLJkicf6iqXiGCJSiZQH1lfGax8CrkT86HBD4cHtYTdBtaQovoEKe1IeS3x1floiuiCEegm3zshPeUaOHMn58+cpLCzk8YcepHaCXNlfBJVymQ6VSqQy4c8PNBSdqsfhLB/nRMqq5IoMZ4ViTsSPX3IYDWeFI85WhnoxG86cDF+fn8uM8LCfJYHBL0skBHMinuQqKQCjLlgT62W8nC84DJFwmBOpjITZ6sOgUHEUpHxqKxOOTsuHH5l9yCUUb+aW0syJuFkiZf0xFRnOCqElUto5kUqkRNT8PNTCAk9nyn5xf5T05cRlGPZEEigcHWSYLfH1ZO24WCIeFIp9TqTM8pVnR6Oinj2F5f7+qPv+LD6rVb20mCCwIpQPJ45B+t9BuvjlYIn4T6jvjlQilQm/5kRCMbHuoQ5P4d+dsOhsDnIVcWdDh5WnXnI83PBN8WVUFashQG12vXAAr+Vjjb7c4zL1dLY2VWZLxI+mqR4+hQKpRCoT/nisl8PqLPXEMfjjN6cEVz+RzOrNWG25nTPx15VdyakqxupOK3lCokQ8HZT0w1ZZfeN/OVuzpU9VqFYraoGn4aKy06BBAx599FHHsdls5vrrr2fEiBGAD+HbbSFrJKHGzaK3WEOiaKUSqUz447EekjkRNyWyw83j2M1PJLPmtQCcrdmizD+CSGMU9f45lrhWSba6QrCQwFmL6C9ZJcViuwf50TV9q+LMSTj+d1C2+61SpQp79+4lzxbfzB5Y0E6J4dvt37fBgKqqWD3NeZUFtciHSkjZ2haRn0+EKR9sIV2CiVQilYnS/EDt/VxILBG3OtxjFLmFhVdtK7N0qlpmJWe0DRFF1mnoUpd6eF/wNjXSOSsOH5VIaWXJz7UXDOx1bXTr1o1169YBsHTpUhfvdefw7adPn2bUqFGkpKSQkpLCli1bSEtL47bhI3j2ten07NmT48ePs3TpUrp370732/sz9e33AB3Lly/nhRdeAOCDDz7g5ptvBuDIkSOO+mbOnEmfPn1ITk5m/PjxqKrK/oMH6Dv6AUfT09LSSEnRfK78Dr1eRlSTCTUvt+SMocYc/Jcm6SdSmSjN6qwiZYKI+9u/e4wot0l21e50hzUAL5tuF7BaUbdtxPrOK+jueQxdoB0+gUuKw+qmUIrBw/ld23LJPue5E1BN1bQyh3I974+QXw1Q4UABOv2lKK7VahhodWOVElswYMAAZs6cSUpKCrt372bYsGEOr3NnJk6cSMeOHfnwww+xWCzk5ORwLiOdg0fTmD7xOaa9NYcTJ04wdepUVq9eTTWdyvCR97D6xx/p2KOPY9fATZs2UbNmTTIyMti8eTNJSZrleM899/DEE09A+t88NukF1q5aReqtXSgsLOTvY8doXLcBy5cvp1+/fhQWFvL888/z0UcfkZCQwLJly3j11VeZMWNGie0tMxlazC0aNwt+XV4ooAATJqpQK6T1SkukHAn4UIQ/1wuFJeI0nKVaLy1lPV4niUJjlSJLfE/UbgeAXrUEYE7E/diKejJD+5yRVrZre8MvX0M/tWWQ3gFatmzJsWPHWLZsWZFQIM78/PPPjrkSg8FAtWrVQIWGdevQrnVrAHbs2MHNN99MQkICRoOBQT1S2PT7TmrXrk1OTg4XL14kIyODgQMHsmnTJjZv3kwHW7j3jRs30q9fP7rf+U9+3vo7e3drq9xuT+7GirXfAtoOg/379+fgwYPs3buXYcOGkZqayptvvklGRkZwblAYck6fRZ7O1RrShWDIT1oi5YlThxoQ/IqdFeLhLKsFdDpyYuqw/fqHqX16Gx2cFIVqVbkYd4X9wKc2qarqNYx3sbIEC52Tn4jt+1ULTKj5eeiiYzyLpXNe3qu1uTiLQT16SmtLowR0hqJLg9W/T2n3r25Dr3WWRI8ePfjPf/7DF198UbqotqqVKtHRjvvgqiBdv8927dqxePFimjRpQlJSEosWLWLr1q1MmjSJ/Px8nn32Wb5euYL6FhPTP/wIk6kAVLi9ezcemPQf+ijD0Ol0NGnShN27d1ee0OsVaLpHWiLlScAntf1ZFxhiJWKxgE6P2RgNQH50gqtF5iSPXrX65idSbBuKDmddItgr6p3mRHZsxvrEXV5zWp3f50r1NZa86stfhg4dypNPPkmLFi285unSpQsLFiwAwGKxcOHCBaeoA1rbb7jhBn799VcyMzOxmC0s+3YdHW+8AYCkpCTeffddOnbsSKtWrdi4cSORkZFUq1bNsc9Hzfh4cnJz+fr79Y42NW7YAINBzxtvvEH//v2BsoZeDzcCpEVCoIykJVKeBNrRL/2oHzKEeImvRbNEVEdYE1drw2W1r3uC1+tbvTvquVsyLu0N0i/M/r3mXERd6RRO3OJ9hzlXS6QUBE+HUL9+fUaPHl1snv/85z+MHz+eRYsWodfrmTZtGrVjbBs02ZRInTp1mDBhAkOGDEE1F9KtQ3t6dr0F0JTI8ePHSUpKwmAwUL9+fUdY+erVqzN8+HBSUnvQqFYCbVpc61L37akpTJn1Fr/++itqYQERqtU19HphIaPHjAlp6PVSWcXBEyKk1UklUp4EcC8PNTcH9bef7Ec+lPBxwjcQuGxGZQG9DqtOe/T0qsWlY3ce+rDqI3yTrzRtsFpDYIDYVoBt2eBzEauXDalUVYXzWVC1usdhq2BgD3HujHO4c+fw7bVq1eKjjz5yyatmnmHdp/Nd0gYNGsSgQYNQs85o7bHRuHFj0tPTHcf2vdntPP3004wfNw6OHdYS4i9NGj9w9108OF7bl049oslsD72uZqSBKR+uaFKappcdVb00jJefC5HR6CpROBtPVO7WhTtlsAKKLCc85GS2l7ZTDTaq+5yIHqtBe1vVqVaX8y4RUPRG39pSmjaUl4c+FF3a7ITFZThLvaRMTflw7iykHfIy+R6Gg+fFhd/xS1wvhQoLUAu8+EHYHR5DEdbHGdt3pFrMcCIdTod+Yl8t5igYSCVSnpTQ+akZx7S9st3TD+zG+tgwV6c9s5P3cpgpEdXZEjFbQK/HYtD2rdapVldLxOlHb9V5t0RcOtTi2qCqWLB4zhus35c3RVXMG6nLcNap45B1xnYtJyE9DYeV2IZyUDJ2mT1WXaZ4Hq7Xt1rhuJchXF2AthIoLfbv3l5vXm7w/JG8CxHS2sp1OEtRlEbAAqAuYAXmCiFmKYoSDywGGgNHAEUIkaUoig6YBfQBcoF7hBDbykP2gODhAbf+71OwWtD/YyTWSQ8BoJ+7zGWc1a5Y1F1boXY9dPUagdnWwcRW9d7x5udhHXu3NpxjD5kRkjdzt+EsnUCJgOwAACAASURBVA6L3m6JFDecZfQun5fJeHeyDFl8HPF/pFS7nhru5YKFH3VY9W4/xexz2tCN+3ySMcKtZEn+J6UWpexYi7NEVLc8vuBHIxz70YQiQoETntpXUABRUf5dpwJQ3paIGRgnhGgBdAQeVhSlJfAMsE4I0QxYZzsG6A00s/2NAd4JvcgBxHlCWVWxLvs/1K8F6uovXd9ezrh53UZoHbD6wyqskx7WLBa7+R4V5X0C98h+KCy4pECgfJb4qrhaIl4m1q16o/e3uBICONo5a8wE4O9qZ4rmDdbciDd5LBav7bHiYZlu1tmi984HtDrKsRMqrgO0n7Jag/uGHsqtDpyxW9LObQt5LDHV48dgUa5KRAiRYbckhBAXgN1AA2AA8LEt28eAPebCAGCBEEIVQvwK1FAUpV4wZFPPZWKZ8zKm7UW9dAOGcwdxaC/qysWXjk+fuPT5XKZruUi3t5pTxy9ZIjGxUOj5oVU9mf4hdjbEYgGrBYsxgk5XrKFqleziJ9a9yeejEilazkmWUA9ngVcF73F11vlMt+XRnvZl8XCx0xmXOjFzYVDiaxWLT8NZqu/fWwnKxqMyKq/hrLyL2kiB8/ccAiWi5uc5HQS9OhfK2xJxoChKY+AGYBNQRwiRAZqiAWrbsjUAnN2Mj9nSAk9sHGz/lYufvBuUywOunWfmaZdT6sE9lw5ysl3O6dyUiJp5+tKDGh3j9aFVN34HCbWhwZVOiSF44pz3lbBoXuhRNfS0qLWdm9psd5tY93F1lvPKtmLb4GGJb7CXYBbXcZmLKhFVVbW2uhMdU3QosGjpokm5OZc+Z56Go4e8bBAVJIqdWFc95Cvt9dXij+FS4MtQK5FzNsXvPOFfzNLu4jDEVkPnZdWeM6rVqu3fYkMfYssnLJb4KooSB3wJPCGEyFYUxVtWj2GCPFxvDNpwF0IIEhMT/ZLrVLUamA/tJcGUg9G54/UDc/pRMsePJmHGfGwDK8TXqI7BJluuTuUCEDv8fnIWzaPK+bPkREZCQQFxQIxTG/KrVeO807Wjjv+Nsck1XAQiq9fEknvRY5tPZZ4mulMy5vS/KbR17BFbfiSueQuMV17tyGc0Gv2+Z544ue6SB3GNalUprFIFVX/pq4yLiaGKrb6zsXGOdIshkqpxcS5tt2PNuYBd7cbXqIEhwbO8J9znEFQrsVViuQjExMRQ1XbtQLb5YnQMOV7OJVSrhr5adVeRLBaOO+ZEtMdZFxUNej06wKrXg9WKHjAYtXz2bsJoMKAzuv6Mi3YhKqT/jbFpUadBo1vZzMxMBg8eDMCpU6cwGAwkJGih9FevXk1kZKSXll3CrKo2N0tdket/unQZ3W5sQ+2EBIw6XRHZPaFazNi7YYPBgGq14qwajHqd47zufCb66jWx6A1awAAu3TNP7Q0U7vdcr9M5ZNSjc5EBbC9LFjO6InNcGla9AUNsVfRmM5Qgs2q+dH/AFrjU3lXqi34HnoiKivL7+S93JaIoSgSaAvk/IcQSW/JJRVHqCSEybMNVp2zpx4BGTsUbAsfdrymEmAvMtR2qZ86ccc/iGwPugk/mkJV+DF1UrH/XsGFdvRQ19yJnV37hSDvz8FD0by5CFxmF9aQ2fJXXpRd8+Sm5589DRBQUFHDh5HEunrZZGxfOw7lzLtfO37QBXUIdAAoNRtT8fOxtVlUVdv4G192AWmAi32qF6EvhNExbfsL020YMc5c60hITE/H7nnnCGOGwjjLnz0HX6kanIItw8eJFcm315WRfcKRb9FFcyM4mx4Ms6sVL1lnmmTPoVM/Whdn9rcxs5uKCOQDk5eVisl3blzZb578JVzVH37VX8flyLno9d/bUCXQFrjKpZjMZ0dGcvbCFwkLN0U7V6cBiRdVrS6LRqVgtZlQ3S8ZsNqMz+Pama3YrazQai6RVq1aNb77RNtCaPn06sbGxPPDAA16v4Qn78JmqqkXyL1q6jFZXNqJ2QgLmnIvaMti6DYv1gSnMz3d0VBaLBdVqJScimlizCZ2qYnaa47NmndXuf4RmrVvNhY575qm9gUOH8/us1T6krNNhtVq04wITREah0+lQz5yEi9lwxdWe/UisVgx6A+h0Jcqsuj3j2QUmqkfGaBLFxPjUZpPJVOT5r1+/fonloPxXZ+mAD4HdQgjnUJvLgZHAK7b/y5zSH1EUZRGQBJy3D3sFA13tetpjYXsgiot9BLalrHk56OKqacfO3qtxVbX/Tp0fFgv8+Tvc0BFyLmgPWEQEREZqS3btcZe+mK9Nrl/IRt36M7rR4y5do2p1yL14aQgrpoo2eW5n93ass19C12+olh4RBVUuve1rFfhu8qvmwv9v77zj7ajKvf9dM7Pb2afnpIcSOoGAKEIQRaWoGBG86IiAcJXrvSq2KwpYLwgKAopXfVFBReqVAUHpCAiChBYIoYVAIL3n9N2nrPePNbNnZp9zEghpJ5nf55Ocvfe09ayZefrzLJASkdqwRlrHmHGQycKS11VK5rR31AvMpRAxl0M0rd/V0yArw5/zTWZnbUr/sHzsAXjsAdiAEFmvC2XYeJVkVnsTlOZRdTqBViU4PFe5ajRN/X2brhnpeW+r8O2KK67glluUEnTqqafy+c9/HoDTTz+d1atXU61W+cIXvsBn3ncYjuPwzQsv4uWFi5FScsopp9DV1cVL81/lS//zI7LpNHf+/re8vOB1Ljjz65QqFbq6urj88ssZO3YsJ5xwAjNmzOCpp57i2GOO4T+O9bstS0nZ9SgZaQSSvF0dOi+1aj35ZIu5s4SIu9UC16OmqzGUiipW1T4G2jtDl6PrgPYW3qXh0ODOK9o2OT1NZgsVp25tS+Rw4LPAC6ZpPuf/9l2U8LBM0zwDWAJ8yt92Nyq9dwEqxfdzm3V0wYNo11Tm1J03oX3nUsRueyMXL4CmZsTYcLEeefuNyLtvRvvK9xEHHoL8vyvxXnoW7T/PVjEWUMIiAu+Kn6B96ydQLKj0XIBUSmVQRTQI+carsNo3unoiGkO+WVkntapiPOlsjFHJQSW05Iol6mFOpYYG5tcD9yffQhx4CNpM5WL0LvhvWLsK7dI/IfLNGzg6OImDmLQzUteVkPM8hAE9nk0fFZqCCu/XX0H2hdXMrp4ZWUBEtf03yyg6ulScod69d6j14v3tBtB1tI+dhFy8AO/Cb6Kd9ytEgztTzpuLd5eF9s0fIRpbrvhjFqd8EXlDQ0xt2FoPxQS0tI2WCTXYR56by7r+wXAONE09k1KqIkSA9HPx+hPXVUV4qNYymnRDBjd7LmPHjeOII47Y4FQ1Ys6cOdx6663cdddduK7LzJkzmTFjBtOmTeMXv/gFHR0dlMtljj32WI49YBpvLF5MT19/fU2S/v5+2trauPrK33Heud9k+i67Yleq/M8vfsXVf/wjYyZP4dZbb+XSSy/lkksuAZSFeuutt6qgcd3nH2WY64l7bFQq8duAEPGh2bb6TdfjCkBfN7KlTd0zz1XPQyqNLBdVdXud8b8F7adBiAjDxaZGho1rvPlWsVWFiGVZ/2LkRMujhtlfAmdu1kFFkfL9lU4N+apqQS0fugum7oV34TcBFDNt71Tb/NYL3m3XoR94CPKRe1Va52P3w+7KHx11wwTwLv8BTD84FDSpjLIa7CriIycqgVWrQmsbrC3HU34Dq6JcUr7TVCpeeBgwuIDppNJvSYiw8FW1gNN7jkJ0jKkXd3nfOBn9qtvxnn4UOfsxtC+eU7e6vOuvgAmTEQcdhhgzTglDXfcF3gBID82AG+zVkIJPedORyxfjXXw2Xn4SHDYDgfQtkRFqXp6dFZnAN9mAUQjElF1VS4zGbaiMvCBDTn5wJvLluer0530V/arbI/t14/3h56p9R38fdIyJnQfPg0wWsef+Q1mBY+N+4eOImSbaCaf6+6u9Ot41j0znAchaV6QuSIZCQFIXEsPC8+rbXD0NCITrqnO9zbTfJ598kpkzZ5LLKcb0kY98hKeeeopp06Zx1VVX1V1gK1euZPHSZUydMpk3Fi/hhz/8IUceeSTvf//7fXIkpSaXQqfO0lmLeXXhIk467XTQNDzPY+LEMNny+OOPHzoQWf8vQvd6Eg62mBBp+O7YSqHTgnsXGUelFCZ3OI7yYKxeoRScCVOGnFq1vumBphbEcDGphndEz1YpUqVZ5mh0s20ObG1LZNtGUI9h23UNUj47C/Gxk+q7eBefrRjC+z6EaB+jble5pNqSBKmt5XKogS5eMPQ6nqcslIglImtVxXwDpl8YUK6rtauQz0cq1QMhUiqp2EMqrQSX58Y15GCZzFRq2MIn76G70T740dhvsUypP1+J/qXvqAc9kk4of/8zNf7XXoK99kc6DvKf96ptN/0B7Xe3Kdp1A9HUjFy9Qlkisc7nUrmKAOknDBq69C2REYTIX68Px3bL1ehf+f6w+8UghHIVBuf4+1/x+nsRn/uG+qF7Tbjv2lWQzQ47F/Lhe6C51RciPdAxRrV6/+e9iA/OpN7iXx/GdeRbUPIuCwIh4irrw2iKZPQIwREH7B8GVT2pBLFhKMszQKTVu6xW64sjDbTsghQarYOLEbkm5fKctDOUCkjHeVMB7TeDRx55hCeffJI77riDXC7HCccfT7VWpaOtjftvuIaH5r/OH/7wB+6+++66hQHg6oq17bv7btx6882IwN0bQSCw4kxwmOdhOEERvHsRAeMN9CF71sHEKQh9E7M+oQENwiybU0Ldcxv6x3kxIVIffyxNt6EjQ1+P+jfcolfry07c/DJk20nx3SYRZE7YtdB9UquptElAfPxk6F6DvPbXSM8Nc7X7e9VL60OWi+FDFE2/FBri+JPVQzDQH7FE0lAu1j+LtAqwi3G+lhapGwlcSrJcDC0RANsXWkGqYbUcnnsYSyQmmAJEXS/BNQNBF8SGpkwFwPvlj5DzX4BaQwyjWvUrrQ0l8EpFFTQ04i+JfPm5+pwAGLqHq6WRjz8UjrEwgLRrQ+se/PYvsjRyQFudWwyhXT75T5jzuPoyGElY6O+FSoSWakRwLpgXzkNft/rtobuQ1h+Qj96n7qfQhu8sHBUAAQbiv0nhjzUoxBTC12i9oatCuk5EwI3ALQJ3V7Wi7uOyhaq301vAoYceyj333EO5XKZYLHLfffdx6KGHMjg4SHt7O7lcjvnz5zN3rrLeugtFpJR87MgPcNYZn+OFF15ASo98LkuxqPq+7TltP1atW8ecuere12q14Vu3x2SIrH+v/7w+IVKr1tvueIMDykIojZQ79zYwHCNvaVP3zpO+MIs0PQ32d52YoJOep/jIYMRjEU2Bd4ezurYuEiGyPgQxkVotJhSCz2LaOxCf9b1rvT3huteuA73d4f6ViCUSZWLpTJgp1b0GUbdE0uGDHgTaA8ukETF3Viocs1PDe+Jh5DW/Ut8Lg+G5G4XIzruBbSNLRdyvfpryg3f654gEgQN3WPA32FatKEFSrSghElg8QQzBrqkXWtehKV8XIjFLpFpRtSTZXL1FvKH7MZEFLwMgn30c779PxfvNxaolCCCO/Jg6fv93IRe8jPf1k5FznlD7OzbuxWfDskXhdYYRIgDe7y5Bui4ywsxlf09McMRe6loVWvzkie616thb/qS2rVruWyK+P7wBMnIeuWYF3lOP4P3krCH7hS4sXyAJLcKMIli7CpYtQkYr/zOhL1wiQmEWvZ/98Qy/DeGggw7ihBNOYObMmRx33HGcdtpp7Lvvvhx11FGUy2WOPvpoLr/8cg7yVzNcsa6HE794Jh/66EzOPu98zjn7bHAcPjXzWM777k/55HH/jugax+8uPJ8fXfxTjj76aD58zNE8+8B9oZJQLg1lmsPJyShdgcCUXvg5amFC3CrYSEjPQ9ZGcCumM9DaoZJkhKbG4nmhVRntFxe1REA9W5Vy3D0W3e4/k7KvBznQpxSIrdwiJXFnrQ/pMLBOqaAejloVGWHwomuceq67V8fN0b6IEIm6tibvAgtfVZ8zmVCjd2zFZIPrrvA1/1xeXdeuDknlA0LrpW6JaPUxy79cE2oxfkM/kUoPfQ/zLYqR9/dCpczAr3+CdsVfQmsmoAFCy8bxNeCBXsR7jkJ2r1ECIhAyfoYatq3cNbqhaJUe8vYb0T4aZjjJij+fXROQA+rtMdIasqzh5ZrRAe83F6l9XphdF9Bi3wOQc59CNLciV6rAq3fbdegHzVBM8vVXYNe9QhqEqKd+NsJZvKAunAB1fDViiUS32TVENqfmceXSUHkA/znJhow/QJDmHLVEVixRbrmGjsyhJRIIEaH+BQK5Ea4Da1ZBa7v63jEGipGlUQO3WjQzzH+WZF8PZHOKHkeloQpfMfnm17+m7hsg7RpfOuPzfPnLX45dOpvNcsNlP1UW6PhJSono64bmVu67+qpwx/GTQEqOO/pIjvjM0YBGOpNl+t57cdt11yhX8Iol6vnq6+G2W25R7d9XLAY/fd2fnTjtUasdwmwoz1PCNJ2GwX6VKRkI4DcpRGS5pBSewNr3mbUQQt3H3nXI8ZOVuzAY1/hJiFykHCCwIOtr3jhKGagH/t2GLgpyaLFwVIhUykpxDPiLbmz+4tkNILFE1odAqy8VlcbQ1hF+97fLgu/j7u9VzMBnHDLQ/PMt6iEPLJHoA5bJxlOGA6vCSIepwNL160Wqw1eiB8csX4ze2cSESf+i6cAOxTCG01BSKUQmiysEAynf55/2A/luJKvr77eF10ullVYYVOIG/mTHURZItaIEWKkQurOCFyNqiWTCGEPU0+M4vvBt76hbIlpVzbEXjDHIgpswOWTauWbFnF0nZMQrl+L9+SqwfWFXj09LVuo5qg2uI+1c5ad3ly9R9ymdUYynUo4rBdGEiFqtLmDlS3PiMaJyyWcaEulbUQDaNy/wzxOxdkolxHA+boSaWynV/AXCJOL6kAjcqJCqVcL7LQRN6QJjmlb77pSI1gvK+nVsxRT7umHVMvW5Zx2sWaniO66rrLiglfmq5bByqbK6HBu5bFHovg3uebGgxqdpQxnb4ICKfYmAQj/93Uipxbs8LxTw1XKo/LhuXfA6muYLcxnMknruokKhbolItYPhJ2e4bujC812AcqAPWSmHdS2uE19iYbVq5V53F65YEmZIBtdcvVzNlUSt+ZJrqCfT/fG5jhpbIFQC+hpTtwOLMvqCrK9jtTPCew5UXHeLdEBJhMj6EMREghe/xa80DoSIkUJedan6XK2ohz/QBoNU3tZ2352lHjrRFHnI8i2hJQLh51dfqP8kr/m1bwHVlGBozPOP1HzoQl0zu0+rYj7DpeAKDdIZHpwyjev2eS+20FTVrG3HLY+1q0Ih0tKmGHdghQQ0VMvqoX78IaXtDfSF7qzXfAYaMDfDiLlZhBY+3raj5lO0ddQD69pSv1Nx4DIImFK1GjaQTGfAMFTiQ3CPhFBZcnVXgzquKCW3du3Gwwsb+of5ixx5gRWVySpLolaJWSIyakE4NWSQAbduNSyOtOv3tVeKA8irLgt/D9xoUbdYuajakbQ0VLBnMur5CWguFdQ/KevWxGA6S3e2WbmrQD2rESGSTZXDeQsWgQoYmZHye5iNkDHU1x0qAaWimt/g3AO9al4cOxQ+AXQ9onE3IBCKjUpzZ5d6Visl6q1sPC9eMGTXqOhpejLN1DRVTFtHo0spdm0RupDciNuoVlXX7FmrUoeDbMe+HiUUGuNWpYKi0x/nkHhSIJyHswiC+16rKgESCPV6f7EGIeL5mVxRV2hMyMj4nEfP1YB1pRLlcnnYbZsSiRBZD4QQSgsPrIrARVP2rY+oJlitqiysNl+IFCJCpFwKs7uisZXmllj1uFz8OtJzw6Bt8HvwUJcKDHkLIz5+t1riHrubfsNVzKZWhQmT0b57GeJY1cpCPvkwcv7zLG5RLQ7coPbAseOWSH9vmEra2q4e1GAcwfgCWoKXqjAYdwFBXUuXK5bGrC4RaXtSc/1jWjtU8SHQv+tTVKf8CdeVys0SWAIDvfWFiOSCl0M30UC/KuI64N2+oFH7BKmyFcOmbfoCupvD+QZU3Qr+falWfUskA9Wq0rQDRSJg/plsKNCDuYrW/pSLwxYUyleeV3/rq0/6+1YrMHYCWjpyTF1YD6ME+Jp+NbAGI5ZWVIgMCy+ofhehZRPAX7ZYXb8WZ0yuE7p2G334Ues4iNloGkOeU9eJWSJ1RIsCg/N6biyYjPSwfeXJ8f9qGRsv7THEvRVVssrFkBn7169fKyp8guc42N6j4lykfSu4XIq7nCqV+HXlMONAJYLEhEE9ttXgvmrsJ+bJ+D0M5iK4b43NRwdHiG8J8LZAvCQRIhuAyDXVmyMGleh1SyQWePbdH4El4jM90dqu9gu0pyALKThf1BKZ9SDyXw+gnXUhWjRl9Z/3+OeMFypCmJ3F3tNZMwYWeGX+1VpWAqBaQew9HTF1L5VJtud+yCf/ibznL/XjJUIF7227zvxErklpZVFLBCIZWkFacUOWS6k4NDsr0ISeflRlkAXj1sOH23V9ptmm3FljmlbRN/UF3PZnsNO6cpOUiqFrwD+PtP6gmIZjKyGQb1GZbHatLgCl/4T3tPWQHdeLtlNYzAgooaBpeMUCsuZbIpms+lwqQJff+zOwdHJN/lwNw4RyTcrt1yhIAXnL1UN+o+xbd5ksE/YO+4rWGW06M2zdAJH01HiGUvAtwoBizEhGXE0yHrSOarROPGNoiOY8Uoda6alOw5o+1OJwHXUfGwWciLifAuboenEhEmGaQfNRPVPDaXaH8u7GZIb6uiJRjT0Sd2htV9ZrlJnXaYzMh9PQCDPG+FXGmNBAuP6zadeUhRONjWq+EAnmPZqBFzuXEiIuVSQRq0z3V/psELBDFDcf+bTLplyCeyQkQmQ9KBaLLJ2yJ8tWr8EDerPNdGfyyFKRgpGhGs20GegH6VFq7qBopJHFQTwEy3Kt1DQ9VqnefsRx7HTmj5GdY/Ee/Xv8on3d9OtpnGkHDR1Q77r4yz1hCrJSQvvBLygdcEj4QmlaXYj0aGm6u7tVXUCE8QUauvzez5U2aNfqzEHrGq+sjkCo+EKk3mm4xReUjSm15ZKqVYhARoLOqbmzw+tHnjzH8Y9pbkUKjRP2vaa+rWYYofAKYlJRzV/XfSFdVkw8yGQLrJUGnuU11EcIISCXRxYLUKlQyOZZlWtVVkmxQHfXJHoyTTDQjy00+vPtKt5i18L4lj+egbaxVCtVDH34pnpRrBszUV3Td6FN7Rwm06e/Z/ggcMyl6WdfRZlgVMA11oMEQXo/TgDgaHpd+/eEhidE3JryGbArVEEgrosrNHVcoyDyXISRIp3Nh61xgr+VUl2oC/w+XIFQiVoiyPh5Y9p0o3Qa3hLRm1pIj5uM60m/D1ljlpeitSY0v35RjV3Rr8Volq4DXkRwDrEeJCAxslV0u1clK9TvRcRirZSpaVoovKP3LToPUiJ1DU0r4umVUGgM1/5mPcWUrVmPtByqeG5qJEJkPbj33nv5m97G33Z7F4tburhhbZE/73UYslzkmn3fx723h00L6VeM7uqi4E/7vA8KAyxq7eKvvVUen7BHGGgHsjupJnsUBuEfd8auKasVrr32Wu65554ND3DVMuRvLqY2fjKzHn88/F3TlLleq3LjuhI33HCD0rQiqY6BK062dfguoVpdO9OaW+MpxYEl4h8v2hsSDAJUSsO7s/xr1Z4Lx6hFYiKe57uemluH1EHMnTA1zIwKxhG1yISmXtRySbkGgwy6WtwSqUMPmZB23q/Vh1yTsmRqFe7Ijecv+YkM1mycUoE/izasPQ5FFvq5c9eDuL5zN/XiVithbMivIbpu7J7cMmkamdz628HMb5/ATZP2Y9GAcmeJdDZuMEQZynBrdEf8/hJ85hLRaNdEepIGAic4JurO8lxszaAnk6fsuOBJ1mWbWZdtbhAOKn7SnW2mV6i4R0+2mZ5MPh4f8Bmc5gsNLXAH5Vv8VFcZ0gZ0d3eH3wPmGIw3ePYaY4CNlkyjuya4tu9y7envZyDV0FJfquMG0zn6ag4lI1MXYnX6febenW2mT0vFYzSNQqRWjX/v7wmTLSJWg7BrVCXU6speVIhoEcvEi7jKA6ERURZi7sSR630k0u/ou3mRCJH1INrV8pFJe9c/ez7z1MrxlN5K0M5bCCgO0ptR/vaykfbXBFEvwODcx9Ruw7SveH2eCkgvWTJ0AanV2dZhx3nHHXdgRKpghZRxPz2oAJvjII46DnYLaXFdt+7Okn5gXTS3xLPBgtbl61bz1LipzHKUpj2kuM+xh6SrKiEy1EdfTYXauhc00m5uqQfWA6xqblUBe387EM+UEsK3RErKDecLkSA7y2vkOULC+Ek4097BrFv+j96//Rma8qrza7VCj38P+xyP1Z4ai6vpMDjAiuYOf7woAVrvh1ao68N9maaY2244vNquMs2qhUHfnZXBiQxUDtG2G9AYuA76Mw3TZ8z1IlosYGsavbajemu5bj3GYHvh8b///e858uPHc/Rpn+dDp5/Bs889V2eSLorR1VlTTKtW2vxNf/0r555/fmS8flaS5w4xJKTnMevZ55j9zGy/lsJ/LgLh1EDroCex9REyl1JpnEyO7qY2ggsJUPvHLBHFuL2gTY/fMTm2vk9kFUo7oC16zeh1oy4rH07wbrsuhZYOdQ7UuxlcN3bfhAhrgRoZv/TCoLwvRAbSOUrpbLzF0TBIsrO2IjzPw46YooV0GLtwfeGRipq4/T0sb+6sf3V6uhlIKyHiIfBem1cPDhZf9KvDhylGq/ladio11CXy5ITdh/wGsGLFClKeWy/gkwCFfrzIG9vf36/cHIbhB8Zlnc6wqZ+iS8u3qJc4KLbz3Veyew1Pj9+dZ/t9QVEchlk2vlB9PSAEHZ/cmfFf3yecHyMcm5c18BAsWNMdvmA+pKGpjCAibrWIgFzhaarVd7ms4ktB37FqlcXNY9RaHBHksw6099W/IAAAIABJREFUtDNw6leZ8eJjtN55o3JnlYqx1ST7yxUKRtgixV0SZmC5QlOZPR0qOUGWCjgRC0pGLDS9I83Ec/Zj4jn7kT20Cwn1oLhXGFRCNpOLe0dGkCFOwFAj7VRsTcera6hxTR8iTMQ/toCO7UnF1CLaufAZ49y5c3n00Ue5+/9u4P5rr+aaX/+SSePGNWQEhZ9lsNpiwKiljD134QWGWiKg7uXjc+Ywe/YzarRGJH08Mu7ocbXG9egD5JuxbRs3MlYR0NZYpOmF4/RQAi7aCcHzvDgDblyds4E9x9KtBVTLFYwxGYyxWZxatS5EABWH6ZoAQVsXz8PVNGzdCOMt0XnyvNBSQQmdip6ioKfX687aEgIEEiEyIorFIp7ncdxxxw3RDAJGb0S1k3VrWJMLs6oKmk6hTTXmq+kGr7WPpxRhSsCwQsTwH4rhFpJJR4TW4slHsmaMqg6eueg5DM+NLPQkob+3rmUCDA74LR9SaZVmHMQMb/kTcvVy9cWPX2i+xi+ffxqECNutRJICJEBPQyUwIPu6Yy+AfOBvCKGR3b0FLRt1qYS0eGmdp8bvxj2PP0l3c9za0tOE7qxgW2EAhGDV2IN5NrsfC4wmNfbAEgEG+vq4c+pBrGuKZ7q1NTtg15QFFiCXx543FycSzyl6XuzFj352/XkV+/lxq2KBajQO4ltB1XQrqw7+cHjtw8eyNKJo1AK3WGsbWpS3jrCaXVU31HUiYxlI5+gJ1md3nSEMtn7a4J6IyIZG7dzzWLduHe3t7aR0nVIqjT5uIp1jxvDo07P5zGc+g2manHXej6j57sLnXnie4//rTI455TRmnv55BotlbNfFq5TwahUenPU4Hz/pZHoGBuju7eNLZ32fkz7xBcxP/AfPPfccS5Yt4/q/3s5V//dnPnT6GTw59wXu/MfDHPXJT3PM6Wdw4n9+0R/deqwzERJlFPrJO1UkErt3HU12tb42SwzSqwtrKVCMPJY568YWvooJoWiml4/ubOjCFAI0JCKlIXRBKnBHBZfWDWWdBW5Zz8VzPYqa7p/biwsRN1J70+jOWg8kW0aQJBXrI6ClpYUvfvGLjBkzhvvuvZdaxCopH3QYlEGP3sx9DqSSHwtV3+w/7EhsmYPePmodY6F3Bb2ZPG21Mi+P25VZr7xMy4svMb3humk3XEAnipqm17cVjTT3jetkfPsnOf6hL7Pr4Dq6s3mclAF46oXr64lpR3+/5x6+BDz+9GymTRwHMg1Cw3lpDpR995Dv269X6D72IL+dfhQHrVjDoQ3jLKUy5B8eJm6zbrVickFNguexormT8VLWH+jCj35DdcFv6odUMmkWtqp6DT0N1agf2QAG+rh/yn6s6alyCqiXTwj+sfv7cWWFld2CPauVMCYC2H6MqhIpcATQdY+i43LvfXdzfCpLq12pp69Ghbyz13Sczi54Y7H6/oGZsEZZGO7Mk5QcnPFB5I2/wy0WuGbf99WPHZRwyx6HsmtmP1ra2oE3ANCCbgL5FpARwdTWgRAR7d5nIM3uE6Rk2Cet1dUQSLR+g+YGTTntutCnXDOalIg16lp5V3Xy9TwNzfNIVSawPPMBPITq9xaNSUjJYYcdxlVXXcUHPnEih777EI76yEc4bPp+nHXhT7jid1eyyy678OPvfZebb74Z0zQ589zv8pvzf8g73n0Ig6tXkspl0RybmtC47aF/cu0NN/KLX/2aznyWM7/zXU4//SQOeM9+rFqxmv88/Vv8447bOfWEj5NvyvPFz5jQOZajfnQB111+GcZOU6EwgETUrTetYf0ZgIqewgOKVZux5QIpUMWFdpUcKIsy0sKmJnTSUtZdpxIBroMbVX5sO86BG2NE64N/D+pfZShE6l0E1q4KXWSuS8pzaK15OMJXJGMuQ6chlhWnXwrNv0YcHnKLSJHEEhkBcqAX7Wsmr938A9INXW/XTTuUM194gJ2zYdyi+8gTGZiwa/37mncdxb89cgt79K2mnM1T0jNo/o1+YbddKXiLmd6t0jrLkZTNXQrd/Pu8Ryj3xN1CrqbRPllZDPM7JvKOScsplMOK6LTr4k5RgV5HQPdAmVfbJtJcq3DmCw8wpaiYkSs0Fq9cTXaPlYw/cjZOJNBMbzfd43VuG/swy3dVmrUxtpeF6et59csfVQtbAa37vsHy//gU4tNnqNUfo1jyRt3CWn7QePjQ8VR0g7/Ya/l/teV4us7cV19ln/FhAL7QlKMn28yB6xbT5ZW4shYGhnXDQw70sWDCGNjvZXpb06qSWEL6wOvIHXIjRUMJiuW76Nw15j6eO6SF4ipVQFZ14zUbQvd4bqxB7uCHeGZflb67wIaynmFlPlwedGD8LhQ7xoX394jjOGX+Y3xQPsHcaZIFBx3F/fNv4un3tbHEi1uYL+pNdOdaWD3lDeZN+QO9XjgGV2i4neq8tdZ2ujN57hdPxywRxdgjloMWMiAphHLXCImme74rXfrHhedwATuqi/o1GhnHpsWuUDHS4KoMpWA7nktTUxM33HADPznn23R2tHPuuedy4823MGXSJHbZZRcAPnnsh3n22WdZtGgR4zo72f2dBzCYrZFtaUUzUtgpmyfnzObKG6/nmssupn1CikpW8q/Zz3DBj3/GJ4/7HF/5r3MpFopUV4TrhHhCo2TUOOig6Xzjwou57fabKaTLuLog1VpEy9h+clmcM5b1FJV2MFoj7tVo00LivNQTQvW+inzHsWNKl6yU465VN95AMdbVWUAqX2aNDApjIRN57jTdRfNXnzRSNlVtNYFKVc8EC+YgEGwC1soaZRxsx6tX09to1DyJ0F30nHKT2SN0JO6XDrZ4c1bL20FiiYyEdBbheqxcPRs7rVZV8xBoSOb941/sC+z2RhhE/vWji9H1Pvy8Ja59+BUuAg5b9Rp/6JjC1XsexzkvqPbl33i/SpVd+ZDad22uFX1anslPKaGSd2pM2mMe3jwIkph06ZIVyk0ycY8079x/gEcXLgbVb5Csa9MyqQkc9aj/x4RPcVTq7+wzoM554Dol8FwhqKWqZPx6iVo6on31ruOR903ANbp54uCJnOBoZMb2YTSXedpZzr7HX4D4zc/ITermee5g36Ovo/+FuTQDsr0TBvoRnstAKktrrUrJ7aZy/Kfh2WdZ6b9gmuvSWpzjMzgFJ62EzntXvsa6/Q+O3QZd92Cgj8wukkxnP3868UCm3avx/v5XIKfm47GxB5ATLSysLaAl18cTB+7EEy9/jCMnWDS16PRET6hJip3KjVCZWIPn4bmBMt89/Ie0O70cOKhcdv9a1IenlwiqNH54/0KuqZVpfxFuOfpO/nlviT0/8A963wW/SX2ad3U/i8BDegbdaeVCa5+6kLXAWmnTofRjFuYnUiqrcS+fsCdz8mMY3zqb8V7odpNCZdYU9BmkxudAQG11maqeUplGgNFSQgiJZxtoKQd9IEVT1cbRdLJOjb6xBhJJcyFDqliiqhuU0k10lAfIOTWKRgY8G9cPpHkSSk0aqXwRBvPMeOc7OWjGYey6197cfbu/lkrgkg/GKSWubmA3S4Rms7qtnXH9A7gpmLLTJJYtXcEbS5ay826tlHUXR8JNN16J1magI2hd69bjKx6Chc2T6BQr+d5Pv81jj69kzpN38akTTuf6my0moKOlbHCypHU7tqa5FAKhx60DKRu1dRHJVBKx7dJvMxMVGpJ4HEZ6bj0kgZRUUlrw+FHJGqBHrtfgedOijNyT4USiPAxZ38MgQGVLZjK4juItVRzW5sez+6AStktaJoAH7ZkCmuGwqrWD5kqVdEOAvVUYDEgHbwuYIoklMgKqehUJGLakoClT2BaKEXQM08NvrxZIS5uKprTicSn1UKc9F0M66NLB0OJacfBojS8N1AVIgFyuEHsWNSlJVSV2inr+fVs2NNHz2EQ75aS8KtmJa2nZWwkPXajxaF0F0juHVo6T9iuBdejuH6TaoR7o2pgadue4eiW1lKrYbtwRYVwE4N577gdgVc3FEYKa0Cl3KreY5sJApYCMPsgCdupfQi3yElfzYSsYmYu//HuWVmJ3r+Pz973APk9XyRpF7p90KNlS2JriE0se56ad3k3eUG3EU3oVDZfc5LWIhn5jnvDQ3fiL9cnXn6bN6SPt+Qs6odGWIubCSLnxlzQlw/O2pHvZd/Icvjz3YbrKA1SF0s10begL/EbLTpwy+w6avBIvVDMYzYpZxN1ZxGMXoFKT/X3GlQeGRAjKmTRSaGQbs3X8YzKug6ZF1iIXKjsr49QYVx5ACrBz0PP0QlYtfB0plJY8f/58xozpZOnqVazsVvN76733cfC7D2Lq1KmsWbuOF5+fB0Cl1F9f037i5AlcfsWP+caFF7HmcbWGzqEzDueGG2/xyRK89JpaxK25qYnBUpmcVJbE0sXL2f+AA/nKN/6Djo42Vq8JEx6EEEMskXS2ISMQhrh8VrdNpFzvACDxHBuhe2iapoSFXcMTImwA4F9nXHkAXXphFwEAKSlH4pkVrSEJpiF5QACBHGkYOlnXVhmDqAQA4a+rHusgHUGb7CPvlOtph7rmDMlANNDI1l11iRDZajD0JqqGgVGTkPZosqs4vtbWagy9MXs3eWScKprPd943Xu2b8hw0JBO7XiOTiT8Yjq6zoHUcmUjAfM7hSiNNVSVCwvJd0rw0fgKG5wuRjKDmM6mMER7XkZd1ISLR+L43m+zYHnIF9fTmM0r7Tk/qRWXoqievltFxdRjo0pnddgh6Vr0sRtrhha4P0SQraI4kow/yzOMDZAthPcJLc8pMqiohVUin0VyXpw84GcPnr7oreWNBX8zN4qU0HD0Xs0RsXaOrXbnq3Fx8boUheWXSkQDsNbtKh9PHfn2vx/Z517rXOaDlVbSUi+61ktfLHJ9ahVcz8Bp4qjuCed9V6SHnlMk6Nbo62tmvK03eq9TjUHvk4sIo1VYgU1TnOsxYxk4rFAOcXOxF+PdnuGDw/rxE2nXZbfw8DK9Gqs0XuFF31jDjk7pAIEnpw63RDkLzKKaHLuwUnXujYdGkkt5C1ne7uFoK3ZEUy2W+d975HPOZ0zj+059m0Ruv8/Wvfp0LLj6Xs776Q0zTxE2nOOn0mWSbJT/92Y+5+IKfc+LH/p2vnfHFesAdYOpuO/Or//k+X/zBeSxdvJxzv/1NXnjpFf5t5ul87MOncN1tysI55vD38PdHHuVU85M8++RcfvbTK/jciTP5xLGn8a5DDmSffSJZiUJEVn1c34RFYhK6ixRFDCd0dxXbU6RzZfSURKQcynozrpYm6HFtGzl0zXdBeR6lbGc9ZmJ7WmQZ2zDRIrxg41dRFx5CAm46tlMpH2ldVK5iDwzE7puIEJjOeWSzmXrMJae7ZLLxLMlQ/xBDpdZmQOLOGgGCFGUti2FXmdA0yEmvzMbZdxw7fXwsa2f1EbBvTwPNg8OcB3mXO4ic18OEb+3LwKIllABdSv77/Wt4Urpos+LXKIxtYux704wbuydrfqO0snW7SngM8hWhUkOBR55cBKshU1ZChLQBOEzt7CZ9zn6svmI+Ldkag+Hgybx3Ons+/E+mP6GEQvs6xfBcXaAZSkuWwP5mGyseLND+aoXBabsjdEla6tR0m/fsfz/2L9eybA+DJz6W513yUt7x+0Gef2+WVw/O8I705VSWLAJg/FQbfY6kMGkMxjL14Oo27G1YTJnZxBx/dHbOwE61YBNqj12ZQY58VGmzbi6esealBCvH7Mb0hZCuwUlXr+C6ffaN7eMInffu9RILpGBvXeNVvcT0Q6bz9MCdQ1J8XU2SyarfMtnwTf3xnD9i7NWGN7+P3Mf2Iu3mmPnwPBisMfGc/RjoX0vxH+F5TuudTcdVBW79Siv77DcGAte+5iGNXCx/P2ohTtJVvGdcocTZB7/OLK+fHluHiDKbb5FIPYUY8IO7AoQuEJ6ko9CooSoaUnqN1owDw21u0km1ptH7QwEkdK/ufwcl8ISEA/bZm1t+/1tcI0dTRd2znlwHR7e/k2N/fxU9uTyi2QXNBmzecfBu3PCX3ynSRYb82gonnPhRTjhRrZK5/1578tAN19A3VkcU0vzi8guopVwMNDocDenBbuzEXTdchxAOGdvlnVccSIYxVPEtZsd/njQwhB3T9EVWjxXUuZpA96TqQRfcEiSG5qAHfRIBhEfbWpdy8yCVvKAqOsiWesjZZWRrhj6vDYI+n4akktZpCpJmVPpVeP4GBcBoT2OvCWN+RlrHLfkuq2HKQKL0NOklHMfBiSR5pEUkpuY4tOoe+roqhXYNO+2MGDxXTrPEEhkWpml+xDTN+aZpLjBN89zNcY2eskNFzzBuicO0rEplNeatQUpJxYm4YnICKaBQrOAsUHEGe00FL9JDKmV4TEgP0Nob14KFobHP7m3orel6JbXjxyiOKIcunvZO5QfPlCV2WtRTZdO+HzazU55U1qjnx2tCsnLtCg6aPXQVPU8HJwUpv3dVDUl2chPZoiQ1vpexSx12HvRN5aIKTk9Z4Lu4iirVdu9nqjT3upRExK/nd0LZdcwTtHWrcWVLHkZmgNbxIYesNhlIiLmzbBlaVDIfV+M8A8aMWRT77eDWf8W+l3NZ7EqRPZZ4tFVtHOngFAeZvKw8RBNzhUtzp3rsW9vCa2lS4s1X9C0oFOipZGFQcZGSbdNdjOtbra8o7S9XlJSchbSv9d2FKWfIi2sj6S+p4ycvVft1rHaZ2lFhXKnIpG5vyDgzTQZ2RoQBXD3uxgn4jiGV1apJF+FbGoMdcUGsN/tJEpHjjVwZTUba4EhPWd1AOiXRI9t06dS3GdJFIuvWZn0fG/DsIa6m+vVsiRZNq/VTYLVMwII8Mrbf6doDRyrlR3dAj6SDZzQbLXINvTkVmzo3qD+KVNJn9QqGcNH8+ItAEhh0AV2Zphop/5qG9BjXPFg/b3PKRaeGCI4XEiLzkxKRlQn9vyJSB6XpAt0JXFZxQ0VCrO5GtZWRdc4sgLwWutI0JMLvCmHYgKgNMTak7VGotSKliBWxbi6MOiFimqYO/D/gWGAa8BnTNKdt6uvk0ZhY7qZ5QNKyMAzN/vaWDCsKIQN0DUEtI1heqFBqVjfs1kcFj68JU0uvWlzj9TeGVrUaMnzZ9WbFZPxicGpLQhNVGn665oCHnRGkGs1nXaC1pOpBNF2T1FYsIOUMfaF3sfNMTYVuD1t6tDVl0CR0rn2e9/+lyLTbFDNd9IZ602p+auoz/topmbLkI9cUeKo7XAtjXVrNya5Ph23ss0XJ/bXQnw1Qyxlohoy5s2S0eK0pLmg9A5zBhbHfWrtd7MjSiJW2HNOv6Wb6bQOM/z8l8FuevIf33llENDSgc5HYQU++YWIWAK+u1bhx7uT691tXLufBFfGamIC354uSYmExY5er6+SaB2NaMKg5fvo1VR+S9t0g7Wtc1thV9r+xm0P/3F/3i0dRzYo6VxKaGNYzYbjQ3OfRts7FDjqpRwu6Iw11o09Nc7+HkCFzyugO6UrgDpVo0gldKpE5zOgumapLS49LuuwzRg9ae1xyAw4jJQM193m0VKojcpyo8NE8cGUFpLrXhh15PpwGLkxc/rr+/MpIO5B8v0dTKWT6muZi+IqgrCenVdD9WhLXDRk+gON5aG6oFAokzYVw7vKRz4ECEe1SrYmosgIpW1NdJTQVg9GCNWKEX2PpyXBcQEulEDle4jU2VWwUIhpUnByeJ3CHf8Q3KYQcYltt2zBN8zDgPMuyPux//w6AZVkXjXCIXLFixQibRkZx3XyeevlalfYXJlOED3DjdyK/N+6j+fndUn3eZ4wSSov78uzSXvQ/t1AOUkVdj7ZsjYktatuaUhM9xUz9euPzBToiPvo1xRyGJilm+lgiq7QJg3TvlOFrknwB1N2uahf21HJQblfnj9IZHfNw3wPaZXBOGe8v5EPXJXt29vGY6wfmi+PpSrlUsgXe8NSLOcbNQ/9Y0ATNLT0s1kMLamytDa/YMeS6Qnisa1fxmHGDE3HtiFUUGatsW02PHvp4dKDVydFrlGl3s+j9E4aZI/8iAT1BoDuYTy2yLVSkw7FJBylrOB199Gs1dhIZeleN48BJAyzsbaXq6OGxQceLzuWslTbTO05j1469MBBUXJ2M7qq6QCmw3VA9dXRfa/aMuNRo2J6SBmmhhm97Gu5Immn03r8VND7rgGOETNVwGta0N2xcPDQETX6RZMXVhz5X9QMAzcHxg41ZL4UUkqpvneQxqDp6/Zq6m0aMVPL/Fmh4q/MRXL8JHQ2B7QlSWpB5BjUnco/ezLl1B0e4aAg0Jz3yfpFzBWNwvTRGdQJeWvG9MfmdhgT7G1EqlWhqii+TMGnSpOAK68VojIlMBqKpTMsgXgtnmuZ/Av8JYFkWXV1dvFXUiqtYM2Zo/6pNgccCRaKln2WRz1F0A28E2zL9kIlvi8VIs/6x/sPULx1oX/SmxvKaV4ZMOXb+TY01kbGuya9mDYAHaU8ng0a3XoROJTAb7bW16X5ID3XLxc7fMkyTwhHgAr2GEip9egU6F73pYzcWS2UVxi9V9711BFoiTKUS3FzdqcfeEAz7tjqaA9rITfhs4YTpsBpb3PcQFShReEgKAXX6yONvRKUhw7GIA5EEE1dffy+pzY1ScO80iFFuvHkao/CQeCPM4Uhw3GYc6ZKWAoSHMUwLpUZkMpmN4pMwOoXIcJIxJtcty7oSuDLYFm2k+Gahpzv46MRLaG1pYvXyJaRzLaAJaqUi0vXI5JvB83AA3dCoDRbQjRRGrgmnWsCtuaTzzQgkTtUGIUnlcmjCwK3V0PPNfoNDgec6SOmgGz4n9/x1AIwmQFB1qqSwcSoV0k2tlOwauVxeuWqkDtLBsasY6WbSTXkqvauolvvQsnl0I43rQQYJRibUSFybVK6VUqVKSvOoVktUiv1k8+10NLfS27sORzpkm5rxah6e6+J4NtmmJnShUak5VKoFstk8eiqF7XngODiVQfKtY/AAr1LD9Wqk0znSLV0Iz8GtDoDQSOWayGSaqZbK9PeuIJNrUvR7NnomTzabx65VKQ10Uyr1kcm1ktE0XFQrbSOTQUtl0TTBYM9KpNBpam6nWiyDV8LDId3Uge7YpFvHUyv10tQ8hnK1gOO4pFJ5PLfk+6Vc0sKlb7BIrqmZWqWGa1fR0mlSmTQ4Lo7tgaFjCLCrLhKbdFMTbqWKUyuRzrehpzTs0iCILOm2ZqrVKlmRhZROzRHoXhlw0bQ0bq2GlA6ukKT1FFLotHVMYLBUoimVJzBtlKMgKNCQytWkaUgpUflaaplZ6XlI6akUUYRflKepegj/7ZBS1jObPM9BRBaPUi44gdA0/1wSTdcBifQ8hF8xLaW/ze/XFbZ5EkjpIoSG0HSEpuO5thq5b50H59GMFNLzF6kKTDy/P5Tw1/aQXkCL/9ILrb60rOd56vrBBEnVpUHT1bxIqTIb6xYpKtZgGCk86eH5/amEptddqVK6aj78Hl/BcZqmqz5a0vPnA7xgyQTd8LvAu2ia38pEQlje6Juw0i9U9CfD85RLTvPn2vM8tKCfmueoOdI0ELoqMvRchK4jfF4BqO9S4kk1p8Jv4KgbKai35JqEYaRUy/0NoFqt0sgnfUtkgxiNQmQZsFPk+xTgrfurNgA9laGlcyJjurqQqbYNHzB2U48gRGPi5tBEzjgyE6a+6XMHLX+i5+zq6kJvWb/g3dAYRkZH7Fu2NUM2WMirAZlUlky+reGIoci1jAnH1Tn8Phm/NXhzbA3s8L52dXWR2QhlY33YmDnSjRT6MH3TNjV00hiG8aYYzMZC09Zj3mrrcdFsAMMsvvumjgnofavHD9k/9VbN9vj9XP/1199fT0/Ft6/vXJqmo2n+GjCbEaMusA48DexpmuZU0zTTwEnA7Vt5TAkSbDdYs2YNX/rSl3jPe97DBz7wAT772c/y+uuvb/jACGbNmsVpp522mUaYYFvCqBMilmU5wFeA+4B56ifrpa07qgQJtg9IKTnjjDM47LDDmDVrFg8//DDnnHPOEFfH5sTmtI4SbHqMRncWlmXdDdy9tceRIMH2hscee4xUKhWzIvbff3+klFxwwQU89NBDCCH42te+xvHHH4+UkgsvvHDI71E899xznH322Vx11VWMHTuW73//+7zyyis4jsNZZ53Fhz/8YW666SYefPBBqtUqpVKJm2++eUuTnmAjMSqFSIIEOwKeXXk9fZXFm/Sc7dldeOfEU0fcPn/+fKZPb1ygAO6++25eeukl7r//fnp6evjoRz/KjBkzmD179rC/B3j66af5wQ9+wNVXX83kyZO56KKLOPzww/n5z39Of38/M2fO5H3vU230n3nmGR544AE6OjYUBUuwLSERIgkSJNggnnrqKU444QR0XWfs2LHMmDGDuXPnjvh7c3MzCxYs4JxzzuHGG29kwgRVj/PII49w//3389vf/hZQWUHLl6u+aUcccUQiQEYhEiGSIME2ivVZDJsLe+21F3fdddeQ30cqSl5fsfK4ceOoVqu8+OKLdSEipeTKK69kjz32iO377LPPDil2SzA6sEMIkTeb77y5jh+NSGjeOli+fDmpN1EctqnQeK0PfvCDXHLJJfz5z3/ms5/9LABz5syhs7OTO+64g5NPPpne3l6eeuopzj//fKSUXHvttUN+X7BgAe3t7Vx++eWYpklrayuHH344Rx55JNdccw0XXXQRQgheeOEFpk+fjmEYaJq22WnfknO7reDN0JzL5Tb6+d8RhMjb6kBmmuZsy7IO3vCe2w8Smrce+vv7r2tvb993w3u+faxcuXLfiRMnzmv8/Yorrkh95zvf2emyyy5rSqfTcuLEidULL7xw6YoVK8YedthhbUII+fWvf31lZ2dn7zHHHMODDz44pfH3gYGBlkKhML6jo2PBlVdemT7llFP2vPTSSxedeeaZpW9/+9s7z5gxIy+lFBMnTqzecsstC/r6+sYPYknwAAAMG0lEQVQMDAzkbdvePG0i1kPv9ow3S3N/f/+8yZMnf3ZjrrEjCJEECRK8Bey000729ddf/0bj75dccskywqb3gKq6Hu73o48+evDoo48eBJg6dWpt1qxZ9TT8K664Yki2wBlnnNHN0K43CUYBRl2dSIIECRIk2HaQCJEN48oN77LdIaF5B0A+n1+7tcewJbGj0QtbhuZR1wo+QYLtGS+//PI17e3t+23tcSTYsdDX1/fStGnTTt+YYxNLJEGCbQiu686zbTt5LxNsMdi2rbmuu9EJB0lgPcEOBdM0hWVZ26z5XavVLhkYGEDX9X1JlLwEmx+e67rzarXaJRt7gh3enWWa5pHAk5ZlFTe483YA0zQFasGu2y3LevOrOY1ymKb5Y+B6y7J2mBRP0zQnA2sty6pt68JzU8E0zbRlWTX/845CcxswYFmW3Bo077CWiGmapwDfRDVy/NdWHs4WgWmaHwZ+CTwB3LuVh7NFYJrmycAXgaAh1Pe24nC2CEzT/DTwA+Bl1OpIn97emalpmp8BvgX8wzTNZZZl/e8OQPOJwM+Bh4F+4Gtbg+YdyhLxtXAD+DqKmRxrWdYTW3dUWwamaRrAz4C7Lcu6r2HbdqWxmaapodaEugTYFfgRsC/QZlnWz7Y3eqMwTfPdwP8C37Isa5ZpmvOAUyzLenYrD22zwTTNg1HK0TeBBcCDwP9alvXH7fVem6Y5FrgB+CEwF3gU+CPwO8uy3PUdu6mxw/hcTdPMWpYlLcuygVdRN2CxaZpp0zRPNE1z6/e82MQwTTMbfPbXYdkbWGqaZptpmmeZpnnM9vaSmaaZsyzLsyyrH7jSsqwPW5b1GEojNwG2J3ohfp+BqcBjvgAZD7wI9G2dkW0+NNC8L/CgZVlPWJa1DvVu/8Q0zbbt7V5H4AEloM+yrDJKMf448I4tPZAdQoiYpvl94F7TNL9mmuZewD3AUv/vs8AngGtM0/yev/+on5cIzV8xTfMA/+cFwLuB21AL+n4P+IVpmhu/2u02BJ/me/z7PN2yrGci9/IvgBOZi+0CDc/2LsDzwC6mad6MWgVUAL83TfOn/v5vqw3QtoAGmncC5gPHmqYZtIvxgAHgG/7+28P7fL5pmjMjPzWhKvw7fEXwMZT78tP+/luM5lE/uRuCaZqfB44GzgG6gEv9v7cDfwc+YlnWqcB/A98yTXOMZVmbd1HizYwGmscBF5im2QksBE4D7rIs61zgFOAwYLetNdZNhQaax6Bo3iVyLztQ9G83z/wwz/YvUZqpCbwGfN+yrE8CZwCnmaY5ebRr5sPQ/GvUCqe3AueYpvkM6pk/GTjONM38aH6fTdPsNE3zSuBrKOsqBWBZ1lKgB/gY6nkHuFwdYo7bkjRvNy/UcPC1rp2AKyzLehLlI38J+KmfpfNDy7KWAViW9SIq2Ny1tca7KTACzfOAi4FfAQ6Q9t0+y1Guvalba7ybAsPQfCnKjXNRsI9lWQuBnfHN/dGunY5wn19EMRKAPEozDWifBey1FYa6ybCeZ/sXlmX9BOXSOcOyrLOBdSiaa6Pc+ioCf7UsqwNYjor7BLgCOAB4r++uX4qKjUzckgMc1S/ShhDRuk7zvxdQL9k+pml+wLKsCoBpminTNH8FtAKbdim5LYwRaP4Z8E5gHxSDHQf8wDTNn/u/jeqg6wg0/y+wu2maH4jsejNwjL/PqNVOYUSafwHsYZrmfsAa1D3+kGmalwGTUUJm1GI9z/Z00zSPtCyr37Ks50zTTKOy01zLsuzRbH1ZllUFHvG//g/wBdM0J/rbFqLiP8cCPzNN8wqUorBoS45xuxEipmmebJrmgf5nEdE+LgZ2M03zCP97N3A98CF/3xNQGosLfCoQLKMBb5HmG4ATLct6APgp0ItKC3y/ZVmbrf32psZG0PyhyOFV4LbRppm+RZpvBE5A3eN/oNKbAY6yLGvU9I7aiPf5KH/fd6LoBvjOFhzy20YjzcHvlmUV/LjH08A/gQsih90EnAesQs3FUX5SyRbDqBcipmkebZrmoygt7CAINRbTNA1fkl+B0sADDdRF+RMBXgE+aVnWNyzLKm3p8W8MNpLmGjDof18FXGZZ1vmjpcjybdznaHvxqy3LumW0aKZv4z7blmU5lmX9L3CyZVnf2gHuc69/isUoZemM0aIQjkSzLzwDHq37f89Fua/2NE1zBjDDLxq+0LKsH/jW2RbFqCw29KV0FrgG5Zq5EDgelbEQPGwOKhtnomVZvzZN899M07wYuBOVCvcggGVZr2wNGt4qNhHNDwTnGw2MdDPQbG9pGt4qNgPN2zwj3ZQ0W5Y1KtYkeQs0S9991Q84lmWtNU3zPlRG2lxU94mt+j6P6mJD0zSPtyzrb/7nU4EzLcs6zP9uAD8B3ovK1NBQro3PAH+3LOvHW2fUbw8JzQnNCc07HM0zgLOA2cBMVILMbyzL2uh+V5sSo0qImKb5NVT7iqcsy7oq8rsBHAp8DjjfsqylpmnuDfwXcIFlWb2Rfeu9dUYDEpoTmkloTmj2aTZNc09gzZaOe6wPo0aImKb576gg4XmogNm9wE2WZb3hbz8A+ANwpGVZgw3H6tYWbgWwKZDQnNDsb09ojh+7I9IcuLe2OYymwPpRqPqOe1GmXQY4NdhoWdbzQJmwYlMEf0fjA+cjoTmhOaGZhOZtVYDAKBAikeyEOajqTCzLmo3qRDvJNM3D/f0EqgI9Z0b6QY2GAHIjEpoTmkloTmgeJTRvc0LENM0J/l8NYkVhjwGaGeaHvwisBCb5+0lUlkNxW5/0RiQ0JzST0JzQPEpp3maEiGmaB5mm+SB+IU0w8REp/hqqZcmnfZ/oMmACqtV3gG9ZlvXHLTfqt4eE5oRmf5eE5oTmAKOKZtgG6kR8E+7nqKZql1qWdW1kmxaR4oOovjB7A5eZpnk2qqlePS98tGRpJDQnNCc0JzSPZpqj2OqWiG+2tQBzgsk3TXP36OSbpnkBqp1DP2oRlg7UzehHFeuMKiQ0JzQnNCc0M4ppjmKrpPiaqly/x7KsV/3vrai1D24APgysBoKGcgVUStwPLcta4O+vAfnGNLhtGQnNCc0kNCc0j2KaR8IWFSKmabajJvkIVIO4yy2/p4+pinA+j+qb/yRqSVMP+LnlN45rMA1HBRKaE5oTmhOaRzPNG8KWdmflgfuAr/qfgywFLMv6JfBBy7IesVSTtb8CB6OWgBzNk5/QnNCc0JzQPJppXi82uyVimuZpqM6acyzLGjDV2sga8G3U0p1XWpa1YpjjvoJaA+PL1ihoIhdFQnNCMwnNjcclNG+n2CxCxM9WmIAKJHnA6yip/XXLstb5+xwOmMDTlmVd7/+WQS3Xehkqf/qswOe4rSOhOaE5oTmheTTTvLHY5O4sP/85yFZYblnWUcCXUet3XBnsZ6mF5RehVhlsM9VyrVXUeggXWpZ13GiZ/ITmhGYSmhOaRzHNbwebzBIxVRfKH6EWT7kbtdTsJy3LOt3fLoAVwEmWZf3T/60Z1Uf/cNT61wcNZxZuq0hoTmj2tyc0k9DMKKR5U2CTWCKmab4feAaV+7wAVbFpAx80TfMQqOdS/wiV6hZgJkrCPwdMH02Tn9Cc0JzQnNDMKKZ5U2FTVax7qOVWrwMwTfMgYCqqqOY3wLv8vOjbUDdlV8uyFgEV4GjLsh4Z/rTbNBKaE5oTmhOaRzPNmwSbKibyDGCZphmsA/wYsLNlWX8CdNM0v+qntk0BXH/ysSzrb6N48hOaE5oTmhOaRzPNmwSbxBKxLKvU8NMxwPP+588BXzBN805Uz5grQfkXrVHUqbIRCc1AQnNCc0LzqKV5U2GTNmD0pbgExgO3+z8PAt8F9gcWWpa1HLb9HvlvFgnNCc0kNCc078DY1F18PSANrAMOME3zF6gOlV+1LOtfm/ha2woSmhOaE5q3H+yINL8tbPJiQ1M1Jpvl/7vasqw/bNILbINIaE5o3l6R0Lxj0Px2sDnWE1kGfA/VdKy6Gc6/LSKhecdAQvOOgR2R5o3GVmkFnyBBggQJtg9s9UWpEiRIkCDB6EUiRBIkSJAgwUYjESIJEiRIkGCjkQiRBAkSJEiw0UiESIIECRIk2GgkQiRBggQJEmw0EiGSIEGCBAk2GokQSZAgQYIEG43/DwoegKgI+lG8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#creating the Metrics\n",
    "disag_fhmm= ntk.DataSet(disag_filename)\n",
    "disag_fhmm_elec= disag_fhmm.buildings[1].elec\n",
    "disag_fhmm_elec.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2     apparent    1.557158\n",
       "dtype: float64\n",
       "3      apparent    1.08258\n",
       "dtype: float64\n",
       "4     apparent    2.284074\n",
       "dtype: float64\n",
       "5     apparent    0.261526\n",
       "dtype: float64\n",
       "6     apparent    1.417993\n",
       "dtype: float64\n",
       "7     apparent    0.287417\n",
       "dtype: float64\n",
       "8     apparent    2.534008\n",
       "dtype: float64\n",
       "9     apparent    0.081595\n",
       "dtype: float64\n",
       "10     apparent    0.10013\n",
       "dtype: float64\n",
       "11    apparent    0.525036\n",
       "dtype: float64\n",
       "12    apparent    0.501027\n",
       "dtype: float64\n",
       "13    apparent    0.534949\n",
       "dtype: float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#compute assigned_energy of appliances \n",
    "from nilmtk.metrics import  error_in_assigned_energy\n",
    "error =error_in_assigned_energy(disag_fhmm_elec, elec)\n",
    "error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2     0.213910\n",
      "3     0.237900\n",
      "4     0.564750\n",
      "5     0.477773\n",
      "6     0.047019\n",
      "7     0.122449\n",
      "8     0.000000\n",
      "9     0.000000\n",
      "10    0.082645\n",
      "11    0.031250\n",
      "12    0.164589\n",
      "13    0.215054\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# make plots of f1 score for fhmm\n",
    "from nilmtk.metrics import f1_score \n",
    "f1_fhmm = f1_score(disag_fhmm_elec, elec)\n",
    "print(f1_fhmm)\n",
    "f1_fhmm.index = disag_fhmm_elec.get_labels(f1_fhmm.index)\n",
    "f1_fhmm.plot(kind='barh')\n",
    "plt.ylabel('appliance');\n",
    "plt.xlabel('f-score');\n",
    "plt.title(\"FHMM\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# comperision of CO and FHMM "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict(clf, test_elec, sample_period, timezone):\n",
    "    pred = {}\n",
    "    gt= {}\n",
    "\n",
    "    for i, chunk in enumerate(test_elec.mains().load(sample_period=sample_period)):\n",
    "        chunk_drop_na = chunk.dropna()\n",
    "        pred[i] = clf.disaggregate_chunk(chunk_drop_na)\n",
    "        gt[i]={}\n",
    "\n",
    "        for meter in test_elec.submeters().meters:\n",
    "            # Only use the meters that we trained on (this saves time!)    \n",
    "            gt[i][meter] = next(meter.load(sample_period=sample_period))\n",
    "        gt[i] = pd.DataFrame({k:v.squeeze() for k,v in iteritems(gt[i]) if len(v)}, index=next(iter(gt[i].values())).index).dropna()\n",
    "        \n",
    "    # If everything can fit in memory\n",
    "    gt_overall = pd.concat(gt)\n",
    "    gt_overall.index = gt_overall.index.droplevel()\n",
    "    pred_overall = pd.concat(pred)\n",
    "    pred_overall.index = pred_overall.index.droplevel()\n",
    "\n",
    "    # Having the same order of columns\n",
    "    gt_overall = gt_overall[pred_overall.columns]\n",
    "    \n",
    "    #Intersection of index\n",
    "    gt_index_utc = gt_overall.index.tz_convert(\"UTC\")\n",
    "    pred_index_utc = pred_overall.index.tz_convert(\"UTC\")\n",
    "    common_index_utc = gt_index_utc.intersection(pred_index_utc)\n",
    "    \n",
    "    \n",
    "    common_index_local = common_index_utc.tz_convert(timezone)\n",
    "    gt_overall = gt_overall.ix[common_index_local]\n",
    "    pred_overall = pred_overall.ix[common_index_local]\n",
    "    appliance_labels = [m for m in gt_overall.columns.values]\n",
    "    gt_overall.columns = appliance_labels\n",
    "    pred_overall.columns = appliance_labels\n",
    "    return gt_overall, pred_overall\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "********************\n",
      "CO\n",
      "********************\n",
      "Training model for submeter 'ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])'\n",
      "Done training!\n",
      "Estimating power demand for 'ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])'\n",
      "Estimating power demand for 'ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])'\n",
      "********************\n",
      "FHMM\n",
      "********************\n",
      "Training model for submeter 'ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])'\n",
      "Training model for submeter 'ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])'\n"
     ]
    }
   ],
   "source": [
    "from nilmtk import DataSet, TimeFrame, MeterGroup, HDFDataStore\n",
    "from nilmtk.disaggregate import CombinatorialOptimisation, FHMM\n",
    "import pandas as pd\n",
    "from six import iteritems\n",
    "classifiers = {'CO':CombinatorialOptimisation(), 'FHMM':FHMM()}\n",
    "predictions = {}\n",
    "sample_period = 120\n",
    "for clf_name, clf in classifiers.items():\n",
    "    print(\"*\"*20)\n",
    "    print(clf_name)\n",
    "    print(\"*\" *20)\n",
    "    clf.train(elec, sample_period=sample_period)\n",
    "    gt, predictions[clf_name] = predict(clf, elec, 120, dred.metadata['timezone'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_rmse(gt, pred):\n",
    "    from sklearn.metrics import mean_squared_error\n",
    "    rms_error = {}\n",
    "    for appliance in gt.columns:\n",
    "        rms_error[appliance] = np.sqrt(mean_squared_error(gt[appliance], pred[appliance]))\n",
    "    return pd.Series(rms_error)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "##the mean squared error (MSE) or mean squared deviation (MSD) of an estimator (of a procedure for estimating an unobserved quantity) measures \n",
    "#the average of the squares of the errors—that is, the average squared difference between the estimated values \n",
    "import numpy as np\n",
    "rmse = {}\n",
    "for clf_name in classifiers.keys():\n",
    "    rmse[clf_name] = compute_rmse(gt, predictions[clf_name])\n",
    "rmse = pd.DataFrame(rmse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CO</th>\n",
       "      <th>FHMM</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=2, building=1, dataset='DRED', appliances=[Appliance(type='television', instance=1)])</th>\n",
       "      <td>34.701957</td>\n",
       "      <td>26.827903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=3, building=1, dataset='DRED', appliances=[Appliance(type='fan', instance=1)])</th>\n",
       "      <td>19.736518</td>\n",
       "      <td>20.009338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=4, building=1, dataset='DRED', appliances=[Appliance(type='fridge', instance=1)])</th>\n",
       "      <td>42.234955</td>\n",
       "      <td>31.054459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=5, building=1, dataset='DRED', appliances=[Appliance(type='laptop computer', instance=1)])</th>\n",
       "      <td>16.902096</td>\n",
       "      <td>18.005224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=6, building=1, dataset='DRED', appliances=[Appliance(type='electric heating element', instance=1)])</th>\n",
       "      <td>21.836486</td>\n",
       "      <td>12.694029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=7, building=1, dataset='DRED', appliances=[Appliance(type='oven', instance=1)])</th>\n",
       "      <td>44.623254</td>\n",
       "      <td>44.580188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=8, building=1, dataset='DRED', appliances=[Appliance(type='unknown', instance=1)])</th>\n",
       "      <td>2.553917</td>\n",
       "      <td>2.553917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=9, building=1, dataset='DRED', appliances=[Appliance(type='washing machine', instance=1)])</th>\n",
       "      <td>0.680000</td>\n",
       "      <td>0.320000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=10, building=1, dataset='DRED', appliances=[Appliance(type='microwave', instance=1)])</th>\n",
       "      <td>56.770583</td>\n",
       "      <td>65.476157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=11, building=1, dataset='DRED', appliances=[Appliance(type='toaster', instance=1)])</th>\n",
       "      <td>12.100002</td>\n",
       "      <td>12.100002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=12, building=1, dataset='DRED', appliances=[Appliance(type='sockets', instance=1)])</th>\n",
       "      <td>99.269931</td>\n",
       "      <td>87.555744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElecMeter(instance=13, building=1, dataset='DRED', appliances=[Appliance(type='cooker', instance=1)])</th>\n",
       "      <td>75.119973</td>\n",
       "      <td>69.173837</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                           CO       FHMM\n",
       "ElecMeter(instance=2, building=1, dataset='DRED...  34.701957  26.827903\n",
       "ElecMeter(instance=3, building=1, dataset='DRED...  19.736518  20.009338\n",
       "ElecMeter(instance=4, building=1, dataset='DRED...  42.234955  31.054459\n",
       "ElecMeter(instance=5, building=1, dataset='DRED...  16.902096  18.005224\n",
       "ElecMeter(instance=6, building=1, dataset='DRED...  21.836486  12.694029\n",
       "ElecMeter(instance=7, building=1, dataset='DRED...  44.623254  44.580188\n",
       "ElecMeter(instance=8, building=1, dataset='DRED...   2.553917   2.553917\n",
       "ElecMeter(instance=9, building=1, dataset='DRED...   0.680000   0.320000\n",
       "ElecMeter(instance=10, building=1, dataset='DRE...  56.770583  65.476157\n",
       "ElecMeter(instance=11, building=1, dataset='DRE...  12.100002  12.100002\n",
       "ElecMeter(instance=12, building=1, dataset='DRE...  99.269931  87.555744\n",
       "ElecMeter(instance=13, building=1, dataset='DRE...  75.119973  69.173837"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python (nilmtk)",
   "language": "python",
   "name": "nilmtk-env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}