run_optimize.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pre Processing Values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample_size = 2000 # Number of Sequences to take from each Run\n",
    "min_seq_len = 85\n",
    "max_seq_len = 200"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://pysam.readthedocs.io/en/latest/usage.html\n",
    "import pysam\n",
    "import pandas\n",
    "import sys\n",
    "import os\n",
    "import subprocess\n",
    "import seaborn\n",
    "import glob\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import warnings\n",
    "from numpy import unravel_index\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_training_data(infile, project_id=None, run_accession=None):\n",
    "\n",
    "  try:\n",
    "    samfile = pysam.AlignmentFile(infile, \"r\")\n",
    "    _ = samfile.fetch()\n",
    "    samfile.close()\n",
    "  except:\n",
    "    samfile.close()\n",
    "    cmd = [ my_local_path +\"samtools/bin/samtools\",\n",
    "      \"index\", infile\n",
    "      ]\n",
    "    print(\"Running Command: \", *cmd)\n",
    "    subprocess.run(cmd)\n",
    "    \n",
    "  samfile = pysam.AlignmentFile(infile, \"r\")\n",
    "  unmapped = []\n",
    "  for read in samfile.fetch():\n",
    "    \n",
    "    if (read.is_unmapped) or (read.flag == 0):\n",
    "      unmapped.append(read.to_dict())\n",
    " \n",
    "  if len(unmapped) > 0:\n",
    "    df = pandas.DataFrame.from_records(unmapped)\n",
    "    df = df.sort_values(by=['name'], ascending=True, ignore_index=True)\n",
    "    df = df.assign(project_id=project_id)\n",
    "    df = df.assign(run_accession=run_accession)\n",
    "    samfile.close()\n",
    "    return df\n",
    "  else:\n",
    "    print(\"No Unmapped Reads For: Project->{} Run->{}\".format(project_id,run_accession ))\n",
    "    return None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_kmers(sequence, ksize):\n",
    "    kmers = []\n",
    "    n_kmers = len(sequence) - ksize + 1\n",
    "\n",
    "    for i in range(n_kmers):\n",
    "        kmer = sequence[i:i + ksize]\n",
    "        kmers.append(kmer)\n",
    "\n",
    "    return kmers\n",
    "\n",
    "# https://neo4j.com/docs/graph-data-science/current/alpha-algorithms/jaccard/\n",
    "def jaccard_similarity(a, b):\n",
    "    \"\"\"\n",
    "    Jaccard Similarity (coefficient), a term coined by Paul Jaccard, \n",
    "    measures similarities between sets. It is defined as the size of the \n",
    "    intersection divided by the size of the union of two sets. \n",
    "    This notion has been generalized for multisets, \n",
    "    where duplicate elements are counted as weights.\n",
    "    \"\"\"\n",
    "\n",
    "    a = set(a)\n",
    "    b = set(b)\n",
    "\n",
    "    intersection = len(a.intersection(b))\n",
    "    union = len(a.union(b))\n",
    "\n",
    "    return intersection / union\n",
    "\n",
    "\n",
    "def jaccard_containment(a, b):\n",
    "    a = set(a)\n",
    "    b = set(b)\n",
    "\n",
    "    intersection = len(a.intersection(b))\n",
    "\n",
    "    return intersection / len(a)\n",
    "    \n",
    "def read_kmers_from_list(array, ksize):\n",
    "  all_kmers = []\n",
    "  for sequence in array:\n",
    "    kmers = build_kmers(sequence, ksize)\n",
    "    all_kmers += kmers\n",
    "  return all_kmers\n",
    "def cosine_similarity(arr1, arr2):\n",
    "    from sklearn.feature_extraction.text import CountVectorizer\n",
    "    from sklearn.metrics.pairwise import cosine_similarity\n",
    "\n",
    "    count_vectorizer = CountVectorizer()\n",
    "    vector_matrix = count_vectorizer.fit_transform([' '.join(str(x) for x in arr1), ' '.join(str(x) for x in arr2)])\n",
    "    tokens = count_vectorizer.get_feature_names()\n",
    "    cosine_similarity_matrix = cosine_similarity(vector_matrix)\n",
    "    \n",
    "    return cosine_similarity_matrix[0][1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_location_data(project_ID):\n",
    "    import requests\n",
    "    import json\n",
    "    import pandas\n",
    "    url = 'https://www.ebi.ac.uk/ena/portal/api/filereport'\n",
    "    params = {'accession':\"{}\".format(project_ID),\n",
    "              'fields':\"center_name,instrument_platform,instrument_model,host\",\n",
    "            'format':'json',\n",
    "            'download':'false',\n",
    "            'result':'read_run'}\n",
    "    \n",
    "    headers = {\"User-Agent\": \"Mozilla/5.0 (Windows NT 10.0; Win64; x64)\"+\n",
    "             \" AppleWebKit/537.36 (KHTML, like Gecko)\"+\n",
    "             \" Chrome/70.0.3538.77 Safari/537.36\"}\n",
    "\n",
    "    print(\"Accessing: \", url)\n",
    "    r = requests.get(url,\n",
    "                  params = params,\n",
    "                  headers = headers\n",
    "                  ) \n",
    "    print(\"\\tMaking Request To: \", r.url) \n",
    "    my_json = json.loads(r.text)\n",
    "    df = pandas.json_normalize(json.loads(r.text))\n",
    "    df = df.assign(project_id=project_ID)\n",
    "    \n",
    "    # Temp Fix if center_name field is empty\n",
    "    df['center_name'] = df['center_name'].replace('', project_ID)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Main"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SetUp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_local_path = \"/gpfs/gpfs0/project/ds6011-sp22-wiki-data/ds6011-sp22-dod/\"\n",
    "bam_files_path = ''.join([my_local_path, \"data/bam_files\" ])\n",
    "csv_files_path = ''.join([my_local_path, \"data/csv_files\" ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "projects  = glob.glob(\"{}/*\".format(bam_files_path), recursive = True)\n",
    "projects = [x.split(\"/\")[-1] for x in projects]\n",
    "#projects = ['PRJNA784038','PRJNA281708','PRJNA634526','PRJNA277616','PRJNA694172']\n",
    "projects = ['PRJNA281708','PRJNA634526','PRJNA277616']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982585\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982584\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982612\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982660\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982826\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982787\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982783\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982588\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982727\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA281708 Run->SRR1982814\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209614\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209591\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209708\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209655\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209634\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209592\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209589\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA634526 Run->SRR14209588\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1926132\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1926134\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1926135\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1909638\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1909639\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1909613\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1909637\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1867792\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1926133\n",
      "\t Reading CSV to frame\n",
      "\n",
      "Reading Unmapped Reads For: Project->PRJNA277616 Run->SRR1926136\n",
      "\t Reading CSV to frame\n",
      "\n"
     ]
    }
   ],
   "source": [
    "frames = []\n",
    "for project in projects:\n",
    "    for myfile in os.listdir(\"{}/{}\".format(bam_files_path,project)):\n",
    "        if not myfile.endswith(\".bam\"): \n",
    "            continue\n",
    "        if myfile.endswith(\"_sorted.bam\"):\n",
    "            continue\n",
    "        if myfile == \".bam\": continue # Not sure what this file is\n",
    "        #print(myfile)\n",
    "        print(\"Reading Unmapped Reads For: Project->{} Run->{}\".format(project,myfile.split(\".\")[0] ))\n",
    "        bamfile = \"{}/{}/{}\".format(bam_files_path,project,myfile)\n",
    "        if os.path.exists(\"{}/{}_{}.csv\".format(csv_files_path,project,myfile[:-4])):\n",
    "            try:\n",
    "                print(\"\\t Reading CSV to frame\")\n",
    "                frame = pandas.read_csv(\"{}/{}_{}.csv\".format(csv_files_path,project,myfile[:-4]))\n",
    "                if 'Unnamed: 0' in list(frame.columns):\n",
    "                    frame.drop('Unnamed: 0', axis=1, inplace=True)\n",
    "            except:\n",
    "                # Remove Old CSV if it cannot be read\n",
    "                print(\"\\tError reading CSV - Generating new frame\")\n",
    "                os.remove(\"{}/{}_{}.csv\".format(csv_files_path,project,myfile[:-4]))\n",
    "                frame = create_training_data(bamfile, project, myfile.split(\".\")[0])\n",
    "                if frame is not None:\n",
    "                    frame.to_csv(\"{}/{}_{}.csv\".format(csv_files_path,project,myfile[:-4]), encoding='utf-8')\n",
    "                \n",
    "        else:\n",
    "            print(\"\\t Generating frame\")\n",
    "            frame = create_training_data(bamfile, project, myfile.split(\".\")[0])\n",
    "            if frame is not None:\n",
    "                frame.to_csv(\"{}/{}_{}.csv\".format(csv_files_path,project,myfile[:-4]), encoding='utf-8')\n",
    "        if frame is not None:\n",
    "            frames.append(frame)\n",
    "        #    if len(frames) == 3:\n",
    "        #        break\n",
    "        print()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Subset Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Runs:  28\n",
      "Number of Sequences: 101,143,062\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of Runs: \", len(frames))\n",
    "print(\"Number of Sequences: {:,}\".format(sum([df.shape[0] for df in frames])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "27\n"
     ]
    }
   ],
   "source": [
    "temp_frames = []\n",
    "for df in frames:\n",
    "    if df.shape[0] >= sample_size:\n",
    "        temp_frames.append(df.sample(sample_size) )\n",
    "print(len(temp_frames))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_frames = temp_frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Subset Sequences: 54,000\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of Subset Sequences: {:,}\".format(sum([df.shape[0] for df in train_frames])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(54000, 14)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db = pandas.concat(train_frames, ignore_index=True)\n",
    "db.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Find length of Each Sequence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "db['seq_len'] = db['seq'].apply(lambda x: len(x))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Remove Rows with Small Sequences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "db = db.query('seq_len > {}'.format(min_seq_len))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "db = db[~db['project_id'].isin(['PRJNA298963'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(53261, 15)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test Train Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "db = db.set_index(['project_id', 'run_accession'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>project_id</th>\n",
       "      <th>run_accession</th>\n",
       "      <th>name</th>\n",
       "      <th>flag</th>\n",
       "      <th>ref_name</th>\n",
       "      <th>ref_pos</th>\n",
       "      <th>map_quality</th>\n",
       "      <th>cigar</th>\n",
       "      <th>next_ref_name</th>\n",
       "      <th>next_ref_pos</th>\n",
       "      <th>length</th>\n",
       "      <th>seq</th>\n",
       "      <th>qual</th>\n",
       "      <th>tags</th>\n",
       "      <th>seq_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>PRJNA281708</td>\n",
       "      <td>SRR1982585</td>\n",
       "      <td>SRR1982585.19285911</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>10933599</td>\n",
       "      <td>60</td>\n",
       "      <td>94M4397N6M</td>\n",
       "      <td>*</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>CTTCATTCTAAAGCAGCGTTATAGTGAAACATTTCATTTAGAAATC...</td>\n",
       "      <td>CCCFFFFFHDDDHIGIIFGIIIIIIIB&gt;HHIIIIIIIIIIIIIIII...</td>\n",
       "      <td>['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PRJNA281708</td>\n",
       "      <td>SRR1982585</td>\n",
       "      <td>SRR1982585.14467585</td>\n",
       "      <td>0</td>\n",
       "      <td>MT</td>\n",
       "      <td>3495</td>\n",
       "      <td>60</td>\n",
       "      <td>95M858N5M</td>\n",
       "      <td>*</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>CGCCACATCTACCATCACCCTCTACATCACCGCCCCGACCTTAGCT...</td>\n",
       "      <td>CCCFFFFFGHHHHJJJJJJJJJJJJIJJJJJJJJJJHGIJGIIIII...</td>\n",
       "      <td>['AS:i:-5', 'ZS:i:-14', 'XN:i:0', 'XM:i:1', 'X...</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PRJNA281708</td>\n",
       "      <td>SRR1982585</td>\n",
       "      <td>SRR1982585.11101432</td>\n",
       "      <td>0</td>\n",
       "      <td>GL000220.1</td>\n",
       "      <td>110225</td>\n",
       "      <td>1</td>\n",
       "      <td>100M</td>\n",
       "      <td>*</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>AAAGTCTTTGGGTTCCGGGGGGAGTATGGTTGCAAAGCTGAAACTT...</td>\n",
       "      <td>@BCFBDFFHHHHHGIJJJJGID:B8ACDDDCDDDDDDDDDDDDDDD...</td>\n",
       "      <td>['AS:i:0', 'ZS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i...</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>PRJNA281708</td>\n",
       "      <td>SRR1982585</td>\n",
       "      <td>SRR1982585.12256500</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>3255989</td>\n",
       "      <td>60</td>\n",
       "      <td>100M</td>\n",
       "      <td>*</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>CCTTAGCTTTACAGCTTTTGCATTTCCTGTTGTATTTATTCTCAGC...</td>\n",
       "      <td>??;D=DDDHHFA?&lt;CA&lt;EF&lt;CFADFHHAFHGCD@HHGHGACGFHGI...</td>\n",
       "      <td>['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PRJNA281708</td>\n",
       "      <td>SRR1982585</td>\n",
       "      <td>SRR1982585.8205913</td>\n",
       "      <td>0</td>\n",
       "      <td>19</td>\n",
       "      <td>5611601</td>\n",
       "      <td>60</td>\n",
       "      <td>30M909N28M897N42M</td>\n",
       "      <td>*</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>CCCCCAAAATGTCGAGTATTTTCTCATTTTCTGGCTCCAAAAGGGA...</td>\n",
       "      <td>CCCFFDFFHHHHHIII2CHHIJGEFEIJJJAFEHGIGIEHIIBHIG...</td>\n",
       "      <td>['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53256</th>\n",
       "      <td>PRJNA277616</td>\n",
       "      <td>SRR1926136</td>\n",
       "      <td>SRR1926136.8578247</td>\n",
       "      <td>133</td>\n",
       "      <td>2</td>\n",
       "      <td>69869763</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>69869763</td>\n",
       "      <td>0</td>\n",
       "      <td>ACTTACACCAATCGATACATCATTTTCAAACGCAATACACTGAATC...</td>\n",
       "      <td>FFFFFFFFFFIIIIIIIIIIIIIIIIIIIIIIIIIIIFIIIFFIII...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53257</th>\n",
       "      <td>PRJNA277616</td>\n",
       "      <td>SRR1926136</td>\n",
       "      <td>SRR1926136.13149211</td>\n",
       "      <td>133</td>\n",
       "      <td>MT</td>\n",
       "      <td>1796</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>1796</td>\n",
       "      <td>0</td>\n",
       "      <td>TGAAAAATTATAACCAAGCATAATATAGCAAGGACTAACCCCTATA...</td>\n",
       "      <td>FFFFFFFFFFIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53258</th>\n",
       "      <td>PRJNA277616</td>\n",
       "      <td>SRR1926136</td>\n",
       "      <td>SRR1926136.11371191</td>\n",
       "      <td>133</td>\n",
       "      <td>10</td>\n",
       "      <td>17235339</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>17235339</td>\n",
       "      <td>0</td>\n",
       "      <td>AGGAAAAGTTTGGAAGAGGCAGAGAAGACCTGCTCAATGTTAAGAT...</td>\n",
       "      <td>FFFFFFFFFFIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53259</th>\n",
       "      <td>PRJNA277616</td>\n",
       "      <td>SRR1926136</td>\n",
       "      <td>SRR1926136.12620085</td>\n",
       "      <td>69</td>\n",
       "      <td>17</td>\n",
       "      <td>29573784</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>29573784</td>\n",
       "      <td>0</td>\n",
       "      <td>CTGGCTCCCCGGGGTCCCCGACTCAGCTGGTGGCTGTGGAGCTGAG...</td>\n",
       "      <td>FFFFFFFFFIIIIIFFIIIIIIIIIIIIIIFFFIIIFFIIFFFFFF...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53260</th>\n",
       "      <td>PRJNA277616</td>\n",
       "      <td>SRR1926136</td>\n",
       "      <td>SRR1926136.11462236</td>\n",
       "      <td>133</td>\n",
       "      <td>1</td>\n",
       "      <td>45693865</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>45693865</td>\n",
       "      <td>0</td>\n",
       "      <td>TTCCAGAAAAAGGACATAAGAGACCTGGTCAAGTATAAAAAATATA...</td>\n",
       "      <td>FFFFFFFFFFBFFFFIIIIIIIFIIIIIFFIIIFFIIIIIIIIIII...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53261 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        project_id run_accession                 name  flag    ref_name  \\\n",
       "0      PRJNA281708    SRR1982585  SRR1982585.19285911     0           7   \n",
       "1      PRJNA281708    SRR1982585  SRR1982585.14467585     0          MT   \n",
       "2      PRJNA281708    SRR1982585  SRR1982585.11101432     0  GL000220.1   \n",
       "3      PRJNA281708    SRR1982585  SRR1982585.12256500     0          18   \n",
       "4      PRJNA281708    SRR1982585   SRR1982585.8205913     0          19   \n",
       "...            ...           ...                  ...   ...         ...   \n",
       "53256  PRJNA277616    SRR1926136   SRR1926136.8578247   133           2   \n",
       "53257  PRJNA277616    SRR1926136  SRR1926136.13149211   133          MT   \n",
       "53258  PRJNA277616    SRR1926136  SRR1926136.11371191   133          10   \n",
       "53259  PRJNA277616    SRR1926136  SRR1926136.12620085    69          17   \n",
       "53260  PRJNA277616    SRR1926136  SRR1926136.11462236   133           1   \n",
       "\n",
       "        ref_pos  map_quality              cigar next_ref_name  next_ref_pos  \\\n",
       "0      10933599           60         94M4397N6M             *             0   \n",
       "1          3495           60          95M858N5M             *             0   \n",
       "2        110225            1               100M             *             0   \n",
       "3       3255989           60               100M             *             0   \n",
       "4       5611601           60  30M909N28M897N42M             *             0   \n",
       "...         ...          ...                ...           ...           ...   \n",
       "53256  69869763            0                  *             =      69869763   \n",
       "53257      1796            0                  *             =          1796   \n",
       "53258  17235339            0                  *             =      17235339   \n",
       "53259  29573784            0                  *             =      29573784   \n",
       "53260  45693865            0                  *             =      45693865   \n",
       "\n",
       "       length                                                seq  \\\n",
       "0           0  CTTCATTCTAAAGCAGCGTTATAGTGAAACATTTCATTTAGAAATC...   \n",
       "1           0  CGCCACATCTACCATCACCCTCTACATCACCGCCCCGACCTTAGCT...   \n",
       "2           0  AAAGTCTTTGGGTTCCGGGGGGAGTATGGTTGCAAAGCTGAAACTT...   \n",
       "3           0  CCTTAGCTTTACAGCTTTTGCATTTCCTGTTGTATTTATTCTCAGC...   \n",
       "4           0  CCCCCAAAATGTCGAGTATTTTCTCATTTTCTGGCTCCAAAAGGGA...   \n",
       "...       ...                                                ...   \n",
       "53256       0  ACTTACACCAATCGATACATCATTTTCAAACGCAATACACTGAATC...   \n",
       "53257       0  TGAAAAATTATAACCAAGCATAATATAGCAAGGACTAACCCCTATA...   \n",
       "53258       0  AGGAAAAGTTTGGAAGAGGCAGAGAAGACCTGCTCAATGTTAAGAT...   \n",
       "53259       0  CTGGCTCCCCGGGGTCCCCGACTCAGCTGGTGGCTGTGGAGCTGAG...   \n",
       "53260       0  TTCCAGAAAAAGGACATAAGAGACCTGGTCAAGTATAAAAAATATA...   \n",
       "\n",
       "                                                    qual  \\\n",
       "0      CCCFFFFFHDDDHIGIIFGIIIIIIIB>HHIIIIIIIIIIIIIIII...   \n",
       "1      CCCFFFFFGHHHHJJJJJJJJJJJJIJJJJJJJJJJHGIJGIIIII...   \n",
       "2      @BCFBDFFHHHHHGIJJJJGID:B8ACDDDCDDDDDDDDDDDDDDD...   \n",
       "3      ??;D=DDDHHFA?<CA<EF<CFADFHHAFHGCD@HHGHGACGFHGI...   \n",
       "4      CCCFFDFFHHHHHIII2CHHIJGEFEIJJJAFEHGIGIEHIIBHIG...   \n",
       "...                                                  ...   \n",
       "53256  FFFFFFFFFFIIIIIIIIIIIIIIIIIIIIIIIIIIIFIIIFFIII...   \n",
       "53257  FFFFFFFFFFIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII...   \n",
       "53258  FFFFFFFFFFIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII...   \n",
       "53259  FFFFFFFFFIIIIIFFIIIIIIIIIIIIIIFFFIIIFFIIFFFFFF...   \n",
       "53260  FFFFFFFFFFBFFFFIIIIIIIFIIIIIFFIIIFFIIIIIIIIIII...   \n",
       "\n",
       "                                                    tags  seq_len  \n",
       "0      ['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...      100  \n",
       "1      ['AS:i:-5', 'ZS:i:-14', 'XN:i:0', 'XM:i:1', 'X...      100  \n",
       "2      ['AS:i:0', 'ZS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i...      100  \n",
       "3      ['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...      100  \n",
       "4      ['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...      100  \n",
       "...                                                  ...      ...  \n",
       "53256                                        ['YT:Z:UP']       98  \n",
       "53257                                        ['YT:Z:UP']       98  \n",
       "53258                                        ['YT:Z:UP']       98  \n",
       "53259                                        ['YT:Z:UP']      100  \n",
       "53260                                        ['YT:Z:UP']       98  \n",
       "\n",
       "[53261 rows x 15 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db = db.reset_index()\n",
    "db"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "stratval = db.groupby('project_id').agg('count')['run_accession'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "db = db.groupby('project_id', group_keys=False).apply(lambda x: x.sample(stratval))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "db = db.set_index(['project_id', 'run_accession'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#db, test = train_test_split(db, test_size=0.2, stratify=project_id)\n",
    "# https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html\n",
    "#db, test = train_test_split(db, test_size=0.2)\n",
    "db, test = train_test_split(db, test_size=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Quick view of Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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></th>\n",
       "      <th>name</th>\n",
       "      <th>flag</th>\n",
       "      <th>ref_name</th>\n",
       "      <th>ref_pos</th>\n",
       "      <th>map_quality</th>\n",
       "      <th>cigar</th>\n",
       "      <th>next_ref_name</th>\n",
       "      <th>next_ref_pos</th>\n",
       "      <th>length</th>\n",
       "      <th>seq</th>\n",
       "      <th>qual</th>\n",
       "      <th>tags</th>\n",
       "      <th>seq_len</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>project_id</th>\n",
       "      <th>run_accession</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">PRJNA281708</th>\n",
       "      <th>SRR1982787</th>\n",
       "      <td>SRR1982787.5323711</td>\n",
       "      <td>0</td>\n",
       "      <td>MT</td>\n",
       "      <td>15353</td>\n",
       "      <td>60</td>\n",
       "      <td>99M1S</td>\n",
       "      <td>*</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ACGGGATCAAACAACCCCCTAGGAATCACCTCCCATTCCGATAAAA...</td>\n",
       "      <td>&lt;=?DFDDDBHFHHJGIGIIIIEFFHHIHIHEHIJGIHGGIGEHGIH...</td>\n",
       "      <td>['AS:i:-1', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:...</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SRR1982585</th>\n",
       "      <td>SRR1982585.3007597</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6144909</td>\n",
       "      <td>60</td>\n",
       "      <td>100M</td>\n",
       "      <td>*</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>TAAAATGCATCACAAAACTACATTATTCAATAGACTTGAGTGTTGC...</td>\n",
       "      <td>CCCFFFFFHHHGHEAGBHIIIIIIIIIIIIIIIIIIIIIIDGGHII...</td>\n",
       "      <td>['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">PRJNA634526</th>\n",
       "      <th>SRR14209634</th>\n",
       "      <td>SRR14209634.3432005</td>\n",
       "      <td>69</td>\n",
       "      <td>3</td>\n",
       "      <td>140490011</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>140490011</td>\n",
       "      <td>0</td>\n",
       "      <td>TGGGAGTGTGGCAGCTGCTGATGCTGCCGTTTGGCTCCTGCGTGGT...</td>\n",
       "      <td>FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFF...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SRR14209614</th>\n",
       "      <td>SRR14209614.2398524</td>\n",
       "      <td>69</td>\n",
       "      <td>12</td>\n",
       "      <td>53498908</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>53498908</td>\n",
       "      <td>0</td>\n",
       "      <td>GTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGT...</td>\n",
       "      <td>FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SRR14209634</th>\n",
       "      <td>SRR14209634.14460128</td>\n",
       "      <td>133</td>\n",
       "      <td>6</td>\n",
       "      <td>118519448</td>\n",
       "      <td>0</td>\n",
       "      <td>*</td>\n",
       "      <td>=</td>\n",
       "      <td>118519448</td>\n",
       "      <td>0</td>\n",
       "      <td>AGAAGATTAACTAGATTAAGACCAGGGGGCAGTAAAGATAGAAACG...</td>\n",
       "      <td>FFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFF,FFFFF,FFFF...</td>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           name  flag ref_name    ref_pos  \\\n",
       "project_id  run_accession                                                   \n",
       "PRJNA281708 SRR1982787       SRR1982787.5323711     0       MT      15353   \n",
       "            SRR1982585       SRR1982585.3007597     0        6    6144909   \n",
       "PRJNA634526 SRR14209634     SRR14209634.3432005    69        3  140490011   \n",
       "            SRR14209614     SRR14209614.2398524    69       12   53498908   \n",
       "            SRR14209634    SRR14209634.14460128   133        6  118519448   \n",
       "\n",
       "                           map_quality  cigar next_ref_name  next_ref_pos  \\\n",
       "project_id  run_accession                                                   \n",
       "PRJNA281708 SRR1982787              60  99M1S             *             0   \n",
       "            SRR1982585              60   100M             *             0   \n",
       "PRJNA634526 SRR14209634              0      *             =     140490011   \n",
       "            SRR14209614              0      *             =      53498908   \n",
       "            SRR14209634              0      *             =     118519448   \n",
       "\n",
       "                           length  \\\n",
       "project_id  run_accession           \n",
       "PRJNA281708 SRR1982787          0   \n",
       "            SRR1982585          0   \n",
       "PRJNA634526 SRR14209634         0   \n",
       "            SRR14209614         0   \n",
       "            SRR14209634         0   \n",
       "\n",
       "                                                                         seq  \\\n",
       "project_id  run_accession                                                      \n",
       "PRJNA281708 SRR1982787     ACGGGATCAAACAACCCCCTAGGAATCACCTCCCATTCCGATAAAA...   \n",
       "            SRR1982585     TAAAATGCATCACAAAACTACATTATTCAATAGACTTGAGTGTTGC...   \n",
       "PRJNA634526 SRR14209634    TGGGAGTGTGGCAGCTGCTGATGCTGCCGTTTGGCTCCTGCGTGGT...   \n",
       "            SRR14209614    GTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGT...   \n",
       "            SRR14209634    AGAAGATTAACTAGATTAAGACCAGGGGGCAGTAAAGATAGAAACG...   \n",
       "\n",
       "                                                                        qual  \\\n",
       "project_id  run_accession                                                      \n",
       "PRJNA281708 SRR1982787     <=?DFDDDBHFHHJGIGIIIIEFFHHIHIHEHIJGIHGGIGEHGIH...   \n",
       "            SRR1982585     CCCFFFFFHHHGHEAGBHIIIIIIIIIIIIIIIIIIIIIIDGGHII...   \n",
       "PRJNA634526 SRR14209634    FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFF...   \n",
       "            SRR14209614    FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF...   \n",
       "            SRR14209634    FFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFF,FFFFF,FFFF...   \n",
       "\n",
       "                                                                        tags  \\\n",
       "project_id  run_accession                                                      \n",
       "PRJNA281708 SRR1982787     ['AS:i:-1', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:...   \n",
       "            SRR1982585     ['AS:i:0', 'XN:i:0', 'XM:i:0', 'XO:i:0', 'XG:i...   \n",
       "PRJNA634526 SRR14209634                                          ['YT:Z:UP']   \n",
       "            SRR14209614                                          ['YT:Z:UP']   \n",
       "            SRR14209634                                          ['YT:Z:UP']   \n",
       "\n",
       "                           seq_len  \n",
       "project_id  run_accession           \n",
       "PRJNA281708 SRR1982787         100  \n",
       "            SRR1982585         100  \n",
       "PRJNA634526 SRR14209634        150  \n",
       "            SRR14209614        114  \n",
       "            SRR14209634        123  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>project_id</th>\n",
       "      <th>PRJNA277616</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>run_accession</th>\n",
       "      <th>SRR1926132</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <td>SRR1926132.13106614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>flag</th>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ref_name</th>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ref_pos</th>\n",
       "      <td>30069923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>map_quality</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cigar</th>\n",
       "      <td>*</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>next_ref_name</th>\n",
       "      <td>=</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>next_ref_pos</th>\n",
       "      <td>30069923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>length</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>seq</th>\n",
       "      <td>GTGATCCCAGTGACAGCGGGGGCCACCGTGCGCCTGCTCTGCTCCA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qual</th>\n",
       "      <td>FFFFFFFFFFFFIIIIIIIIIFIFFFFFFFFFFFFFFFFFFFFFFF...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tags</th>\n",
       "      <td>['YT:Z:UP']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>seq_len</th>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "project_id                                           PRJNA277616\n",
       "run_accession                                         SRR1926132\n",
       "name                                         SRR1926132.13106614\n",
       "flag                                                         133\n",
       "ref_name                                                      16\n",
       "ref_pos                                                 30069923\n",
       "map_quality                                                    0\n",
       "cigar                                                          *\n",
       "next_ref_name                                                  =\n",
       "next_ref_pos                                            30069923\n",
       "length                                                         0\n",
       "seq            GTGATCCCAGTGACAGCGGGGGCCACCGTGCGCCTGCTCTGCTCCA...\n",
       "qual           FFFFFFFFFFFFIIIIIIIIIFIFFFFFFFFFFFFFFFFFFFFFFF...\n",
       "tags                                                 ['YT:Z:UP']\n",
       "seq_len                                                       98"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db.head(1).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "project_id   run_accession\n",
       "PRJNA277616  SRR1867792       1095\n",
       "             SRR1909613       1079\n",
       "             SRR1909637       1054\n",
       "             SRR1909638       1059\n",
       "             SRR1909639       1038\n",
       "             SRR1926132       1112\n",
       "             SRR1926133       1050\n",
       "             SRR1926134       1049\n",
       "             SRR1926135       1034\n",
       "             SRR1926136       1076\n",
       "PRJNA281708  SRR1982584       1009\n",
       "             SRR1982585       1063\n",
       "             SRR1982588       1040\n",
       "             SRR1982612       1076\n",
       "             SRR1982660       1064\n",
       "             SRR1982727       1070\n",
       "             SRR1982783       1042\n",
       "             SRR1982787       1094\n",
       "             SRR1982814       1055\n",
       "             SRR1982826       1094\n",
       "PRJNA634526  SRR14209589      1534\n",
       "             SRR14209591      1512\n",
       "             SRR14209592      1509\n",
       "             SRR14209614      1509\n",
       "             SRR14209634      1523\n",
       "             SRR14209655      1495\n",
       "             SRR14209708      1491\n",
       "dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db.groupby(level=['project_id','run_accession']).size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# K-Mer Optimze"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accessing:  https://www.ebi.ac.uk/ena/portal/api/filereport\n",
      "\tMaking Request To:  https://www.ebi.ac.uk/ena/portal/api/filereport?accession=PRJNA281708&fields=center_name%2Cinstrument_platform%2Cinstrument_model%2Chost&format=json&download=false&result=read_run\n",
      "Accessing:  https://www.ebi.ac.uk/ena/portal/api/filereport\n",
      "\tMaking Request To:  https://www.ebi.ac.uk/ena/portal/api/filereport?accession=PRJNA634526&fields=center_name%2Cinstrument_platform%2Cinstrument_model%2Chost&format=json&download=false&result=read_run\n",
      "Accessing:  https://www.ebi.ac.uk/ena/portal/api/filereport\n",
      "\tMaking Request To:  https://www.ebi.ac.uk/ena/portal/api/filereport?accession=PRJNA277616&fields=center_name%2Cinstrument_platform%2Cinstrument_model%2Chost&format=json&download=false&result=read_run\n"
     ]
    },
    {
     "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>run_accession</th>\n",
       "      <th>sample_accession</th>\n",
       "      <th>center_name</th>\n",
       "      <th>instrument_platform</th>\n",
       "      <th>instrument_model</th>\n",
       "      <th>host</th>\n",
       "      <th>project_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SRR1982584</td>\n",
       "      <td>SAMN03495628</td>\n",
       "      <td>GEO</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA281708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SRR1982585</td>\n",
       "      <td>SAMN03495629</td>\n",
       "      <td>GEO</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA281708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>SRR1982586</td>\n",
       "      <td>SAMN03495630</td>\n",
       "      <td>GEO</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA281708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SRR1982587</td>\n",
       "      <td>SAMN03495631</td>\n",
       "      <td>GEO</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA281708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SRR1982588</td>\n",
       "      <td>SAMN03495632</td>\n",
       "      <td>GEO</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA281708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>416</th>\n",
       "      <td>SRR1926132</td>\n",
       "      <td>SAMN03394587</td>\n",
       "      <td>Weill Cornell Medical College in Qatar</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA277616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>417</th>\n",
       "      <td>SRR1926133</td>\n",
       "      <td>SAMN03394586</td>\n",
       "      <td>Weill Cornell Medical College in Qatar</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA277616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>418</th>\n",
       "      <td>SRR1926134</td>\n",
       "      <td>SAMN03394585</td>\n",
       "      <td>Weill Cornell Medical College in Qatar</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA277616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>419</th>\n",
       "      <td>SRR1926135</td>\n",
       "      <td>SAMN03394584</td>\n",
       "      <td>Weill Cornell Medical College in Qatar</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA277616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>420</th>\n",
       "      <td>SRR1926136</td>\n",
       "      <td>SAMN03394588</td>\n",
       "      <td>Weill Cornell Medical College in Qatar</td>\n",
       "      <td>ILLUMINA</td>\n",
       "      <td>Illumina HiSeq 2500</td>\n",
       "      <td></td>\n",
       "      <td>PRJNA277616</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>421 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    run_accession sample_accession                             center_name  \\\n",
       "0      SRR1982584     SAMN03495628                                     GEO   \n",
       "1      SRR1982585     SAMN03495629                                     GEO   \n",
       "2      SRR1982586     SAMN03495630                                     GEO   \n",
       "3      SRR1982587     SAMN03495631                                     GEO   \n",
       "4      SRR1982588     SAMN03495632                                     GEO   \n",
       "..            ...              ...                                     ...   \n",
       "416    SRR1926132     SAMN03394587  Weill Cornell Medical College in Qatar   \n",
       "417    SRR1926133     SAMN03394586  Weill Cornell Medical College in Qatar   \n",
       "418    SRR1926134     SAMN03394585  Weill Cornell Medical College in Qatar   \n",
       "419    SRR1926135     SAMN03394584  Weill Cornell Medical College in Qatar   \n",
       "420    SRR1926136     SAMN03394588  Weill Cornell Medical College in Qatar   \n",
       "\n",
       "    instrument_platform     instrument_model host   project_id  \n",
       "0              ILLUMINA  Illumina HiSeq 2500       PRJNA281708  \n",
       "1              ILLUMINA  Illumina HiSeq 2500       PRJNA281708  \n",
       "2              ILLUMINA  Illumina HiSeq 2500       PRJNA281708  \n",
       "3              ILLUMINA  Illumina HiSeq 2500       PRJNA281708  \n",
       "4              ILLUMINA  Illumina HiSeq 2500       PRJNA281708  \n",
       "..                  ...                  ...  ...          ...  \n",
       "416            ILLUMINA  Illumina HiSeq 2500       PRJNA277616  \n",
       "417            ILLUMINA  Illumina HiSeq 2500       PRJNA277616  \n",
       "418            ILLUMINA  Illumina HiSeq 2500       PRJNA277616  \n",
       "419            ILLUMINA  Illumina HiSeq 2500       PRJNA277616  \n",
       "420            ILLUMINA  Illumina HiSeq 2500       PRJNA277616  \n",
       "\n",
       "[421 rows x 7 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pandas.concat([get_location_data(x) for x in projects], ignore_index=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_seq_len = int(db.seq_len.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "28 84\n"
     ]
    }
   ],
   "source": [
    "kmin = int(avg_seq_len * 1./4)\n",
    "kmax = int(avg_seq_len * 3./4)\n",
    "ntopics_min = 12\n",
    "ntopics_max = 40\n",
    "print(kmin,kmax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration  1\n",
      "\tTesting: K-> 30, Num Topics-> 36\n",
      "\tElapsed Time: 43.9136164188385s\n",
      "Iteration  2\n",
      "\tTesting: K-> 66, Num Topics-> 38\n",
      "\tElapsed Time: 31.34852623939514s\n",
      "Iteration  3\n",
      "\tTesting: K-> 50, Num Topics-> 24\n",
      "\tElapsed Time: 28.341092824935913s\n",
      "Iteration  4\n",
      "\tTesting: K-> 41, Num Topics-> 14\n",
      "\tElapsed Time: 25.123136043548584s\n",
      "Iteration  5\n",
      "\tTesting: K-> 83, Num Topics-> 37\n",
      "\tElapsed Time: 20.390226364135742s\n",
      "Iteration  6\n",
      "\tTesting: K-> 47, Num Topics-> 28\n",
      "\tElapsed Time: 34.23981308937073s\n",
      "Iteration  7\n",
      "\tTesting: K-> 80, Num Topics-> 31\n",
      "\tElapsed Time: 22.56856417655945s\n",
      "Iteration  8\n",
      "\tTesting: K-> 28, Num Topics-> 21\n",
      "\tElapsed Time: 38.1758828163147s\n",
      "Iteration  9\n",
      "\tTesting: K-> 32, Num Topics-> 25\n",
      "\tElapsed Time: 34.802687883377075s\n",
      "Iteration  10\n",
      "\tTesting: K-> 42, Num Topics-> 21\n",
      "\tElapsed Time: 30.048315286636353s\n",
      "Iteration  11\n",
      "\tTesting: K-> 36, Num Topics-> 38\n",
      "\tElapsed Time: 43.74851870536804s\n",
      "Iteration  12\n",
      "\tTesting: K-> 58, Num Topics-> 40\n",
      "\tElapsed Time: 35.93093490600586s\n",
      "Iteration  13\n",
      "\tTesting: K-> 60, Num Topics-> 36\n",
      "\tElapsed Time: 32.369874715805054s\n",
      "Iteration  14\n",
      "\tTesting: K-> 32, Num Topics-> 27\n",
      "\tElapsed Time: 37.9310405254364s\n",
      "Iteration  15\n",
      "\tTesting: K-> 77, Num Topics-> 30\n",
      "\tElapsed Time: 23.958709955215454s\n",
      "Iteration  16\n",
      "\tTesting: K-> 79, Num Topics-> 25\n",
      "\tElapsed Time: 21.16655921936035s\n",
      "Iteration  17\n",
      "\tTesting: K-> 58, Num Topics-> 27\n",
      "\tElapsed Time: 30.300694704055786s\n",
      "Iteration  18\n",
      "\tTesting: K-> 72, Num Topics-> 35\n",
      "\tElapsed Time: 26.707971811294556s\n",
      "Iteration  19\n",
      "\tTesting: K-> 36, Num Topics-> 31\n",
      "\tElapsed Time: 39.983216285705566s\n",
      "Iteration  20\n",
      "\tTesting: K-> 50, Num Topics-> 20\n",
      "\tElapsed Time: 28.39588212966919s\n",
      "Iteration  21\n",
      "\tTesting: K-> 79, Num Topics-> 40\n",
      "\tElapsed Time: 26.00676941871643s\n",
      "Iteration  22\n",
      "\tTesting: K-> 34, Num Topics-> 25\n",
      "\tElapsed Time: 37.8917019367218s\n",
      "Iteration  23\n",
      "\tTesting: K-> 75, Num Topics-> 27\n",
      "\tElapsed Time: 22.391425371170044s\n",
      "Iteration  24\n",
      "\tTesting: K-> 61, Num Topics-> 34\n",
      "\tElapsed Time: 32.61837935447693s\n",
      "Iteration  25\n",
      "\tTesting: K-> 74, Num Topics-> 29\n",
      "\tElapsed Time: 22.165712118148804s\n",
      "Iteration  26\n",
      "\tTesting: K-> 52, Num Topics-> 15\n",
      "\tElapsed Time: 24.234911918640137s\n",
      "Iteration  27\n",
      "\tTesting: K-> 76, Num Topics-> 13\n",
      "\tElapsed Time: 16.29258704185486s\n",
      "Iteration  28\n",
      "\tTesting: K-> 29, Num Topics-> 18\n",
      "\tElapsed Time: 31.724104166030884s\n",
      "Iteration  29\n",
      "\tTesting: K-> 37, Num Topics-> 34\n",
      "\tElapsed Time: 41.51498317718506s\n",
      "Iteration  30\n",
      "\tTesting: K-> 81, Num Topics-> 40\n",
      "\tElapsed Time: 24.74760413169861s\n",
      "Iteration  31\n",
      "\tTesting: K-> 56, Num Topics-> 24\n",
      "\tElapsed Time: 28.821160316467285s\n",
      "Iteration  32\n",
      "\tTesting: K-> 34, Num Topics-> 12\n",
      "\tElapsed Time: 28.637648344039917s\n",
      "Iteration  33\n",
      "\tTesting: K-> 64, Num Topics-> 15\n",
      "\tElapsed Time: 18.884990692138672s\n",
      "Iteration  34\n",
      "\tTesting: K-> 78, Num Topics-> 17\n",
      "\tElapsed Time: 15.271913290023804s\n",
      "Iteration  35\n",
      "\tTesting: K-> 49, Num Topics-> 15\n",
      "\tElapsed Time: 22.571495532989502s\n",
      "Iteration  36\n",
      "\tTesting: K-> 28, Num Topics-> 31\n",
      "\tElapsed Time: 40.36688709259033s\n",
      "Iteration  37\n",
      "\tTesting: K-> 41, Num Topics-> 19\n",
      "\tElapsed Time: 28.022223234176636s\n",
      "Iteration  38\n",
      "\tTesting: K-> 42, Num Topics-> 38\n",
      "\tElapsed Time: 46.30999541282654s\n",
      "Iteration  39\n",
      "\tTesting: K-> 47, Num Topics-> 14\n",
      "\tElapsed Time: 28.671597957611084s\n",
      "Iteration  40\n",
      "\tTesting: K-> 39, Num Topics-> 39\n",
      "\tElapsed Time: 46.8013231754303s\n",
      "Iteration  41\n",
      "\tTesting: K-> 42, Num Topics-> 40\n",
      "\tElapsed Time: 46.86448407173157s\n",
      "Iteration  42\n",
      "\tTesting: K-> 31, Num Topics-> 39\n",
      "\tElapsed Time: 47.58927774429321s\n",
      "Iteration  43\n",
      "\tTesting: K-> 75, Num Topics-> 39\n",
      "\tElapsed Time: 28.591898202896118s\n",
      "Iteration  44\n",
      "\tTesting: K-> 42, Num Topics-> 29\n",
      "\tElapsed Time: 39.07671546936035s\n",
      "Iteration  45\n",
      "\tTesting: K-> 72, Num Topics-> 21\n",
      "\tElapsed Time: 20.402565956115723s\n",
      "Iteration  46\n",
      "\tTesting: K-> 31, Num Topics-> 22\n",
      "\tElapsed Time: 32.26232933998108s\n",
      "Iteration  47\n",
      "\tTesting: K-> 46, Num Topics-> 32\n",
      "\tElapsed Time: 37.613850831985474s\n",
      "Iteration  48\n",
      "\tTesting: K-> 58, Num Topics-> 16\n",
      "\tElapsed Time: 22.21425199508667s\n",
      "Iteration  49\n",
      "\tTesting: K-> 29, Num Topics-> 16\n",
      "\tElapsed Time: 34.19434070587158s\n",
      "Iteration  50\n",
      "\tTesting: K-> 49, Num Topics-> 14\n",
      "\tElapsed Time: 22.948264360427856s\n",
      "Iteration  51\n",
      "\tTesting: K-> 81, Num Topics-> 15\n",
      "\tElapsed Time: 13.008006811141968s\n",
      "Iteration  52\n",
      "\tTesting: K-> 82, Num Topics-> 19\n",
      "\tElapsed Time: 15.147243738174438s\n",
      "Iteration  53\n",
      "\tTesting: K-> 45, Num Topics-> 36\n",
      "\tElapsed Time: 39.43368411064148s\n",
      "Iteration  54\n",
      "\tTesting: K-> 54, Num Topics-> 18\n",
      "\tElapsed Time: 26.950051069259644s\n",
      "Iteration  55\n",
      "\tTesting: K-> 62, Num Topics-> 19\n",
      "\tElapsed Time: 59.77190113067627s\n",
      "Iteration  56\n",
      "\tTesting: K-> 75, Num Topics-> 14\n",
      "\tElapsed Time: 44.497663259506226s\n",
      "Iteration  57\n",
      "\tTesting: K-> 68, Num Topics-> 40\n",
      "\tElapsed Time: 118.87326550483704s\n",
      "Iteration  58\n",
      "\tTesting: K-> 55, Num Topics-> 28\n",
      "\tElapsed Time: 134.0945417881012s\n",
      "Iteration  59\n",
      "\tTesting: K-> 30, Num Topics-> 23\n",
      "\tElapsed Time: 128.4014253616333s\n",
      "Iteration  60\n",
      "\tTesting: K-> 50, Num Topics-> 25\n",
      "\tElapsed Time: 138.68492555618286s\n",
      "Iteration  61\n",
      "\tTesting: K-> 58, Num Topics-> 13\n",
      "\tElapsed Time: 32.15292000770569s\n",
      "Iteration  62\n",
      "\tTesting: K-> 81, Num Topics-> 20\n",
      "\tElapsed Time: 28.17275881767273s\n",
      "Iteration  63\n",
      "\tTesting: K-> 55, Num Topics-> 16\n",
      "\tElapsed Time: 55.18068027496338s\n",
      "Iteration  64\n",
      "\tTesting: K-> 36, Num Topics-> 29\n",
      "\tElapsed Time: 116.61451005935669s\n",
      "Iteration  65\n",
      "\tTesting: K-> 73, Num Topics-> 14\n",
      "\tElapsed Time: 62.98215413093567s\n",
      "Iteration  66\n",
      "\tTesting: K-> 51, Num Topics-> 18\n",
      "\tElapsed Time: 49.04584550857544s\n",
      "Iteration  67\n",
      "\tTesting: K-> 71, Num Topics-> 32\n",
      "\tElapsed Time: 84.62914681434631s\n",
      "Iteration  68\n",
      "\tTesting: K-> 81, Num Topics-> 25\n",
      "\tElapsed Time: 66.66202926635742s\n",
      "Iteration  69\n",
      "\tTesting: K-> 62, Num Topics-> 14\n",
      "\tElapsed Time: 82.29737520217896s\n",
      "Iteration  70\n",
      "\tTesting: K-> 41, Num Topics-> 32\n",
      "\tElapsed Time: 154.99244379997253s\n"
     ]
    }
   ],
   "source": [
    "import random\n",
    "import gensim\n",
    "from gensim.models.coherencemodel import CoherenceModel\n",
    "from gensim import corpora\n",
    "from pprint import pprint\n",
    "import time\n",
    "\n",
    "def clean_seq(sequences, K):\n",
    "    tmp = []\n",
    "    nucleotides = [x*K for x in ['A','C','G','T','N']]\n",
    "    ##### BEGIN TEST WITH DOUBLES #####\n",
    "    #from itertools import combinations\n",
    "    #combs = list(combinations(['A','C','G','T','N'], 2))\n",
    "    #temp = [ \"{}{}\".format(x[0],x[1]) for x in combs] + [ \"{}{}\".format(x[1],x[0]) for x in combs]\n",
    "    #temp = [x*K for x in temp]\n",
    "    #temp = [x[:K] for x in temp]\n",
    "    #nucleotides = nucleotides + temp\n",
    "    ##### END TEST WITH DOUBLES #####\n",
    "    #print(nucleotides)\n",
    "    for sequence in sequences:\n",
    "        if any([True if repeated_nucleotide in sequence else False for repeated_nucleotide in nucleotides ]):\n",
    "            continue\n",
    "        else:\n",
    "            tmp.append(sequence)   \n",
    "    return tmp\n",
    "\n",
    "db = db.sort_index()\n",
    "indexes = sorted(list(set(db.index)))\n",
    "N = len(indexes)\n",
    "\n",
    "coherence_vals = []\n",
    "KS = []\n",
    "ntopics_arr = []\n",
    "done_list = []\n",
    "avg_seq_len = int(db.seq_len.mean())\n",
    "for i in range(70):\n",
    "    \n",
    "    #K = random.randint(int(avg_seq_len * 1./3), int(avg_seq_len * 2./3))\n",
    "    K = random.randint(kmin, kmax)\n",
    "    ntopics = random.randint(ntopics_min, ntopics_max)\n",
    "    complete = (K,ntopics)\n",
    "    if complete in done_list:\n",
    "        continue\n",
    "    done_list.append(complete)\n",
    "    print(\"Iteration \",i+1)\n",
    "    print(\"\\tTesting: K-> {}, Num Topics-> {}\".format( K,ntopics))\n",
    "    start = time.time()\n",
    "    mydict = {}\n",
    "\n",
    "    for i in range(N):\n",
    "        key = indexes[i]\n",
    "        seq = db.loc[key]['seq'].to_list()\n",
    "        seq = clean_seq(seq, K)\n",
    "        arr = read_kmers_from_list(seq, K)\n",
    "        mydict[key] = arr\n",
    "    \n",
    "    text_tokens = []\n",
    "    for key in mydict.keys():\n",
    "        seq = mydict[key]\n",
    "        text_tokens.append(seq)\n",
    "    # Create Dictionary\n",
    "    id2word = corpora.Dictionary(text_tokens)\n",
    "    # Create Corpus\n",
    "    texts = text_tokens\n",
    "    # Term Document Frequency\n",
    "    corpus = [id2word.doc2bow(text) for text in texts]\n",
    "\n",
    "    lda_model = gensim.models.ldamodel.LdaModel(corpus=corpus,\n",
    "                                           id2word=id2word,\n",
    "                                           num_topics=ntopics)\n",
    "    cm = CoherenceModel(model=lda_model, corpus=corpus, coherence='u_mass')\n",
    "    coherence = cm.get_coherence()\n",
    "    end = time.time()\n",
    "    print(\"\\tElapsed Time: {}s\".format(str(end - start)) )\n",
    "    coherence_vals.append(coherence)\n",
    "    KS.append(K)\n",
    "    ntopics_arr.append(ntopics)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "coherence_vals = numpy.array(coherence_vals)\n",
    "KS = numpy.array(KS)\n",
    "ntopics_arr = numpy.array(ntopics_arr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "pickle.dump([avg_seq_len,KS,ntopics_arr,coherence_vals ], open(\"optimize.p\", \"w+b\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "avg_seq_len,KS,ntopics_arr,coherence_vals = pickle.load(open(\"optimize.p\",\"rb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import axes3d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(10,6))\n",
    "ax = axes3d.Axes3D(fig)\n",
    "ax.scatter3D(KS,ntopics_arr,coherence_vals, c='r');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy as sp\n",
    "import scipy.interpolate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Coherence')"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdUAAAG+CAYAAADFip5uAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOydd3wcZ7X3vzPbi7SrVe9dlmXLVVISp/dKGi30+tIucOFeIHR4ubQL9+UCN/SWCwQCIaQnTpzuGrmr997LFml7mXn/WO9GUmRbtleWbM+XTz4hcTQ7mpl9fnPOc87vCLIso6CgoKCgoHDmiCt9AgoKCgoKCucLiqgqKCgoKCgkCEVUFRQUFBQUEoQiqgoKCgoKCglCEVUFBQUFBYUEoYiqgoKCgoJCglCf5M+VfhsFBQUFhXMRYSU+VIlUFRQUFBQUEoQiqgoKCgoKCglCEVUFBQUFBYUEoYiqgoKCgoJCglBEVUFBQUFBIUEooqqgoKCgoJAgFFFVUFBQUFBIEIqoKigoKCgoJAhFVBUUFBQUFBKEIqoKCgoKCgoJQhFVBQUFBQWFBKGIqoKCgoKCQoJQRFVBQUFBQSFBKKKqoKCgoKCQIBRRVVBQUFBQSBCKqCooKCgoKCQIRVQVFBQUFBQShCKqCgoKCgoKCUIRVQUFBQUFhQShiKqCgoKCgkKCUERVQUFBQUEhQSiiqqCgoKCgkCAUUVVQUFBQUEgQiqgqKCgoKCgkCEVUFRQUFBQUEoQiqgoKCgoKCglCEVUFBQUFBYUEoYiqgoKCgoJCglBEVUFBQUFBIUEooqqgoKCgoJAgFFFVUFBQUFBIEIqoKpw3yLJMJBJBluWVPhUFBYULFPVKn4CCQiKQZZlwOIzX60WSJNRqNRqNBrVajUqlQhCElT5FBQWFCwDhJG/1yiu/wqpHkiRCoRCSJBEOh5EkKf7vY2KqVqvjfykiq6BwQbAiX3JFVBXOWWLp3lAohCAICIJAMBhEluV5oinLcvyvGHMjWVEUFZFVUDj/WJEvtZL+VTgnkWWZYDAYj0ZPJIoL/zwmxuFwOP7nKpVKEVkFBYUzRolUFc45JEmaF5HGBNDr9dLe3o5eryclJQWLxYJKpTrp8RZGsoIgzEsXKyKroHBOoqR/FRRORKwYKRwOvyH6HB0dpaenh9LSUsLhME6nE5fLhUqlIiUlhZSUFJKTkxHFkxe8LxTZ2D6txWJRRFZB4dxBEVUFheMhyzJjY2OYTCa0Wm1c1CKRCK2trYTDYdatWxcXw9ifB4NBHA4HTqeTmZkZ1Gp1XGSTkpKWJLJ2u52pqSlKS0uB1yNZjUaDSqVSRFZBYXWi7KkqKCxGOBwmFAoxMDBAaWkpOp0OgNnZWRobG8nPzycvLy9eqDQXrVZLZmYmmZmZAAQCARwOByMjI8zOzqLVaueJ7GLiGIuKY6lkWZYJhUKEQqH4n89t31FEVkHhwkURVYVVy8J0byyqlGWZwcFBhoeH2bBhA2azecnH1Ol0ZGVlkZWVBYDf78fhcDA0NITb7Uav12O1WklJScFsNscFdW5GZ67Axs4nGAwSCAQAEEUxLrJqtfqkhVQKCgrnD4qoKqxK5vaexkRJEARCoRBHjhxBq9VSV1e3pEKkE6HX68nOziY7OxtZlvH5fDidTgYGBnC73RiNRnQ6HaFQ6A2tOjFOJLKxl4G56WJFZBUUzl+UPVWFVcXc3lNg3p7nwYMH8Xg8VFRUxCPNhSzWp3om5+L1ehkdHWVychJRFDGZTKSkpGC1WjEajSf9nNj3K2ZIEfudFu7JKigoJBylUEnhwia2VxmJROZFc7Is09PTw8DAAFVVVfH90cVIpKjGcLlcjI2NUVFRgcfjiRc+eb1ezGZzfE9Wr9crIqugsHpQCpUULlyO13vq9/tpbGzEYrGQkZGBVqs94XEW7n8mEkEQMJvNmM1m8vPzkWUZt9uNw+Ggs7MTv9//BpFd7BjAvKKnWLo4VmQ1d09WEVkFhXMLRVQVVpTjFSMBTE5O0tHRQWVlJampqbS2tq7IBJrjCbUgCCQlJZGUlERBQQGSJMVFtq2tjWAwSFJSUlxkY1XLC4+xmKViIBBYtPBJEVkFhdWNIqoKK8bxrAYlSaKjowOPx0NNTU1cjJYzCk0EoiiSnJxMcnIyhYWFSJLE7OwsDoeDlpYWQqEQFoslXl28WNS9FJGNfZbBYIhXFysoKKwOFFFVWBFixUiLWQ02NDSQmZnJmjVr5gnGSonq6X6uKIpYLBYsFgtFRUVIkoTL5cLpdDI8PEwkEsFiscQLnzQazaKfvVBknU4n4+PjlJeXA8zzLVYm8CgorCyKqCqcVU6U7h0ZGaGvr49169ZhsVje8LOrPVI9GaIoxlPBxcXFRCKRuMgODAwgy/I8kVWr3/j1jF2zWBuPLMtIkoTf74//N4rIKiisHIqoKpw1Fus9hahjUmtrK5IkUVdXt6iYwLkXqZ4MlUqFzWbDZrMB0evgcrlwOBz09/cjy3I8VRzzHZ57TrG/L4xkFZFVUFg5FFFVWHYW9p7OFYKZmRmampooKCggNzf3pCPczmfUajWpqamkpqYCxAcD2O12ent7EQQBq9WKVqud15Izl+OJrM/nm1d5rIisgsLyoIiqwrJyot7TgYEBRkZGlmw1KAjCccXkfEStVpOWlkZaWhoAoVAovp/qcDjwer3zItnFqoJj13yuxeNCkZ075k4RWQWFM0MRVYVl43i9p8FgkKamJvR6/SlZDa7UYr9a9nI1Gg3p6eloNBq0Wi1FRUU4nU4mJibo6upCrVZjtVqx2WzHncCzmMjGsghzRVYZ2K6gcHoooqqQcBYu1HMX91h7SVlZ2QmdkRbjQotUT4ZWqyUjI4OMjAzg9Qk8o6OjtLe3z5vAYzabT0lkw+Fw/GVIEVkFhaWjiKpCQpFlmYmJCdRqNSaTaV66t7u7G7vdzpYtWzAYDKd87As9Uj0Zx5vAMzw8zOzsLDqdbp7InmjMXYy5Ihv787npYkVkFRTmo4iqQsKIpXunpqYwmUzxfdKY1aDVaqWmpua0HYGWEqkqC/zrzJ3AA+Dz+XA4HPEJPAaDIS6yc1+A5rKYyMbm28b+XBFZBYXXUURV4YxZ2Hs6d490YmKCzs7OuNXgmaAs1q9zOtfCYDBgMBjIycmJj7lzOBz09fXh8XgwGo1xkT3eBJ7FRDYUCtHe3k5WVhYmk+kNwwGU+6ZwIaGIqsIZcby5p7HeU6/XS21t7UmN8JfCSu2prrb0byLORRAEjEYjRqOR3Nzc+Jg7h8NBT09PfAJPrLrYYDCccJZsMBicN/N2biS70LdYEVmF8xlFVBVOixP1noZCIfr7+yksLKSysjJhi+hqE7fzCUEQMJlMmEwm8vLykGU5Puauq6sLn88XHw5gtVoX3ROPPQMnGtg+V2RjvsWKyCqcTyiiqnDKLEz3zl0UR0ZGGB4eJjs7m6KiooR+7lJFNdHzVC9EMV9szN3s7CxOp5OOjg4CgcC8CTzHu+ZLFdlYulgRWYVzHUVUFU6JE1kNtrS0AFBcXLwsn30hittqQRCE+ASe2Ji7uRN4ZmdnAUhPTz/uBJ7YcebOkgXeMEtWGdiucC6jiKrCkjhR72nMarCwsJDc3FyGh4fjaeFEcr55/57LLJzA09DQQHp6Ol6vl5GREcLhMMnJyfFI9ngTeABFZBXOKxRRVTgpx5t7Kssy/f39jI6OsnHjRkwmExBdCJdDhBRxW91YLJa4oUckEmFmZgaHw8HQ0BCRSASr1Rr/60xEVhnYrrCaUURV4YScyGqwsbERo9HIRRddNG9xW64q3ZWMVBVOzMI9VZVKFY9SgfiYu1ifbGwCT+yv4425ix0r9hkLB7YrIquw2lBEVWFRTjT31G6309raSnl5edwiby7nY6S6miLkRBdinQ1ONOaur68vPoEnNhxgMT/oxXpkF4psbAKPSqWKVxcrKJxNFFFVeAPHS/dKkkR3dzcOh4OtW7ei1+sX/fnzLVJVODmnKvQLx9zFJvBMT0/T09ODKIpxkU1OTl6yyM6dJTs7O4tGo4lHwsoEHoWzgSKqCvNYaEEXW4R8Ph+NjY3YbDZqa2tPOvd0ucTvZMddjkVTEfOTc6bRc2wCT3p6OhDdXlhsAk9MZE80HCCG0+lEr9fPe/lTZskqLDeKqCoAJ+49jVkNrl27Np6+OxHLJUInSyuHQiGampoIBAKkpKSccASaQmJJdEp64QSeYDCIw+FgbGyMjo4ONBpNfM/2ePdYlmVEUZy3Jzs3kgVFZBUSjyKqCsftPY1EIrS3t+P3+0/JalAUxWWzEzyeqDqdTpqbmykpKcFkMuFyueIj0JYyneVEKJHqyqPVasnMzIxXF8fG3I2MjMybwGO1WklKSorfs4UFdIuli+cObFdEVuFMUUT1Amah1eDcBcjtdtPY2EhOTg5r1649pcXlbEaqsbaesbExNm/ejE6nIxQKzVuAF05nMZlMcZE9nqetwqlxtounFo658/l8OJ1OhoaGcLvd6PV6IpEIWq32hG5PC2fJLhTZuRN4FJFVWAqKqF6gxKaUzDXBj/37kZER+vv7Wb9+PcnJyad87OWq/oX5kWos3avX66mrq0MURSKRyBt+ZuF0lrmetn6/f57dnk6nW5bzPt9Z6Yrk2D3Ozs6OP9sdHR1MTEwwPDx8ShN4Fg5sjxmeAPOMKBSRVVgMRVQvQGK9p/v27WPbtm3zrAabm5sRRZG6urpFeweXwnJV/84Va5fLRVNTE6WlpfFoZannNtfTVpIk3G43drudlpYWwuEwFoslnkrUaDSrLv270gK2GKvp+sydwJOeno7Vao1P4Ont7cXj8SwpW3E8kY0NbIfXRVaZJasQQxHVC4jFipFii4DL5aK5uZmioiJycnLO6HOWU4QkSaKvry+e7jUajWd0PFEU4562RUVFi5oUWCwWwuEwkUhk0dYOhSirTVBie6pLmcBjNpvniexiLLYnO1dkBUEZ2K6giOoFw4msBvv6+hgfH59nNXgmLFf6NxKJMD09jUqliqd7E81iJgV2u53x8XEOHToUb+1QKovfyGoTkBPtpS6cwON2u3E4HPMm8MRaeE7Uj71QZBe2pCkie+GhiOoFQGxfaKHVoCRJHDp0CJPJlFCRWo70byySNhgMVFVVJfTYJ0KtVpOWlsbAwAA1NTXx1o65Vac2m42UlBRMJpOyaK4iYi+QJ0MQBJKSkkhKSopP4ImJbFtbG8FgkOTk5LjIHm/ffTGRXTiwfeFwAOV5Of9QRPU85kRWg9PT03i9XtasWRNvuE8UiUz/yrLMwMAAIyMjVFZWMjIykpDjni4LWztilcV9fX3z9upsNttx04gKZ4eFLTVLZe6WQGFh4RvG3IVCoXn77ksZcxc7n1AoRHNzM4WFhej1+jf4Fisie+6jiOp5yvF6T2NWg06nE6PRSFpaWsI/O1F9qrHqXp1OR11dHX6/f9U5Kh2vsrizs1OpLF5hElXQtXDMnSRJ8X332ASehcVtixET2UAgEPclXmxgeyxdvDDyVTg3UET1PGNh7+lCq8GGhgbS0tKoqanhtddeQ5KkhBffJCJSjaV7i4uLyc7OTthxl5PFKovnRjhzK4tTUlJOu7paYWksV5W0KIonncAzV2QX3mdJkuJR6cJINiaysc/RaDTxdLEisucGyrf6PCKWXopEIm/4Ao6Pj9PV1UVVVVV8MYhFlKtJVGVZZnBwkOHh4TcUTp1ro98WRjgLF1/gpJNZEn1OFxIx8VpuljKBZ67IHu+85oqsMrD93EUR1fOE4809jVkNBgIB6urq5qWmlqtK93TTv+FwmKamJjQaDXV1dW8QmdUeqZ6MxRZfh8PB1NQU3d3dqNXqk/rZwurqCV3NrFQ/78IJPOFwGKfTid1uj/fJ9vb2nnTMHZx8YLsisqsPRVTPcRa6vixmNZibm7uo1eBqGtE2MzNDU1PTvHRvIo67mlGr1fMmsyz0s9Xr9XGRVSqLT52lVv8uN7EK8lj9Qn19PcnJyfGXqdhAd6vVisViOW4UC4rIngsoonoOc6Le0+HhYQYGBqiuriYpKWnRn18u4/tTWchOlO5d7Ljnk6guZDE/W7vdPq+yeClTghSinG7173IjCMJJx9ydLGOxmMjG1oO5Iruwulhh+VFE9RzleOneWMm+Wq0+qdXgck6TWQqxdG/sXE+2p3i+i+pCDAYDubm55ObmzqssHhsbw+/3EwgE4j2yS50gdCGxGu0cF2PhmLuFGQutVjtPZE9kqRgjJrKBQCBe+BSbwKNSqeLVxQqJRxHVc4wT9Z7OHX92vBTqXFZSpGLp3lOxRbzQRHUucyuLDQYDLpeLtLQ07HY7w8PD8bYOm822aMXphci5IqoLWZix8Pv98fad2ASeWIHb8UYZHm/MnTJLdvlRvnnnEMfrPZ1rNXgqfrgrEanKsszQ0BBDQ0Ns2LABs9m85J+9kEV1IbGKUovFQnFx8bzK4ljF6elWFp9PnA9Codfryc7Ojk/giYlsbJSh0WiM3+vj7b0vRWRlWUatVmMwGBSRPQMUUT0HOFHvaSAQoLGxkaSkpFO2GjzbohqbghPz7j3VhX6ponohLAYLf8eFlcWhUAin0/mGymKbzYbZbFb2185RBEF4g+HI8Vy9rFbrScfcxZBlmfHxcbxeL4WFhYASyZ4uiqiuchabLBNjamqK9vZ2KioqTstq8GyK6uzsLI2NjWc0BUeJVKMs5RpoNJpFK4uHh4eVyuIVJNHP79wxd3P33p1OJz09PXi93vgEHqvVesIxd7Isx/dcFxvYrojs0lBEdRUjSRLT09PxfZO5VoOdnZ3MzMywdevW407ROBnL1VIzlzNJ9y5E+RKfPnP36RaLbpYy+kzhzFluQ4q5e++xMXdutxun00lXVxd+v3/evZ67dsw9t9h6M3eWrCKyS0MR1VXI3HRvU1MT27Zti/+Z1+ulsbGR9PR0ampqzuhBXi7zhxjhcJiWlpb40PMLdV9vtbFYdLNw9FlycnJ84VUqixPH2XJ5ijF3Ak9szF3MOjNmChPzpw4EAsetx1iKyM4dc3chi6wiqquMhb2ncxkbG6O7u3ue1eCZsJzp30gkQn19PYWFheTm5i7LZxyPc7Xqc6VYbPTZ7OzsvMriWCGMUll8Zqx076wgCMedwDM+Po4sy7hcrvj9PtEEnoUiO3dgOxA3orjQZskq345VxImsBmNzHRdaDZ4JyyGqMeMJn8/Htm3bzijde7pcKF/e5WKuZ3GsstjpdM6rLI5FscnJyUoG4hSIRCKrqkhs7r2ORCIkJyej0Wji+++nMoFnYeHTXJGNzZK9EAa2K6K6CjhR72ks4svLyyMvLy+hD2KiC39i6V5BEDCZTCd0R1ouztcv6kqiUqnmednGKovnOgAFAgFmZmaOa06wEqzGorbVnEWRJAm1Wo3VasVqtcZfqGZmZpY8gSfGYiIbDoffMLD9fBRZRVRXmBNZDQ4NDeHz+bjkkkuOazV4JiQyUo1V9xYUFJCXl8fevXtX9QKicPosrCz2+XwcPnx4njlBrH3neC0dFypne0/1VIhEIm/IOsR8ieeOuXM6nTidTvr7+5FleV4/9JmI7C9+8Qvuvffec/55UUR1BVn4UC1mNWgymZZFUCEqqpFI5IyPMzQ09Aaf4Zhgr9YF5Fxmtb2s6HQ6tFotVVVV8yqLF7Z0nM3K4tV2jWKs5u/EUs5tYdZi4QSepZqOLCayf/vb37j33nsT9wutEIqorgAn6j2NWQ2WlpaSlZXF3r17l+2LKIpiXNRPh1i6F3iDz7DSU3rhMFfAVktlsSKqp85ikerJWDiBZ6HpiEqlmieyJ/rdz5f1QhHVs8yJrAZ7e3uZnJycZzW4nBHfmbTUuN1uGhoa4unehaykqK7WBfV85UTXe7HK4tge3XJWFq/WZ2A1i2oizm3h1sBiE3hi9zs5OfkNn5eIe1ZUVERSUhJNTU1HgLAsyzVnfNBTQBHVs8RCq8G5D9Ncq8Ha2tp5f7acbS+na/4wPDxMf3//ioyVWwqrcTFViCKK4hsKYY5XWXyyyOZEKKJ66pxOpHoyFpvA43Q6GRsbo6OjA1EUefLJJ7nyyisTel1eeukl0tLSNiXsgKeAIqpnAVmWCYVCRCKRN6R7Jycn6ejoYM2aNfEUylyWU5xO9diRSISWlhZkWT7pWLmVilT9fj+jo6NYrVbFfu8scSYCtlhlscPhYGJigs7OTjQaTXy83alUFq+WAeULWc2iejbOTafTkZmZSWZmJgAul4usrCx+9atf0dfXx2233cbVV1/N1VdfzcaNG8/Jdi1FVJeZ4/WexqwGZ2dnqampQafTLfrzq0VU3W43jY2N5Ofnk5ube9IFa7ndmhZjenqatrY2MjIy5tnvxYzmj3eNFc6MREaFGo1mXmQzd+zZ7OwsRqMxHsmeqLJ4pU0WjseFLqoLsVgsfOxjH+OOO+7gU5/6FPfddx8vvfQSP/7xjxFFkfvvv/+UjykIAjfccAOHDx8+CPxKluVfJ/zET4AiqsvEiXpPvV4vDQ0NZGZmsnXr1hMuSCqVKiEVuouxVOEbGRmhr6/vhOnehZwNX+EYsizT09PD9PR0/HrGImW3243dbqelpYVwOKzMHF0GljPVunDsmc/nw263z6ssjkWyc31slfTvuUXsBbioqIgPfOADfOADHzjtY+3evZucnBwEQbgZ2CEIQpssy68m7mxPjLKqLAPH6z0FGB0dpaenh3Xr1mG1Wk96rJXcU41EIrS2thKJRE6a7l3s2GcjUg2FQjQ0NGA2m+NeyMFgMH4OsSKZwsLC+MxRu91OX18foijOcwY6Vxa71SoYy83cyuK5ZvEOhyPuOJacnBzPSqzGa6SI6uLERDURxKZgybI8IQjCI0AdoIjquUqsGGkxq8HW1lbC4fApWQ2uVPo3lu49XSen5Tzv2LV1uVw0NTVRVlYW36M5kZAvnDkaDAZxOBzxognFtOD0WCmRP15lsd1up7+/H5/PR2dn56ryLF7NL0QreV6xYetnisfjQZKk2P67CbgB+NYZH/gUWPmn7DzhROneuW5DS9mPnMtKiGos3bt+/XqSk5NP69jLFanGjhsbJ7dp06bTtkPUarXxoomFpgU+n2/efqzC8VktQjG3sji2r26z2RJeWXwmKJHq4iQqUh0fH+euu+6K/WM98BdZlref8YFPAUVUE8CJek8HBwcZHh4+7Vmiyy2qc4VvYTR9Jm/2y5n+bWpqQhCEhI6TW8y0IDappbm5GY/HQ2dnZ3w/9lysSlwuVmPTvizLJ60s1mq1cZE9W57Fq1VUV6r9LYbX602Ic1xJSQlHjx6N/eO6Mz7gaaCI6hmwsPd0odVgU1MTOp3ujBb/s7Wn6vF4aGhoIDc3l/z8/DNeYJbjvD0eDxNON6GIBa/KxK/+3ohOoyLVqCHPZqQ4zYRahBS9iEGjQiUKBCMSM74wnmCEqiwzJt3SHvm5I7KKioqor6/HZrPFi2RiqeTYfuxqiNRWktX2+y8mXomqLE70ea0GJEla0RdFj8eTkPTvakAR1dMk1nt65MgRysvL5z0QDoeDlpaWuNXgmbDc1b+SJDE6Okpvb+8ZpXsXkshINSLJ/PKFFv55ZJQBlwRyP7IsIcsSAiBJr+9fxyt/AQGInoKMVgSdTsunrinjnTU5p7xgCoIwL+oJBoPY7XZGRkZoa2uLL8g2mw2DwbCsIrNa0q0xVtv5wNLOaWFlsdfrnedZHBvevbCy+ExQRHVx3G73ioyJXA4UUT0N5vaeyrIcj8hirR1TU1Ns2bIlIebhy13wMzs7y/j4+BmnexeSKFE90GfnSw830OfwQSRC9IgyKkCjFgmGJEBGjoSReD0VqRJABrRqgUhEBpWITSPw3aea2dVl5zu3r8Fm0oLbDYEAJCXBKfjQarVasrKyyMrKmrcgd3V14ff7SUpKikeyy+Vvu1o4V0V1LnPHFcYqi2PDu2OVxbGRZykpKac903i1iupKz3n1eDxkZ2ev2OcnEkVUT4G56d5YMZJKpUKSJPx+P42NjVgsljdYDZ4JyyWqsXSvSqVi48aNCV8Uz/S8JUnmq4818Y9DI4hSCGSQkBGO/U+nFalIN9M44kILhGQwqAX8YYmwBNKxCNUfAgEBlUqmb9pHjkXHrtZBbht08j+1Ji7u2A+CgGwwEL7lFjgWiZ4KCxdkSZLi+7HDw8NIkoTVasVms51wcodC4jhT8Zqb/i8sLESSJFwuFw6Hg8HBwSWPPEv0eS0XKy2qsZ7j8wFFVJfI8XpPVSoV09PTjIyMUFlZGU8PJooznSSzGLF0b1VVFa2trcsSZZxJpBoMS3zmwUPsaJtCI8hk2UwMTvvIT9FxebZAemYmTzSOc3hoBl0kwLeTJ/CuWcv3j7jITtaTlazjwIArmkkQABkCoajADzn90XacGRfv3xng01mFfCQrhMrlRP3ii4Tf+tYz/t1FUcRisWCxWCguLo6Px5qenqa7uxu1Wh1PFa+mod6ny/kQqZ6MuT3N8MaRZ7HKY5vNdsKe59Uqqiud/vV4PKddxb/aUER1CZzIatDpdOJyuaitrV2WNF8iI9VIJBJPZdXW1qJWq5etcvN0bQoDoQj3/GoPzWMe1mcZ0ahEjgy5sBo0XFGeRhYuXIEIfVNeNuOkR2Xi2+4Mvnmwni9fejnf2TuBKMANa9N4rnUKUZaPpYUBQUYrCggiqJEoCdn58bCFfR4j/1kgkD05BJIEx1n0TnehXjgeKxAIYLfb4wUyJpNp3n7sucaFIKoLWXhPY9NYYj3Pscpim82G2Wyet2asRlFdDjP9U0HZU71AOFHvqcfjobGxEY1GQ0FBwbLtmyVKVGPp3pycHAoKCpZ9ETwdm0J/IMhH799H05gXZInmUTcyIAogiAJ/3T98rPpohg1qL18zO5jQhvj6pIVvhEv5jxk777ukgN/s6qd/ehJBJSIhgCwjICPJEAjLiIKMqBZplvRUajwccOi40W3gyzkF3InAci8tOp1uXoGMx+N5w7zR2H7s6e7dnW1Wm6iebUP9hdNYYpXFAwMDcWODlJSU+Fqy2lhpsU+ko9JKo4jqcThe7ym8bo6wbt06pqenl7XHKxGiOjY2Rnd3N+vXr8disSTozE7MqaZ/Z2dn+fJDB9k7FOD2DZlcVZHG/7zYzcRskB++eR1ZyXrax2b52uPNhCMy3eho8KnZbIzwrXQXXxlP4ovNYYJSPzq1QAAVhVYdl5fZeODACAa1iC8oESIauXqCYVRqNS1+DQgSEcnH1wd13P+r/bx9ay61hRayknVIMrSNzfJQW5A/9TQzPBPA5QshCgJrMs1kW3RU5yRxRZkNi+HUBVAQBMxmM2azmfz8/HmuQLG9u1jEs1KGBSdjtfapruS1Ol5lsdvt5siRI/FB7atl0MNKR6rKnup5zIl6T8PhMK2trUiSFK+WdTgcyyqqZ9JSI0kSbW1tBAKBU7JGTASnEqkODw/z21e6eLY3SEW6EbUoct9LPfROe3nThixUokAoIvGb3X1oBPjCjcU88NoI33bm8snRQS7DSa66nIagjEEt8v071tA15eFnL/eTMurmizeU8uMXezHpVZTYDBwdcSOJAlIkjCxJCKJAWFAjEqB7eJLvjTlAVB1r1RGRBRk5IiMIbpCibTwgMzxpR0BAFlWIKhXvuqiAT1xRRLL+9L9Wc12BIPrMORwOJicn6erqQqPRoFar0el0qybtulrOYy6r6ZzmFrJNTExQXV0dd+9qaWkhFAolpLL4TFAi1cShiOocFqZ7534pZ2ZmaGpqeoPV4HL2kcLpR6qxSTjZ2dmsXbv2rC8woiie9LrEHJzaJ/082B5AliO0j7vpmnQTisiIAjzRMMYTDWOkGDU43EHet8FEVZaZ77y5mh883cJPRvN52FjIsCfM7RuyeLZ5nPte6ePrt5Tjv1Tit7sH0apF7r2xjP98rpvDI25EQUAQ1IiCED1HKQxSkEgYZEFAFgREQQ0qAZWoQiCadTaoREJAWJKQJaIRriwjRMIkq9U8sLuL51sn+clb17Mu58zdYSC6d5eenk56ejoQTSv29vbidDqpr6/HbDbHI55E9VKeKqtJwGKs5nmqKpVqyZXFZ8u9a6UjVUVUz0NOZDU4MDDAyMjIolaDy1Gdu/D4pyqqK5HuXcjJ0r8x0TenZvI/hxzYDGq+dNNaTFoVn/9nCykGNd+5o4oJd4AnG8fY0TKOCBweD3BJIEyGwci9t6zlow8cYdAd5Jb1GXxgWwFrs8z8cEcn973Sx2evKSEQkvhj/RAHB1yEJTDrNHhDEjdWplGRaeKhQyOMz4bIStLgC0m4AhFkWUAUBKxGNZeXpqBWiTzdMIr3WE9saaoBGeid8lKdZUYUBRpH3FxZlkLrhJv3//EQv3znJrYWJP7a6/V6rFYrZrOZvLw8PB4Pdrt9Xi9lbD/2bBnIK+nfpbPYeZ2osjjm3rXc05QkSVrRgQPhcHhVpMETwQUvqov1nsYIBoM0NTWh1+uPazWoUqnw+/3Ldn6nIqorme5dyImqf2Peq1VVVXzhyT7GXV4+d10pZp2a3+zqxxMM8c3b1mDQqUiRNBwacFKYYmBbiY2HDg3z9ae7+crNa3i6aRy3P0yWRceO1gm2laRyUXEK77son/tfG+T+fYO4AxEEQSQkSdQVpvCvVxfxp/oRnm2dZGzGzzdvqeDXuwc5NORCEFSsz7bwySsLaR1z80zLJE80TiITrRq+vTqDzkkPbeNurl+TylXlqdy/b5ANOWauq0zl+bZp7tqUya5uBx/9yxH+9qEaStOXr01g7n5sbEpLbLRdf3//WTWQX21R4WqMnpfKYpXFc6cp6XS6+H2dW1l8JkQikRUVtdX4Yna6XNCiGrMajEQib0j32u12Wltb540VW4zVkv6NRX5ZWVkrku5dyGJ7qrIs09nZyczMDLW1tTx0eJyXOyZAlvj+c51A1PRBrxb57a4+NuVZaBt3M+sP8tWb1lGcZiJNdPPnRjdf+GczgXCEuzZlceembL78aCvff7aD791Zxc1V6fTbvWxvnUJA4O1bswmEJR49Os6f94/wgUvyKLLp+e3uQT79j2bCksBFxTZMGhUvd9r5+lMd/OtVRVxdYaNz0oNKEAhJMilGNd+8tYIH9g/zRMM4t63P4N11ufyxfpgSWxiTVuThw2MIgoxKCPCvf2/i7/+nBqP27KTVFkY8Cw3kdTpdPIo1mUwJe0ZWo4CtxnM6XeZOUwLi+7FzK4tj9/V0LTJXck/1fBJUuIBF9Xi9p7Is093djd1uX5LV4HLaCC71+OPj43R1dS158PlClmMBWpj+DQaDHD16FKvVytatW+ma9PD9Z9upzDBy/dp0PAGJB+oHSDJp2JJvpXvSwwP7h5AkibJ0U7yytjhFyzu3ZvGzXYOoBbi4OAWzTs1Xbq7gi480873tHXz26iJax9yICAgClKebWJ+TRDgi82TTBHq1yI1r00k1jzPuDqLXiNy2Pp3KzCSuXpPKfS/38bUn25GAjTnJ/MsVhfz3cy38sX6UkATvrs3F6Q3zZNMkIhKCJNM94aY41YAkRdCoRd5bm8fP94zwgx3dfPPWioRe26Wy0EA+thj39fXF97Bio+3OJEpZjYviShfeLCcGgwGDwUBOTk68sthut9PV1YXP54t7Fp/KfV3pPVVYfdmO0+X8fOpOQCw6DQQCQFS0YjfT7/ezf/9+ZFmmpqZmSY34yx2pxmwQF0OSJFpbWxkeHqauru60BPV0TRqWctzYeTscDvbv309RURHl5eWEIjL//o9G9Cr4wg3lXF2RTvPoDBLw9Vsr+diVxXzppgqM2ugEmgG7j0//7Sivdk7h8ks8cGiM7CQdNpOG/3imnWGnn8xkHffeWMHkbICvPNHBjD/MV28uI9eq50cv9jI2E+DddblcV5nGow3jfO6RVmaDET51VTEpRg3feaabo8MzVGaaeXddLhEZkGBdthmrUcM9lTq2FVv464ER/npgmP19DpAjqAWJz19XhNWgRqtW8e/XlODyRhibDXDrWhv/ODDAa32OhF/f0yG2EK9fv566ujoKCgoIBoO0tLSwf/9+Ojo6mJqaIhwOn/KxV9uCeD5FqiciVlmcn5/Phg0bqK2tJS8vj0AgQEtLC/X19bS3tzM5OXnC2o+VfAlZDYKeSC6oSPV4VoPw+j7fqVoNno1IdTHRnpvuraysPO0FJJamTfQXKnbcvr4+xsbG5kX9P36xm7bRGTblJvFk4xiTs0Fe63Vw87oM0kxakGXue7mHSCTCt++qBuC+l3v48YvdJGkEghGZb7+5EpUg8KVHW/j20+18/8515Fj0mA1aHJ4AFxfbWJ+TzBeuL+XLj7Xzgx3dfPf2Ndy5IZOdXdO4AxHeujWby8tsbMhN4jvbu/jBjm4+dEkBf3ptiMIUA1nJOv5yYJRkg5ocQeBjl+UzYPfzyNFxCi1a3rY1nx8+38OA3cc9NTn8atcg/nCEbSUpPH50nB/eXUV9v4tvP9PBPz9Si0a1et5hBUEgKSmJpKQkCgsLiUQi8f3Yvr6+eCo5ZqV4oudjNQrYajyns4EoivNGFkYikXjf88DAwHEri1dS2M4ni0K4gEQ1HA4v2nsqSRLt7e14vd7TshpciT3VM033nuz4iUCSJCYmJkhLS6Ouri6+KNf3Ofjdrl6MaoHeKS8Nwy7CkgwyPNM8wbOtE2SYtYzNBPjAxflkW6JtIt9601ru/WczXVOzlNn0ZCbp0agEvnJzBV97opXvbm9HlsEbCFNbYGVnl4OLCq3UFln5t2uL+fYznfz4pT4mZgOoRJGyDB2PHB2nMtNMdW4yX7+5nG8+1cEvdw5g0once0MpKUYN//lcN7/dPcgH1utJ9Udw+cMgy+RZddQWWtmSn8xTTRP89G3reLxhnH8cHuOz1xRT3+fkicZx3luXy3+9PMTfD43yrtrchFzb5RCM2HxYm80GvF4cMzIywuzsLHq9Pr5vt3DW6GoUsNXaUnO2mVs5DMevLF7OYsuT4Xa7zytRXT2vzstELN07t7o39mXzeDzU19djMBjYsmXLaVkNLnekOndhmJvura2tPWNBheU5f7fbTXt7OwaDgaqqqrigzvrDfOHhJrLMGn7z7k38/r2bqS1MQasS+NotFXz5pnKuX5POmMuPLEk8fHiYZ5rHkWTY3++kZ8pNmVVDz7Sf//d8F2FJpjzDzCevLKFr0kv3pJt/u66Uz15TTGmagV/s7Gd8JkBVdhLvqMnlyNAMY7NB7r2hjK/dXE6ORcd/v9DLqNOPWaeKCvixplRBEFCLIv96TTE2k5a/t/v5+c5+AqEIV1WksrfXxbDLz50bs5gNSuzpdnDHxkx6p31MzAa5piKVlzqmKUzRsy5Dyy9e6WHWf+pp1ZUiVhyzdu1aamtrKSsrQxAEenp6qK+vp6WlhbGxMYLB4Eqf6qKsxpaa1SD0scri8vJyamtrWb9+PUajEZ/PR3NzM0ePHmVgYIDZ2dmztld+PvWownkuqrHUx2JmDsPDwxw9epS1a9dSVFR02g/7ckeqMXw+H/v370ev17N58+aEeQ0nWlRHRkZoaGigtLR03uB2gK8+3sKY08Mnry5Br1XxatcU+3odvKM2ny2FKWzKt9I95cGsU3PvjRXkpxj5za5+PvdwI//zYjdl6UY+XpfKO7dkUN/n4Gev9CDJMm3j7uhEGqLCrVYJfOaaYhDgJy/1EghLtIzNcswGGEmWMWjU3Ht9KSqVwA9f6OHZ1kn297u4uSoNSYIfv9hLWJIwa9V88soiXAFoHvPw3roc3rUpA41a4Kn9/axJ1VOcamR76ySXldiwGlQ81TTBHRujlZpPNo2zITcZu8vNH18bSth1PpsIgoDRaCQ3N5fq6mrq6urIy8vD7/fT1NREV1cX09PTTE9Pn5XvwlJYjdHzajyn2MuT0Whky5YtVFRUoFarGRgYoL6+nqamJoaHh/F6vcsmskr69xwg1nvq9Xo5evQoF110UfzPwuEwLS0tAAkZzH02RDUUCnHo0CGqqqriaZxEkahh4rEe2WAwSF1dXdyUIMaTjWM83TgKssxXH2vBpFUz6w9i1KoZcfrZ3jRO75SXrgkPn7u+lItLbFxUnMLzbZP84pUeZEni9g2FaFQerimzIKu0/HX/EE5viKODLt5UnUm/3ctvdvVRZKukwKrnY5cV8v9e6OUbT7bTa/fx/ovyeK5tkp++1MsP7lpLulnHZ64u5j+e6eT+fUOsy07ivRflsSbTzE9e7uPhI2O8fUsONqMGUZCJRCTWpWqx7XqJK9DxykCE9zz/EjeUb+DX+4bptXu5vjKdh46MEZFkthRY2N46hYyAUSPyv3v6eXdd7ml5BK8m5s4aLSoqYmRkJL4fG0spxlLJKzXabjVW/67Gc4oR21PVaDTzKotjwx66urrw+/3xyuKUlJSE9bUq6d9Vztx0r1qtnheFuVwu6uvrSU1NZcOGDQlxEFnO9G9MqEKhELW1tQkXVEjM+ceiaKPRyMaNG1Gr1fOOO+Tw8Y0nWimx6fnIpYXcvSkbjSr62RnJOl7rc/DrXf082zqBQSMw5Q7iC0ZTpU3DMyDLpJm0/PTFbhrG/ciyzFu25LAxz8LhQRel6Qbed0kBn7m2FKNWzU9e7CUQjlBXZKUiw0jPtJdtRSncvC6Dz1xdwmwgwi939gMyazJMWAxqIhJsLbCgEgW2laRwRWkKjx4Zo3fKw292D6BGRqcWeXh3N4LTyU2ZIiFUvDIe5rLwJHqNyEvt01xbmYYow89f7efwoAtJltmcl8Q3bynH4/Xyx9eGz/SWrTpiJhRzU4p6vZ7h4WH2799PY2MjQ0NDyxrtLGQ1RoWrWVQXu16x+xqrLK6pqSEnJwe/3z+vYvxklcUn43xL/55XkerC3tNYu4gsy/T39zM6OsrGjRsT+la0XF9cn89HQ0MDGRkZGAyGVTtabmpqivb29jdE0bEIOBSR+Pd/NBIJh/j8DevJTNbx2JFR7N4Qn7iihOuq0rG7g/zbPxpRAelJWu7fO8AjR0apzk1mV+ck76zN46Z1GXzn6Xb+cMiBTqujVNbTNjaLXg0Ddi+DDh+FNgOfvrqEbz3dzl8OjLApz0LXpBe9WkX7hBtfMEKhzcC7anP4333DvNg+zciMH6cvTIFVx98PjVJXZCXdrOV9F+dxZHiGH73Yy/hskNtKdKBP4ummCd5uTaZIE6FcF+b5YAo3+zxcUpzF3l4Hl5Xa0GpEWsfdXFxkxekL0T/tIy9FT12+kQdeG+C9F5370eqJ0Gq1ZGVlkZWVNW9Cy9xoJ1b0tFzPtSKqp8ZSrpcoilgsFiwWS7yyOOZZPLeyODZRaanVxOfThBo4TyLVE/WeSpLEoUOH8Pl8XHTRRedEmmFiYoJDhw5RUVFBcXHxsi4OpyuqsizT1dVFb28vNTU1UUENBuHYG2vsheZHz3dzqN9OfoqB7c0T/OrVXv702gDl6WYyk7VMzgT40Y4uAqEw33hTJd+/ez3fu7MKs07Fq52TWPVqblibjlmn5mu3VlJo1fKrfeN884k2zDqR796xDpNWww+f68QXirAp38It6zLY3jLFj17soTBFzxeuL8HhDfGn+miUeHNVOutzzPxh3xBPNE5wfWUa995YhixL/HpXP7IsY9apecvmbMZmAmSY1FyUo+HGShuiKPDkjB5kiWuMPgZlPV2GVC4vs+ELSXz7mU50ooBANOK9c0MWdm+I+n4Xd2/KYtbt4S8HRhJ6D1eaEy3IsT7KvLy8edGO1+ulqamJAwcO0NXVhd1uT+g2ymooClrIahbV0yGW5i8tLWXr1q1s2rQJi8XC1NQUhw4d4tChQ/T29uJyuU64xpxve6rn/B2O9Z4uVow0PT2N1+slPz+ftWvXrvoHOtbeMzg4+IZ073KlzU5nTzUYDHLo0CEkSaKmpgadSoX44IOo7r0X1Re+gPDkkwjA3iE/v9vVQ6ZZg9Mb5OnGEbY3jxOWZDonPXzjyTY++pcjNI26KM8w4/aHkeVomtXuDpJm0DAbCHHvIy0MOXwYtSret9kGArh8AT60rZDiNCOfva6UsRk/vz0miLWZIAoywUiET11VxPqcZG5bn8kL7dM0jcwiCALvqcvDHwqjVcG763JIN+t4R00uDcNu9vQ4ARi0R9sMglKEiAQWvZrLy1J5CRueSQeXBcbQivDCrJbn26aQZZkkvZofvaWKVKOalzqm2ZyfTGaSlu3NExTZjGzJMfDn1wbwBE6/Eni1RWGncj6xaKe4uJgtW7awadMmrFZrfCE+fPgw/f39Z1x9ulqrf1fbOcVIxPMUm6hUUVER3wYwGAyMjIxw4MCBeGWx2+2ed2/dbndCItXt27ezZs0aysrK+P73v3/Gxztdzun0b8wIf6HVoCRJdHd343A4MBqNcZu21Uws3Zuens6WLVvmPeSxaHI5mrNPNVJ1uVw0NTVRXl4ev67Cyy8j7NkDhYUgSYjbt9OlsfLzg14q0g18601rEQT41lNttI97+NJNFaSZtTzTNMYzzRPkWw10T3n46uOt5KfocfpCGLQi37+zimlPkO9v7+DLj7fwxRvK+d/DdgQ5QrJOzf17B6jKTmJ9TjJv3pzDQ4dGsDHDa2MSKgEkSeAvezp49+Z0blmTzP4BJ7/ePcAP76zkkaNjqEWRYBgODsxwaYmNG9am80rnNH+qH8JmUrOjbYq6Qgv7+50cGg9RUgQ3r8vgpU47z1dfyW3r0tn6Sj8vd9qJRCSqs800j84SjkhcUZ7KYw1jOL1hblibxp/3j9Dv8HHXpiy+9kwvDx0e5f0X5yf8fq4EZyLyC83jA4FAfED77OwsJpMpbkKxFIezRJzTcrGaRXU5mLsNAG+0yWxqamJmZgan00leXt4ZfVYkEuFf/uVf2LFjB3l5edTW1vKlL32pSpbllkT8LqfCOXmHY+neWI/c3HSvz+fjwIEDCIJAbW3tstnwJZLJycl4urekpOQNi8FyFkMt9dixEXitra1s3rx53ouK0NEBKSkgiqBWM2ZO5SN7Z4mEo3uY/zw8wpceaaFpZJZLS23M+sPs7bGzvWmCiwqt/OTt1fz+PVt4/8X5DDv9uLwhClL06NQiazLNfO/OKgxqga893kqfPcAHa9L56i1rmPYE+PGLXcgy3FSeTI4RHukMMDIT4rPXFHPb+gwOjEXon5VxTk9yXXaE8dkAP9jewWt9Tt5ek01pmoE/vjaENxhGFAQ+cEk+dm+YH7/Yh0Wv4uOXF7Amw8Qrg0HCkkxhqpGKDBPPt9uRBAG3P0JIkrhmjY0PbstHRmBPr5OrylORZYGd3XauKk9FLQrsaJ2M+hBn6PjDnn78odXRfpIIEiVgOp2O7OxsqqqqqKuro6ioCEmS6OjoiFvuTUxMnLQwRhHVpXO21seFNpkXX3wxPp+PF154ga985St86EMf4i9/+QtjY2OnfOz6+nrKysooKSlBq9Vyzz33ANyR8F9iCay+O3wSYsVIi6V7x8fHOXToEGVlZfFm9eU2Z4DFJ7IshVi6t7+//4TVvWf6O0iSTOf4LPt67OzunubJxjHaRmfwBMJLOnY4HKaxsTE+XWZh/ykZGeB2AzAlqXmvaiOzEQmzWuLl9gn+dmCIzolZQObl9in++/lOHqgfQJIl7J7ovNSJ2QAvtk+iRub6yjSaRmb493800jXhIcWkJd2sJxKR0Kll8iwaKjLNfHBbEUcGZ/jTrg4GB/vJTUsmEpEpsBnYkm/hrVuyyTBr+Fujk/zCIm7dVs2mHDNNEz5yzVCmmua2Ui0ub5h/HBoFoCLDTEmagWlPiLs3ZWLSqblrUzaugMyeXhcA11WmMuoK8Ls9gzSMzKBXCXiDEXKteopTDezqtpNl0VGeYeSVTjtmnYptxVZ2djvwhSLctj6TKaebh4+c+uKxGlkuAZtbfbpx40ZqamrIzMzE7XbT0NDAgQMH4hmphc/wahSw1XhOEI3yzvZ5CYLAunXr+PKXv8yll17Kr3/9az7ykY/Q19fHu9/9bvbs2XNKxxseHiY///XMz7HINzEWZqfIOZP+nTv3FOZbDUYiEdrb2/H7/W+wGoz1kS6nr2VMmE7lwfT7/TQ0NJCWlsbWrVtPuCidTi+sLMu81D7Fz1/poXN8Fm8gBMhR9wMABARRRBQFCm1GNhdOU1eUwnWV6STPqUz1eDw0NDSQn59/3BSNdMMNiF1ddA/ZeZ+hlmlEPnZJPoJ3ksMzRl7tsnP92gzef3E+O7um+e3uAbKTdNQVpdA44uL+vQP8fnc/apFoj2pxKjdWZfLDHR185bFmsi16Buxe3lWXy+NHRrlvzwT/lZvF9ZWpvNYxwuMtTiZLbOzvt7Mpz8LR4Rn29Tm5pCSFD24r4PvPdfNk0wQ3r0tnwh29DskmM2srS3G73dQN+dneOkmp3k26NYkRpx8RiX67D4BNeUlkmVQ80zrFdVWZXFycwu/2DPJ82zSXFFlI0ql5uWMaXzDCZaUp/Ll+mDFXgCvLbfxmzxA90z6uX5vOK90OdnZN0z3lQ46E+dWrfbxlczY69epbaE+FsxXpiKKI1WqNO4mFw+F5o+20Wm28qlgpVFo6y7W1tFQ8Hg8Wi4W6ujouuugivvzlL5/yMY7zDK5IivKcENUTzT11u900NjaSk5Oz6BzRs2HOEPuMpfa9Tk5O0tHRwdq1a+NeqyfiVCPV/mkvX32shdd6ppAjYUCK6ikyyNH/L8kgIBOWZbp9XrpG7TxcLyKo1Oi1GrYWprAlS4c55OBN26pJs1kX/SxZlmn3CPxt7Zt4yDdEMBRCluGnO6NFQ7LgRRQEXmib4MXWSSKyjNWo4saqDC4rS+WK8lS+9VQ7bn+QiCTz81f7CIbh8jIbX7qpgi/+s5m+KTc3r8vkLVtyydFH+PGuMX74XAdvLorwnq3pfO35UXZ1TXPL+gw+cEkBX3ykmfv3DrIhN5nN+cnUFVp4+PAIvVNeRl0BrqtM4/k2O0eG3WwtsPDBK4w0/KOFXRNqVBNuguEI69NEXuqwc2W+jpJsG5fn63iozUfD8AxFqcaoM5MU4X0X5THpDvBcW9SR6ZKSFP5cP8LuHjs3rk3n/n1D7Oyy876Lcim2GXj4yDgOX5gry1PZ2efmoUOjvLtuRV6oE8pKCFisMCY9PR2IvqjGjONnZ2dpa2uLm1Do9fqzfn4LWY0paVj5KTGJ6FPNy8tjcHAw/s9DQ0MAK1Jmv+pF9URzT4eHhxkYGGD9+vUkJycv+vNnQ1SXKnqSJNHV1cXMzEy0anaJjiSnIqovtU/y8b8cJhI6JqYCSJGogJo0Ar6gjCiAzaxhfCaARoCwHAEpQiQiQCiA2w87m9282iogimq+v28fBp2GnBQD+SkGZv0RMpK0dEx4mHD5cHn9IEVI1omUZ5mwe4KMzgQQBIENucnkpRjY3+9kYjZAVrKWiCRz/95+7t/bDzIk61X88O51IAj88pVefvxCF692Wumb9iAgU2gz8FzrBBtyLZSnGbir0sxDrW6KUlKJuEL4ghEEIfp7qVUiH7mskC8/3saDB0f40LZ83ntxHgcfikavb9mcxV0bs2gZdfOn+mE25iVjNWq4c2Mmfz04Agi8eVMW161J49MPNfNUm4Nb/S4K9V6SNCKPHRlGo9Eci4REDg+6uGZNKhlmLbu67VxRbmNNppE9PQ7evCmLrfkWdnfbeVdtLuuzzTzRNElFppmPX16E3dvJr3f1ctfGaJr5XGW1iIVerycnJ4ecnBzq6+spKCjA6XTGnb4sFks8kk2E8cupshJp1qWw0hF0IkS1traWzs5Oent7yc3N5cEHHwR4PCEneIqsvjt8DFmWCYfDi/aehsNhGhoacDgc1NXVHVdQ4exGqifC7/dz4MABVCoVW7duPSWLr6WK6nMtE3zir0ew6VVcU5lGtsUAsozVoEJAJiNJy7duX0uGxYDTF+ae2jwEQSTTrCZFr0EjyGjFqEGuJIUhHEYK+Qj5Pcy6nLQPjPL8kV7qOwZ58kA3HQNjOGdcEAqgFSTc/iANwy6GnH6kYx67R4dcPHF0jPGZANnJOq5fm8EXbyzn4hIbyDKiIDHjD3P/3gFEAb59x1q2FFg40O/A6QnwxRvL+d6d6yjPMPGjFzo52DdJpTnAlWU2nmia5unmce7YkMlNVRk83TRB69gsJWlGblibxo7WKXqmvEzMBI+ZgEisy05CrRJ470V5jLr87GidAuDaNamIgFqAOzdkYTNpua4yjQMjAWw5xaRYkrms2EzzuJ8jQzPcUKgm06zmpfZJALaV2GgamWXGF+ay0lSGXAH67T4uL0vFFYjwcucUL3XaEUQBm0mLKMI9W3OYdnn4w75T8wReLSIWY7WdD7w+2q6goIBNmzaxdetWMjIymJmZ4ciRIxw8eJCenh6cTuey11zEWI1tPrDyYp8IUVWr1dx3333ceOONrF27lre97W3IstycoFM8JVbfHeb13tPYZJm5X9iY1WB6ejrV1dUnfeNcDaI6OTnJwYMHKSsro7S09JQXoKWI6sF+J5/5+1GsOpE8q469PXZGnV5uWZfBH967hS/fUsmYO8yvdvbx1ZsqSDPpeLpxnHdtSWPSGyYvRU+qSYOIRJ5FhwhY9CKCDGpBBmRkSUaQZWRJQkaK/l2KGtn7QtFeToHYQxX776AwzcD1a9LQqwX+/Nog//6PJvZ0TXFxiZX737eFD2wroGfKw7891Min/tbAoX4nZWkG1CL84tU+Zvxh7r2hjBSdyJ8aPMyokhly+JFlCb1K4M6N2bznonzSTBp+8UovoYjE2zZnk6xX84tX+/h/L/SQlazFZtTwx9eGkGXYlJdMdY6ZfxwewROM8ETjJJIsEJbg6PAMALdXZyIg81jDGCpRZHNBChKQZlRxz7ZyLs430jnlZ+eBJgp0HiII7Ome4qIiKypgT6+TTXlJmLUif6wfRiWKXFWWyoEBF3ZviPIME9sKzPxhdx/DzpUbvXWmrEZRhfkp6diIs9LSUmpqatiwYQNms5mxsbF4D+Xg4OAbeigTyUpHhMdjNeypJqJP9ZZbbqGjo4Pu7m6+8pWvJODMTo9Vd4clSSIQCLxhkLgsy/T29tLa2sqmTZvIyclZ0vFWMv0bawXo6+ujpqZmSfunp3L8GA5vkI8+cIhQOMKUJ0TnlB9fSMak1/B08wQ/eamHTXkWvnhjBSMzIX6/Z4DNBVY8IYk/HRhHkAUaRmaYdIcISgKjMwFMWpEZf4T0JC0RGTLMWkTAqBURiT44OnX03qhFMB4rtpEiMpIsISJj1kQFecDuZf+AA/cxw4MUgwabUc2+Hgdff6KN7GQdt66P9rKNOr1kJGn40k0VfOv2KjzBMF99vJmGpjb+9fJcJAR+um+KEaePD27LJxyR+NXOPvQaFR+5opghp4/HGsYw6lTcviGDfrsPCYkv3VDGu+py6Zn2srPLjiDAu2vzcAck/rhviCcbx7mq3EaORcvfDo4gSTI2k5Yry1N5uXMap1/iwcPjaJDxBCVQabhpYwECAmNiGpuKs8kyibzQMsZwbyflqRp2dk7h9odBBn9I5tNXF3H3piwkSY5HyO+qy0EO+fmPZzpWfevX+YRGoyEjI4PKykrq6uqoqKhAFEX6+vrio+1GR0fjmbJEsFpFdaX3VCORCBrN+WPbuWru8ImsBmMOPoFAgLq6uje2dJyAlYpUY+leURRPaf90MU4mqt9+ugNvIMw7avP4/A3luP0h7tyYyf3v3cI9dQW82m3nf/cNkJdiOLa/6eCphhFS9AIRGQqTJLKM0X7eqiwTMlHPXmSZiZlA9O/uADLgCUaQJBlJkgmEo+cUisj4QtEismyLlhSj5th/A5cU28i36HB4gozPBBCR2ZSfzDdvW8unrirB4QnxH0+38+CBQaqyTHxwWwEuX4gv/LMZURD4tyty8PjD/Kld4uHWGQLhCJIkUZVt5rbqLN5Rl8++vmhV7dYCK5eWpPLIkTHaxtzsaJtCQEAtCyTp1VxaYqM01chfDw4TDEsUpRnYVmLl5c5pzHoV770oj7dtyWbQGWBPrwOA2zdkEpFk/tbqpWfKx50bM/GGJPb2OEg1aajKMbOzexqT2cQVazIYcMuk5hZzSZGFaW+ELz7SjD8cfUEcmHKTYRCpyTKwo2WCQDBMqknL27dksrM9OuD8XGS1RqqngsFgIDc3N95DmZeXRzAYnGccPzU1RTh8+k5Yq1VUV+t5nausiit5ot7T6elp9u/fT0FBAZWVlad881ciUp2amuLgwYOUlpbG+2XPBJVKdVxR3dM9zRMNI7x5czZv3pzDX+sHybFoeUdNLiqVyNu25nJrdQ5PNk7w6QePMmJ3k6wTSDNr+MGbyihMgjEPfP1N67Aa1fhCET68rZBAWOaemlzSzBqsehWXFEV7aNdlm7EYVahEgdrCFIwa0Krhtg2ZFNn0jM8GSDGqubLchoTEq51TDDj8VGYn8Y3bKrl5XSZ7e+x85qFG7t87gMsXwmpQoxYFuiajs1S/e+c6VILAVx9robFvgupcC1NuP4f7HdxRlcKtFWYODrrY0TrJ7RuyWJNh5ne7+nF6g3xgWwGiKPCd7V24vCE+emk+s8EIjx4dQxQE3lOXi90T4pnm6F6oXiMgy1CcasSkU3FxcQoFKXr+cTg6vi0rWcf6nCR6HGG25pp465Zsci16drRFI80ry1KZmA3RPu7hshIbAgL1Ay6qC9NBlpkJwr9fXUhOkpoXW8cZfeABrho6xGwwwovbD0AkQrJejRwJ893tHTSPzp7Rs7ISnA+iOpfYaLvCwkI2b97Mli1bSEtLw+l0cuTIkSV72i5ktYrXSkaqsezM+fT8rOgdnluMtDDdG0ud9vT0sHXr1njZ/KlyNiNVWZbjFWg1NTWkpqYm5PiiKC76O0Qkme8+0066UYNWJfKFfzYz6PAiIrO9eSIabQIXFVkRBAFfIMAP767iX68pZdId4h/7unjfRTn4IzLbm8f54LZCeqZ8IENNQbQN5UOXFuINSXiCYW5am0bzyCxvqs6mONXIoUEX76wtoCLdzFONE9gMGtZlJ9E76eXlDjsIIndszObKilS6xt18f3sHs/5QvCXF5QsiCjJXVaTxn3etoyTNyP+83Mvf9g+yOSNa6PRwu49DA062FlgQkBlyBbim2MTG3CR+v7efUZeff7mqGF8ozG939zM2E0CQZQKRCHdvzOKayjQuK03hqaYJJt0BqrKT2JyXzKNHx9jX6+TFdjuFNj0NQzOMuvwIgsDbtmQzOhNgZ5edUERiYjYISOSl6BAEgesr0+ia8tJv93FRkRWdWuSVTjtZyTpK04282D7N95/tQqUS0WtUbCxI4bq1GQy5ZWSjjerCNEpVPh6bkGl94SX+Uj9IjkWHRSPxiQcb6Jv2JuS5UUgMMeP4srIyampqWL9+PUajMe5p29DQwNDQEB6P54Qp/NUqqqv1vM5VVuxKxgQ1Voy00Gpw//79qFQqampqzqjH7GxFqoFAIG6PeKbp3sWOv9gb8RMNo7SPzzLhDvLH+iH6p72YNCr0Wg1/2DfI97Z3MOUO8KPnu7BoZSIy9Ex6SApMUWxVsXcc8mxmanMNPNM8QUWGiQ25Sfz14BDvuTgfrUrkyYYxPnRpAQ3Ds6SaddQVp/CX/YPcvTmbykwTv9vTh0mnJlmn4sCgk+aRWTblW7miPJVwROa51gkKU4x849YKUo0aXu2epm3cw7psM/9513qurEjjicYxvvV0K3kpBnKStbzW5+TZbj/JOhWpRg0gcev6LN53SQEHhjy82OPhU1eXoleJ/PcL3WQm6Xnrllz2dNv5+uMtWI0acpJ0PNs2SSAs8c6aHEQE/ro/2rb2ztpcfCGJn73SR75Vx+evK0GjFnjwYPTPawosFNv0PHxkjL8dHGVsxk+WScWr3a6or2+ZDa1KZEfbFHqNyEXFVvb1OQiEJdZlmxl2BfAEJd5Zk40nFKFheJbLS22oBZkXZRtGg553Zco4BB1Pz6Zi9wvcVKzjLSUyfq+X9/zhIAd6JhP2/Cw351ukejK0Wi2ZmZmsXbuW2tpaysrKAOjp6WH//v20trYyNjYWt1GNsVrFayUj1djc6/OJFb3Di5k5jI2NxX1wT6dSdiFnQ1R9Ph+9vb2UlJQkJN27kMVENSLJ/PyVXlKMGj50aQGfvrqEiCzz0SuK+cGb1/PxK4o5MjTDt59qx+Pz841b15Bv1fHXfb0kmc28/9IyHL4we3pnuK7IgCDAPw6P8oFthXiDEZ5vm+Q9F+fTPOpGJQhcUpLC3w8Oc11lGukmLT/a0cXUhB1Zhtd6HRi0Km6rziLZoKZ1bIbq3GQ+vlFLVbaZP702yNceb2PSE+KGNRlcXpZC69gs33iyFZUoclFhCjO+CNubJxhz+ajONmLSivhDET5xZRH5KUZ+8FwnazLNXFKYxDOdHgbtPj5+ZTG9U15+vauPgwMuJFlCJcLXby7nI5cXMuUO8uiRcVLNWm5dn8HuXgedkx4yk7SYdCL+cIT3XZJPRpKO29Znsq/XSc+kB0EQePuWHCZmAzzROM61FancWm7E7ouwp8eJWa/mkmIru7rs+EMSV5Sl4glJPNMywQvt0wBcUmzlpqoMknQqXu2yYzFoqElR8bJHRzAis0EXZL3gpt4J5Zlmbqwp58qa9Xzl+mLkcIAP/6WZb/19Dy2tbUxOTs7by1ttBU0XmqjORRAEjEYjeXl5VFdXU1tbS25uLn6/n6amJvbv309nZyfT09OEw+FVKaorKfbn29g3WEFRjYnpXKvB5uZmRkdHqaurO64P7qmynKIaS/dOT0+Tl5eXsHTvQhYT1efbJuib9vChSwu5rTqb+l4HVr2aS0qiFcbXV2VQmWWmf9rN7dWZJAkBtqZJTPphJKSnOs9CeYaJZ9qmSdLBdZUZvNwxiVGr4uqKdJ5pGqc4zUiORcdvdvcx4wsRjkh8d3sn4+4goYiEMyTwLm831b5xxt1BMpN0/Oit1VRkmPn5K7082RNmfCaAIIBJryYiybRPuNlWksr37qgizaxjR+sEe3sdpBtF6rK16DUq2id83Lw+C4tByw92dHFbdSYpRg3ffaaDW6tSyTCJ/PcLXUiSTLpZy47WcYYcXu7ZmksoLPN44wRVWWYuLUnhicZxxmcC3LExE6tezf/uG+THL/Xi8UfQiAI7O6Mi+KbqDJL1ah7YP4wsy1RkmlCJUdOMe7bmsCZVS55Fx+NN48iyzHWVaXjDEru67dF9Zp2aBw+MYlALVGYaOTo8g0oU2FZs40C/E3cwwnVbC5mVVewd9SHY7eSkJyEL0RSxcOx/JZkWPnBJAaFQiL+3B/jcC04eOzrG/oOHOXjwIL29vfh8vmV5zk6X1SbyK0lsP7aoqIgtW7awZcsWbDYbdrud6elp2tvb6evrY2ZmZtVct5WMVBM19m01saKvTTFBdbvd1NfXk5SUxKZNmxJaXr1cohpL9wKUlpYu65veYqL6hz39ZCXpuLjYhssf4kC/gysr0lCrXo8YwpGoLaHL5WR8fJy3Xb6eFKOWJxvGEIA7N2Yz7g7TMhHirk05yJLAPw4Nk2rWEpZk7v1nMyOuAP4QDDr8VOcmAwJX5Bj41sQeIoKKdr2Nrws9XOwZ5vd7Bni2eYIv31ROdU4Sva4IwzMBbq3O4o/v28pnryslJEn88LkuvvRYCyNOPxcVWVmXpmbKJ9MwGaGu2Ea2Rc/Dh0fxhyMY1Cp+9nIvuVYDgbDE91/ox6QVcflC/OdznXh8QYwqsBjU3L05m1vWZ/BsyzidEx7eXZeLShS4f+8geo3Imzdn0Tnp5dDgDB+8JJ9b12XwapeDfrsPg0bF3ZuyaBp1c2Rwht/tHUSWQUZgd68DQRC4aY2VQbufo8OzlGeYKLDp2dE6Rce4G28wQkSS+Ow1xVy/Np1Jd4jWMTdXldsIRmT2dttZn2chz6rjCUMBY9fezMvTAoWpRhqHZ3mlazp+315om8JqUvPZqwpAlvjx3ik++2qAHVPJzEY0zMzM0NvbS2NjI8PDw6tCZC/USPVkqFQqUlNTKS8vJyUlhbKyMvR6PUNDQ9TX18fvodfrXTGRXclI1ev1KpFqIpFlmcHBQRoaGli/fj0FBQUJ/3Kq1eqEi+r09DQHDhyguLiY8vLyZfmMuSys/m0dneXQgIub12WgEgT2dtsJSzJXVrweKQ9M++iacJNmENk35Ke0vAKDTssNVRkcGZpl1OWnriiFdJOWnQNezHo1eTY9z7VM8NDBESz66IvNZ68t4YaqdDzBCB/cVsR1a9LYNerDIIX5YKiXA2IKD6ry+cJEPdeWWvn7wWE+/OcjNI26qUrTUJ5u5qnGcb76RAuZx14CZEFAVKlAEGgedlGZY+Xz15djNWp4udNO/7QPZJkpdxCXP4gkyxzodxAMR5j1hxlyhahI1SMAd8128gXnEUZdfh6sH+KdtfnYjFp+tXuAZL2at2zJ5NDQLLu7HLzW60SWZUxakavWpHLHxkyMavjzS+0Ig4NcX2EjK1nDL3cPsLvbwVu3ZLMuy8Q/j4wRiMjUFSZhM2p4/OgYgiBww5p0+uw+vrO9G6tRhSgKNI3OUltgxaARebVzmpI0IwU2PS90TCEIArdUZdBr93PfniEQVXzu2hKqsoz8ZtcgXZMeBp0+jo66uemYcf8P7lzD+2qzmfX6+Uv9IO9/qJvDM0aKi4spKSmZNxYtEW0fp8NqS/+ulghwIbIso9frycrKio+2i93Drq6u+H7s+Pj4G/Zjl5OVjFSV9G+CaWxsxOVyUVdXR1JS0rJ8xvEqZ08HWZbp6uqKVyTHBisv93i5hb/DXw8MoVVBSbqJ1/rsPNk4SppRTZrp9ek825uGEWWJt27OYjYIe3vtAFy/NgOVSuDZ1knUosANa9Poc8l8+sGj9E/7EEWR69em8+O3VaPXqDjQ7+TddXkYNSK/39PP+y/OJ9Wo4Se5F3Pd0FFu9vTxT20ez6y7iso8KxpRwB2KtqLcXKLhu3es5eNXFjPi9HPvI808cmSMy8pS+dHNebynSkdpRhKPNNn53nPtTLmDXFpiY0OeBQSBVLOGNJMWlQCZSVpUgsx15SkEwjK1YTvbghP8zVCMxZbE9f5hHmscY8Dh5UOXFjDo8PFowzg3V2WQbtbwP6/20Do2y+3rM/AEIjzdNE5yXzdv8/XS4JRoeGonul2vckNlOg5PkOxkLXduyOIdtbnM+CPsGvCjEgRuWZ9B87iHnkkvaUlaJGRUInz7tjWsyTDxcsc0OrXIJcVW9va78Iclrq1Io2faT8+UhyvLbZi1Ii1jXu7alEVWso7PXFOM1aji+zu6+PvBEXRqFddXRp8tQRDot/vQa0T+6641rM0w8JPdE7zQHR3gPXcsWnp6Oi6XK972cbbSjKtRVFfT+cRYGBEKghC/hxs2bKCmpoacnBy8Xi9NTU0cOHCArq4u7Hb7sr60r2SkqqR/E0xZWRnr169f1uqvRKV/Y+leWZbZunXrvIrk5S6Gmiva3mCERw4NEwhLfPXxFr7/bCeDDh9TniD/58+HebJhjOHhEfb02Nmcl8R16/PJturZ3jwOgM2koa7Ixovtk4QiMqkmDZIA7lCE/7h9LbWFVvb2OtBrRG5al8GubjtOX4i31eTSMDxD09gsn7yqlBFZx/01b+LDBSo2J8n8xm3lZ6/2sSY7mX+5ohiHL8SvjgY4POjksrJU8qwGBEFEJUYj678emeDyzZVsLEhBFgRsRj0IAvv7nazLTeL/vqkStUqF3RsmSa9hwh3EpFezp8/JmlQ1f3XpuUrlJFkO8//CubwzNEC6KsJPX+pmfbaZS4tTePjQKD99uY+p2SCSFC22enddHjUFFv55ZAznK3u50RomWxXi9/pSnK1dPNs0hlol4PSFmQ2EKc8wUVNo4eU+H7P+MNeuScOoEfnf14b48Qu9mLUqQhEJtShyzZo0RlxB2sbdXFWRij8UYV+Pg8vLbOhV8FzrFC5/mIgUXfQzkqIV4ha9hq/cWIYoy7zW72JTnoUkffQ74Q9J7Ol1sK3YRq7VwOevLaUiRcV/vTrKwQHXvGdkrg3f+vXr56UZm5qaGBkZwe9PvB3iaosMV+PYNzi5eImiiMViobi4mC1btrBp0yasVitTU1McOnSIw4cP09/fz+zsbEKvubKnmlhWVFTPRtifiNTswnTvwi/G2YhUY8ff0TpBICJxw9p0Pnd9OW/dnIMowCeuLKE6J5nf7xngHw1TzPgjXFqWhiAKXL82g/YJH4PH5oPeUJXObCDC3w4O8/Od/ZjV0UHmBakG7tiYzWwgzMv/fJk723eiFyT+cXCYG6syyEvRc//eftZmJfGm6iyesat4qPhi+sMaEEBA4IaqDK5bm84P716PWQPf3d7Fpx48StuEh09cUcBntuq5rMjM/pEgH/9LA3/aN8ilJSn8+t2b+Nk7NlFbnMqD+0f5n5d6mHYHkWQJpy+ILIPTEyYYgQFniAy1zC8o4MOaMUZlHQ9oCvj0WiPjMwH+5+U+dGqRiCyzr9fBjVVpXFFmZU+3g95pL++9KBdJkvmjkItGq+GDBjsjkoYva9Yx7Zf45BVFBCMyf3otanT/zpocQjI82WrHqFWxITeZltFZbEY1n7uumLAEL3VOc3GRFaNW5IWOKdZkmMm1aHm+YwqTTsXlZans7Jrmhzu6QaUi16LjgfphvKHos5mdHG1XQoZDg052dUczC6/1OQmEJa4qjxagadQCt1cmI0XCfPJvjYy4FhdJrVY7L81YVFREOBymra1tXkVqol4GV5OIrVbj+lONCNVqNWlpaVRUVFBbW0tVVRVarZaBgYF5L0pnuqeu7KkmltX35CWYxdK/0YklJ3/Tk2WZ7u5uuru756V7F3I2I9VHj4yQYdbw0cuKuLTExqDDR5pJwyX5Rm7J8bMx28jOAT+CLLO10ArA1RXpaESB59uivY/VORbSTFoeOTJCrtXA2yt1BCPwcvsUa21aysMzPDEmY2lp5NaJFnZ12xmbCfDBbYWMuYI80TjG3ZuzMWtV/PXAEDqthm/fUUVllpmfvtjN7m47uVY9/2ejgTSjmilPGKtBTcQ5RnV5MZ+5sYqPXFZEGBkQ6Jrw0D3pITNZx79fV8rGAguT7iDJWoH/fst6StPNaEWZNLOaiCThk2AiLDIjanjIZ+HqwAjP63N5dMZAsl5Nfb+DVzqnqc5JOtYDLfD+i/NJ1mv4+av92Ixa7tiYyU4xjYYJHxsFN6lykDFRz9s2pLOt1MYdGzLZ2e3g6NAMuVY9tTk6Xu6Z4cnGcer7nYhAaZqRdVlJVGaaeK51Ao1a5LJSG/t6nHiCYa5dk07XhJf+aR83rk0nLEOfI8gnryjiE1cU4vAEeKA+2hvrD0etDzflWylNM/DTl3v52av9vNQ+SUaSjjWZJhgeQTh0mMbWEUAmHAryqb814Q2e+NkTBAGz2Ryf2LJlyxZSU1NxOBwJiYBWW7p1tZ1PjDMVL51OR3Z2NuvWrYu/KEUikfieent7OxMTE4RCoVM67krvqZ5vkeqKdt2ejQdfFEX+0hrgl+0HGZh2MzHjjzsNGTQqSjKSqc61UFNo5caqDHSa6MMVCARobGwkOTmZmpqaE34ZzpaoTs4G2Ntj566NmYiiQCgSHa1Wl2egu7ubNRUVfLpczQf/eJBkrYj52IxOi1FDbZGNlzunePdF+YiADERk+OC2fLAPsibDzPaWcW7TOrljoon/yt7GgYwybg9M8VQkxEMHhvjsdeXUFlr428FhnmkawxuS0ahUWAxq1mSa+cota/jO0+389/OdhCKl7OoJMe2LcFWJmUPDHn7XLOE3uaktjLa2lKaZeGddPr/e2cfXHm/jo1cUEonIHB10sjnfwpFBJwf7nXzhhnI++1ADWcl6jBqRUWeAsAwg0KNJpgcLyDL7+50UphjQq0WmPUHesTWHV7vtPN00SWWmiY9cls8Pnu/hwYMjvH1rDru7p7nPU07prJtpQYtBJbJ7wM3NGyPcvSmLfb0OfrVrgP+6ay03lBo5MBrkT/XDVGUayLLoeaVjmnfU5nDzunR+9GIfhwZcXLcmjedap3ily85V5TYePDDCjrZJpGOVxFq1wPqcpGN9vRk80TTB1oJkJtxB3EGJuzdlUZZu5OHDYzzaMEYEuLTYhtA/gPrQQcIGI/vsKWxQ+7j5skp+8MogX368jR+9uQpxid+nmENQbMBDIBCID/eOpeNif74UE5PVJmKr7Xzmkqjzir0omc1m8vPzkSQJl8uFw+FgcHAQWZZJSUnBZrNhsVhOuH6t9J7q6brlrVbO+0i1YdjFjs4ZBodH0YU9+N0z1GVrWGsFz4yTWccUf9/dyr/9tZ7LfvgyP3+lh6GxSQ4cOEBhYWF8esWJWO70b6z6d3vzOJIkc0VZNGJuGXXhC4UpNEVYt24dBqMBWY5+4WYCEYYcr6eFrq1MZ9YvUd/n4OmmcezeECpR4OhgNDq5aX0Go84AjfYw22aHSI/4eVSbh4UwN7u62d1jp8/uwaLXEAjLeMIy37mzik9eXULbmJsH6ocwalR89ZY1rMtO4qcv9nB4Isy1RTpuLzfw07dvZEuBhf/dN8DnHm4CAT5/Qzlb8q38v7espzoniZ+90sdvd/ezKTeJr968hktKUvn7oWgk956LCmgeje5ValQCGUlaBAEsuuhL0GWlNjLMWuy+EJ+4sogUk4b/fL6H6yvTKEs38rNX+7GZtNxQmcaTjRMc6HdycZGNybBAvWDlvRfl8dlrS+i3+/jV7kHUosAnrijC4Q3xu72DPNvtIyJFhwzctSmbuzdlgSDwROM4NYVW0kwanm4epyjVSHmGkR1tk5i0ai4psfJi+zQvdNi5qiKVUFhme0s0Y/D2LTkUpOj52St9PN4wTkVGEpVZZtQqkbfX5HBzVQbI0RFyv9o3jN9qo1FlxYmGy5hmsy7Ae2pzeb5lnP9+sfe0n6+FEVBBQcE8M/nOzs5lL5ZJJKvVuWg5ie2pl5SUsHXrVjZu3EhycjITExMcOHCAI0eOxF+aFmYjVtr84XyLVM/7J6/IZsRk1GHQqvjO7ZWUphlpG5vlc9eXU2DTM+ML8flriyEUQPDN8KOnGrjxF4cgvWzJb1BnI1KNRCI81TRGoU1Pvs1AIBDg+cM9qAS4saYSlToqLkeHnCDLaFQiTzSMxo+xMc9CepKW7c3j/O3gEJvzktlaYOWljikiksy2YhtJBg3bXWrUaanc6minWWOjxxnkjrJk1KLA1x9v5YWOacrSTfiDEhEZrixL48aqdB49Osq+XjsGjYory1PhWGrXoNdRXFyM1aTj3hvK2ZxnxRuSUIkCvmNpS5NWzedvLEevEgiGwty0PhqJf6AmC1EQ+fNrg1y/NoPSDDNPNk5y5xoDE+4g1blJuPxhMs1advXYKcswoREFfrijh3dszUGWZb67vYv31OVg1qr5zjOdpJk1JOlU/PeLfTzSMEGqUYuMjN0TZFNeMvdszWZXl52HDo1SnmHi2jVp7Ox2UD/k46ZyCzaTmgfqh7EZtVxRZuOFDjtuf5gb1qbTMualz+7lxrXpjDiDHBx0MekOEZFhQ04yH7+skC15STzeOI7bH0ajFvjM1UX4Q2GmPGHetCFj3n1vHZulONXI3RszecGj51sjRl5wqjAisUmIGu/fXJXG9Wts/GFPP3+uHz7jZy023HuumbzNZmN6eppDhw4tujivtshwtZ3PSqDRaEhPT2fNmjXU1dVRWVmJWq2mv7+f+vp6mpub44VrK3m9lD3Vc5Bkg4a3rjHQYQ+zt9fJRy4vwukL8+jRUT5xZQl2X4iGkRnevCmLaU+ItSngd8/wwT8e5OHDI0v6jLNRqDTlDXNowMUlJTYcDgdtbW0MekUqM80YdK9n8Y8OzWBRy1wjTfNK2wQzvYPHjiFw9ZoMmkfd+EMSH9hWxPVVmTj8YTqcElq1yDVr0qkfnGXigx/h+opU9HKExzddT1ftFcgIzAYkPnxZEd++Yy3pZi2/3tlHRJL44LZCytOM/PTFbnZ2TvGrnf0UJKuoTlfxZNsM9+8dQJJkuiY8HBlyclFRChpR4CuPtdA+HhWHZ5om8Icj2MxafrerH/8LL5H5Pz/i9pkudnXb6R918OFLC3H4I7RNBSlNM9E47CYzScv4rB+rXs2eHjs6tYCAzM9eHaA0zchsIMy3nunCH4owE4jwwP5RgpKMURdNxf7LlYXctDYavf529wBvqs7k6gobDx8d4wc7unm1cxpRltCIMleWJvHuujx6HX5e7pzmzg1ZRCISjzWMc92aVHRqgaeaJ7mkOIUknYr/ebmPtnE3mclaRlx+IrLMPTU5+AJhHm2IVmNnJesxHbt/B/pdcaEadfrpsfu4ojyVe2py+Ld1JjoDKvbOiGwUZtCYDMipqSAIbM23IEfCfP/ZDp5oHE/oszfXvKC2tja+OM+dO+rz+c56b+yJWK3VvyuJXq8nJydnXjYiVrjm9Xrp6Oh4gx3m2UCJVBPM2XrwL8tVsS4/nQf2D1NkM3B1RSpPNY6RpFdzw9oMnmkaxxaeJj9Fy5hH5qryFMI+N/f+/RCffPDoSY+/3L+HKIocGA0hSxKtQ9M8fngAdVoB/Y4AG/Osr/+Hskxj3xQbJ3u4rX8/IUHFS395FiaiC21NoRUZKEg1kpeiZ0uehVSTloNj0S/SjVXpyMCOIR+Gt93NNRtyeGVWw3ee6ybbYkCnFugYd6PXqPjApQX0T3t5qmkcjUrkCzeUoxUF/vvFHkw6gW/cXs07q5O4odLGEw1j/M9LPfz05R5sJi2fvqaU7925jhSjlm891cbBASd/PzTExUUpfOGGCibdIR48Mg5padxldGOWw/zluUYikoxBo+LQRIR+uw9BgElPCL1GxOELIQAjrgCuQISwJHNkeJaIBAjgDUlUZZmxGdSEJZnb1meRbtbyvee6STGquWNDJs+3TfHNp9pJ1qvRiAIH+p1kW7R867YKREHgt69NUFdooTLDyF8PjGDWq7is1Mbz7dOEJZmrKlLZ1TnNU80T+EIS3pDERy4t5P0X5zPlCbGnx0GhzcAV5TaeaZ5gzBXg1a5pXL4I24qtvNI5xaNHo/dqV48dQRa4pCh6fy++qILbi01EgC7RzNTmWji237m9ZQKLQU1Vpp6vPNYaTy8vB7HFee7c0VixzNy+yuV8yTwZq7X6d7UQy0bECteMRiNpaWnMzMxw5MgRDh48SE9PD06nc9nvoyKq5ygatZrPXV/KVEDgmeYJ3lmbj0oU+MtrQ1yVK6IR4TW7nk9eXYbdG8Ks11CQooeAm+2H+/j8w00rev6CILB/NIReJdMx6efRDh9ffaIDSYoQikSQpWh0M+Tw4wjJVIedFBoF1gftPJNcDEeiLwb1fQ4QBFzeIBEJVCqBayvT6XFFh5BnJevZmGvhudZJwhGJiCQjyZBt1fOfd6/jlupsdnZNM+TwcVFRCjUFFh48MMykO4BRI6ATJWQgI9lEkkmHSlTxrq1ZvH1rLq90TjHs8PORy4oxaVWkJ2n55m2VGLVqfvBcJ0gyH9xWSGWWmRtSwjxlKmJQZcQkSNwljXEoaOTrT7Ri0kZ7XS8vS+X7d1UhIHBxkY0b10ZTp9esSSVJq8KgEakrTEYC0kxa1uUk0TruwaBTUZCi56HDY/jDEmkmLX89MMqO1gmMWoHOSS+PNoyRnaSlPM1I37SfIYeft65Lpsse4KHDY3zwkgI8gQh/PzTK3ZuyCUckHm0YpzLTTFiW+cuBEcrSDWhE6Jn2sjk/mfwUHY82jCPJ8I6tOahF+MPeQR46PEppuolPX1XE5aUpPHhgmBfbJtnd7aAq20yqWRt7CBiWtCTrNNglNT8+6MQdCDPq8nN4aIYbKtP5wnVlVKTpuPefzTzdPHFWnsvk5GSMRiPr16+f11d54MABjh49yuDg4ElHoiWaCzH9GwhLPHRohH77qbfXCIKAzWaL9zhv2LABs9nM2NjYst/H1eqoJAiCVRCEHEEQTrmY94IQVVEUqSu0cPmabP7ZMI5Bo+KWdRns7nEw5Qny9pp8jo7M4g9JXLMmje3N47yrLh8QUEX8PFLfxf/uHVix8x8Ym6bDEeHaihT+/MEa7nv7BtZmmlEJAg8dGuG7z7TjD0ZoGpkBZDb4ogvqrb5BJrVJHPBpCIUjPNcyQUW6GYcvwpFBJxAtYEIQeb41Gt3csj4ThzfED3d0sb1lksxkPS5ftET/zo3Z6NQCfzs4jCAI/J/Li5AliZ+92MV/Pd3AZEDg1uosOic8/HZXf/TkZZltpTZiS9zOrikix76XaWYtb9uaSyAcIcWoJvWYI9Q7y0zoZIk/kguA0e9BFsCkEfnRm6u4OEfDK13TaFUq7t6UzStd02wusFJblMLLHXbuqckh1aTlQP8MN69NIxiK0DQyS5JOZHQmQPeUH2SYcocYnw2CHI1kvUGZDLOGVKOWAYefiszo3NZf7R4kGJG4vCiJxxrGcXhDXLc2nedap/AEwlRmmXm6eYIfv9SHTiWiEQW+cH0pl5faeKVjGrc/wl0bsxhy+qjvc5Ji1PCWTVkcGZ5h2hvhXXW5iKLAxy4vZGOOmV/vGmB0JsjlZbb4M+AJRDg06OKKchsfrrExOhvh/z7dyeONE6hEgesq09BrRL54fSkVaTq+8M8WHjr0+p76chITsbl9lXV1dfEiv7kj0cbHx0+55eN0z2c1sZwvFcGwxK0/e41vPNbEHT/fx5GhmTM6L41GQ0ZGBpWVlfPuY29vbzzlPzo6SiAQOONz93g8y+amd7oIglAIfAJ4BdgmCIJKEIQPC4JQupSfvyDSvzEDiM9eV447oubhg/2UaxwYNSIvDIQpSjVi0qr4yYvdx3pY4YH6QarzkglLIAe8fPeJBnqm3GflfGPIskx/fz8P7WlFkmWuXJuNIAjkWg04fEFqClP4yGVFHB5y8Z/PddAwPEO6RibbOQFOJ3Ujrdgifp6O2Njd42DGH+ae2lysRg07jvWsZiTpKLUKvNA2SViS2VxgxaRTU9/n5Jb1mfzrNSW4gxIvtEVTjLeuz2J39zR90x4yknTcsc5Gw6ibw5Myb9mSw4cvK+LNm7PZ0TLBy/0+JFnmVzt7MWrVvGVLtAjo97v7kGWQZNjeMk6STsX4TJBnW6Kpz6RN67nT5KFesPLsjJbfmdaQZVTjDkmMzgS5Mk+DQaviT68N8NatuRTaDPx6Zx//59JC1mSa+N99Q9xRnU6aWc1TzZM4/FI0Qg9EEAWRolQDFr0qOrZLK2LUqpAFMZpOdoewe0MYtSJPN08y5PSRn6Lj781uTFqRzGQdP36pB7UAKhG+8kQbzSOzgEx5moGv3VRGWJJ5rnWKN1VnEoxIbG+Z5JJiKznJOh46PIokw+Z8C7IkoVOJlKZF39TVosC/X1eCRa9CAuaud/sHnIQlmW0lKVSl6/jEJRmMzQR4sX2KLXkWrMaoV7Neq+LmdenIkRD/96k2frGzf8VsCg0GA7m5uVRXV8+z4GtoaODAgQN0d3fjcDgSnmJcjdW/yyn0P36pl5FpF+UpasxiiF+82rfkn13KtYrdx7kp/0AgMK86/HQ9p1eTqAqCEGvW/S4wCLQDRlmWI8CdQNlSjrO6nrxlIladuz4niW0lKTzdYqfRl4ROI3J4cIb/+3Q7nmCEaW+I1/ocqESB/ikvR4dmkAGVSiTkneH2+/Yy6jw7E0HC4TANDQ243W76gskkaaDs2OI75Q4yNhOgOjeZm9Zn8vErSzg6NMvhQQfri9OJfOyjSOvWIV5Uyw3rszkyGeDpxlGyk3VsyrdwzZoM9vc7mXZHTbtrs9TYfWH29zvY2TWFNxBBFEWuq0xnbXYylRlmnmgcIxyRuWNjFkatigfqh+jr68Omijr6iILADVWZALyjNo9Ly2w81eHhT/vHaBl1856L83lnXT63b8zimaYJnm4a49XOKXonPXx4WyEb8pL5U/0gdk8IVCpuvecakrQiv7GuIy1Jx7fuXIdJp+GfR0YxqOHuTTkcGnTRPu7mU1eXMOsP8+fXBvnQJXmIQoT7dg4w6Y6QkaQFASrSjfzHrWuoyjYz4PCj16moyDDiD8t4wxICUeGRkJElcAcigMyUO8SA3Y8sSzzT7mLMFcAbjPBU0zhGjYAsw83r0rh+TRq9016SDRo25SbxdPMEaWYtNQUWtrdMEAxLvGVzNoNOH3t67Px6Vz96rZpAWOLBg69X7apEgbAsY9Gr+PWuAZ45lsbd3W0nw6ylPN0IwNoMAzeuTUeSBVrHZ+mZ8kYPIMs81jBOmknLZcXJ/Oylbr76ZAfB8PLtjS1FtOda8G3dupVNmzbNa/loaGhgaGgoIdNaVmOkulxCX9/n4I97+yES4vPXl3B5aQqv9UzjCSxN4E71vOaOtotVh6empuJ0OuOe0729vbhcriW9LHm9XoxG45I//yyRDmwH7EAs7FcBs0v54QtCVEVRxO/3c+jQId62LomgSs/2likyzFp0KoGKDBM/v2cDaSY1BSkGfv2uTeg1IjUFFi4qtBKWAEHEMz3MjT9+hfbxN0asgiAk7I07NgovLS2NsopKdnVPsyZFjC8UTSMukGF9TjIQTeHWFlrxhyJkJeuQq6uRP/hB5He+k+trS1CrBLomvVxbmY6AwPVVGSAI7GiNLtjlKSIZZh0PHxrhF6/0sSbLjFYF248t6HdsymZ8Nsi+XjtmvZrbqzM52O+kbTrMQ+1BUs16VKLAL1/pAVlGEAU+dVUJBRYVz3c6yEsxcF1ltD3pvRflU1ds5fd7+vnjvgFK0gxcXpbKR6+I2v39aV80zW7QqcmwGAhIEm/dmku6Wcet6zOp73cx7olw6/pMbCYdf9w3QHGqkbs2ZfFq1zRffqyNiKxCp1ZhNar4l9pk7ipV0zPl5ScvdfPWDam8dXMmk+4Q7RM+TFoVd2/M5po1aWhVKgREsi26+Ag9QQBRkBGAZC0ISFxabAXgilIba7NM7Oyyc8PaNNSiwN8OjXD3pmxm/BF2tE1y54ZM3EGJ59qm2FZiJd+q5/d7B2mf9PPhbYXcWJnKM00TtE+4EVwuDh7pxR2U+OhlBdQUWvjDviHue7mPxpHZaBr92DMgI3NkeIZ8mx69RsU3n+7gtT4HR4Zn6Zr0cvfmLD55RSFv2ZjBY4eH+MCfjjI5e+bpuuNxqiKmVqvntXyUlUWDgNi0lra2ttNyB4ILR1QnZwN87K+NSOEQCCIOb5j1OWZC4TCNI0ta/8/YTSlmJFJWVhb3nDYajYyMjMx7WTrefqwkScvq/X6KxBbwFmAL0chUFAShBNAQFdmTckGkf0OhEC0tLeTn53PrJdVcWpGF2aDhG7et5c5N2XRMuAlGJO7amEPbuIchp5+b1mdyZMjFO+rySNKpybPqkBFwjQ1yz692M7FggUpUW83o6CgNDQ1UV1eTm5vLa30OvIEwFSmv36ro/qCaQpsh/u8qs8yAwMHWEYSf/Rzxl79C6O4hxaQhz2pEQqb2WCVpZpKOzXkWdrROEI7IiILIlRVpdE/7MOlUfPHGCq4oT+flzilm/CHqClPIteh55MgoMzOzFItTJOtV/PGIg9mgxJduquB9F+dzeNAVF2KNWiQnSYsgCDi9QaY9oWPXSeBTV5Vi0qmZ9gR5+9Y8BFEgK1nP7RuyeKVrmu5JD43DM3RPetCJIvX9DiC636tXi+wcCqFVi9xTk0vnpJedXdP0TLiirkWCyPfurOLrN5XjDUg80OznrkvX8dUbSwhHZP7v9j4eOjJOpknNjRUWQOTxpglSjFq+fms5qSYNozMBIseKv+RjxVoSMrNBGbNORX2/iy35yTzZNMllJSn4gxJPNE5w87oM9vY4UYsCG3LMPN44Tn6KgY05Zp5sHCcYltmYk4Q7EKEiw8QVZSm8syaHVJOaX77QRXj7c7zYMEyqEKE2MMG/XVvCXRsyebXLTkSCsvTX3+g7pwIMOgLcXp3Jd26vpCDFwI9e7OWXO/tIM2m5sswGgsBbtmTzgYtzaRi0c9evD3JgwHnGz+hCEiFiRqORvLy8+LSWrKws3G43DQ0N8WrUpUY/q7GlZjlE9Ve7BwiEJZAiaER4oX2KIpsBpAgdE54lHSPRFoVarZbMzEzWrl1LbW1t/GWpp6eH+vp6WltbGRsbY3p6elm2Jb75zW+Sm5vLpk2bEAThiCAItyz1Z+XXT+he4K1AFvCfwAvAf8my3LaU45zXkaosy/FURGlpKRkZ0QrRT15dgjui4cX2yfhC/fjRUa6tTCfJoOLRIyO8qToLAYHnWye4bUMmw64ARekWQMIxNsz77z8w76E4UwMISZLiD1xtbW18n+Gl9in0aihOFuOf1zQyQ1V2EoL4+sLRPenBrIIOTzSCEZqbEX/0I+jpxRMMIyJwZPD1qSa3rM/C4Yuwr9eOKAqMuvyAQGVWEhaDhlurswhFZHa0TCCKAnduyqZ3ysuOw91Ur62kJEWPLwTXVGZQkm7i5vWZbM5P5g97B+if9tE75aF+xMfFBWYissx/PtseXQCOEZGi0d9rfY74v7t7Uw5JOjV/2jfAb3b2kZmk5U0bsnmt18mQ00eyXsMNVZk0TkmMzfi5ek06uRYdP32ph8PDfm6rzkJG4OnmCcoyjHzyqiK6Jz387JV+KrKS2VxgPWZvKHJFmY2bSvV8fKOG9akq/nlknK892Y4nGOaujZlkJenQiHBFWUr0/iIDErOBCDLQODJDtkXHXw+Ocn1VGju7HJSlG0nSq/hT/RBv2ZyNyxfh2dZJ3rI5C5cvzM9f7efplkkMGpHxGT/+sIxeq+JjF+cw7AnzWzmPI5KJayxh1EePIvp93FOTQ44lOv7uJy9FnZciErzcM4NZJ7KtJAWrUcM3bqlgY24y094wWjV45vgBt4+7USGjIcSH/niEX+8aiL80JIJEL46iKGK1WuPuQLFq1NHRUQ4cOHDS4eyrsaUm0aI64w/z8KFRkMIk61VszLNweGiGZIMGk1ZgYIlVwMu5/ywIQvxlqbq6mtraWnJycvD5fLzjHe+grq4On8/Hc889h9frTdjnfvazn+XIkSPIsrxJluWnT+MQelmWPwzUAO+SZblYluXtS/3h1fXkJZBgMMihQ4cIBALk5+fPexvbWmBlc1EGTzSNY9apuWZNOq922fEEI9xclcmBfhf+kMRlZam80DbJ5eVp6NQiWRYdokYPIT/tfcPc93JP/JhnIqp+v5/9+/ej1+vZtGkTGk206ESWZZ5vHSfZoCEkR3f9ptxBxmcDrMuZs7kvyzSPzlAbnCQz6OaBrM3IKVZApmfXQabcQXKsBp5sHI9a7REtlMmy6nmqcZxup8Tu7mgP5ZFBF55ghEKbgY25yTzdNEEgHCZPNYNFJ3JwWs3BB57iyJiXZCnAofZRgqEIghCNQM1akf96roNfvNqHSSPyjs3pfPaaMnqmffxmVx8Ajx4dwRuMcO2aNF5om4ymswGDVsWbN+fQMDzLkNPHhy8t5PaNWeg0Io8diVay3r4xC5Uo8OiRMbz+AP5gkAgi11el8+HLirlrczYvd0xzZGiGuiIr77koj319Dr78eDuvdDl404YsaguS+fvRKXaNCmytruRDV5Zh0IAkC+gEmUqTn89emk6qWc/RkVnuqM5AQsSqUyFIMhFJJhiWGZsJEAhLdIxHTSj+WD/EnRuzaBnzMBuIsDE3iccaov3QKQY1e/scrMtO4vPXl+Lyh3n8mFFDdaqWm9VOXvBoEQWR6yzhaN45GL3XIzNB7tyUzeY8C3/eP8J/7Zrk6Kg3mrJWR7/CGpWALySRpFMzMRvkC4+00Tg8Q9eEh93dDm7bkMEP7lzLRQVmfvpiFx/689HjTrg5HZYzMpxbjVpbW/uG4ezt7e3zjAsuhPTvsy2ThCQZORLmstJU1mYnMTkbxOkNkWZUMzaztFT/2TTTn7uv/txzz7Fjxw5EUeTZZ5/lyiuv5Nprr+W+++47K+dyEr4sCEKeLMt2WZb7BEEwC4LwKUEQtCf/0fM0/et0Otm/fz95eXlUVla+QfAEQeBDlxUy4RN4rc/BbRuyiMgy25vHuWldBmoRnmoa403VWfjDMvW9Dq6qSOfI0Cz5ackgqpA9du57vo2JmWOFOqeZ/p2amuLgwYOUl5dTXFw875p0THgYnwkwMePnvsN+XowJ0Jz9VIBBhx+XL0y1PMvbppvp0tk4qMsAWWZ3JBmVIPC2mty4AUHsGtyyLov2CQ//7PSTbdHysSuK8YdlXjxWGfymDVnYfSEeeqUJi9nEm2vyaZv08tNQDpVhF5/ztjARFvnn9sNA1Lj/M9eWMezy0z3h5Y4qC0aNQE2RlTdvzubFtkmeaR7nyYZRLiu18ZHLi8hM1vObnf2Ej127bSU2ZFnGpBHYWmAlWa/hmjXpvNI1zbQniM2oZXOGmhfbJ/n6Iw04ApBnNVDf58QXivC2Lbnkp+j5zZ5BvIEIt6xLpyzNSO+0n015Ft5Tl8u/XVvCdWtSeeToOP/72jA/eXkAGZFPXllIEDV/aQ2gIsKbSwU8QZnhKReVqWrcQYkUkwadOqp5kiQRikh0TfrxhyOMz4RoGZ0l1aTmd3sHKEw14AlK/PvDrcwEIogIrMk0sz47mW1FVp5sHGfaEwSjkTfbggiyhEoQUM/OgMEAJhOvdtoRiLbM/Pu1xXz66iImfRHCEoTCEqFj0X/jyCwdEx7u2ZrDd+9Yi0mn4tvbu/jh890kG9TcUZ2FQaviU1cVcc+WTA72TXHnL+t55MjYOVUYNHewd2w4e0ZGRty44NChQ0xMTBAIBFbVnNdEi+pTjaOADFKEy0tTKEk1wrGB9ikGDeNLFNWVrJQ2GAwkJSXxox/9iP379/PAAw9QXV19xse977772LBhA4Ig/F4QhJTTOMRdsiwPxf5BlmU38CFgSQ/UeRWpxtK97e3tbN68mczMaDXqYjNVr6vMID/dypONY2Ql66kttPJsywRGrYrLylJ5qWOKzGQda7PMPHNMbEMRmbJ0M2i0IEkEnBN85qGoMcSpRqqxsXI9PT3U1NTEJ4bM5eX2SZAlvn7rGmw6+Nkrvfxx3yBGrUjBnP3U5mP9qVWXbuCqyQ4yA7P81VCMpFazW5VKdW4Sl5elkZti4NEjI8jHno1r16SjVonMBuFDl+RRmRU1dH+yaYyIJFOSDBkGgdcmICsriy35FoSIjCyIfMnXwiZmuGq2n3+OvG7eX5RqQCcKyMjIsoB0LM349q25rMk08/s9/QTDEu+ozUOrVvGBbfkMOXw8f8wF6MnGMUQBPEGJjoloQdibNmQhyTLPNEUju4syoxN2+t3w/9l77yg5yqvb+/dUdQ7Tk3POUTmBJJSQEEHkDM4YGxvjjP064AyvE8Y2Bhsb24DJOQgFlAPKeXLOsXtmeqZzV1d9f9RoABuMsLHh3vudJa01anVXVddU1XnOPvvs/dmlBdy6vIDxYJSXTg5ilCU+t6yA8UCUvx3qY1/7GG2eIAl2A6cGJmgeDiBJgk8vzmFNRTIb6kZocwf4/Dl5nFOcxBeX59PtjbCtV+WcuVVcUJXKkcEILqOuZawoUaKKhst8mjim4TAJxoIKoHGwy4vHrzDqi/LyySEMAhDwgwtLmJ+nV/7eYJTr5mWiqjEeO9QHksT2tAoQEqgxfu5LIXD2YjTZMGVh5yDZrven5+XEYxCCZJuBDfVuvv58A6d6J3jq6ABJNgMrS5PIS7Ry1yXlzMyOY3RKaarnTeYKp/omMMuCLKfMd1+q5zOPn6Jv/F+vWj/IyvDtzNmNRiMTExP/cXP29xLvZ/IaD0Q53DOJpihYjTIFyTay4y0goHc8RLzNgNt3Zt/3gzYof7PwQ3p6OsuWLXvXz5177rlUV1f/w98XX3yRW265hba2No4fPw4wAPzyvRyTEEICRoUQeVPzqUII4QQMmqadEWvuA0+q79fNGIlEOHbsGKFQiPnz57+Fpv12CU+WBB89K48md4S2ET8XVqczGYqxp22UC2vSCSkq25tHuLAmneHJCG5fhJpMJ3X9E9isVlQktGiI/fWdHOwYfU+V6mloWlEU5s2b9472WtubRyhMtjIz28Vn5zq4eXEOY4EIoajK3rY3iGi1AxMk2UykzqpE+tIXuco+Sas9lRevuIXhgMLioiQkAZfOyqTDE5zurQajsan0Kkh16JDzxTMyGPZF2Hisnd7eXq6al02vN8zrbaP87LVWDDIoQqZbtoOm8YnBI1gljXt3tKOqurG3oglKUu08fWqc/qkVs0EW3LAgm6iikmAzkhGnf+cFeQlUZjp58kgvPaMBNtQOsawkCYdF5vljOuSbHmdhYUEim+qHaevsZtCnAAJZEiwsTKAs3cnCggRePNGPNxilJNXOBVWpbG1yc9/ubsrTHPz00gpSHGbu3tbOeCAKQnB2Yfz09z/9m5uT62JtZSob6920DftwmA1oQnCgXyGmqnjDOkXQG45hEGCWwB9VcZllEqwyTrNEmtNIUbIVs1Hiu2uLEcDOllGunZtFWFF49vggqU4z66rT2NM6xrFuLy82jzMz28XnVxTQFDbw4z1D7GkbZcQXZUXZGz6+O1s9BBWNT85P4dtri9GAH29qpXnYz5qKlGnWsiQEA94w6U6LbobwSjN/fr2Hbc0eagd8XD8vix9eVMbHF2RwpMPNxfcf4K/7e6ZtEf9PDZPJhNPpJCMj423N2Zubm//lmcp/J97PpPrbnbojkaYqxNsM+piLxYDdZGDQG8ZlNTLmj5xRpf5B2779KxKFW7Zsoba29h/+XnLJJaSlpSHL8unv9Edgwb9waE8A96KTla4FHgWeP9MPf+BJ9f2I03BvVlYWFRUV/3CRvFMVecXsTOx2O6+eGqI600luooVXTw1RmGyjNNXBxrph5ufFk2AzTkPD7kCUeXkJCIMBm1GgBjzc9uTJtzVDf7vwer3T0HRZWdk7XtBjgQjHe8aZmxuvfwdJZmamfgEmO4z8amsbzx7tB02jvn+S6ilDbq20lGWfuIRMl4Vn2gNICObn6wjIOcVJpDjNPHNENwp44nAvAg2BYMNUpTgnJ44kq8Tm5nEqKitZXq5r5N63s52+8RC3n5NDkhrkYSkbbWgYV0kBn1qcT/OQjz/u6WRL4wjrZqbzzbWlmA2CBw+OENXZQWxpHMEgCUYDUXa3Ti0KhOBji3LxBhV+va0NhMb1C3I4vyqdA51j9E7NBV9YnYo/GmN7+yQbugVZ8RZiKtPV640Lc4jE4Jkpq7gLq1ORJIGiweeX5xNvNfHVVYUEwjHu2dZBOKrywJ4ekh0mipKt3L+ra7oPdc2sVFwGwS/X1/P44V6ynQaEgC8vTiEjzkiCVcZqkIhqGuGYwCg0/JEY3kCMNJsuZFGR7kBRNLY2e1hdnszWZjeqqrGqLJktjSP0jYe4ZEYayXaZ3+7sJBDVuGF+FmcVJnLbinw6R/38bmcnJqNMlssCQEzTeKV2iDyXgaJEMzOznPzs0nISrEY0TeP5E4O8cmoIRdV4tW6Y4ckwNy/J5e7LK1lblcKmhhH+sKebNKeJVWVJCCFYW5XK1XPSCYUj/GJzC1f96QiHusbf9Tr++/gw9TBPs3/fzpw9OTn5LTOVnZ2d/7I5+3uJ96uaHw9EefrogH68msrgRASPPwxC6OYSvjBxFgOKqjIReveFwwdZqf4nHGoGBt6iInYZ8J40ZjVNUzVNuwf4E7ACWA48qGnad890G/9HJ1VN0+js7KSxsfEtcO/fxzslVafFwGVzctjb5WUypHBeVRrtHj8tw35WliXTNx7m5VODlKU5OdLlxShJxFsN+MIxJIOZaEwDJcbw8DBb2wP/tFLVNI2enh7q6+v/6bGejj2to6iqNp1UJUlQP6jDoV9aVcI5JUk8erCXh/b3MB6MUvWmHqssS9y4MJvJcIx0l4k4iz4HZpQFl87KpGHYz47mEbY1jbC2Ko1ZGSa2t4zh9vpobKhnbVkiPRMx6gd9yAKSHGb8isZFMzKYW5nNdeeU0mJJYueVN6N++tMsLU9lYV48G+tHSLQbuXpuFgk2Ex+Zk0zvRJTHDvfS6Q6wq9XNxTPSKUqx85d9ndPWb8WpDmZkxdHuDnBuWQrJDhMXVKVhMki8dGKQUCiE4u6mMNHC1o4g4ZjuxTo/P571p4YIRmJkxVs5d0pictAb4onD/WhTkg6vNbgByE2wctPiHBqGfPz8tTb6JsJ8+uwcvrKqEEnAfbs6UTWI27WDOf4B3DGZ2REPd8UasZkkXmud5Ib52YwFY6ytTAEN4iwSUVUQUzVMBkH7eJRsp8SG+hHmphvZ1TpKZboNm1Hmr/t7uHJWOiaD4K/7ezAbdPauLxyjLM1JXpKOrpxdmMC31uhVqBJTuf3FRu54pYmH9vcwNBFmZaFj+gF9sn+SsWCU6+dlUZnu5JFD/Xz5mXqeOTbA3FwXNZlOLCaZjy/KYX5ePGgwNBnhRxta6PIE8PgiPHNsgPI0G19ZnoPXF+ATDx/jq8/Wv69Epv9mvBP79+9nKquqqrBYLPT09Ezbob1f8nt/H+9XRfjY4T5UDVCnSFmIaZ5EisPMiC+C02IATWM8+O5J9YP2Un2/k+rtt99OTU0NM2bMAD0pfvm9bkMIYQUOo4/WfBnYJN7DiugDT6r/6uotGo1y7NgxgsEgCxYs+KeqHP+s33nDgmwUYWJj3TB2o4xBSPzw1Ub+sKsLFZW/Hehlf8coGnDnpmYmwzFO9E2QnuhAQSIv0Yoa9PLISe877iMWi3Hq1CnGx8ff9VgZHET+ylfY+cAzuGJhimP6ELcQEo1TDjElKXZuW1HE4qJEXjw5CGhvIS6BDpkCjAaibxmvWFWeQqLNyF/3dWOSZa6YncXKAjuhmMZjexopLCxk3dwC4m0mnj3ax4snBmkc8mEzyDQN6iv6ZZXpFCbbeKg1QCCmr8DTXBaE0DBIEoapUZ+ZGTaW5Np58cQA9+/qwG6UuWx2JjcvyWc8oPD00TeUhE73Jy1G/ZJ02YwsL0lmZ4ubo7WNFBYUMCMnkVAsRmWy3lO+YnYmvrCuaQxw9bwsDLLgtzs62Nk6zuWzMlhZksz62mG6RnXK/jnFSSzIdXFqYJKqDAezc1ykOMx8bFEOTUMBttUOMNzYxj5zKkZNZdJoxdHfw4pUidrhMKlOvbLd2zbK0qIE/JEYyXYZWQiC0RgCjaGAisUgMRiEBIvEQ/u6WZohqB2Y5HjXKFfOSudEn94H3djgxmmWaRmepNP9xljB6+26stf/XlrODfMyGQ9E2FA3ggD6JiJ4QzFiKjxxuJ/MODPrZqTxjTXFfHNNEb6wQiiqMhGI0D6ib7Nx0MehLi/ralK5aXEOveMhbn++ke+83IiiaNyyNJ8F+Qn88rIKziuNZ1NtHxf+7gC/2d6B7wzVeT4scaZVodlsJj09ncrKShYsWEBeXt4/mLN7PJ73xSv5/UheMVXjsUM6f0aLKaQ5zeQnWjk61c5JspsY9St6y0LTGA++ewvwg+6pvt8ONY888ginTp3i5MmTaJp2saZp70kAe6p/+iXgWWAfcALoBDaf6TY+8KT6r8T4+DgHDx4kMzPzbeHevw9Zlt+xh1Kc6mBBUQpPHxvgnu3taGgEwzEumZnGvNx4JAE/WldGaZoNh0lmSWEimqqPUyAbmYwoaEqUEfcotf3/qLTk9/s5ePAgiYmJ1NTU/PMLWFEwfOtbxBqb2OnKZ463B+NvfwuhEJIk0TjoozLdiSwLJAluW1GI3SjpLjV/B18d6BxDEoKQovH0m2TwzAaJFWWpeIMK8/NcxFkMxIkQpfESh0cEksmC2SBz8Yx0TvZN8siBHhYXJXLjomwaB30c6/EiS4Kbl+YzFojw+MFeOtwB1p8aoibTxfBkhCcO6ze+EIJLq1y4LDLNwz4um5WJw2ygONXByrJkXjk1yPBEmE63nxO9XjJdZl5rGCEYiaFpGgvTJaKqRqeaiMFiY0+bG1lIjId0RKA0zUF1ppOXTg4QVVQS7SZWl6fSOOQnK97M5bPSuWHKoPyPe7qnCVqypFewI5MRwor+wFxWnEhFuoPHjg9xr6UMNJWrzR5ahZ0TcgLL02XMsuDlU0NcNTuTYb9CbqIVoySRHmchpmmkOo2omkY0BoFojK6xCLlJNtxBDYPDRa7LyCOHB7D4B7Eb4JnjA6Q5zfzwolKcJolfbGtn1BfB44uwrdnDspJk8hNtXDIjnWvmZSEEFCRZea3Nz7c39fD9V5vpHQ9x7bxM5Kkk4jQbCERUZmbFMeRX+J+XmrhnWwe/29VJisPIVXMyWF2ewj1XVVGd5cTtj6JqGrtaR/VFgRA0DvmwGgUzM6w8sKudC353gMcP9/8f02/9V8QfTkPFf2/OPjo6ytGjRzl27BhdXV1vMWd/r8f07ybVg13jeEMxNFUXfChKsVGZ6aR1JIiiqiTaDYQUFZOsL1C9gXdPqh90T/XD4lDzpkp0BnAjsErTtAqgAihEh5LPKP6PSqp/D/emp6ef0edkWf6n0OwNC3PRZBM3Lsjizksq0NBXfR9dlIOqQetIgEtnZuCLxFhSnERVphO7SUIymHBP6uLrscA4fzr0VrutoaEhjh8/TmVlJdnZ2e9+oMPDMDDAsdwqvJKRGibB50MMDuGPavRNRN4C8xplCUlIGGSJe7a1EXvTQ29/+yjl6Q7OLU9l/akhesfegPJGfCGEJOgfD9LY2ADA1fNy8IXVaUWkmdlxqJrAZJT4wvIiVpenkRpn5m8HelBVjdI0B+dVprK+doifbW4hzmrk62tKWF2ezPPHB2gcnERIEkahYTHKoEEo+saK/7r52UhC8MThXp443IvdJHHzkvypynOI1tZWXIYYc3JcvNbo5pkjfbh9UdZWptLjg9YpxZgr5mQyGlTY3qxDvCP+MAiwmWQMksBhNvCRhVm0ugNsbXTTOuJjX6eXJUWJDE9GeP74IKA/VD9+Vja+iEadHM9HAq1cHOgmUQvzjLMYc2oii/Od7OsYJzPeRHGyjdcaRji/KoXaAR/zc114fArxFiMOs4SmgUDjWM8kEvDSKTcxJCYjKn+sixGM6WQilxzF29/OR2sceH0hvvliA3dv6wAhccVs/fpWVJWnjgyQk2DhJxeX8z/nJHNWroPGIR052NnioWnIh6pp/GlvNwk2I187t4DfXFXJ5bPSOdg1xuBEhEyXmcmpPps/HKNl2Ed5mp15efE8e3yALzxVx082tdI5GuRzS/P52qpCfnJRCckWwY/XN3DR/QfZUDeM+iEaVXm7eD/EH/7enL2yshKTyURXV9e0U8vg4CCRSOSMtvd+JK/NU05SqDrRbkF+AqWpNiIxjZ7RIAlTpgrRmL7I9v4f0FP9EHqpRoDXNU3zCyEMmqYpmqYFpsZqzig+8KR6pivKaDTK8ePHCQQC7w6h/l2827jLueWpJMc7aR7yU5LmoCjFzmuNI+TEWylLc7C1aYS5ufG4LAa2Ng6zujIVf1ilLM2JJBsIRmMQDXGsa5RAJIaqqjQ2NtLb28uCBQtwuVxndqB2nVG7xZyJpsbYaM0mhIRms9Hp1W+QyjeJPnSPBpkMK5xbkUrzSIDnjusknb7xIF2jQc4uTOSGBdmYTTIPTjnDDE2G2ds2RnW6gzZPiEHFTnx8PAWJZmbnuHjxxABjgQi/2d6BxSgRVjQ6PH6MBp292+EJsmMqgd24IAeLQTAwEeYz5+TjsBj4+Nl5pDhM3LujnWhM41BvgKGJMGVpdl44OcDQ1FxvksPMBdVp7GjxcLBznItnpjMj20VFmo3nj/VhszspLCpk3QxdR/fFk4MsLU7k+gU5mCV99AZgRpaLkhQ7zx8f4GDHKAc6vMzOjqfFHaB2Sv90aVECFWl2Hjvcx1/39+Kyytx8dg7LShJ56dQwvVPjJOlxZsxGXcN5xtxSDFnpXJQuU6fZ6Z5UWVHgQAiJ9bUjXDk7g2G/QoLNiMtqYDwYwWGWcZplfOEYmS4zArCbZSxGCZMsGPZFyIjT5S4/tjCbq+ekU+uO0aEmUZHp4pa5DiQlTPOwDw2NJw/3cbh7nPW1wwxMhLl+XhayJEixGbCbZYQmWFWWTPOwnzteaeaLT9fRMRrkxgWZWIwyVqPMooJ4NAQ5CRbqB3x88Zl6/rC7i19tb8cgCb60ooAvryrgrovLSbIbqRuYwCxLjAWiRGMqhclWXBYDQosRjYT5+rO1XPXHI+xq+c/IzL0f8Z8Y8TGbzWRkZFBVVcWCBQvIyckhFApRW1vLoUOHps3Z3+k58+8mVXVqjl6NqWhT+6hMd5CfaEMAXaNB4q1TSVVVQdPOiKj0QfdUP4RJNQRkCCF+A6wWQpwrhFguhDgjhxr4ECTVMwmv18vBgwfJyMigsrLyPV8E/wz+BTAZJK6Ym83RPj+jviirK1LoHg3SMuJnVVkyvWMh2t0BlpUkc7h7nMp0JzazjMUkg0EnBWiA4vfy+IFODh8+jNFoZM6cOdPqSGcULhfqxz/OTmMqecExWkzx/HDm5YTiE2kbjWI2vGETBnCq3wtoXDYrk6VFSTx1pJ/2ER/72kZBg0UFibisRq6fl82Jvkle7xjlheMDCE1jVUaU9DgTLzZ4dUKPpnHtgiwmwyp3bmimazTIl1cVEW818uiBHjRNY3FhIqWpdh492EMgEqNhcJKQoiJJYtolxWqS+fyyQvrHQ7xQ6+HlxgmKkm18dXUJBiH4y5t8aS+blYGkaRgkjYuq05mcmGR2QpTJiEarTydXzciKw2qUUJQYNy7IwWaSmZ1mYG/7KB5/BCF0FvfgRJjf7eogO9HKV88tJMlm5PHDfTosLgSfPCuHYDhG60iQq2ZnYjUZuGF+FmajxF/2daNpGhvrRghHNYwGiedjycQuWsfK8+ZiNxvY3DpJvFXmnJJEtreMkptgJt1p4okj/aQ6TbSMBEl3men1hilLtTPgDeO0GLAaBaFojPJ0BzEV8pNtFCdbefJIPwvy4pmX6+SRg30cGVbJzc4mhEy608ysNBMHO0b52eZWHjnYQ5xZEInGiMQ0hv0Kr7WMs6wkkc8syePea6q5em4Gw5O64MEzxwbY0ugmGFG4f1cXTrPMDy8s5TdXV3FuWTI7mj20jQTISbASmhKPSLAZGQ0opDst5CVZePD1Hm57uo67t3VwsGucGxdk8esrKvn80ixGJ3zc8vhxbvjLMfZ3jH3okut/em5WCIHT6SQ/P585c+Ywe/bsaXP2o0ePcvz48X8w9f53k9exngl8YRUN0DQVl9VAvNVIWpwZo0GiZzyEayqpRhQNOLOk+kH3VD8stm9vChsQAMzAp9H7qz9DH685o/hQJ9XTfqINDQ3vCe79+3g3+BfgqjlZqLKJbU0jLClKwmKQ2NIwzNnFSViMElubRlhZnoKi6tDqkqJEGgd9GI0mNGQkQAlN8MedbRQVFVFUVPQv3dhd511CszWZVYVxfKnSTn3MzPdfaaDJHaIw4Q33FNCF9dPizKQ6zXx6ST4uq4lfbW1jb5uHsnQ7SQ5dVeu8qjQKku08uKeTrQ1DzE4zsGRuDTcszKNnPMTBXp25XJLioDjVRuuIj4tq0lhYkMiVc7KoG/BxtNuLJAk+tTiPsYDCQ/u6+fX2NgqS7CwtSuS5Y/3TRJua7DjOLU/mtZZxvGGVTy7OI9lh4orZmRzsGJuWJewdC00xXDUaO/vp7OrkorMqyYq38vJJfWyg3R0gEI0hJAm3X4faFqTLaBpsrNWh6vl5CcRZDHgDMT69OA+rycBVszNoHQlwqEvfV1a8BbvZQExTKZoSpo+3Grl6TganBnzsbR/jxZNDzMl2sqYihR2towxNhrEaZc6rSOXUYIihySjratKIxuCrzzUyOBnBF9Xo8+ps0abhABrQMOxDEvoIjCegkJdk5VT/JHOydfh4RlYcJlnw082tfGR+NjMy7fxxbxffeKERo8HIdy8o5etrK/nTjTMoSrJgQKDEVH61vZNPPXKMe/e7kYXg+nkZAFgMEp2eACaDxMcWZmMxyPxxbzc3P15LhyfAtXMzcVgMJNpNLC9JAqGPJLWP+PnqMw38cks7/7u5lXA0xjfXFPHDi8r49toSbEaZ/Z1jGA0CWYCiaiwtSqIm0wlqjPahcW565Bg/PRT+jwj1/6vx3xbUf7M5+/z586dV3E6bejc0NDAxMfFvGW78ckubzghQdR1vf0QlEI0hC0FmnJm+8dA0y98XUjDL4owIZh9kpfqfGKn5V0PTNE0IIWmadkDTtKuB2zVNu1zTtIs0TVugadpdZ7qtDzypvtPFfxru9fv97xnufbt9vNtqOi/JxqKSNLY2e7AYZc4uSmRP2ygSgrMLE9nbOkqKw0Rxip2tzSOsLE0mEoPSdBdIkj7tqcHIuJehyHuoTv8utk+pKM1dNpsl59Rw+5oS2t0BBnwKma43pCdVVRfWr5nqsTosBr64spA+b0iHfouSpt8rS4Kbzs5mPBQjqsJHzynHIMssLkqgNNXB+qZJgtEYI5MR+sZCIKbIT8DqihTS48w8fKCbWEzvpS4vTWZz/TAxFb62poSbluThMMn8bmf7dF93bVUaqqphMwjKUnWI56IZ6aQ4jDy0rwdNVXnsUDdxFh0afaXBrY84mM2sq0mn3ROkrn+SR/Z34zQbsJskXp3Syk20SCzIj2dz/TDhqMrwZAh/JIYkienKa2lxIpkuM08c6UdVNV7vGMMXiWEzyjx6sG/6elhdnkyWy8yf9/Xgj8a4dl4ml8xIQ5IEz031W8+vSsZokNjcMsFLJ4fQ0PfzhWV5FCRacZplbluRjxCCZcWJSEJClgWToRho0D4SJMEqc7RngopUG88fH+SCqhQmQwo/3NhEnMUImkYkphGMxthQP8LARJjXGt20eUJ8bFE2f7pxFt9eW0R2vAVvWCWiqHznhToe2d3IK8e6Odg5zpWzM7iwOpX/vaSMm5fkThOLHtzXw327umgcnOQ3OzqItxr48UVl/O7aGi6dlcaxHi/tniDJDhPuKY/d7HgLE+EYyXYT+YlW/nqgj1ufrOPube1sbfZwSU0a911bzccWpNM7FuLjDx3jk48cn2aifpDxQQvqWywWMjMzp029s7KyiEQidHd3/0vm7KP+yJSVmwaxGJokiKmwd2rWOyPePI2MCHSxfbtRwhd6d9byB1mpftjgX03TVCFEmhDiVuCAEOLzAEKIG4UQZ5/pdj7wpPp2cVog4TTd/b91g1wzL5uRoEZt/wQry1MIRVVeb/ewsiyFkKKyr2OMlWUpdHlCSJIgN8HCZCiKZDAxBcyg+Mb57fb2d93XO8XWxhGyXCYypgb+F+QncPXcLAD2dE5SP9UnbHf78Ydj1GS90a+tyXJRlRGnW6C9aRHh8/nwDXZinBp18UX0FaxA8PGzc5mIqGxoHOO3O/TV8OKiRDY1jjA4EcIoS3x0US49oyFeaxhB0zQiSgyERJLDTLrTjNNi5FNL8mgf8fNq7RBoGo8f7MEkC4KKytYmvQer27Vl0z7i58kjfdT3+1iSIbMk38HJIQW3T2crLitNwmk28NihXk72TXDlnExWlaeyv2OU0alq9aKadCYjMXY0u3lofw9GWSLJbuS5Y7oghiwJrp6bSa83zK7WUZ4+OkBuopVr52VSOzjJsV7de9ggSVwxU1fTKki0kpdkI8Fu4tzyZHZNVatxFiOLcu0c7A+xo2WUlWWJaELg9ke5ek4mQ5NRIopKcbKN2v5Jzi1LJKpAVYZux2eQBWNBlRhQN+hDljT+dlCfN3T7FHa1eZmVk8Cdl5SxMDeOTbWDfOmpWv66v5eCJCvLihP1XqrDRO94iNJEEzfMSiLeYeWV5gAPHRnBIoPwjdDXP8D4pJ9Xa4eJtxr5yboyVpUms79jlO+tb6HfG2JlaSIWo0Sc1aiLVKhQkmIjGI1x5+Y2vv5cA99f30REifGdtSX86KIy7ji/hHibgX0dY0joLPJoTOX8ylSqkwWaEuVUj4eP/uUIn370BMd7P7jk+mES1D9t6h0XF0dpaek/mLOfOHGCnp6ef2rO/tf9vbpHkhrDZpIAQarTzN4OPammO82MTIbR0BfXkyEFq0lm8gwq1Vgs9v/3VAEhxOmVxc1AIrpJ+Wlseibw7vqJU/GhSqqn4d76+npmzpxJRkbGf3X/q8tTiHfa2do4TEW6k0yXha2NI1SkO8iIM7O9aZilxUkYDRLbm92cXeCizxvGbjXrCi6AFouw5WQ3k2cwI/b3MRGMcqhjlPl5b9WA9gajmGSByyzx/fUN7Gh2c6pP1/utyXxrT8IXVrCbJB492Ef7iJ/h4WHaOzoYEMlEYxrxNiO/29lBZKqiK09zsiDbzuaWSWr7Jvnk2bl88uw8DJLEw1OG4QsLEqjKdPL44V6eOtLH3vYxFhYk0DsWZFuzzkg8uzCRuTkuHj/Uw8a6YY72eLlyVgp5LpnHDvQQDE+NrpQmkxVv4blj/ThNGhfPy+O6s4oRaLw8RT4yGSRWV6TQMOgj0W5kbWUa51eloqKxaWomtTzdQWGyjWeP9nGgc5wr52Ry+exMmoZ91A3oC49F+fHkJ1h49FAfQ5NRrpmTyZryFDKcZh491Ddtf9bnDYGAEV+YUFQ/L5fUpOluOCf06jik6FrGFWl2Prskd8rBZ4TqTAeFSTaeOzHItfMyGA0q+piSVSakqCTbjSQ5jNhNEnEWGbNBQpYkHBYjURUWFSSQ4zJyvNfL5voRblyQxWeX5CBQMQrocAf4zOOnuHdHB3dtasVokLhhpotzCuP40boyytJsGCSBLMs82hDhhzuG+dHGNnrHQ1xWYiJRDvHRBRl8dGEOAHaTzDPHBrn1qToe3t/Lr7d3kpNg4XsXlPLbq6v57JIcRgNRBiYjyJLgSI+XQFQl2WHC44+Q6jBRleHgqWMDfP7JWv53cyv7B1TWlCdz/zVV3DA3ldpuDzf++Qg3/e34B5JcP4x+qqer5783Zy8tLUWSJNra2t7WnF3VNB4/1Ium6WM0dpOE3SRzVkE8TYMB/JEYaXFmYoDHF8ZpNjAZVrAaxRklVVVVP9BK9cMC/74p0tHnUzuA04xfE3DGF/IHnlRPX/zRaJQTJ07g8/lYsGDBB3KyzUaZi2dlc6DHhz+ssLIsmYZBH4MTYVaUpVA34McXVliUn8DO5hEyxDiSEKTH25AkGZtBr/6UwDj/u7nlPe9/Z4sHJRZjQUH8W14/1eelOMnCV5ekUp7m5Dfb23mtYYjcBCvxtjcg4d7xIF2eIJfOzsJlNfLjV+vpHR6joqKSDQ0eytLs3LaymL7xME++aXb1nMI4Yio4zRKrypJJtJu4fFYG+zvGONnn1V19FufhCys8eWSAJcVJ3L66mIp0Bw/v62UiGEUIwaeX5qNpGn/Z10V+kpU15clcXOZgPBTl2WM6M1mSBBXJJqKqyvy8RFISE0hymFhanMyWhpHpkQ9dvEIjL9GGURakxVmYm5vA5vrhKU9UuLAmjRF/lDirzLoZ6awqTyHeauS54wMwVa1cMTudiVCUFKeJubkuDLLgunlZ9I6F2NXqIRBR2NjgpirDgS+ismFqpEivVlPY2TrKy7VDHOgNkBVnoNUTZDygcMWsdLzhGFua3FwzN4PhSYVBb5h5uS5erXNz+Uwdwl6Y72JkMkpFmh1/KEZJig00SLYbKUm2cKBjjEWFCVw6I4XdrW4+/+Qpfrerm5xEG/dfW8OPLiphcYGL19tHGZgIIwGv9wQZ9EVZXztM05CfTyzK4YHra/jyygLirUa6J/Rz+PqAyp72cXYfb+Kh/V0UxBu4+5JCvrGmiOx4M6/UDuGLKCTajLS5/ciS3jf1R1QWF8ZTkGTj0UN93PLESb71YiOKqvGd80v4zvml/O8l5eQlWjnSM46GhiTrfb51NWmsLk9Giykc79KT63+7cv2g4d+3i1gs9raJ3mq1kpWVRU1NDfPmzSMjI+Mt5uzrDzQQVDQ0TUNTY0RVlbI0BzOzXahA45CPFIcJhGDEF8UxlVRtJhnfh5z9GwgEPkxEpdNYfB2QAZwNIIRInPr3GYtIfCiuvNNwb1paGlVVVf+xldOZsBSvmJOJIozsbvWwvDQZIXRh++WlySB0eLY8LkogoiHiM5mbF8+ILwKyiYACqDGU4ATPH+15y9zomcRrDcPEmWVKU96ARCaCUbrHQlSk27HIcMeF5SwuTGLAG8YgS2/5Tq+3jiIELC1wcVWxRCAKT7cq7OsYZ2gypHty5rhYVZ7CCycGaRyaRNU0nq8bw2SAyYjKgc5xAC6ZmU5anIUH9+i2bKqmz1Wqqsr51XrP8bPnFBCMKjy0rweAFKeZ7AQbEUVjZVkKRlkiyyGzrCSJV04NMjIRor29g2N9kxiEoHkkMN27XTcznXBMZUvDCJqqsb52EKtRonloctrc/PzqNLwhhQaP3qvUVH1R5rIYMRlkTAaZi2rSOdE7QYdH1wwe80cBiXBUna5MF+bHU5Rim1Y08odjfGR+NvNy4nnp5ND0w+jimjQE8NjhfkoSjXxibgIxVePlU0OUpzuoznDw8kn957JUG88eH+TqObq5e6cnQHW6ne0to5xblsThngmWlSRSP+hnbl4cAxNhIqrGWQUunj46yLFuL4lWI4qq+0t5fBG2NLnJdFlwWY3EVI2zC+LJT7Sytc3H9zb38PDBXvISrSzIc2GQJQqSbAz7ouQnWrlydjqjwRiPnvRy//EQmiZxXlkC4x43Rm8PFk1nTs/NctIy4ucHr7Zy65N1PPh6DzMyHdy2vIDvnl/CnevKsRpkJsIKwYjKk4cHaBvxY5R0I+zMOAvVSRKb64e57ak6fri+hWePD3JOccJ05VrXM6on18dOcnwKdv9PxocJ/j0dZ5LoT/uNvtmc/eWWoJ5QYyqaqjEeiFGQaKY4xYYsCZqGfKQ4zDqfwxfGaZHxhXX3Gl/ozNCyD+pc/ScUlf7V0N54kD6JLvgwC10E4jCwFXjhTLf1gSdVr9f7X4F7z9SarTIjjsrsJLY1u0l0mJiV7WJHkxuX1UBpso0Np/pRJSNOi8xLJ4eYkxuPL6xit1lACBKdZgSC0MQo9+3qOOPjC0VjbG0YYiIUmZ6bBKYIClCRZkfVVAyyxKryZITQx1ju29kxNYyvsbvNTWmKlaHuVhZW5nPbqmKahvz8aW8nGU4T86Zg5U+clUeK08RvtrXz0olB2jwhrqqKJz/Jxp/2dhEIxzAaZD65WGcHP3m4jzs3NOOyGUmwmfjL3m5UDXISrFwyM4PtzW7q+ic41DlG20gAp1ni1dohlJh+XNfPz0ZF44Ft9Rzsj+AOqlxQnUbPWHBavD0/ycaMzDjW1w6yq9VDlyfAuhnp+CMqu1s9AMzMiiPTZebAgEJEifHUkT4SbAZ6xkJ0e3Tm8drKVOwmmRdPDhGNqbxwckgn94QUtjfr2xECrp+XxWggyosnBqnJdFCUYuO6eRkEo/pnABLtRuKsRpQYrKuMJ9VuYElxIpsb3YwHFK6ek8F4SOW1BjfXz89iLKhwrGeC8yqS2d4yRpbLQlhROdk3SZxZZnfbKJkuM6+3e8lwmej0BDnU5UWg0TkaYjQU48rZ6dy5JpdSh8YTh/v4zGMnef7kMGcVJPKlFQV8e20Jty1MwCQJjELQNRrgs0/U8oP1Tfzg1WZA4+urC7lqTia/vrKSqozTOsEavz8wwm+OBNk4aOXQQITVRU4uzVe5bZaBi0psePwRYqpG45CPP+/roXssyK5WD+MhhRvnZ3FxTSon+rx866Umbn+hASEE315bzJUlJn59VTUzspycGtAFKVRNYzSgs6XX1aSgxRSOtQ9z458Pc/N/OLl+GJPqv1IRGgwG9nX5AA0hVCShK4HZFB+tTQ2kOyTq+8ZxWaSphZhu89fvDdMy4mc8cGbCFB9U+P3+D02l+iZFpVuBezRNKwOuA4o0TfuDdiYV2VR84Ek1Li7uvwL3nslYzem4Yk4W7WMKm+uHMcmCEX+UG/98hOaRAL4o/PmIh8lwjOZhP3/Y3QkCEBJCkgmEFQSg+MZ4cM+ZJ9U9rR6iSgyrQeKera26Kgo69Gs1yRQmWaerupN9ExgkweWzM9ja5ObuLa00D07SNxai1BmjoqKCuDgXZxcmsao8hYmQQma8vrIFsJllvrCiiMGJCH872ENVmpUFWRbdhzQY5eEDei91fl4Cc3JcPHt0gEBE4Tvnl/HJxXm0uf1sqtMTz5VzMkl1GrlvZwf37eggP8nKbSuKGPCG2dw0qhuOyzHmpUgcG1TY0OKjIt3ORxZmkxZn5rnjbzBx181MZywQ5eH93eQmWLlmTha5iTY21A7qEKMQnF+dRq9P47FDfQz7InxqcT5mozQtBmEzG1hblcaBznGeOTbAaFDhE2dlU5aq9z1Pn9fqTAdZLjOBaIy1lamAvkg4pzSJjXXDeHwRdrWOMRaIYpQFezv19srlszKIqRovnBykPN3BzCwnL5wcIifeQmmqnSePDrClyYMmYGOjm5gGg5MRAlGVaAz6x8MIodE9GsZhNhDTBDazgevnZ5LhMPLM0QE2b6/llrFTnCXGdQavprGvY5Qfb2xhW5Obx05NYJAFP7+8grsvr+SSGam0uQOM+COElBgPH+hjX/soz58Yom7Ax/XzMvnjDTP47NJcZAkOdnl1qc0geIyp5BUUc2QwitUAV5cZqUqS2d7k4evPNbChfoQ52XFcWJ3KDQuyufuKShJtBmKqhj+scNemVo4Ox/AGozQM+siON3NeRQqHurx85dl6vvtyM3872Mf8PBf3XVPN9XNO91wP8+lHT3Cy7/1Prh8kpPlO8a8c02mCkqaB0FRS4iwIIREyx1NRWUlxioNub4S21mbsRjjW5eFo9zjBYIhJfwBf+L3zOv6bEQwGsVqt7/7G/06cTqqZwGwATdM63ksyPR0f+JUnSdJ/pVH+bgIQb46LatIxmMz8flcnR7rHkdBwmeFTZ2djMUrMy43nW2tLkSTBwlwXWcYYwagKsomQArIEQtOYGB/l1Vr9Yc/EBNL69UjPPw+9vf+wzw11wzjNgi+sKKTD4+fPezsB3VC6OiMOoyyjavqi4FjPOOXpDj6yMJdPLs7j9fYx7trQiCTB5UsqMZne8Gcd9UcwG2SOdI+zrXFk+vXKDCepThOKqjEz04GmaRSnOlg3I4PN9SOc6vOixFRCioqGRnaClbxEK0uKEpmR5eTRgz2MBnRBik+enUf/eAhfROHLq4qYl5/A7Jw4Xjg5xOhkkNbWVj6ytAQhNCbDCh8/Kw+DLHPJzAyahwM0DOgJa062iziLzGggynULspEkwflVqXR4gjRNOfSsKE3BKMGm+mFqsuI4uyhJF95v9UxrnV5YnYZBFrxaN0JluoPqrDiumpOBxx9la5NeraoahBUVgaB1xD99Xq6enQFC8NjhPh492Etxso0LqlM51BugfyJKRpyZpcVJvNboxu2LcO28TPwRldtfaKB5JEBMhbQ4MxdWpSKExCcWZlOeZsdskLh5SQ4IwTnFSRQm2wgrKp84KwujBC+eHOLa+ZlckRpjV8TGzUo5+4nnEsMofyiPcvXsdAa8QX6/u5Mhf4zCRDOjgSgZLjMGSYe3z69IYW15Kq0jfu7Z3skTR/pJthvJjNPfU5nuYDyokOE0cfHMNAa8YX69vYNbnqxncCLC9fNzuHJJDV86t5TrZsYDGkYJjvZM8PnHT/HUkT5+u6OD8WCUr51byBeWF2CSZV5uV/j2y80IAV9bVcinFudw79VVLC5KoGnIN125Dk+GuXhGGtfNy0CLRTnSMcL1Dx7ms4+fnFa/ej/i/5ZK9fe7OnWCkqagAQYhsBgltjS6UTUoy3ARUiA5pxiLyUizO4hTCiOhsSBNIqqo+AOBd93PBxWapn1gJKm3idPJsxP4mhDiNiHE+UKIFUKIZUIIy5lu6ANPqv+tOFP4FyDRbmJVZQY2s5GvL7CxONfOZARWlKVxTkkyp/q8VGY4KU+x09vaw7f3PKxjirIBDY1YDBAQ83v56aYWGBvD8IlPIP/oR8g/+xnGj34U0dg4vb9gRId+F+YnsLAgkUtnZrCpfpinj/QxMBGmJsuFkASqqjEWiNA1GmB2TjwAq0viuaLEyFhYxSgJApE3FladHj/Heia4fHYGM7Jd3De1SADY3uRmeDJCutPMs6fcjEyNs1w7L5vMBAv37mjnvp0d1A9MsrQ4kdYRP6+3jyGE4OalBURiKg/t6wKgze0HNCTAYdYH0D+2KIdQNMaO7jCVlZUIg2nKA1Kbfs/KsmTizDIvntQ5ADH0ZIcGiVM6pueUJGEzSdMEIptJJs0uEY5pXDpLbxdcNCOdqKKxsV6vnl02I0XJNkKKynmVusF3TaaTslQbL5wcJBLTONrtxe2LUpZmZ0PdCKNTUFmKw8T5VansaRtnPKjwibNzuHRGOhajzMuNelV15Rw98T59dIBOdwAhYMQX5do5GZxXmULfeIhlxYlUZzh48tgAN8zPIqZq7Gkb45IZaexsHWVBXjxpcSb+dqCPjyzIItVu5Jdb2jnq1QCBEPpdvkd1sdejsqwkkSSbCUkSFCcYaHGH+NGrLXzykZM8c3yA2TlxfHxRNh9blM3Xzy3CKAtcFgOhqMIvtrZz82Mn+MYLjSiqyjfWFHPj/Cx+e3Uls3NdOgogCf60r4evv9DAk8fdPHFynOIUOw9cX8Oti9NJtQmePT7EqQEfuS4TdlljcWEC/3NeEQ6jhiQ0AmGVb7/SxOOH++kaDXG020tmvJmLa9JoGJjkGy828t2Xm3hgj24cf+9VVVw7O4XjnW6uffAwn3viJA1NvRieew7DH/+IvGUL/As2bB9G9u97TapHu734IjE0VUNMIWzjwShl6U7GQwoneifITbQigOePD+KZDGEhyroZmSAg0RWHpmmcamjm4MGDb2vO/kEqYX0IVbhOH5ATaAPmArcAX0NXVIo/0239P5VU34uiydpSFyHNwKTs4qK5+YRjGq+3eVhRmkw4BvvaR1ll8dNnimMyIYWKyUEsqMiyUU+wCLRYhL4hN01Prkf096NlZqJlZKDFYsi/+930vrY3uwmEoywuSgR0Td15uS4en7J5mpkdhyQkNFXleI8XNMHsnHjGxsZobGqiLD8LaQp+vv25Wo72jAPw/LEBLEbBhdXp3L6mhLwkG794rYVDnaP85fUuKtLt/ODiCgySxEMnvEQUFbNR4tblhYz4FHY0u7l6biZfXFlMYYqdB/d2MhlSyHRZuHxWJrtbR3nqcB9PH+nnrKJEVE3l0UM9xBSFwEgP8zItHBmBsWCMP+/twijLGGWhz5ICJoPM2qo0DnWO0T8eZFvjCJOhGGaDxMap0RmrUWZFaQr7OkbxBqNMhhRG/LrQQ/tUhZkVb2VenosNdUNElBjhaIy+8RAS0DQ0tVIXgqvnZDAaiLKt0c1LpwZJdpi4ZWkeigZPHx2c/n0sLU5AQ8NllSlKtuEwG1hbGk+9O0pdv862PK8ihR0tY/xhbw+lqXaMBpmBifC0etGf9/Vw0+IcYiq8cGKQT56dQ8OQzrCdn+viySP9rKtKI8Vp4r5dXTr1UIP2sIwdlTtcw3wvbpi0mJ+Hho18/olaWtwBblyQzecWJPCLC3O5sCpF150GjvVO8vkna7l/dyd3bmrFZTHw88vK+cP1M/j66kKMssAfUQhGVb7/ajMP7uvmwb29HOvxcumMNP54wwxuOisHNFhfN0wkppJgM9LhCbKkLIPybP3aLEm2MjAZ5QebOvnKk0f55vN1hBS4Y20xd11axozMOF44McgPNzSjIbh1WT43Lszm3mtqWFWWTNOwj5iqYTVKjAUiXDojnc8tzQUlwuvNQ1z1VAu3tRppkRxIJ04gb9r0Xm5z4MPJ/n2vSfXnW9qmftIwGwRmWRccWZgfj8Ms83rHGNnxFjQh2N44RIpVxSTrs8caYDKZQNMoLK1g3rx5b2vO7vV+8EIdH7bFj6Zp30KXJrxlyjruQk3TFmqaNvguH52OD/zK+2+d1DOFf0874SRFhkiOd3CwL0hJip2seDNbm0YoS3OQ6TLrcoa2MGYlypakUs4bbSEsm0E2AuL0HxTfGF8dcqG9+YYym2FsbPqfL50cINEqU52hqyNJkuCr5xYTZ5FRNTjaPY4QAlVTOdrtJcFmxBD00D8wQFVlJQd6/FiMMj+9rJKUODM/2dDEn/Z0srfNw9qqNBwWAzazge9eUEaCzcTPNrcRUmJ8blkhaU4znzk7m36fyh/36JVntyc43WAoSbEjS4LPn1PIZDg2Pbt6+exMEm1GnjjSR3GyjS+tKOSimgy2N7rZcrCW5ORkPr60GNC4b2cHh7u9XDsvkzUVqexoduP26RXI2qpUDLLgpZODPH2kj7I0OyvKktnTOjqtXXpeVRpKTGNL4wgvnRwgogqKk6xsrB+aZlivm5HORFBhZ7OHTfXDTEZizMpxsbXRzcTUzHB1ppPyVBvPHO+naSjABdWpZMZbpohFHjqnPFc31LoxCIE3pHJ4SuZwRbGTBIvMwwd6iKkaoaiKJvSK+nsXlHJhdRq7Wsfonwhzw/wsmkYC1A/6uHZeBkd7J1FVjeXFiTx/fIhZWU6SHUbu393FwESEmCboGg9TmRnHzWfnIBllfuhNZX/IwgVVqTjMMrKkz0E/sr+He/aN8tRJDxvqRyhNs/P7a6u5dVkuOQlWtjd7mAgphJQYzx4fpK7fx66WUbyhGLeek8/XVhZSmeFga6OHzc1uzLJeEo/5I8zNdRGMqjjNEsuKE6kb8PGjjS3c9OhJXjo1xPKSRH5ycTm/v24G18/LYCgInqCKWYZDjV34BjtZm28g3mrAJOtaxXe83MwDe7ppd/s52DVOit3MuupUmoZ83P5CIz/e2MKvt7eTFW/hV8tSuErrY3fQxLr+NL5hqGKguQvO0Anmzffwh/BhfcbH5AtFpxWUNE0lqGjEVBUQlKU5mJcbz5HuCR4/PIAajZBi1VhSlMhkODblR3x6P9qU0pj0FnP26upqLBYLvb29BAIBamtr6e/vJxT675nSf9gq1dMhhFgK/A44JISwCSGKhBC3CiHOWCbvA0+q8N9JrGcC/yqKwokTJwgEAixauIBLZ2dzrM/PZEhhRWkKTUP+qZlVfX51LLeQJZ5W9sQXMGe0A7sSwmKQdehO0xBoxKJ+GoIq/UY7BAIQiSC8XtSVKwFw+8LsahpmcWECkvTGeTDKEqqmz2s+tL+He3Z0MR5UON4zRpELYqpKRUU5MWRebx/l7MIEchJt3HVJFctLUni1bhhVU1lenDy9zXibiQtrMojGVAxvSvKzc+JYmW9hW5ObP+zq4IE9nczMcpKbaOW+nR1MhKIUJNtYV5PO1iY3tf0TjAUihBUVTVOZnRuPySBzbpEdq0Fj97CBpKQkUhxmZqfKHO/zkhln5sKadNbN1PWbXz45OH1MS4uT2NY4gicQ4foF2ZxfnUY0prJ1qgecHW+hJtPJxrohXjk1RE2qkUtmpOHxRTnQqS9OqjPjKEy28cKJfl44MUBNhpOPLsgmElN5tW6qlywE18zNZCKsYpQlVpbqUo5XzErHbpL56/5e+r0hdrZ6WF2ZSk6ClYcO9hJRYpgNMuvKnXSOhfnFlna2NHmoznAyFlQ43O3l8plpJNqMPLi3h6VFCVSmOXj0YB8L8uKpyrDzl/29xFkMGGXBA/t6GPEryAb9AXjD/EyWFsZTNzDJ4W4vP7i4glXlyWwOOfhZXRCE4Dvnl/CH62r46KIsJsIKO9smiKkaFlniVN8kZSl2xgMRTLLMtXPSqcpwsrPZw52bWtnXOUZRso04i4FZOXEsKUxE1TTy4s2Uptp56dQgX32ugVufqsMbjPDFFYV8flk+f7iumhUlSbqIgCbY0+rh/l1d9IzpCwZVhQurUkiwCF7tiHLPsSi/2DtKIKLwsQoDX1vk5Ow8Ozua3fzw1RZC0RifOyePj0xVrmsrUzjVP0kwqpEZZ0aTZK6Rh/le8hhGJcILYxLnBav4+c5uxs7AG/R0fBiTKpz5c+4XW9vRE6qGAGRJQtFkJEmQ5TIzMyuOYFRl46k+UixgNUgkWo2ctlUWwGmGUyD8j888k8lEeno6JSUlxMfHk5+fj6IoNDY2vu/m7O8U4XAYi+WM25T/zfgl8BcgCkQ0TWtDh4HPGOb8UCTV/0a8G/w7OTnJwYMHSU1NnZZGvGxWJjHZxK63zKy6WV6ij7Rsc2usOn8BQdnIocQCFkUGicomkAwIJF1lSUjEJtzcuPqrYDQiIhFiH/0o6kc/CsBTR/qIKgpR9a0rt1a3H18kxg0LsvnYWbkc7Z3gV4f8+CMqc3ITyc/LQxIS+ztHCUZiLC/Tk6fZKHP1vCwkoSEJiR9saKSuX+8FjvkjPHWkl/wkK2ZZ8P1XGhmcDCNJEufmmylKtrOxbphsl4nbzyvhSyuLmAzH+MPuTjQ0rpmXRVqcbuv2/VcakIDqDCevnBygvr0Lr3uQq+dmUz+om44LSSIaAzSNrHgLBkmQ6jCzuCiR1xqG8U8pvpxbnkJU1ciIM1GTGUdugpWqDAebGoY5fVrWVqfh8UeJxFTOLbAwK0snWq0/9YYf6sUzMxjwhvGGFK6cnUFWvIWF+fFsrh+Zfrgk2fXerpDeeMg5LAaumZtJw6CPP+zuxiBJXD4znU+clc2IL8qLUyM2s9LNZMZbONztZXFhAt9aW0xOopWHDvQihOCji7LpGAuysd7NZ5bmoajw2x0dxJmNhBWNl2pHSIkzYzMaibca+Mm6MnITLDx5ZIBZ2S4+dVY2tX0TfOulJl2gXggy4sz4wwo/2djCIwd68fiieEMa+QlmLqhKod8b4ne7OvncU7V0jYZYU5HE6vIUvryigHPL9WsiL8HKgDfE/25u45OPnuCXW9tJc5q544IyvnN+KXdfUUWS3UgspjOU79rUys9ea+OJI/1sb/YwOzuOuy4pY2V5Cvs7x7ljfSvHeidYXprE9fMy+FSNmW+uKULVNCYjKqomGCSR9NRU1hTacZoEBkmgxjR+vLGFB/Z2MeILc6J3AqtRYkVpIsf7JvnS1gHus5Ry95ANEzG+L7WzOAkefr2T836zj/t2dZ6RSPzp6+H/xNA0jWeP6WYSqCpGSecZ2MwGYhoc6PKyt32UWDRMRbKJ+fkuBicjuGxGhABlCrk5LZsaiL5zYjyt++twOMjNzWXWrFnMmTOHpKSkfzBnn5ycfF+ryw+TQfnpEEIYAFnTtN2AommaMvVaRNO0M15h/D+VVN8J/u3v7+fUqVPU1NSQmZk5/XppmoPq3GR2NLtJtJuYmRXHziY3CVM/72h2UzKnnCyXhc3z11JxdhkaIBnNgIaqCpwmmVgkSNtEkB13/5Xo5s2on/88GAyoqsbzx/pxGjV2NI1Mw5QAx7u9CCExK8fFpTMzuGNVNhZZt5h76tQoL5wYwBdW2NowTHqcmar0N8zLnz7ahywJvntBGVajgTtebuCpw73cv7uDUCTG11aX8P11FURiGt97uQGPP0rPhELveABZEkyEYwQjMfKT7Vy3IJt97WPsbPFgNkhcNz+boYkwHn+U71xQymeW5hNWVF6qG6e8ooLzazJIdZj428EeTvR6OTESoyDJxtFury6SAVwyI4NQVGVzg943PdE7AZpGOMobSbQqneGJMMem+sMlKQ5UDVxWA6l2A0LABVXpNAz6aHdPsYdz4gENiyxQR7vp7e1hdZEDf1SdNnh+pXYIoyQRUTTW171hKr+qTJdPbBj2s6IsiXibgepMJ4uLEnn+5BADExEGfQojkxGEJCFLAoMk8emzc/H4FZ482s+i/HjmZsfx1NF+vMEouYlWGoYCHO6dYG5uPAJBfqKNH1xYQjSmcc/2Dm5bnk9pqo17d3bRMxaiPMOBPxJjLBgjJ8HCl1cW8OsrK1lemsieNg8v1w5hNggW5Dq5bFY6P7m4lPQpdq/LIrO+dphPP3aKzz1Zy6t1w5xTksjPL6vgjzfM4Ib5mSiKiqZpDE6Eue3pWn6zo4Ofb2ljPBjlG2uK+OXlFVxUk0rjkI+XTw0jCUGmy4zFKPGJRdnMyXEBGnEWme3NHm57pp4d3QpPHOknoqh8ZnEOZxcmsKF+mK+/3Mb/7nXjVwTfPb+IH6/NYWGmmW1NHm5/voHBiTC3LM7ilqX5/PqqSpaVJLLFa2AQM/PTrBSct4TPXljDnRcVQyzC77a3svbe/fzp9e7pBdn/bfHAnu5p7W5Z6LKiAOFoDKfZwO93dXKwbYRkqyCmaWS5LIQUjdMgVySmomkQU/VKNxB596T65pBlmcTExH8wZ+/u7ubgwYPU1dUxMDBA+F8gkL05PoxJFTAAJ4UQ5wEOIUQ8cCkw9F428v9UUv17OENVVerr6xkeHmbBggVvO4h8xewsOrwxOj0BVpalMOKPUNs/waryFNz+KKf6Jzi3IpWm4QAIoY9PmE262pEkMBtkJCFQvCPc/Lcjb1FZ2t3moWtkkstmZhJWYjx26I1RmyPdY5Sk2nGYZTq7ujCEvRgkieJUB9kJVh7a38OnHjlK7cAkM7LjOL0w7xsPsrPZzdrKVGbmxPPzK6pYWpLME0f6OdQ+ymWzMsiOt5KfZOeOi8oIRFTuWN/Cn4/7iLcauOOCUkLRGD/b3EJUUblkRgYV6Q7+tLuT+oFJHjvQi0EIVE3DQIzR3jZWFDs51B+iezSIUZa4Zn4W7SMB7t7SRrIFvnqu3lt96YTO8i1ItlGT6eTV2kHcvjAvnRykLM3BaCAynUQXFiSQYDVOa/2+fGoQWYLxgIInoM9uripPwWKUp91rtjQOgxCEYiAlZOF0OokTQYpd8OKJARo6+9nR5GZ5bJgF4SFeOtLLuFvvmcoSpDpMgHjLA/tjC7OxmQw8dGSEh457sZklLqhKZXfbGCd6vZSlOVhdkcyG+hGaRwJ86uxcVOB765vpHA2S5rSABlfNyeD6+Zns7RjnQKeXb55XxHhA4WevtXLtvEyy4s1sbhihdsDPBdVpfGxBFmOBKN98sZG/7u+lZciPqglmZsaRZJV45qSbzzx6ki88WYt7MsKXV+Tzh+truOuScsrS7Hj8EUCwq8XDN15o4I97unn62ADpcRbuu7aa/zmviHm5uvxhz1gIoyw40jPBZFihNNVOKKKRHW9mdk4cm+rdfPmZej7z+Cn2dYxx9ZwMHrh+Bv+zpohMp4UdvQodngAzs+KYkR3H58/J538vKcNmkpgMKSgxlX0dXlwuF59bVU5Rsk13jgLu3d3N/ZtP4B7sY8wXRBKC0lQb2wej3LZ9mM2Nwzx/cgh/JMb1c9LIi5O557VmVv9mH/ft7GT8PcDC/0r0jgd5YE83X32unjs3tdI1GvyP7UvTNB7Y08lpubDYFJxrNQpU9KmEQCjM8kIHiwsTaB0JkuQw6WIzUb02Pa0+FpvCgv3/JKmeCXnq783Zc3NziUQi1NfXn5E5+ztFIBD40KgpnQ5N00LAT4GvosO/64GvADe9l+18KJLqB9FTDQaDHDp0CJvNxsyZMzEYDG/7uQtr0jCbLWxrGmFBfgJ2k8S2JjcL8hNwmGW2NukShgYJjg0rrC7XLeFU2QSqinsyjMMsoykRQl43n33sxPS2H9zTSYJV4sKadM6vSuW1hmGah32MByK0ugPMzo6joaERgyxjS8tjNAznlqfyo4srufvKGjJcVjQNNjeMcOsTJ3nmaB9/eb0boyxx+Wzd2cZqNHDVbH0WUhKwoW6IY73jABQlO/j00nzGggrhmMYXVxQxI9vF55cX0jTk5y/7upEkwW0rilA0uOOlBoJRhW+dX4JZFty3rZn8/Hw+tqQEh9nAw1MG5EuKEjHJ4A8rXFFiJCveyrKSJLY0DjMxJZ22bkY6Hl+Ue3e0E42pfH55AQk2w/Rcr0ESnFuZwpGucZqHfWyqH+asQt2tZX+/bsZttxhYXprMrlYP/WN+nj3SS0miGYdJZn29B5crntzcPD62tISAovHYETdRFVYMneSySAdRTfDkK0cBGJoIc6JvkuIUG7vbxqYVf1xWAx9flE3nWJQhX4wvLCvghvlZZCVYuX9PN76wwvXzMkmymbh3Ryf37erSf/9IrC5L4c5LyomzGvnl1naWFyeyqiSJ504M0joS4KvnFuL2R/nuK80MTISpzorDJEtsa3ajAndfUcmSwgSOdnvpHA2REWdmRWkSX1yYyM2L0jAZ9NEiRdO4Z1snP9rQwt8O9NE05OPc8mTuubKCjyzIRtM0trd4CCu6ZdyGuhEMksAzhRysq0llbq6L3a0evv9qMz/b3IbTInPb8gJuX13EvVdXUZBkZXwKSdnUMMIzRwdIdZiIqCqSgKoMJyf6JvjiU3X8ekc79+7sIhBR+dzSXFaVJbO1yc1tT9XylWfraXUH+MKyfO6+vIJZ2fHs6FH44e5xjvUHWJMr89EywZcWJ5Fsl3nw9V72tY+xpCiRdTPS+J/zivnhBUWIWJTf7Whl1T2v8+MNLbS73995zL7xEJ/62wnW/vYAv93exsmecZ450sslvz/Iq29CON7PePb4AOGYhqaq6E1R3QlJ0/Rpgi73JHZZRxkqpkzvQ9EYGvr8twACkRgSmi5y8i9Uqv8sTpuz5+XlMXv2bObMmUNCQgIej2fanL27uxufz/euUPGHVEwfTdMaNU1bg16hXqtp2tmapv2jsMA/iQ9FUv1vxJuTqtvt5ujRo5SWlpKfn/9Pk3qCzcSqqkz2tI8hBCwuTuJAxyjRmMaykmQOdOivL8xP4NiwwsxMBy6rAclg1hmbks7gFYDiH2PbqS52t7g50j3O/jY366pTMcqC6+Zlk2g18NttbezvGENDIj7qITMjg+zsbH0/wIL8eAAy4y2M+qMsKkjgC8uLSLQbeexQH4e7xoi3yDQOThJTVaJKjLu3tGAxSHx/XQUJNhM/Wt/E3w700Dri48E9nbjMMgYJfrujnVF/lMVFSVw6K4ON9UNsaxzmZJ9OilFUlQuq0kg3BlmebaBtQqPBo+AwG7hidgbHeyc42efl6aMDRBQNhMZ4WL+5LpmZQVhRp43F5+TGk2w3cqJ3grWVqeQk2FhTkcrx3gkGJ3QW4uryVISAP+3uRIlp3Lggh7MKEznUHyak6L/Li2rSUGJw98Z6/IrGJ5YWT43peOn36tspTXVQleGgeVxhQWSEiowk8kywOtLL1lg8Ww/W88TBLmRJ8KWVBeQkWLl/d/c0+9hlMYDQZ0ftJhmjLHHrOXlMhlT+uLcbq1HmkpnpDPoiNAz5+fTZOZxTnMDGxhE6RwN8bVUhEyGFX2xr5yMLs1iQ5+Lh/X384rU2FFXCatIZ43Nz4/nF5RVUpjt45EAvX3y6jj0d42QnWLl4RhoxTeM3Ozq5Y8cIfzo4hNko8/PLyrnr4jIuqkmhbSTAqcEpa8ARP/s6xrGaJAan5pFvnJ9FVryF9bXD/HBDC3WDPmZkxbEoP4EvLMvn5iV5CE1gM8l4g3qVfMfLjdy9vYMOT5Cr52TwzTVFFCfbefb4ILc9XUfLSIBz84x873zd5ebc8mT2to3R6QlQlGKnKMXBpxfn8MvLK4i3GRnyRTBI+iImyW7iq6sKWVaaRFhvvlPnNTJuTKIoyUplvH7t2IwSe9pG+cbzDRzqHOdIzwTjQYVLqpNZlGvnmcNdXHz/AT760DE21A0Tif3rvT9fWOGe7R1cdN8B9neMkmST+d011fzssgoSrAa0aJjbn6vjpZPvCRE8o7h7a4cu9oCG1SjrEwRCH6XRFIWzcmxcPiuDxuEgVpP+/wPeMEZJ4PFHsRglJkMxrCZ5ykv3nyfVf1d5SpZlkpKSpqHi8vJyDAYDnZ2dHDx4kPr6egYHB4m8DXv7w6T7ezqEELIQYp4Q4jZgHXDJlAjE5e9lO/9PJVVFUWhtbaWjo4N58+aRkJDw7h9EHx+ZiMoc6x5nVVkKYUVjb5uHVeXJKKrG4wd7cVgMBBX45bYOEu1GkGQ0IYGmEYyqU/CsQJkY4ea/HeWHrzRgk1XOq0oDdFGDL6wopG88wFOHe3GZBctml08f4752DzlOQaJdd6XZ3eJmMqxwUU06K8tT+MG6SvISrdhNMpGYxk83t/DZx47znZca6PAE+MKKQmqyXPz0sipWV6bx3PEBvvFsLQZJcOelFXys0sxYIMr3XmlgPBjlhgXZVKQ7uW9nB/fvaqc63U5VhoMXjvfT75nkYytnkJNg4S/7uogoKmur0vQxkZ3tPHO0j1VlSaQ7TezsjaJpGjmJNubmxvNq3RARRT8fJoOEpqnMz4sHYHVFKhKCzfV6/zPZYWJGVhwt7gDLSpNId1lYW5VGSNHY267DtnbCFMdDu1elJjOO8nQHF1SnYTRIvHTyjdGyjDgzmgZJsRAC3Uj6IwySGgvxXHuMAz1+ZqcacHe3cnW5FV9I4Xc7OglEYjywt5sUu4E4s8w92zsIRBQKk+1cNTeD/Z1e/ri3m4cP9OIwGVA1gVGW+PTiPLJcVn79WivxWzbyBeswLUMBfryxRdd2FoJQTHBWYTy/vbqKObkuHj7Qy327ulhSmEiC3Yg/oiI0gaJqpDlN3HVJGecUJxBSIBbTmAgq/Oy1dna0eqgf9BGKqpxfkcy1czKQheDJI/38fncXsZhGTZaTklQ7ty7LIy/JiiSgMMlGXf8k332liZsePcVvd+g2cPddU81911RzzdwMusZCNA37MEjgDSokWI18dmke2fEWDLLAKAle64ry/fVNNA5O0jLsxyAJzipMpHs0yNefr+ee7R38eX8vbl+Ey2emMyfHxVNH+/ni0/X8alsHO5s9rCpL5isrColpGr/c1s3Pdo/wUmuI2dlOfnVJITfUxDEZDPOLre08f3yQyjQrV8/J4JZz8vjNlZWkWOB4p4evP1vLl3YE+M7LTexs8UzP8b5beINRfrW1nbX3HuBPe7tIdZqQgG+sKSZ5bIiXn9jOoDfAVwo0qlJN3PFyw7Ru9bvFmRB81tcO4Q1Gpt6rEWcx4DTJU3QjHQ42G2WWlyZhlCUOdI6T7TLTMhwgLc7M0EQYl8XAeDCK1STr1o6a9k+///ttUP735uzZ2dmEQiFqa2unzdkHBgYIBoPvS6X69NNPU1VVhSRJHD58+C3/d9dddyGEaBVCNE31SN8x3qT7Ww08iC6mHw9kASVA7ns5rrfHPP/L8d+Af1VVZWhoiMzMTObNm/ee9rmkKJHUBCdbm9zcck4+iTYjjx7o0VeLCDY2jCDQUDVoGg7isMg6tV02oiohNE0gCYHQNCQ1SmB0kPpoBFSFyZCCxaEnyupMJ+cVWtnYFiQ7yYls1OUGe8eCdI6GWFtgnPouGi+dHKQgyUpVpk5Q2twwTJcnwJdXFXF2YSJHesZ5ZH83TcM+jELjRK+X3EQrqU4LC/MT2NE4TESBiKJblmU7Jb5zQRk/Wt/ID15p5Oo5WQxPhFFUlTizgVuX5VLX1MZ9btjYrTGnSuKTZ+Xxg/VNrK8d4rJZGSwsSOCVU8PkJVi4eWkBO1vc3Lerk+O9XmbnxHPxjHS+90oTu1o9xFkM9HnDmGSJnS0eZmS7SLTrbMatTSNcOy8Lk0HCKOsLk/wk/QasSHeQ6ZTZ0uRhfoYRj8dDfnoiLeMeiqfcfVxWI6vKktjc6ObK2Rk4zDIHOseJsxjYJWdz/WALDqFgFYJbVsTzw1q9ov3oOWUkWg1MTk5yYUmEF5t8fO/FWkZ8Kl9ZmkYkHOb+Q+Pcu7OTr68u4pIZaexp9fBao4fcRAvfOb+Ee3d08qfXe8iKt3B7dphvnYzyYyWDbyjNFEkWWtxgkiVuXZ5PhzvA+roRxgIRbl2Wz4xMJw8f6KNu0IfdZODLKwsQQp/h/dPrPfx5Xw+qJihPNnHTogw6JwU7WzxsrBtGQ2A3GYipes/abjbQ5g6Q5jSRFmdmR7OHLY1ufdxCCC6fmc7ls9KJKBp/3d/LzlYPaNA9FuKO9c2cXZhAbd+knqgrUwlFYuxoGeW1RjcGSaBpuiRhSbKF5w62sX8wwq+3d4LQvWivnZdJIBLjpZNDvFw7TEzVKEu1s6YihSS7kaZhP/fu6OT19jHsZpn5efHMyXExN8/Fowf7pkhkOhksiJmL5xdTkDXBTza2YpCgfijIF544wTm5VryKgWFflM8uzSXZYeKVIx1sONnL88f6MBkkZufEMzPbRUmqnUyXhTiLAW1K8L9l2M+BznF2tbiJxFTSnWa+tDyfX2/vwGGW+f4rTdgiAUaJ4yxpnEV1tVSdtZhvB0x89Zk6nrl5HqnON2RB3+nZ824V4Q/WN+sNVA1AEIzEsJtlvV2iKpSk2tnROs6MTCfz8xLY3TrGvJw4DveMU5piZ3AijMtqYCKkYDNKhBQVg6QRiLzzxMN/UiP5zebsp0d2xsfH2b59O3fddRdOp5OsrCwaGhooLy//l3JAdXU1zz33HJ/5zGfe8np9fT1PPPEEQBW6lu8WIUTpOzF436Sm5ABOapr2yfd8MG+K/ycqVa/XS2NjIzabjdLS0vf8CzTIEhfPyuZwr59PPnKc0UAUb0g3B15UkIAAvntBGReX2VE1je9fWM7SkiQw2UCNIAmhX7wCHZYJTqB4R9BUlVdrdRgpHIlQX1dHVrwVIcn0eUN89ZlT7GgeYUvjCEJAVZK+qjzcPU7PWJCLZ2YihMDjC/O3Az1UZzpZWpyILAusBplBb5jKNDtLS5LZVD/M5x8/ybdeqOMnGxrJTrBw56XlJDss/HhTMzt6FMrTHNxyTgHdYyF+uqkJswy3LM0nGFG4Z1MDM0tyuX5+Dke6vexrH2Vmtou5uS6ePdrHziY3m+qHMQi9CjXIgmUlyTgNTCsoVWfGkZ9k5eWTAzy4t4vcBAsrS1PY0+aZZj6vrUxjMqTwescobp9OXLKZZfa1n3aYESzJtdHrjXCie4zCkjIOdXmxGiQOdI1x+va4qDoVoWm8cHKITQ1uJsMqH1uUg1+VeHL+hUQvvgTlUzeRVVOiMyeFoGXYjyzLxMfHc92SUmZlO+n0xihPMeOIjOGMebm4zMHRnkkePdTH9iYPPeNhzEaJiaBCNKbypRUFJNiN/GxLO4bGRr4hdzMkTHzRUEOn7GR+vISiwvpTw5xflcoXluXTORriK8828tSRARQN8hJ0XeD7d3XROOhjXU0acRYDKgJZQJM7ws939FM/MEnXaBCr0cClM9KYkeVkV6uHuza38ce9PcRZDFwxK5OvrCzix+vKiLMYEOj96mePD3LTYyf54YYWdrZ6mJXl5L5rq/jU2dmYDRJPHhmgdmCSTJeJslQ7Ny3O5fsXFmOfghZjmsYf9vawod5NnkvGJAuMBkGy3cRLtUN89bkGDnWP0zkaQNM0qjLstHsCfPHpOh4/3E/bSIAhX4SiZCtxZpmfvdbGjze2cKBznB0tHjJcFtbVpHKyb5KvPlfPz7e08ctt7WTGW7j/uhq+tDKfeIeVF5oDbG3zkuGQsPiHSFAnWJ4pMMmCFJvMyiIXI14/f9rVztefreXGvxzh4vsPcsnvD/Hxh45x58YWTvSMkxlnQhKwpiKZ+3Z1oqoq8RYD5yRAVMgosRitqpne+DQcrc18ZWUB/mCQb73UOOUS9c7xbsnr4QO9+COK3ktVY6BpTIQU3L6o7guM4ONn5VCWHsef9vUwJzcOX1TFZJDwhWM4zDJDk2FcFiPeYBSbUSYQiWGWpfe1p/rvhMFgIDk5mauuuoqjR4+ybt06HA4H3/3ud5k1axY33XQTtbW172mbFRUVlJWV/cPrL774Itdeey2apoU1TesAWoEF77QdIUSGEGIZUA5IQoj5QogcIUSKEMIhhHhPefJDUan+p0LTNHp6eujv76eyspLu7u5/eVtXzM7gTzubmZVp46KaNO7c0ExBsoOr5mRyvOcY+9pHWZ5tY1NbgI11w1w8M4M9bR5Uoxk1EkIzWXBaTEwEwmjRMIoSBoORDbX9nFfsYLCvm4KCAjYeHMZlM3Hr8iIe2t/Fr7fpfRZJEvzhRIgHTh3GF1bQNPj1tjZ+va1tqkrW8IUN3LO1HYdFZmvjMOlxRr5zQTk2k8w1c7P46aYW6gf08ZWcBCvJdgt3XVrFA3s62dE0QtMTJ/H4I8ioIOk6uDWJsCzHwLbuGEeHolxYncbOZjd/3tfF7Jx4PrIwmy8/XctvdrRRmGxjRWkKf9zbxeGucebnJbAoU+a1rklah30Upzq4sCad3+1oBwQ/WleOw2xgc8Mw25pGuHRWJjVT9m6b6oZpHPCBEFxYlc4zx/rpdPvJcpnINwewGwXHx2QijW68IYVr5mbx5JF+DnWOMT8/gWSHiRWlyWxrcmM1yszMcnJOcSL1g5NsaBll5fwKvb94sA9NCPITrPx+dzcZLgsFSTY0YCKkYJBl2kcjRGsySZBCrMl1MjAZ5eVTI4BGRYqZjyzI5s7XOrlzYxvfv7CEb64u5o71TXxbKSRZCxOT9P5XUizEx/PjODctl9/u6OD2Fxq5fl4mVRkODndPEFEFaU4z18/PJD3OzPMnBtnUMML6hiGsspEb5mXqhK8THezsDLGt2QNCkGo3IkuCC6tSkATs6xgnO96ML6Lw+z1dPLCnCwSYZMHXzy2iOsvJid4JHj/cR5vbhyQkmkb8PHNskJpMJ2FFxSDBvLwE2t0BfrOjE4tRIqZqmGTBXReX4YvE2Fg/wvOnRohpKmZZ4vPn5LGoIIGDneM8dXSA3+/uAgSry5O46excRnwRnjw6wAsnh1BVlZwEK3dcUIrJINjc4Oapo/2c7JvEYpK5ZWku5WlOLqlJ44kj/Wxp8gCCGZlWfOEYZxUkkGQz8v31LTjMBgZ9MX59LExV6gSD3iiBiMQnqq2UZFhwzE7ljo0djAUUPnl2Dhqwo9nDqQEf31pTRFqcia8820CC1cAjB/vQNF1z+tZlBdTvq2XDUIzVFi+HIha+G8zhR1YPWfEWlhTFs6V5hCcP93Pd/Kx3fHb8s6Sqqiq/2tLKlFoDQpKwGCRCikY0FpsmLTnNMrcuy+Prz9Wzs9lDos00ZRGpowZhBSxGCW8oRpLDxHggitkg/in8q6rqOxI0/9NhNBo577zz+PjHP46iKBw6dOh9s4Hr6+tj0aJFb36pFx3KfacoBL6JbkSeDTwK1AIykAQ8C/zqTPf/oahU/xPwr6IonDp1iomJCebPn4/D4fi3FEKKUx3Myk9mNBBlVo6LeXkJ7GgewWSQWFyUxO5WDwZZZmGuk50tbtLjzJSnOZDMdjQlghryMemd0BOsXhoRnRgmGAzy173tVFRUYLLaOdzt5ayiJOblxfPrq2dy89I8hCQxKyeB8iQjxalONCGYn5fANXOzmJEVh6ZBWZqDOIuJ2n4v62uHCEVi9I2F+faL9dy3s507NzbR6fGzpiKFC2vSeb19lM8/eZK/7u/CapSQhEb/eACDpPGjdeXcvCSfE/2TPHyon5tXz6Qy3cGf9nbi9ke4eWkBoz6Fxw728OyxQVRAjal8cXkRaypSSHUaeeZoHxoac1JlrCaJl6dEGrJdFlRVI9lhpCrTSV6SlcoMB5vqR1CnlHDOq9Rl7LY2jnBueQrrZqRjMki8cqKfhvp6EuIcrChJ5GDnOM8d66cmM44r52RNWcn1c9pw4rJZaWiAL6Jy9Vx9/vjaeZlYjDL37+piLKCwqcHN2YWJ/M95JdgtBn76Whsef5gdTR7aPUE+viibeLuZ3+wZYNAXIzExgZXVOdOC99WpZpTxAa4ukfH4o/xgfRMWg2BxYSKjwkiLcHBttIcf+E/gl43cXh9DiWn85OJy7GaJB/Z2c6R7gguqUrltWT5CwM+2tPPLre00D/nRNMhPsGM1Ch491M9Xn61nZ2eQIb9CRZqdG+Zlkuo08fzxQb77SjOvd4xTmmrjU2fncN+1NVw3L5PT81ahqMZPt7Tx/VeaeOpoP33eMBdWpfH1cwuZlxPP3rZR7tnRSe94iHm5Li6bmc5vrqrkmtkZRKIq4aiKL6Ly1wO9eIMKK0qTMEoCqyxA07hneye/2taBQdL/LUuCRJuRLY0evvVSI0MTYfISrGiahsMs0zce4lsvNXKsZ4JZ2XGYZP1aUVWNH77ayoP7uhkPRmka8mM1SCwpiudYj5cvP1vPTze18dPN7SQ5TPzqyip+e00Vl85Mo9UTYTgkkGQ47JE52h/gwV0tdI+FuLLCTnWyTKbTRG3/JOdVpJBgM3LnpjYUFfwRlSyXGYdZ5pNn5aBqGn/phxQtxE2TDfw4Wg9C8KNQFo1DPva0jmLWoty9tXWaXPd28c+S6i+3dUzb+wkhMAhBqlNXGhKA2WRANhj55kvNNA37uXFBNrXDQTLjzDQN+bEYJXwRBd1zVTAZ1ivVUFTF8i5J9b9Zqf59BIPB6Z6qwWDgrLPOIi8v7x/ed+6551JdXf0Pf1988cV33PY79LD/GZxwFPg08H3gGmAN8BPgDnRB/RfO5Dudjv8rK1Wfz8epU6fIyckhOzsb0E/0vyu7dcXsLL77zACtI35WlSdzoHOMI93jrKlKYWvzCEcHQiwvjGNPxyTbmka4fHYmjUM+hMmMpkR0soFBEFYhyWFi1BchOjbM68gs6vKCBpEYLJ2SFpQENA35sJsNfOO8Eurq6vhjXYiceCvfOK+EE30TPH20l7MLE/ja6mIaBn3ctbGRJKvMRxbmM+yLsLfVw2uNw2iqhkkS+MMxFhXGUZXh5JEDPWyqG0ZDJccuuGh2AQ8f6Obnm1q4rtzImmInm1sn2Vg/wm2rivjy06e4d3s737mgjLI0O6/UDiOhcnFNOq/WDvLyqUFuWVbAZTMzeWBPF7X9E1gMgtXlKayvHea6eSHu392JzSThmYwwPBkm1WlmbVUad29t53ivlzk58SwvTeGh/T1oaFw5OwunxcCCHAe720a5Zk4FwYkxVhTZeLXBw0Q4xjXzspElwSUzMnhgTyen+iaoSLNhN8kYJEFE1YX5AeItRj55Vg6/3dnBr7a1E45pXD4znXibkW+uKeL765v54ast+CIq5Wl21lQkMys7ju+90sjvj3jBNs7vd3eRFW8l3iLzdP0kjoVZrJ6XiCtxmHv3DHLrU3WoCCpSrPR7w7wk55BTmsOdVYXcs2+QX2xtx2mRmQypZMRZGA9F2dzoJqZpfGtNEU8c7ef1Ni+a0B19FhXEs6w4iWO9Xh471M+ATx9jcfuj+MIKy4qT6POG8YVj5CZY6PCE+NHGVl1/N6aRHmfm22uLiCqwq9XDq3VDhGIaEoKGQR/xViMzs50c7hnHLgnyk2wc7vZyoMuLyyrjDSjkJVr54vJ8jvVN8lrDCL/b1UlMhUSrzI1VdmaWFbKxfoRX64bZ3zGGLASfOCubc8tT2N3q4amj/fx4YysAs7Li+Pq5BZzsm+Rvh/r4xZZ2JAksBpmfrCvDZjLw3PFBtjSMsKluBGmKlb0oPwFvKMoLJ4ZYXzuEBmTGm+kaDVCd4aQ8zcHzx4cojpfISU3gcNc4eztUVE3DZjZwfFih0d1L7VAISUgc7vCwcYoUV5hs58rZ6fxsSztlKTZ+vqUdb0ihdyzIp+fnIpsTSFcUvmFJ5I6dffx4QwsWA3xtdRF3burgrk1t/Pqqqrd9brxTUg1FYzy0r1tfTAJGWRCNQTAaQ2//qcRiMqsrU+gaDfK7XT2YZBU0ODXgRwhBsllmeFKfSdY0DU0TGGQ9mSaZ5P8o+/ffCZ/Pd0aV6ZYtW97ztrOzs+np6XnLS0D/O71f07QgejWLEKIcmIPeX+0G9mqa9p58CT8Uler7GYODg5w8eZKqqqrphArvzfrtneLCmnQsNitbG0eYnRNPkt3I5vohilPsFCbZ2N0VIMtpoCLdwaunhpiV4yLDZUa2unTlSAEuswFJ0xjzRylJtoCmokwMc/emBv68t5s0l5WKdJ1wMx6MsrdtlBWlKZgNMrv7FAYnwlRnOvnxhiZ+/GojaBqHOz1c+ceD/M/zdfhDOsFhV+sou5rddI8GSHeY+OjCHFaUJXOyz8vdW1r52aZmRrwhKtLtFCXZ6PVp7Gge4TNnZaHEFP5cF2ZFdTZnFSTw8P5ueseCnF+ZSu3AJJ98+BhNg5MIVaE6w8HHz8plTUUq25pGGJwIs7IsmXirgeeO6WIP51enoWkad29ppdvj51Nn6xXZa1MPtIUFibgsBjZNzf+NByNTY0gCi1HQ399PjStKVIX93X4QAptJYJAlhKaRm6gbHa8qSybBZuSpI30AbKx3E4mBWZb428E3Rs2WFCVwVmECjUN+KtJ1MQ2AvEQbt68uZtgXZSIY5YpZGQghSIsz87Xl2RhliV9uawfgf9YU8c3zSliQG89f9vfx2JFBhNGKJEnENIHZIFhRYOOmGWbirTI/75B4qtZDcYoNIWAipPfEzqtK4e4rqlhZmsTWRg9feKqWfe1eytPt3LI0j8JkG08dGeDWp2r58+u9aBrcWOPk4/NSyXKZefnUEPfu6sIbVFiYH89HF2bzu6urqMpwElE0ZAGDEyG++kwDf97fw7ZmD5oGN52Vy0fmZ2GUBI8f6ec3OzqJKCprypP53Dl53H9tje6/GlBACLpHg/ztUD9JNqN+fWqCRJsBbyjG/Uf9PHSgF0mAEtNwmg1YjRIPvt7Dz19rIyPOQmGyXpXIEhzvm+DBfT3kJ9m4dXn+NLTsj8R49FAf3mCUG+Zn6tZmAtA07t3RyZ/39eAPKbSN+DHKEqvLkhmaDPPjDS185Zk67t7aQU6ChevKzXxuaR73XVtDmsOEzWSgIs3BSCDG8aEoExGdbZ/lMpFqE1gMgvPzDdy7owM0jc7RAGFFpX8siKaqPHR0kE2xBGIVFRQWpJGfbCccjTE3J46yNAeXzUxha/0Au1tH3/a58U5WdJ97shZNOz0ZoGI36TXOiC8ypTQroQmYnePi7MIEEIJwVMMoqRhkDU2D0YDCgFdXN4rEdMU1WQgCURWzQdJ9nt8hPshK1e/3/8dGai6++GKeeOIJhBBmIUQBOoP34Du9/3TPdEpM/x7gBnS4+A7gz0KI7Hf67NvFh6JSfT/gX1VVaWpqIhgMMn/+fIzGt5oKvB/7cFoMnF+TxcbDLXziLI0VZSk8d6wfty/K2qo07tvZQas7xLoZ6fxscyuHu8a5em42v94WQRhNyFoUtz+q4zqaxFhIIyPOyMBEFGV8kNFIPH41hU31Q6ytTGND3RCKpnFWYQL37Wxne3cEoWlsqB1E0zTMsqA0zcHAeBCPP0qy3UBpqoOesSDHukdRNYEkYGgyzCMHe9FUVYeUDAKX3chESKFx0I/ZAElW3az7niEfRclW+rxhvvNiI8UpNiRJ4ofrm/TDRocBv7SyiAFvkCeP9NM4OMnlszN5rWGE54/1c8uyAi6oTuOxg32clWRmhtNMSaqNxkEf51Wmsqo8hUNdY7zWOMzV87IwyoJV5Sm8cLwfjy/C44d69Ype0Xj29UYW59pYtaCabQMNbKgfYubiRHa3T+jziELwWv0Il83OwGiQuGxWBn9+vZujPV5ePjXEnJw4KjOc/O1QP0d7J5iTrbOl46dmT9s8Adrd/umHvtOsjzEYDBIPvN7D/6wpIiveQqrDRLJNYjwkEYhq7O8c46LqNL60qpCH9vewvnaEl7UhCpJsfGxhNn/e38PvD46ybkYaP7w4md/u6GB35yRCQFG8gUurktjUFuDhA31sqB2mINk+NWZkQNU0GocDmA0Sy0uSkIXgRN8kkgQhReXZBh8zMzU0JDQNchIsJNmNHOgcY2/b2LShw+KiRD51djbt7iAb6oY53O1FQ8MgJA51jrEgP56iFDst7gCZcWYsRpkXTg3z/KkhrAZBKAqry5NZW5nC7rYxdja7OdKjjzKVpNj53LI8lEiY5472sL99nKiqYjdJ3LQ4h9nZcWyqH+GFE0N8u7sRSQgunZnGuupUnj85xKb6EXa1jIGkEWc28P0LSzjRO8lLJ4f45ktN2AyCsKLylVWF5CdaeeHkEK81DLOpfgQEfGJRNmurUvmEorKt2c1D+3uJqjAaiLKzVyItN8yWJjdDvgjfWFPMnBwXYSXGl5+tJ81p4a5Ly2kdCXDHK01UZji5/8gEqgaVSTKXl5jZPQjtIyq3ry5ic8MIf9rXS+94iMx4M61DPvISzOxqHeP8ylQuqk5jV8sY/7u5hUUF83XG+pvi7cy46wcmOTBFvkNVQQjGg4ouphSL6d28KUWlV+uGqO2bZFamnWvnZfLIwT5q+7wIESMciyIhcJolghFdAELTIKpomAwSgX8i5/hBVqrvx0jN888/zxe+8AVGRka48MILmTVrFps2baKqqoqrr76akydP1gMK8Pl30e6V0JcxnwCe1TTtj1Ovf08I8Qf0mdX7z/S4/q+oVEOhEIcOHcJisTB79ux/SKjvZ1w1J4ugZuD1dg/nlqegAlsbh1lSnITVKNjRPsGC/ATSnGZePDGgj+M4TEiORCIqCA10ty0Vjz9KTAVZ0tBiMZTAJBFPP/dtOsXVDxzgmSN9OM1GvvNiA5vrhxFqDKOkTYttR5QYp3ondBs1TWPUH+X19jF6x0LIAjLiTCTajFOuFXoSthgkogoM+6L6UPnUHO2wP0ZU1VCUGC3DAQJhhagSo2HAS1acEaNQyXYZ+dmlFUiS4HDXGBfPzCTBauBvB3pItBlZWZ7M9uYR3L4w51WmYTYI9g8o9I0F6PKE0NAoSNarwvMqU/EGFfZ36Kv71RUpaAKeONzL/vZx1tWkkWEXHOiPkl9QgCQEF1anMegNc2o4zIbGUSrSHMzMcvJK7QDRKXm2NRWp024xvojKVXMyuaBKr+r+/Ho3IUVlLBBla9MoZxUk4LLoPbXu0SCapvHwgV6sRpn/WVNCRFH59suNHOwc59XGMdrGYnxsUTYLC+L528EB7trUiscXnhKJEMiSxKg/gj8S48frylhZlsxLp4b5/NMNnBgMkuEykeE00TYW47m6MRYlRzk/38BYIMKBznEkIXH57HTuubKKq2en0+4O8NsdnRzvm6Qyw8FPL6ngu+cXU5Zs4GDPJMd7vRgNgtIUOxdWp3H7uYXYzbLO8JUl9raP8bnHa3n0UB9He7zkxFu5fVURF9WkMeyL8ODrvbxaN4zLYmBNRQq3ry7i22uLiTMbpkYxNLY1uXn4QC9GSSCEPpdalGyj3e3ny8/U8/u9fTS7oyiaRlWGE7NR4tfbO/n++ha8IQWEjhQYZcHLp4Z4Yn836xJjXFUeT0xTiSoawUiMg51eVpUl86OLSnFZZPyRGLEptTC3P8InFmVTnekENCTgof093L+rixFfhKPdE2jAjfOzKE+1sac3ym1P1fHiySFK0xxkxOla3C+eHMLj143nozGN+3d3goC6QR82k0yyw8h3Lp6BNTmT3Z1+FmQYsUz2ck2JxLnFcWxscPOX13uYlWnju+eX4DTL3L+7CyEEH1mYReewDs//fcRisX9Y1N/61BTT9fQYjRDYjJL+gBA6FG4xGbCaDZzs96NqcKzPx293dJJoM5DsMCPJBjQhiIYDeIMRBifDb3QOBRglQTDyzkn1g6xUA4HAv01Muuyyy+jt7SUcDjM0NMSmN3nvfvvb30bTtCJN08o0TdvwLps6fdbcQIoQwiKEOC2mbgbeE/z7oahU/53weDw0NjZSUVFBYmLif3x/8/LiKUhLYEuDmxWlKczO0j07r5qTxeL8OLa3jjMeiLKuJo0/vt5Ny7CfK+dkcN+uCKrZgRb2EdVAQqAJjZGAgkHSbZ2QDaiqghb04g9NoMkWIkEzGHR9T8GUtueUuo+m6jOPFoNEUFFRVLAaJVIdZvwRheHJMKqmsz41/j/23jo+rvvM93+fc4ZBM9KMmBksM8aOIeAkThxs2qRtUm5KW0wh5TZlZkzbNEnTcBwwx8xsSRYza0YaDWv4nPP74yje3Xu3d9tt2s3d331eL79etiydAc15nu9D749Aap6ygjZHAgjasUrVGIbaknkGOZ4AOY0iCAiCxIAnTVluFuPBBO2TEe5YUsjT5ye5rimfO5cX89ujI7SMh7l9SRH7umfY3ubhHWvLuao+l72dXr62oweDBNlWE690znBdYz6LSxwUZBnY0+FlfY2L/Cwji4uyONQ3Q5ZJokbnx9aUxx/Peun2RmkqsLOmKgfXqVFe7goRiCt89JpiBEHgy9u7OdDr4/qmPAw6kVsXFfDwqTFq82xUuy0AvO/KMr68vZdHT48jiQIy8JaVxagqfHVHD1/d1cPtiwtpnYzwttUlLCqx881bG/jBfm1oCGBRvoEbF+Ry44Jc9nb5eOz0GB99pgNFhXtWFbOkxM7PD4/w/f1DXFHhoMptwSBpO4OCANW5Vt68vIjWiQhPn5/gd+1pQMBu1LGhUE+fL8nTF6Z4qdVDnt1IJCXjsupx2/R0eaLcv62LwiwDMxFtGGXLgjy80RQnhwIc7JtFRsWik7h3dQkbarI1VvO5KQZ9MQRBYCqc4EDfLA35VnSigCDCksIs/HMpHj09ziNnxhFRMegkPnFNFUUOE8cG/OzvmaVlXMuyl5RkcUNjLqU5Jh4/M8GJoSCqqmA1SDTkWXn/laVcHIvw5IVJBmfjGCSVt6wqZk1lNi8d6+eVgSAH+kEWRRZYVO66qp4d7TM81zLFy5e8iPPB54HN1UyEkmxv9/Lgzj7MeoF4WuVdV5SyotzJy5c8vNI9y6E+HyDwpmVF3Lo4n1Qqh1PtA/yhPUlKUen1RvnYsx1YJZiTIUcHfzo2xFA4Q0ZRcNuNbKjR0JEVOWYeeKGL8WAcVIXSvFwqanIR5BQbhFmODKrEkxmKLSqinOSda0r48aFhdndOc1NzHosLTPzq8AC3LMon2/KvB/v/VTT9sdNjTIcT2g0MmAwSWSaJ6WgaVdEUlBIZFb0E+VlGPOEk799Qwexchi5vhONDITKy9pmyGI3MzYWQMzGmMiZEg4W0ImtlYFFgLvmX2cj/3Znq64WopKqqLAiCCXgJTaD8u0CLIAg3Aing1N9yvddFpvpfKc2qqsrg4CADAwMsX778nxJQQXuub1xRQrcvyXgwzvUL8pidS3F+NMi1ddnICOzpnOaqhlzsBpHfHh3kULvWNBeNVk2sXBU1WLYgoggCMiKCqNeCpSjOn15FREVGTacgk0JQMqhKGkHQdFpNooDdKJFRFCJJGXU+eCbSCqP++GVmqCAIpGRNvUILxgKIApJOR7nLitusQ5REGtwGPnNDIzctr8aSnYdgzUZVFNRUHCUWYHRyGotB5MlzEywvy8ZtNfCH48NcVZ9Lrs3Ak+fGyLMZWFftYm/3DNFkhg21OWRUrST3+S31bG3OZ3A2Rt/MHKIgcF1jHp2eKGPzkPKqXCuZDNQ5YGFjPdctLMVqENk9v8urE7Wf8c7JlDr1LCx20FyURV2elW0tk2TmJW6iKRkQCMbSl7/WkG/j5sX57Ov2sa/bx1V1LvLtRgqyjHxlaz0WvY5HT4/jMOu5vjEX0IhOn762GpNeh6LCoD/Nwd5ZFBXWVWVTkGVCRXtPTw76iaUUvnlLPbcsyuPkcJA/nZ0gx6Lnu7c1cPviAs4MB7n/uU7aJ8PzzkyTRIskFfwZA+/eUMvbluchqzDs19Yl3Ca4rtbJ926rp8plZiqUIqOoxDMKQ7NxmgpsLC3VDtVuswFRFHjk9BgfeOISvz46ylQowc0L8/jSlhqub8plcCbG42cnGQ8mKHEYWVrq4DPXVbN1YR4CIIoiybTCjw4M8esjI7RPRphLytTmWdhY42JgZo7vvDLAR5/t4ORQkOIsA29stNJUaGdbm5ePPtvJY2fHSWZk1lRkkW838uipCR7c0Yt9ZBAXSWStUUp/VOFk1xR3ryjkzcuLkGWNAJTMKBzu99NUYONrW+txzdOlAI4P+vGEE7x9dYmmh6sKiMAzF7W+8Eggxt7hJGkFvnFzPT99YxPvLchgSicQBQF7Jo4p4MOmB4fFwIc3VmrkLVVlYEYTLVAUFbfVwDMXPXzk2U7OTiR4qiuOoig0FVjZ1T/Huf4pbHMT1GbreOrcJN7gHBa9RGQuxs8PD/87n/Fvg1cyI/Pdvf2oiowkiKBoX0tF45pEkypryjSCSkaFiVCCrc15bKx1c8eSAsqzzSiqwD2rSnjj8hJMRiOSJQtVlpETMTJzIY71+0BVEUXhddtTfT1hCgVB2AxsVVX1GPAhoAcoB1qBXwNDf8v1/q/MVNPpNJcuXcJisbBixYq/6bT1WggY37GkiB/t7eaVrhnetroUl9XArnYvH1tfSJNLz45LHsYCMWJphaHZJFlmHY2FNro8UdSEgeXFVs6NhEBzyaiCQJbFgKyoxGUBEBEkHYLegM2op6HARttkhIwi4DJLpGQ0ebbM/M0qaFB1eZ55ajPqKMuxEEqkmQxpcILCLBOheJpAPI1B0kqEY8EkiqqSY9XjMkOuTU9TYRZ7OrwsLMmma0pHKhZCjkUQ0jEiYQmz1crvjg1xz6oSfnRgkJMDs9y5rIhfHRmmZTzMbYsLOdo3yzPnJzg7HEBQVcw6iUq3heJsE4+dHmdPh5e6PBtX1+fyxNlx9nZNc+/qUo70TiOIkNJZMJu1MvFV9bns6pgmEEuTbdFj1GvP3WaULivz3LmsmG/t7uVQj48rqrLZeclLXZ6VvpkYuztn2NqcB8Ddy4o43DNLMKmwtNRx+feZbzdyTb2bP5+fJBTP8KujI7xjTQk2o44/nh4nLavcd0URe7pm+M2xMV5o9SArKsF4hk9triKRVvnT2XG+tKOP+jwLE8EEBp1Ins3ARDDJTw4Oc+/qYr62tZ4f7BvkxEAAURK4vjGXO5YUcKhvlu2XpvnKzj4UVcFpMfCBDSWM+eMc6PHx8+NToE6CAOsqbCzLFemNSJwZjXB+LAiIFDtM3LYkn6XFWRwd8PPk+SkiKRkRgV0dM4zOxrGadIQSafJsBpaWOuj0RHjk1Bh/PC1c3l2+b10pdpOO/d0+dnfNkJIVBLT+ptum5+7lhTx9wUM4kdHWsMIpXkkKLC4xUeI0Mh5MoKjaEM2IP8619W4UFV5s8/CUWIgE3KgPskkfZnfExN5BiV2DnagqFDuNfGhDBWdGQuzr9nFqOIgkalWd+6+pIhTPsK11igd39pFrMzATTbO50c0blhSwo2OavZ0zHB/wAwJbmvOocGlqOCNTo8xK1bzLPMtWQ5jTMym+IzpYUW7lKzt7UVWVG5vy2FSXwxde7mFZiZ3PbK5mJJDg98dH+NXREVRF5oPrS1hR5uSBF7t5tjfBd25t5N35c3x+xwDf3N2HZ06h2Cbx7Pkx3rqiiKpcrWf4b4PqOx9rm9d1FpBlGZOkklBFgqqEqmpl4nhKAVWrHBkk8TJBLZlRePmSl0212dy6uACAmxfm8+4/tZGeC0EqBnKadDKGZHXiiySJp/6yr/vvzFQTicTle/x1YB8BHgJQVdUL/AJAEAQH8BgQAC79tRf7vy6ohsNh2tvbqa6uJj8//2/62VcngP/ehWeXzcDm5iIOtw3y1pUlXNeUyxNnJ3iqRcdYOEMsrXJxLMTyAgMt0xmWlDj46DU1/MuTrUwk7PR6w1TnWuifjoGkIoo6oikZnU6PKAiokgG3w8LdK4p55NQolyYilOWYGZmNMRuTQRAw6CRWlDmozbfROh6mdSKMUS9RkGVGURT6fXOkZBWTXiLLpMMfyxBJKRQ5LZTZRWbnUvQHMtTk2XCY9VwYD3L6hW5EoD7XzJe3NpJMy9z32EUiog41FkSNh4nr9PROw8aUTHWuhSfOTvCTNy3kmfMTPHthgq/e3EChw8hLbR4cJokbq4zsHEpxasjPlTVuNtS6ONQ7yzvXluMw61ldmcOhXh+ZuRC+OYX1tTkc6w/gDSfIz9JYv9sveXmlc5rblxax45IHu1GidyaBfy5NjlXP8jJt4ObZixPMRJPMpRXevbaUJ85P8vSFKdZUOHHbDHjCCSIpGZtRzy+PjPDFG/RUuq3MpWS2t3tpLrTTVGDn+VYPLePa7uSpoSBvXlHE2gobNdYkk7KDXx8bJpGRsRt0TASTXNvgZmlpE9/ZO0CXJwqoXFOXw72rS+jxRHn07DjfeWUAaZ7/fG1jLjORJHu7fJwaDrChOocss45ISsaq1xOJyzx+dpLN9S6uX5DHcxe9CBLIChwfidI/DQVWiVhKwayXWFBoZ9AX45eHRxEEba0iyyTxwOZqdDqBU0NB9vdoHFwBFbtJh8um522rS3n45ChT4RQuqwYT+NL2XuxGHbG0jF7SNFSTGZXTQ36ea/Ggqtqq17LSLNZUZjMbnuOVbj9HBvyggsUgckWFE6tRx7mxEH86O6nJBIrQIMRJyTI70w4OpO3kC0kkUauCmo0SU6Ek397bz1V1btZVZ/NK9wyKoiIjsqdzmresLOaHb2jim3v66fHGEEQN55eWVe5ZWUxhlpHfHB9DL8Ku9mlaxsKsrHBwQCijWohzgz5MUoXf6aqQgHOjIURUFhVn4bYb+MbuflAVFhU7CCW0NaKlpU56PFEEVaHLE2VDjYuPbKrgSy/38ujpCT6woZzmoiwujodYUmTjrUuy+eLeSb6y7SL3r7bjcrkuB68DPT5aRgOgquhEAVmAhCKCrO2ZoqqokkTmVUEAFQRBZGfnzLwMnPb1I/1+grE0Wxfma3SrjIzOZEZOJ5DUDAIKqWiAHo8OfZaBTEbGeOgg0smTqGYz8i23oNTV/bdmqsB/W0D/DywfLSt9VahcAXSqqoYEQcie//dfbcJ/Ant+7aTe/08Poqr/oZLB/2rj4+OMjY2xaNGi/9Lk2Llz51i4cCFG4/+Z1fnX2MlBP2976Dhb6+1MhZKcGQ4iigIVDpFgQkEvSfzyrct4/Mw4L7ZN8ZM3LcIbSfD1nb3I4RmKrAJjgQToDehMdi2D1unQ6Q0sLXVyfjREuctCRlaZCCdAELHoRG5c4KKx2MXxAT+nhgLaqVYUsBl1LCl1UOQwc6TPx8xcireuKuPWxYVsa5nk8TPj3LK4gDXZCaKKjh+eDFCTZ+UrWxsRBbhwqZNfXEwwG4lTl2vhq7c0YtJJXBgL8vUd3cipOHLYp/VqHLkY9TpuXljAtlYPb1lRTCSRYXv7NDlWPbNRDQz+vg0VlIsBfnI+Tq7dxNdvW8CQb477n+vgXWvL2bqwgAsjPr62ewCDJLKoOIv3b6zkfY+3cuviAu5drXGsv7azm2FfjDuXF/PQsRHuXebi8Yt+bl5UyNuvKAXg4liIr+3swSCJLCvL5uNXVzAdTfKp57uoy7PyuRtq+NH+YS5NRfnyjXX84MAgc8kMH7+6ktaJKDvap/n2bY1UuMyMzWcoHZ4okgg3LchnVYkFKR7gmX6V1okQty3Op296jvbJCCad1hediaZYX60JIBzt95NllrhjcSHDszEO9vkRUFEFgQ3V2bxpWRGzcyl+dWSEqXASQYQrq3J42+oS+qbn2H7JS5d3DhXINut55xWlNORbOT4Y4PmLk0TSWgZZahdZ6BJZUGBm51CGTm8CvQRpGQRBpdptIRDLMDuXYnNjLtlmHWdHQgz54yiKiijCyjIn1zXlkm8z8PuTY7RMREBREEQNRlDlNuOLpvHH0tTnW8ky6WifjBBPa7hCASh2mqjJteCLpuj1zl0uuwuoOC064imFZEZBVFUsaoaooNO2zAQBgySyusJJldvM8cEAfTNxUFWKnSY+sqmCbu8c21o9hOKZ57uOdQABAABJREFUy3zbLU156HUCuzp8qKgsL9PumUqXibfUG/BJ2ezvnqHbG0NBwCSo5Khp/IKBlCDiMOmQJJFALIX2CkBRZHSCVoWV5geyer1R1lXaybNphKtbFuXzlhXFPHlugm2tXm5c4GZnxwwSCpUuC1+/pZ5trV6euRTgN29eSLklzeTkJOF4ivuPZUjPH2w0U0FOky+k8apGnKJMUNC0mDOyilGv4/f3LEIviRwd8PPzQ8Pc2ORGpxM51u/HH0shqSpFxjTDoSSZoBdVyeAUMoRUHYIoYiio5ewVItkvPoealwepFMLcHKlPfYozU1OsXLny7/aFf6upqsqGDRu4ePHiP5r7/lddXBCEk8BHVVX939ZuBEFoBW6bxx3+Vfa6yFT/szdWlmU6OztRVZVVq1b9l09Xr8Wu6qu2pjKbCncWL7d5sGUSlAoy0zobt5UpzFny+e3pac6NBrllcSG7Orw8fX6cj11dw4JCG+2KwmQyhs5uBb1J22fNMnF0wIdRJ+KyGjHoJEYCcQw6HfeuKcdtM/DYyRGea52lwZOgJMdCSlZxWPWsKMsmlMjQMRHmWP8sqiAiCgKPnR7jsdPacrnNqKdrdIa5ORvdviiiCB+9qlrj3gJt0xkCsRRbFxWwq93Lzw8Ncv+1NSwtdbCgKIt+r0osk40cnUUOT5N2FLCtZQoVePzMmCY6rUIqneEz19XwyMkRDvfO8o4miWsb3Dxx3sOYP0al20pNroVXuqbZUGFBCIxjml+fePe6clxWAyvLnezvmuHuFdpu6E0LC/j6zl6ePDdBXb6NNaU2uqcT7On0csfSQuwmHUtKHOTaDHgjKW5eVIAiy+SYJN66sog/nJzg0VMTnB4J8qblxVS6LXz1pjq+vXeAb+4eAFHD0lW4tHJUqdOE2SBh1Ik0FWSxq9PH9nYFo6SSlGFLUy53Li1EJ4pcGA3yk0PDeKMpjBKUOE3c2JzHjQvy+M2xER4+NYaqaqIMb19TwvZ2L7s7Zzgx6MdhNuCfS9NUaEcvCRwbCHBuJMSi4iw8kRQ6SaQuz8qoP8aPDw7htOgRUZlLq6wtz6Io28rp4SC7hhPsGIogoFKbo2NrpYRozuLUZIoTw2FtY0MU6PVGWV7moD7fyrA/hsuqp8Bh5OJ4mLMjIRAUUEXq863cu6qIeEphZ+c0F8fDqIqKKAgM+eKUZBuxGCTiaQWnScJh0HQ/j4YCWqBUtX1Uq0EiIysEYhlUBFRVQUEgjA5B0HqhtXkWjDqJU0MBjvT7EYAci54Kl5lLkxG++HIPNzbn8441Jfz22CjBWBqdKGDUidy+pIDrG/N46NgIJwf9CJJItcsMgsKGmhySGZnu6RjrKrPJJoXPP4cvrCkWpWQV31ySbIueRUVZHOmf5ZbmPN66spjRQII/nhqjYyqMpMqsLHOyptJJJCnzYpuXUqeJNywtZH/vLDvaZ1hUZGF1RTYPnRjj5FCQm5rzeKVnlp8fm+DP71wKwI92TZBOJ7WDlaoiqDI2SSEqSOiVOIgmnA4LsZhCKgMIAmsqnRh04uXyebHDwL1rSpAEgTctLeTTL3Qx7o+SUUXKhRQDogiCjnBGQRUE1EwaJRFj8twQzrw8sFq1P9EoYnc3OBz/gWf759k/Q0jlr7SHgA8JgmBF4wSn0LLTNwGdaFPBf7W9bvLvv/QGx2Ixzpw5g9PpZOHChX9XueK1DKoC8Lbug6A38qWzT/Lxo38ipUB3UGDzojLy7QaeuzBJlknHjc0FHOv3MxqIc9/6SkRRZGV9MT9+ywqWlTro9UZZVZlNY0EW/rkMe7t9rKjI4W1ryhAF2NHmwWU18oVN+dy5MJsBX5RXembQ6yS2Lizk3jWlfPiqSrLMOowGHR/YUMmnr6vlrhXF6ESJLJOeEqtKUtBzaDjCWChBLKXwi0MDdE5FiCQyvDKcoCnPzLvXlnPP6jJODAY52ONDQODOZcUkZYGygmwEsw0lk0ZNxZDQYAECKtc15LK1OZdoMkO5y8yWhQX0TkeZiiqsr3KiE7UBLtDWXsYCcY62DZKwFZHMaPum6fmy1/VNeYQSGU4Oaus2S0oclzOUu5YXI0ki19c5SGQUdszjD2OpDJGkjCSKnBqcRZQkRFHk6tocVpZnsaNjGotBx5YmjVblshr42s31uO0GZEWg2zPH2eEgqgrHBgNcGAvz5hXFfO6GGn795mZWl2eRVLTDyq5OH+99/BI/3D/AL44OI6HyofXlLCp28MT5KT7xbCfHB/1MhpJYDBJOi57jgwH+cHKMa+vdfGh9BZIo4ZtLIYoClS4zH9xQzvdub6Ak28SZkSCBeIbyHBNbFuTy87uauWNJPuFECt+cFqAmQklNx3VhPmaDiEEScNuM9AdkfnQhxcMXApwYCuM0wLsX23hDsxOdJPB8q5ednTPoJJGFRXa2NOVy/zWV2E0SIiJZZh293ihf3N7Hd/cP0jIWxm0x8J61Zdy3rpT6fCuDs3Fm5rTJ5UhSYSwiE0vJmnC8QaTQocdh1hNNysTS/5qxZhl0OM06BAH0Arisevpm5midiJCSFbLNElkmkUAsTTIjc/81lSwry+LFNg8/OjCEQSfywHXVrK3KZlurh48+28H2Sx46PFEKsowsL85id7efbx0P8YP9gzx8cpzFxVl89KoK3n5VHSl7FoooMRVJ4ZtLUZBlothh4nD/LDpBa9n8+OAQT5yboGsqzOICKxU5Fn50cIiXLk3zjjUlNBXY+M2xUX5xZJhwPI0wn6FeXe+mPMfME+cmEIG7lhbQNjrLrs4ZtncFaZuMgKposAdFW3Opybdqmqg6reoWkzXJt3kpIe6aR2ueHQky5k/whnlJP4C2yQiTwTibG9z4MiJRvQlRZ0RQVRRRmzwWBAE5FuBJsQSSyX91XrIM/439zL9GCu+faaqq/gHwAJ9BAz58BXgKDVn4yb+VqPS6yFT/knm9Xvr7+2lubsbxGpyqXsugyswMbzz0FN+/Yym7ypbxoUs7WOAb4ZRYzDuB25cU8uujw7RNhLh1cQF7Oqd5/PQon9tSz9aF+Wy/NM1dK+DulaV84eVOHtzRg8mgY1WViwujISKJNDc2F7CsLJvv7e3jyy93sbbcSs9MEhWJTbUuJkNx/nxmjCfPjmHQiaQVgU9cW8MV1S5kWWFbyyR6Ce5bZGL1onrSqsSHnmzBadbk317pnuELL3ZRmmMmnlK4Z5W2onLrogLOjQT446kRVpQ7WVRsJ8dqYCaSQGdxIKdiyHMhdEYzybRMXZ6NlrEgX7+1kT2dM+y45OXuFcX8+cw4pyaSrGqQWF3p5Eifj3tWl1Cqn8MgqXTHrXSfmqDYacITTrC3a5p3r9OCU5HDyK4OLxtq3ZpAuqwNdxh02jmw0K5jTWU22y952LqogO1tHhJphdWV2ezsmGZTvZvyHG2d5pqGXE4Ph8goKhOBOFVuzTGNzgeH9TU5DPlifP/AEG6LnnAyQ4nTzOYGLQBPhpKcHYvQkCPx2ZsW0DkV5ZXuGU6NhEBRaSywUOW2sKnOxaWJED85NMxLl6axGgQ+e30t5Tlmtl+a5sW2Kc4MB1ARKLAbefsVlZwcCrKny8feLh85Vj0z0RRLSx1Uucwc7vfzw/1Dl0EYuTYjH9pYwYW+CVpn0jx+dgIFBaMksbU5n2sa3MRSMr86OszQrDZRHUgJ7B5OUemQCUQ0Pu2yQhM6nZ5zoyGODPiRVW2ncWNtDldU5QAqvz02ii+awmrUMRtL84dT4/POUMCsF1hYlIVJL3FpPEQwoWAzSxQ5TESTMp5wAlnRiqqiqrAo7afQbuCY6iSUkJFEgbSiMhNJIopQ7TIRmEsTTGRYUeagymVhR/s039k7iChosIosi0Qglub5i1O8a20Zmxvc/PjgEDs6pjHqJN66spjVFU56Jvw8dX6S00MBEETGggl+sH+I8UCcyUgaTZBIex3eSBJvKI5ZL7K81EEspXBhNEQyI6MTFKpyzWxtLuDhk2M8fnaSVEbhjsWFfHNvHycG/Gxd4CYUT7OzY5pr6t3cs6KIb+4d5JVuHzc05bK7Y4bv7O5jZi6lBTJBO1wggCQJhFOChiZMq5iNItNzGYw6rbeq1wn84vAgqYzKSCCBQYLW8TDxlMzSEgePnh6j2GHgXVeUsrzUyXdf6UcwmFBTc1zemRNElHSclzIlfDGWwBAdBUVBLShAXrwYurpeG1/4N1o8Hsdisfy3PPZfMlVVPyMIQi5QB5iBMVVVe/4r13pdBlVFUejr6yMajbJy5UoMBsNrct3XNKjq9dhTMW4fPsOTJYt5e/9Rbh86xdfdd3G8f5ar6nN59sIkT50b5xu3LuCOJYU8dmaMjqkId68o4djALN/Y0UM8oyAgIkqwrsbFhzdVc7jPx08PDvDt3b08cEMd371jAd/e3cfRoTCSJPGBTVVc26BNtA755vjOnj48ES3r+f4r/dS1TWHUSfTNxLhnoZ11yxoQBJGnTo8SimX4/JZ6anJt3LG0iEdPjbG93YNVp5Jt0qoAoijwvvWVfOLZSzx5bpxgLI0/mkTOpKjPs9KVyCITmSaTTnGs38+7ryzjoWOjdHuirK3O4UCPj7esKmF9jYujfbPEUjKbG/M4NhBg2/FO1lY6WF/j5mCvpuH57TsW8GKLh8O9Pu5ZXYpRJ3J9Uz5/PDXKkC9GlyfCXErBYpB4qc3De5bnADJvWl7MqaEAz5yf4EC3j1WVTt5/ZTkf80T51eFhvnFrA6IgsK3FQ7bNiEES+da+Qe6/porGPCuPnZ3AadbzztVFmAw6zgyHePT0OImMyngowbsfv0S+3YAnksSkE6nN0TMRTOK06OmeniPXoueKSif7e2f59AtdbK53E03JRBIydXlWxgNxHtzZx62L8tmyII9hf0wrsyIQjKcZ9MV4x+oSNtbk8IP9g3gjaS27UBSqc63c2JzP9/cN0O2dA2B6Ls0fTo7R4NQmvBFSVGZrWc4Ll7xsu+TFIEBahRuactnanMf5sTAHe3wcH0sAKhaDiFGvo8ImoyRVLswI5FpEsswGDvf5OdjnR1YVJASai+ysKHeiKAovtE4TSWWoclnQSwLd3jnCCRkVrSw8l1LonY6hqgp6SeSaEhNrO45zzlHKPqOL1gSoZJhn7WMQ4dqGXHqnY/T7YjQXWLnSbWFP5wznR0PzQzra9xr1Am+cz9L+fHaCz77YjdOsIRIb8m3MRpP86MAQ9flWFuabGA5lyLboub4pj+HZOD3eMIGEjFmvm+fqChQ7tf1PnSRQYDcwHkwyFU6QkTNcXauVjl9onebEYJB/2VBBMiPzzEUPgqpi0avMZRQQ4J5VJZwbC/PI6XE+vbma5iIrz7V42FjrotJl5kB/ECTpMnwlJWvlXzmjrfBo2+cK8bSoYQgz84dHSSQaT5HKyKQUFbdZomU8zOG+AKIwhqIofHRjGTpJZHm5gzuXFfL0uQzCnF974wTdPKRfIhIJsvON72drehKMRuRFi1D/TprR32PRaPTvpin9I0xV1Rlg5u+9zusmqGo9OZVkMklbWxs5OTksW7bsNa27v5ZBNQCk1q3jXW2v8HjNOnYVLuRuwUOeCba1TLGhxsXtSwt56NgILeMhblqYz65OLw8fH+bq+lziaYVERqUmz8Znrqtj+yUvL7VN0VyUxVV1uWQUlV8eHuRrO7pZWOygwxsl324knlH41eEhhnwx7lpezJ7OaaajaT6wsZLaPBvnhoMc6ZumK5RCFESOjmcQWz0sLMni5UseNtW7Lot5G3QSsqqiEyApw/cPDPP125ox6kTKcsxcVZ/L7nYPqDJvXl7AM+cnqMm1EoqnGU/NIcei6BwGer1RihxGXmzz8J615Rzp93O4d5bNTbns6/ZxfDDAdU155JhUWnwKb95UTrZ3DFmBlRVOanJtbG7K5djgLKcG/Wysc3N1g5s/nxvn5UtTtI6HaCy0s7DIztPnp7ipzobbBBUuC+urc9jZ7kVWBd64pAi7Wc97riznh/sGeO6ih9o8Kz3TMd57ZQXLyxx8fVcv39g1wNrqbHqnE9x3ZRk2kx5FUXGYJIKJDFua3DQU2OibnuPoQIBkRkUWVV7uT/ByXw+qqmDWC3ziulqq3VZuXVTAE+cm2NPtQ1VU1lVn85FNFQTjGR47M8GzLV62tXpQVIE3LS9iRamDZy5O8cwFDy+2eVEUFZNO4pPXVjIeTPBKp4/v7Ru8HFRuWZjP1oV5nBwKcKB3lj0DMVQg125kfU0OV1Rm0zM9x2+ODl8upe/u8nFpIkyuzcBoIE6hw8jm+lwGZmO0jIU4ldJ2mx0micX5JopMGWpsOvaPZnAYdRQ7zYz6E7RPjaOoGrtWL4kk0jLxFESSMk6Ljk1Vdpy6DN1BidMjQWxGHYIqsG8iyQX7ApaIMSRlfnUMKHSYuHdVEdsvTbOz08eSIhtbF+Syq3OGtskooiigEyGjQonDyB2LC9ndNcNvjo5SnWthQaGdk0OBy6thzYU2rmus5PS83FyXN4ogiNQVWEmkZWaiSQJxGUEUiafky2tYE8E4Iio5WUaC8QzBeFrTMkXgYJ+fMqeJdVXZtIyH+cJ2bSpeQCOQvfuKCro9UXa2T7O63Mkdiwv48zlNtu6elSU88GI3vzg8zLnRIIoiI2FEELWA+mqgq8+z0DMd056MoGNhsZ1Lk1G0dFbku3csINdm4Pv7Bol6QnxmfS7xaJjxcIY/dqRJyjKPnB7HbTdSl2fljiWFvNjqIaMzoioZDJIOl0XHVEJAyST5RVuQLR+86rL/Uv8Cj/ifYa8FovD1bK+bnipAIBDg3LlzVFZWUl1d/Zr/0l+LoKqqKiMjI/T29uJ46CEqHvgI16l+djZuZO6rX+PKEh2j/jinh4Nc25BHns3A42fG0Esim+rc9Pvi/O7kGLV5dhaXOBiZjRFLydyzuoQFRXZ+fWSIHm+Ua+pz+ZdNVXRNRXni3CQry7P50uYSvnhVATc05bG7w8t9f7rI3q4Zbl1cyPVN+VS5rWyusaNk0jjNeu5dVYJJL/LoqTE+/XwHsqJyQ9O/riHNRpPs75rm2oZc3rbEyeBskt8cGf7X92ueIduUb+HOZSWsqXJxdGCW966rQGdxoGYSoMKxfj+b6tza8jwqVS4zezq91LitFNlEDvTO0tvbw7UNefT5kpwcnOXFVg9GnSawDrCwaF5Hdb7vajPq2Fjj5kjfLIF4hntWlXJjcwEmvcCenuC8IgfcsjifjKLithupnN8LvKIym421Lp6+MMkfToySazdxdb0bt83IN29tYnVVNkf7/UgiGHQiqiAh6nT88fQEuTYDb15exOoyBwV2I5FEhneuKeahN9Zxf3mcHIMGoUhlVL66s5cXWj3oJJid07LMXJuRk0NBvvvKABlF5c0rinBZdcjzQ/kXRkJEkzKfvLaK2xfnk8qopGSVREahZTzMilIH/7KpHJN+/tYUBF5s9/K9fYNMBBNMR5KYDSIbqxzYjBJ/OjPBB59q58cHhjDoJD53Qw2/vKuZt68qJp5SuDihMW0TaQVPJMnCIjsOs6a/uqLcQaHTzOHhGI93p9g7kkYnQrVTpNyUoDZbew6lTjNvXlHM5no3SVllIpREEgRCcZkXO4I80hrh5FCAAruBD20o51dvWcgnGw3IgsA+JZu0qvKxRDefNU0yl5T5/v4hjczU5KZ1MsJL7dMogMOsVUpq3GbedUUJc0mZnx8epshhZFGxnYHpGMcGA+RlGbn/mkqWl2XxXIuHjzzdwc72aaIpmSKbnsV5Bsb8c2xr8TDgi8+v7WhlV72kgSe0gSGBqVCSwFxS63eikcpEAUYDcY4NzBJNpDBJGhZ0S5ObihwLD50Y5ap6N26rgV8fG2Vzg5s8m4FHz4xTlmOmId/K2dGgBtJXVU2hCm2FRJREzAaBvKx/v4GQyqgY9SIq2mcy12ZgOpri3GiQzQ15lJcU0dDQQMLkJiXLvLHJhqCqfHVHL0c6RglFY5gMOkRLFoKgicZbjTrNwUsGBj0Beqejlx/vv5um9D85qL5uMtXh4WGmpqZYvnw5JpPpH/IYf29QlWWZjo4ObdBo5UpEUUS55x7eOxZi76+O8spAiGa3xKlpiafOT7C6wsldK4v52cFhPvdCJz0zcxh0InpJ4lPX1SIr8NGn2/jpgQG+dfsCPnltDQ9s6+Tbu3v58k31HOyZQZ3PEAZ9c/hiWeQYBN5zZQUl2WZ+c3QEURI5MeinKtdKtTXFiy1TeOLw6c1VrKnK5vZlRVwYDfKNXT0gwJe2d3HXshJuXVLIi21TKKrKHUsKiQc83NKs48V2H4tLs3CY9ezt8FKSbaRvJkY4kWZTnZujfT4SGa2Pl9EZSSQTGIwmArE0VoPIrnYv1zfl86ujw/R451iWJ7F9KIUpt57rHFaeaZnmpwcGcZp1bG7M5c9nJxjyzVHptnJdYz6PnB5j1B+nLMfM1fVu9nRPU5RlorFQ44Re15jPjnYP11ZZyJcznB8JIQgivmiKvukotXlaFv6edeV0TEWYDCV488rSy5Bzi0FiUVEWJwcDOC16fn54mIdPjuO26RkLJrh9cSFj4RRziQx/OjdFfZ6V9ZUOpJ07ODkuErSX8JWZUziv3chjAQtPnJ/i+dYpkhmF964t46o6F7s6Z3j6/CQff7YTnaiJGnxlax3ecIqnLkzw4K4+XBY9s7E0i0vs3LWsmH09Po72z7K/1w+KtobypS11OC16jvbPsrNzhh7vHIIA5Q4dlS4zb1ldxp/OTnB8IIBJLxJJynxjzwDl2SbiKZlAPMNNTfmUu01cGA1zsHeWvd0+QKXGbWVBoZ36fCvPt3i4MBamLs+GzSAx6JujdVpGUWVEAaYjSQ52ThHLQCSlsrrSyd3LizDoBL6/t5/hQIoKlwlfJMV3XxnCZdMjCRAUDKxNeunXZ/EjUyM3FNn4xopyfnVshMfOToAKNoOEXicSjGdYWpJFldvGUxcmGTozwfqaHDqmohzq86MCFTlmylxmTg0F+cmhEdZUOLmi0smZ4RBTkRSiIDAVzRCIZ4hnNMFvQdD2e7V+pqbkohdBJ4qa1ug8DEYbnVE0qJGiwvyijYpKKgOl2UZ2dvjY0uhmdi7JL4+O8PY1JXxv/xA7O6a5Z1URPzgwzC8PD9M1FdVADqpWNcgyQTijzJe0BRaXOOj2RBEEbY/cYdYzEUzwqmtaWaHNj+zr9iGgsrlR6+2nZIWXO2ZoKrDyxrV1bF6S5uu7evnN2VmKOgPEZQlRZ0RRVRRVJaMoCPPleTWT4tFT43z9lobLvux/okLN68FeN5mq0+lk5cqV/7CACn9fUH11Cjk7O5vm5uZ/d8pbUurgitoCXro0jazAG5cXMzIb43C/D08wCSj0TEd507JivnFrE8mMzONnxnFa9HxwYyUDszGeOKsh8j6/pY6ULPOpbR10e+N85OpqvnP7AhQFvn1gnJbJOcYDcf50ZozqXCufua4Wq17kR/v6eXDfOAfHM6ypzGFNVfbl53e4z4dBJ/Kd2xewrNTJY2fG+PwLHeztnGZDjYu8LBMCAlsbc6jPt/O7Y8P8/OAgRVl6Pn5NNWkFdrd7WVScRbbVwLGBWa6oyUFndSBk4qDC/u4ZrqxxcWo4wMJiOyadyLazgzQ6VfSCwJGBIKmMjH7emX12Sx3XN+Wjl0T2d2ttjE11bnQC7O3UsITnRoMIqkAsLV+eDL5lcQGiILC9cxavL8D2dk1iz2XR85ujI5f3I406rS8lSTp2dU4zNKv1JRNpmafPT9BQYOdXb17EF7fUs6Q0ixF/HBXY1urliy/38K29/SQzCj3Tc7znyS7e5y/hRFY5VyqzNJozlOx9iY9tKGF5mYN4WkVFoG96jlhKZmtzHp/ZXAUCxDIKBr1IYC7NxtocfnjHAmpyrfPTsxCOZ5gMJ3jnmhK2NGl9coNeywK/vKOXR06P0zEVIRTLsKg4i1sW5RFLqzx6bpr3PdHOsf4Ai4qz+OkbF/DTNzZxc3MuE8EE3mgKVYAOT4TZaIa6PC0zyDHr2VjjIp6WeezMBJ9/qYfzY2GKHSaWlmZxTYOb8vnVog01ObxjTQkbat2E0yKhpDbgc2Y4xCee6+JDT7YzHEiyssjIl2+s47dvXcTbVhcTjKXxRJI4zTrWr6nlB1cVcn1dNruG5vjUC110eqIYdRI6ScvM3rKymDsW53N8MKiJUFRlo5MEXumeZSqSpKHASo3bzEggTpcnwpYFuVTlmDk24OfEYBAEaMy30ZhnxayDhCIgCKLWrXx1X1ZV5ldZVNKyJiKhn2divzqMKggSCOJ8hUyYJx+BrKqMBJIYJIFdXT5Wlmcz5o/TMh5mXaWTF1o85FgMZBlEDvf5UOQ0gqqpZ1j0AgYBSGu72yoCNW4LM1ENI5qQFYqdZqIpjUUsiBK3LSqg0xNhT+c0ZTlmWsbDnB8NaWSxuSRvWFIIgNOi5wtb6jDqdYxGBR64vhaDwYSiqiTTMtGYBthXEUBO8UrX9OX7478bUfj/MtV/gmVnZ792Q0R/wSRJ+qsgE/+r+Xw+enp6WLBgAU6n8z/8nn+5qpq39nk4703xtmuycVp0/PzgMKoATUUOOqc0JYzaPBtbFxZo/c06N6src7i+IZcXWqdYVJyFzajTpiPTKqU5RlZX5mDWS3z3jia+ubObP1wIYuuMopckHrihHodRwByFgblsnmgLkFIgmZbxhpPkZxkZD8Q51u/ntiWFVOfa+NTmGg71+fjFoSEyGZkrqzVmsiCKCKh8aFMVH326lWgixTdvaaTKbWV5mZPd80SjddU57G738IH1FRzu8wPS/GlexRdJkpFVjvX5aHaJtHoT3NGUx/KSOAd7Zjg+qJVcBUVgNpqiIsfCmkonh/t83LumFIdZz9pqjbB0Y7P2HjUX2enwRDk+MMvGWhdOs47rm7SF+2cueoimVK7IzbC6MIvfnp3l+YtTvGl5ESeHgkyGU7xtTSk72rVA+YH15XjCSQIJmfs3lyAKIotLsjg2MItOEvnGLRoI4/A8NvCaehcFJhnPiIfDAT2SInNEzGXIYuGtyQ48ndOcHw1yQ6MLo05kZ6ePc2MhtjbnsrvTh9Uo8Z4VRexqn+Gnh0Z4odWDSS/RPxPjjiUFuKx6dnbO8PPDw/xaFMjIsLLcwceuqmDYH+fAPDA+Mz+d6zDrqM+zUWlK8OilGOFEBrtJR9tEhA882U5Fjonx+TWe+9aVMRNNaYNcF6dQ0A43y8s0EtIblhTyzT19+ObSNBXYCMTSPHNh6rJoilmvUbd0ksCliQhJWeUjmypYU5lNrzfCLw8P44upmCU4O5ngwp/baCywMuCLYzFI3L64iP09s3y3RZPbu7bBzdnJOXzRFAIqV9e7uLYxl98eG+WXR0ZYUZbFlvne6t5uHwadwBWVToZm5+j2xKjJNXHTgjxODfl56ZIXVAGLQaI8x0xG0TRQE2kVVQCdANoIlaBladonHL2kle0FUduRzSgqOknEZdGRyKhEEmlKnUbuWq6BOZ45P0k0kUFCQVEV4ikNlrG/x0ee3cC+bh+rKpwowOdf6kZVFJhHDYqoyKqAosJMXNUGtDIpRIMmUoCqotOJZBSN78t8cBeAT23r1gZ4FZUhf5zfHB8FQJUVjBIEY+nLuFXtwAlXNxZw94pinjw/Rec0yACiFjQFAbL0EIlG2XVhiC1Lyv5by7+xWOx/dKb6ugmq/wz7WzNVVVUZGhrC5/OxYsWK/yOJaVVFNlfUFnCwa5hTT7QSjGtKEW9YXMg9a8r4+s5unr4wwcY6N3evKObkoJ9fHB7ih29o5h1ry+jyRvjeK33IKjjNBu5eVcjDJ0b51u4evrClnhyrkc9cU86nX+wlmMxQm2cnk4zTNTBMRUUFEZ9MRtU0MtsnI3z0mTbuXl7CoG8Oo15TbgGt/HRllYuHT4wSUxV+fXSYr+eYEUUBRVUvwyBenb4ETZz9qyNBjg/62VDrZvslD/G0rPVcjSbkVAKDwci50SAWvcRLrRO8f20x57xTvDIQJRjXeoYWo44Hb2viSy93caDHx/IyJ9c05HGkP8CpoQAba93c0JTPkT4/Pz88iKKofHBjFd/e08sLLVOsr85GVVVuX1zIvm4fJyYSLC1xsK65jGAwyOI8Hc9cmCRXirGte47ibDM3NeezrjqHH7wywI8ODCEKAk2FWTQUaOXk3ukoB/tmuX1xIdW5VoKxNId6Z2kqsHJ9qda37ZzORvKH+WH0LAPWXJ5W8vi2axXqOS9NhTbeubYcVIUrq3P49bFR/nxuEkkQ+MCVZayvcWm94X4fvz8xRiKj4jTryLMbuLI6hw01OXxjdz+9MzEQ4OxoiM+91MOSEjsdnoi24rQwj7mEzJmRIMcHA8jzk6RvXlnMDY25TIaSPH5ugtbxMKBlZ0cH/FxRlU2Fy0LvdIxqtxW7UeJQ3yz7emZRFAWdJHJNvVurVtgNfHvvACP+OOtrctCJAoO+OQ71aQICogA/PzzC8y0eIokMcymFj15VSX2WQq8nzIGRBK1Tc6CqVDkkSiwK37q5lr3dszx5bpJzoyHsRpFPX1vFhbEQu7tnaZuMcN/aUg71++dLvCouq4EFBTbOjoQ4OxJkXXUOy8t0HOydodcbQ5REciwG8uxGgvE0Pd4YqvBqAJ0v96IJ3IuCQEbVhqRESQuiJr0GU0jJWlASBYHEPEvbbtQzHkrx/f3DNCZ95JssREUjqjy/Ry+o89mrOq8yA6fnpQuVjAxKBuYFMTJoVKZ4SlOTUQWtlKxkUgz5tAw4I2s1aU8oqa0riRILCq0sL8/mUI+PuVSGB7fWIwBHB/w8cW4Su0niZ4dH2NM1w79sLOeh42OYTAa+fGMtgiCwqCSLrk4Dqpy8LEKgAs0lTk6MRHm6dYZC/JeDajgcxm63/1OHll5PMP1/hL1uyr//DPtbgmomk6GlpYVUKvWfBtRX7WPX1JDWmTHrJb6ytZ6GfCsHe30k0jLvXFtOWlZ47PQYJr3E+zdUMBGI89S5CYw6keVlTuZSMoIg8rVbm7ixuYB/2VRFx2SE7+zpIy0rvNQxSzgtsKnWzYgvxude6kHnLkNvtvGHkyNUua08cH0dP3vTIpYWO3j01ChH+2fZUOMiy/yvUlRH+meJJjLcd2Ul8bTCgzt6iGe0Q8SfTo9hkLTA/scTo6gqLC5xUJJtYuclD9W5VoocZk4MzrK01IkoSAj/Zpc7nZGJpOD7hydBVdg/GGMokMCiFylxmihzWdhQ6+bssJ9wIkNzURYFdgP7urQBpfp8G4UOA11TUW5YkE+h08StiwsZ9cc5PxJEEAQcFgP1+TbSssqKcicmk4mCggLuv3EhhQ4jf7gYZioqc4U7w9DgAGo8zBe3VNNYYCOjQKcnwv3PtfOHk6P8+MAgNoOO2lwLF0ZD/HD/APG0TLklxWAY2iIWLk7N8dYFTop1Mhs9XXzJPIFZJyKK0OWZ45HT4yRkyHeaScsqRp0Om1HHL46N8b19A/R5QpwaCpLKwFV1LrLNBn57fIz3PXGJjz3TSd9MjHddUcJDb1nIO9eUaASdSzN4IkncVgMmncSWBbncsSQfULHpNV3Tx89Oct8Tbfz88DBt42GWlmTxpRtrua4xlx5vlJ8cHGJ31wzFTiN3LCngU5ur+fKWOsx6CaNeh8tq4JXuWb64vYf3/LmVIV+MRcVZ1OfbWFPpJCWDQZT42KYKPrO5hlsX5eOPpQklMyiqyi8PD/PbMzN0+NJ0TicodRi5bVE+0zGVbxyc4js72zk36CGjqDjNGqv4uZYpNje4+fz11cTTKl/Z1c+h/gC5dgOVLgv+WFpbBVtVzIYaF8cG/GzvmCaeVinNsdCYbyOelumenmMqrJW4QUQQRFRNvFALoqKAgoBFpwkvZGStpGsz6pBVlUKHkc9eX8NbV2qIzc31Ln53czkP+w5xU3KcbqOLftlIjprkmno3BVlGREmc79OKyPPlZCWTQUnEQE5o2aOSQVUySKoMsvZ30CoNAmDQCfOi5BJOi46KHDM6cb5SJAp86cY6lpVkMRpMcH1jHrk2I26bkfOjYfJten52VzMf3FDBRCjJJ7d10+WN8Ymrq3HbtLXDFaUOBJ0B0BRqJAFM83McOlWm3RNj+fLlVFRUoNPpGB8f58yZM3R0dODxeP5Llby/1f6n91RfN5nqP+Ok9NcG1Wg0SltbG5WVlRQWFv7V119W5mRJoYk+b5Qqt5V3rC3jcy908ULLFHevLGFrcwEvtHq4tiGXpaVOrq7PZVvLFKOBGGdHQzQUZNE7PcfzFyd534ZKNta5ScsKvzwyzAPb2hnyJ1hTbOLWKonmLCtP9yZ5cFc/TYU2gvE0D1xfhygIuGwGPn19DV98qZuOqQhH+30sL3eyslzL9F6+5KEsx8zmpjxKss18ZXs3fzjnY2tjNqeG/Ny9vJBss4FfHhngzEiA1RXZbGku4HfHhuifnmNDrYunzo5x76oSre8pSaQVFZNOYGV5NudGQxRnm8gy62kdj/CpDUV0+1Web50iMJfi6vpcdrRPc6x/lhub87lmfmBpMpSgMMt4ef1iYXEWqqqyttLBk2d1PN86xfJyJ/G0zKAvjkGS2N05wzUNuehEAbNe4v5ra7j/uQ70ksTmVU2YhAyBQIDe0V76phOsLLHQVOTkwkSUvZ3TpBUVQRD57j5NM1Wdz2x2DaZh0I+iKkio9KV0XLzznTTmW/nRzl4U3xxfu7mRYwN+9nZOc7Tfh92oZyaS4vNbaqnNs7Gj3cuLLVOcH4uAonJDk5t7V2mUnNaJKL88OspsInNZSSYwLxDgi6bIsei5sjqHXm9UW7+5OIWiaP23O+oMLCrPp8cv8/CpMUYCcURgIpigxxPl6noXg745QgmZujwLU6EkP9g/iH6+GmE26Pjs9dXU59sYDyT49t4BZudSFGcb6Z2O0jIRvgx6yLMbuDQZpTzHxIXxEBlZ5TPXVmEx6jg7rFG3LqQVdCKsq8nmxqZ8bl1cwFPnJ9ndMY2qqDTkmbirXkenN8mekSSffamH+jwrc6mMBpZXIJaUuaExlzuXGnni3CS/PzmOrGjg+Wq3mVhKZiKUYCKERuAXREx6iWyzgVhaJpzQyq6qIKAqKjLagJheL5JQoNplxm3Vc3Y0hCCAN5Li23sHUBQVQVC5OBbmZ7NBSoxFHDQWkqsmuEqd5bhs52C/phmrqAKiqDGLkSETjyAKKqogaTJu2nwTovhqyVnFqheJpTMkVBFJ0lPpMtPtnUOQVQJzKsFYRjsYCNq9IwgCB/tmkYCNtVprpndmjt7pKO9cU4xOFNlU56LSbeZzL/fSWJTFG5YWXPZBDQU2RIMJORHResqygkkSmAolWFhkp3U2xZmRINU2kaysLKqqqlBVlWg0yuzsLO3t7SiKQnZ2Ni6Xi6ysrNe8TDw3N4fb7X5Nr/l6sv+Xqf4v5vV6aWtrY+HChX9TQH3V3rzARlzV89zFKerz7VxZncOLrZP4IkneuLwYt03Pb44Ok1FUbl9SgCiqnBkJcfuSIr55WxO3zdOX9nZowzrXNuZx++ICBn0xLHqRdbkp9Ho916xo5Ht3NFORY+LiWIjSbAuV7n9t/k9HUnR7IlzT4KbQYeJbu/rYdnGK9skwo7Mxbl5YgAA0Fdq5b30F3TNJHj43Q5ZJ5OaFBVxV76bIYeaJM+PICmyqdWPSS+zq8LCh1o0qiMTSsra/J0mgKtiNOs6PBbmyOofJYIL3rCvXHNbEHBvnJcCO9M1S4bZoi/E92oDS1fW5iAK80jXNiUG/huDTiRzonkGWNbWU25cW0euN0TYRZm/HNJGUwt0rihkPxHmx1XP5dY/Mxi6f+r+8vYdgSqSkuJjWqA0EkTcsdLMgK8EtpSmseoFSh5Fv3lzHg1vrcVl0ZBngwRvK+fEbF7Co2I5eEllZkc3FsTBf3dHLfU+00eWJct+V5TQU2HjPujK+e3sTekliMpTEZpIIJTIYJJE7lhSyrNyJAJgNEnu6Z/nUi71sa5vmiXOTJDMKH76ylPeuK8Fl1bOtzctDJ8ZIyQqLCu0sKLTxyWuraS60gQq5NiNT4SQ/PxfhS7tH+MOpMYLxDO9YXcKHNlaQl2Xk6YsePv5cJ13eOW5oytUGiN6ykHtWFaGgIqsQTWb40vZePv5sB1/e0YM/luZ968v53m1N/OxNzZQ4jehEkfXV2RQ7TJwdDvDwqQlGZuNYDCJtE1FQodJtISErFFglqt0Wnjrn4V+eaee5linOjAQx6kXqC2z0zib5dVua8vJSPnlVCXaDSNf0HBlZ4ab6LB68sZLqXAuPnZ3kV0dGmI2lURFwWvQoCAz5E5h0Ivl2IwgiRr2eYqcZp1nPzFxSK92aJMqzzegE0OkEqlwWzHqRUCIDKgz545wdDSOKAldU5nDHkgIcZh02o8RtiwupL7RzKaTypL2emCpgUzNkZIVbUmPaQU+UcFhN5NiNSKKIzmhB78xH0BnnA66CKGhAiLJs03xHF/SSiIAyL0iu0jsP8qh0WVCVNOp8xosg8o7VJWRklcO9sywtc5Bt0bLPne3TWPUCV9W6Ln/OTw4FySgCX9hSe1nYHaDKbUHUm7X2zfwhKhFPMhmMc8uifJR0io7J6L/rqQqCgN1up6KigmXLlrFkyRKysrLweDycO3eOS5cuMTExQSKR+Jv94X9ksVjs/w0q/U+x/1NQVVWVvr4+IpEIK1euRK/X/4ff959ZZbaBW5YUs/3CEFsW5HHv6jLOjAT546lRPrm5lvesq+Bbe/p46Ogw50eDiKKkQdJTMqIgcM/qUkb8MR46NkKR00SBw8ShXh92k8ZSfXFY5BsrirQbwahDBUx6iVF/jK/t7ObTm2uxGnVsuziJKAi8ZUUpNpPEzw8N8djpUQqyjNiNOtbX/OsNem1DLke6PbRNRbmxKReTQRtwuHtlCT98pZ8Tg7Osr3GxqS6XfR2TvOvsizTYF3OqB8qzREbDmgvxx9IgCDitepIZlT5vlHqXkdOjUd610URdnoVDfT5uXVLIVfVu/nBijOHZGBUuCyvLnRzonuZon0iVy8LikixeuuTFG05S6DBxdZ1bU9s5PY5vLs2SkixunT9sPHNhgmVlDspyzDxzYZLyHCvvWVfG917p59PbOtnanM+B3lluXFhAXbl2qn/63AShpId7l9hI+UY5Op5mNibz0fXFNJXlcnYkSNtEmDcvL+INywpJyQpPnpvkxZYpBAFeaJnCpBdZVe6kdSJMMJZhfY2LsUCMnxwY5BnHBLk2I60TEe5aXsStiws4Phhgd8c0z7ZogzalTiPhlMLSEjuyrNI+FaUwy0ihw8ipkSCHB/6197WpTpvETWVUHj/ex9HRBBlFQRJFeqfnuK7RzbuvKOFru/uJJGTMBondndMc6ffTkGelbSpMvt3E57fUEEnInB0JsqPdSyytIAoCvzoywqOnNfh/PK1w26ICrmt0YzXq+Naefrq9Ua6udxOMZzjYO8ue7hkUWcFtlbhvZQ4Lq0vomY7yx5PjvNzmRRTgjiUFvGFpId2eCA+fHOeHB4YQUNGLIncuKaRtPMhL3WFOjkRYkS8ya5OYiGr3p0EUWVqiCdC3jIc5PxZCRkRE64dOhDQtYFEUERCIphQiSW3gRxTAN5cillZoyLdz04Jc9nTN0DMd46s31VGXZ+Wh42OEYxm+cGMNzYVZJDIyX3wpTjowx1WxEfr12bxgKEU2CKhhjVYWnQdmiJLEggI7A74YcSEHMTFHOh5GEEVUQWDYn0AUNCJ7OJHSDgKSSEbOkBFEJEnCF00izoMeVEGg0GHCbTewu2OGYCJDIi3z5R09ROIZxoJxSrNNHBsMsKbCSTKjsLNjhhub81lcnPXv/I9eEsk2iHgFUFSFLBTigo5ESlP9cZsFOjwRtlTq/uL0r06nIzc3l9zcXFRVJRaL4ff76e7uJp1O43Q6ycnJwel0/pcmiKPRKHa7/W/+uf9b7HUTVP87y7+pVIq2tjYcDsffTXESRZH3X1nCnk4vfzozxieureH2JYU8dW6CSxMhlpc7Kcs2sbdrhnyHke9ubeRIn49tLR6WlGZxRaWL+6+t5YEXOvjunn4cFh2JtMI7mw3EDbk8fNbLt/f08sD1dRzo8dE/E+MjV1ejqiq/Oqztw35oUxUHema4piGPnPley8eursasE9nTNU2ly4JO+vdFirSsOdezI0HuWSVjNkqsq3LxnHuSJ89OsLbKxZbZHnapTr6qbyCWUpgSVJa7JYZURXMoCDiMEj2eCHl2A4f7ZlldauORi366vVE21rp56Pgow74Y62tcPHpqlIO9Pt55RRnXNeVxajiIiMKnrqsl2yzx8iUvOzqmec/aMvSSwJuWF/PLw8MgSpdh4+9ZV0qXJ8IP9w1w86ICPNE0n9pcQWOBje/e3sSvjgzxXMsUCAKqrHJhNITZIPFim5flZQ4qSgppG5zk0HiCYruO3gkfnSNejk/K5Nt0bF2YC0AipXCkb5Zip5HblxTyQouH770ySIHdgDecYE1lDh+9uhKAk4MBHj45ysXxCAZJIJqUmQwlWVnuZFf7NDpRZF11NsOzMR47O8ljZ1QURSXPbuDNK4pYWGjDG07y7X2DRJMCZr3IoT4/xwcCVLktDMwkcFok3r22io6pCId7Zzk1FEAQwKiX+OKWGmrzbHRMRXip1cv5MW2AKZLI8Mz5KRYU2WkZC5GWVT59bTW5NgPd3ijbWjz4Y2kkUWBbq4dtrV4kUUVW4JoGF3cuLSTHok29PnR8BLtJwh9X+MbhGa4YzVDhMjHij5NvN2Az6nj+opdTQwHevKKYlWVOxoJemB8SCsTSfPyaKvpnYvzh5Dg7hlIIgsDSAiOLchQ6/HBmKMjhPj+qAGa9xMpyTRe3fTJKny+G22JgZbmDcFLm9HCQHIuedWUmpmMqp8fmKMk247Lo+fO5KTzhJFajxLf3DpCWFeJpmSyTjpfbpjk7EmJgJsZYIM5HNlWyLmNnLhLjiwMS0wmZm5rzSCRl9vXOIitasOzwRFFUbXc2LRuZ8EnIc4H5xoWkAUJkrceLqN0XM7EMqNrgUjCuIuoM2p6sKOG2Gnjv421EkzIoKmOBOEUOE5pgjUA0IfPbY2M8cnKcfIcBRRD4yFUV/6EPqhUTeHVGkFNY1AxWeY5pycIjp8epyzXTMhpAWe34q8q6giBgtVqxWq2UlpYiyzLBYBC/38/g4CAGg4GcnBxycnKwWCx/le/8f/CH/0H2HwXVcDjMpUuXqK2tJS8v7zV5DLdFx3vWV/OzPe3c6Ilw2+JCDvX4+OVhTb5rNJBALwk4zXpKs83cvaKES5MRfnlomGq3jTy7kc9sruGjT7cTScm8dYGZTSsWIAgC/oCfF/tDfG1nN0O+GM1FWWysdV2m+Xx7Tx9f3tGNogrcvuRfy9eiKGA16ZAEgRF/nF8eHuSDG6sQBeibidIzk+DKcivHR6I8cmqM92+sQBDhruUlfHdPL/u6pmnri6LanAyYnNTOzaDobZyf0UpeqBrTNKMIdHqi3Nicz672ad7QWIxBhKN9s9y9spiHT45yuM/H268oY0V5Nof7ZrhnVSl6UcNUOq16qt3azbm+xsXB7hnetKyQLJOeVeVOfiVoQx8VLg3IbTfp+fjVVXxlRw9/PDFKeY6VVeVOANw2A29ZWULrZBf5WUb2ds+ws3MGBG2q9/xYiPNjIUCrVExE0kzM61EoikospfCeP7XQkKPDnxSYS6b4wg1NGh6xJoeX2rz8+cwEKgLecILTwwFWlWejKBCMZVhYbMdu1DLGHe3T6ESQVXjrymK2LspHEgQePjnGzo5pXDYDoXiaHx8cRkSbFjXqBD6ysYzlZVn0eed46qKX7mmt9CoAvmiSu5YV0Vxo54f7B1FVLfh/95UBNtW5KMs20+2NUppt4qbmPNqnIpweDnKwdxYVyLMZaBkPU+W2cGooSCiR4SObKllTmU2/b47fnxhjLKAB5/f3zLK/e5Zcm4GZaJKGfCufv76GgXEvR4ejHBkOcmxIxqqTeO+6MhYVZ3F2JMTjZ8b5/r5BUGFZmYP71pXwcvsMe7tmONLvI9dmJJTQBAwEQaDFm6BrVqLGbQQhgSSJlNoEZERODwW0QCVqGWsgnmFv9yyaoJyIP57h5Z7o5XGl2bkU8bSCL6qJsC8utiOKAscHAtiNehrzbczOpejwREjKCiICPzk8ws/m7xlZzWjs6Fbv5VWjyhwLdpOO9qkI967UKhAj/jifeaELQcpFjoVQ5AwJWZifpBcwSCqxf6OjKs/vy6qKgiiJiJLEzFxag5IMBbi2MZf3XllGMqPwwScvsabCycevrmRoNsYzFzycHw9x7+oySpz/sdLM8io3JwJxlEQKIZPBIafw6awcGwxyfb0Lj2cObyRJkfNvh9pLkoTL5cLl0ipd8Xj8coCNx+M4HA5ycnLIzs5Gp/uPw8v/G1T6H2T/a1CdnJxkZGSEJUuWvGYnJ1EUURSF91xZwTPnx/jDiVG+dnMDDfk2Ds4j9z46n1n+9NAQO9q93LyogE9cU80nn2vnh/v6+dotDTx3cRJF0FCBbUE9twkiogDLciWKi4r55ZFhBAHeva788umwuTiLT19Xw5de7sKg0xFNZXj1mJDKyOzvmmFNZTblLitPnhvHpBN595UVbLs4hdkgsrXOjsuZxQstk6yryWFhcRarK7IpzTHzu+PDYCtk40w3R9y13B7q50Q8zOncalLqq1hwtDH+eeFpWVXpmI6zsMDE8YFZ3rW2jKWlDo70+bh3dSlX1+dycjDIicFZ/nx2HLtJIhDPMBaIU+6ycNviAg73zbLj0jRvXlnM7s4ZVETSisq2Fg9vms9WGwvtbKxzsb/Xj0ESSSsKhvlM/IlzE9hNBr53+wIEAU4MBvj10REWFduosqaJY2BPX4SbF+Zz+5JCerxRvru3n2vq3SwosnNpPMzJQT8JWcGqU9l3oY9Ntdk4shwc7vORZZbYurCA/d0z/OCVQRxmPeFEmto8K1+4oQ69TsAXSfKVHT1MhVOIwJ/OTvDcxUkcZj2ecJLl5U7uv6YKAYEX2zw8fX4SUYBEWuXHh0YoVWKYRZVe1crK0ixKzUlaZuHhkxP8+ewEKRmKHUY+d0MN3kiSPZ0+drZrIBKjXmRTrYsV5U421Lj47r5+2sajrK50Ek8rHB/ws6/Hh6KAXiewp2uGft8cAz4tc3vXmhKua8xlNBDn+YtTnBwKAtDtneP7+wZZUajXFJJkhRKnmUg8zTf2DLC42M51DbmIooAgCph0AhfGQvzuBLxpeSHNhXZ+cmiIybCGPKzOtXBDUx7pjMLDp8Zpn4qBIKATwGyxkGuRCCUjZBBYW2ygudDOUFhlb2+AEoeFO5cWIisqvzs+isOiyfqZ9Do++0IXbpue793eiM2o47fHR0jJCl/fUkdNno2ZaJJPbeuiLNvEG5cW4Y2kONjrY9ifYHGxjfIcMyeHAgQTGTbX5zIZStA6EcZqkDg9HGBoNk7vTBSzXuJrtzfws8Mj9I97kRNzIOrZWJfD0X6/9pnUCeglHdGUjCCJiKgIkoEv3FDDwuIstl+a5vhQgBubtbv2aL+fSELmhgW5CIJAlduKXhKwGg3ct67sL/qgdVcu4BetXmRFQNEbyLhcEFWwG0TOjoZA1tM7k6Ak5+8PbGazmeLiYoqLi1EUhVAohN/vZ2RkBFEUycnJweVyYbPZLvupWCz2d5d/n3nmGb7yla/Q1dXFmTNnWLFiBaCR+RobG6mvr6e1tbUFOKWq6vv/zpf5N9nrJqj+M8q/rz6Goih0d3eTSqVYuXLlXzxR/Vfs1cBtMUh86vp67n/iHPf9qZVoSsZtNxKJZ2gosJFvN3Ji0M/jZ8ZYVuag2GnmAxuq+MG+fh54voMhf4KbF+RQVZDNTw4M8eujQ3xoQyUq85qL2ivi6fMTfOKaaqT5IHJhNIROFMkySnz55S6+srWR6lwbx/r9RJIZbmjOp7koi3g6ow34CHB62M9NDTkYJbh7RTGnh/z86sgQP37jQtKyQiKtkFYUrtMHeU/HTnrXvJOXbFW81d/OMXct9bkWuqcjl3FvFp1Ax2SEihwzZ8fnuLbKxvlJP23jYTbWujk7MsilyTBLSrLItuh49NQYoXiGB66v4XuvDLK7a4b3XVlOSbaJ1ZVOdnV6uarBzcttXlZVZGPQiWxrmeKKqmxKs7Xl/86pCLk2E30zc3z55R4+dk0VvmiKlokIb1tdhsWoQ1VVjvT5sBklthSnKS8p5sH94xQ7jLx1dQnpjMofToxS4jTy3ivLMEgSJU5tWb8u14IkwO7hGIfGZ7DpZ5iNqbxnmZUVpQZuampkX+8sf5yfWu2bnuOrO7tZUebk/GgQbzjFx66uYmlpFq3jYZ6/OMWQP4aIwPnREO9+rJU8u5GRQJwSh5Ev31RPprWNUwdbeCa7gahgQFIVhGiI6pJs3ri2nD+dmWBHlw9UmIkm2dUxzY3N+Vzb4ObcSJAsqw6zTuRPZyf409lJbEaRaErhlkX5vGlZITpR4KHjY+zv8XFldTZOi56BmSh7u2bIKNoU44tt0wz64uh1AqeHQywusXPvqhJODQU40OujdTKCoijU5Fr53A016EWRPZ0zbGuZomU8jE4UuG9tKeuqc9jZPs3Ll7x8+oUeRFSyzDo+tKGcLk+UA70+jvQHMOoEUjJcUZXNhmoXnZ4oF8dCdHnnUNEqLsfH05yanCUtqxgkEYOQYV+Xl8HZJBlVpcRh4JkLHnqno0yGU2yozmFft49ALM2+nllWlTlAEJiJJPjFkREUVeVjV1eRbzdyoGeWYX+CtZVOynJM7Ony4Z/nOu/q1Abr9CLU5lqQFWibCBFOagN7j5ye4B1rSjjWb2VP2xhKKsahXh+iIKKqCsm0iChq1KQcq5FQWiDXbqS5KAtVhUO9PmrdFoqdJlRVZVfHNBU5RhrzteA3MDPH6ZEQH9xYSY71Lyt31efbQNKjCioZs4WUIJFjkSh2GGmbiiJJAn2zSTa/xkQlURTJzs4mO1ujuaVSKWZnZxkdHSUajbJ//37y8vKIx+N/d6ba3NzM888/z/ve977/7f+qq6tpaWkBWPJ3Pch/0V43QfWfZYqicO7cOXJzc2lsbHzNg/mrmSrAzQsLePx0Pq3943zm+lpqcq185OlL/PrwEF/e2sAHNlbxsXn27zdvbWJtdTbPnjcy6E+wrNjOO66sRRBgKpjgmQuTFDvMFCrw0NFhKl0WNtS6ePTUOL/SS3xoUyWheJrdnV421Lq4e0UJX3q5m69u1wLrznZtjaa5KAtBEHjbmjLC8Qw726eRBNhcn00mFsaok3j/hkq+8nI3T5+fYNAXwx9N4DKJXBRyEe+/n5vOjfB7YzG6W5dTeEErv0mIKKq2KRhPyfROR7l1cSEvtnnIX+TAahA52u/jAxsrsepFDvf6WFScRYnTSNtklDuWFLG0zMn6mhyO9M3y1hXF2Ew67lxaxKmhAD87OEgsrXDXiiKcZj1tE2F+cnCQb93axNH+WbzRDA9cV4Wsqvzi8CAfe+YSNoMOp1nP9fPC5GeGg3R4omyp0OPIK+aRFj+ecJKlJQ5+8MoAg74YgbkkqyuyOdgzS1mOmZ8dGiLbIvH5LXXYjTr6Z+b4xeFhRvwxRKDFJ6DXBbEok7xwKYnNIPDhDZX0+1OcGvTzp7MTKIqK1ajj3EiQUDxNrzfKiD/OTc353LqwgG5vlN2dXro8Wml3PJjkcy92syzkodNRRVwycHdqBH8kzmGhnjMng5T3JBn1x1lR5uC2Rfns6phme/s029u18naezciDW+twmPUMzcb59dERhmZjiILIS5em2d7uxW7UEYpnWFys0YxKs838+ewkPd4Y1zW4KM020+WJcmLITyKtQehFBMYCcbYuzMcbSXFswI/TLDHoi/GRp9u5vjEPl1VPWtEOfxlZ4TfHx2ibjHDbony80SQHezVgQjCe4ZVuH1ub81lZ4eT7+4aYS2YQBIGTQ0GmI0nybEY80ST5WSbeuaaElKwJDxzq8+Mw66l0W4gnUgzOxomlVRwGkcHZGD3TcSIpDYt5dGD21W0X7XMwEuTMSAgErcyvEwUeeKGbjKKSSMkgwomhICeGtaHdujwb19S72NaqQU++eXM9uXYDnnCST7/QRXO+lQVFNvZ0+/jy9l7ys0zozRby85yM+6Mo8QgGQSWlKMRlPTqdnlBaQJAMXNvgRhBgYCbGaDDBe9eVEoyl2d/jYywYZ025k/3zn8VnL0zhtBh42+qS/6MPshgkdDo9GVUhmVYQ0xkKHRaSGYVlxXZaJ6P0+pL/cKKSwWCgsLCQwsJCbadXUXjppZfo6urilltu4frrr+eGG25g1apVf/PAU2Nj4z/oWf/99v+roBoIBIjFYjQ2Nl7uCbzW9m9LzIIg8NWbG7nt57NcHAuxuiKbe1aV8Ntjoxzq83FVXS73ra/gB/sGePbiJIFgmJFgkiyjjt6ZGNPRJPl2I3etKGEqlOCx02OUZ4lEkvCFm6qocltJZBSePj+J3aRDVSEjq9y5rJi8LBMP3tzAF1/u5osvdxFPKbxvfcXlQ4QgCLx5VQkHe2eQVZgIpcjTaW5nUbGDqxrcvNDiQVFkbqqUqCkp4MeHxzhgKePqdy3hyT9d5OXRBNc15fPHkyMsKrbTNhHWasDze3oimnTZJU+cNVU5HO/384GNlVxRlc2xAb+GIJyKIgoCxnlVlhub8zjQO8sr3T5uX1JAhcvMyjInZ0ZDrKl0XRYe/+CGCr6zt5/fHRvm0lSE6lwry8sdCAjU5Dbzs4NDXJqKIogK9z1+iRyrjvGAFlR2D2fYNTx4GfU24o8jSRo60WnW0T4V5tSwtpsIKuurXQTmUtiNOmbnUowF4qytzMZtM3Cw18fZMQVpnupz9yInlpSfel2CUavEaBDWVGZjkES6PBGOD/ovE4p6PBFeUFSC8QzdnjnWVGRzz+oSOiYjHBuYZXc4D1UHJfIcuWEvd/gGuHtZMV+L5F3Och0WPblZJj5+bQ3L+nz88vAIsgLeSJpv7u7nikonU+EkI/4Eb1xWxJamPLq9Uba3T9PliaKXBFonw7RORmAexF6WY6Ys20xToR2dKHBqKEhtrpnaPCtnhoP89NAIoqiiyLC+wsLdi12EVAsvtHnY1uZFURTcVj2fv6EWp1nHjvYZdnR4OTEYQBQENtVo98GBHh+7Omb41t4BFEXBatLx+RtqyLUbOTMcZH+3jwGf1vP2RVI8dX6KHIuO1skIBVkmPrO5mgKHkedbPPRdmOJdV5TRYIkih2N8tyVBvkng4+tcuHMc7O6b4/lWL+9fr1VAJoMJfndiDKdNx5pKJ8mMyonBACBw59JCBAEePzdBSY4JSVR56PgoGVXBpBP56aEhlpc6ODMSRELlXzZV4LIa2NpcwLMtU7zY5kUniVxV56bIUcIvDw0QTaQRRZFCp5WmQjuqCof7/ayr1rK6PZ0zSILA3s4Zfnd8DC3iq5waDnFqODjvXQQ+dV2NpqX7n1hhjo0Br6L1ijMqBVlGLoyF+fwNNVzc1kX3TOKfyv4VBIErrriCK664gmPHjrF9+3YOHTrE73//ez772c9y8ODB1yzBGRoaYunSpbS0tBwGvqCq6tHX5MJ/pb1uguo/svyrqipjY2NMTk5isVj+YQEVtEz13/ZtGwrs3LuuikcOd3NNvZvrm/I50u/j4RMjLClxcmW1i1MDfp4+N4EiCGxuyOP2JQV8+vlOvre3j2/dtgC9JPChTVX0zVxiOJzkhgUFVM3vpN61XCPCvNg6hSQKbKxzUzQ/wJCXZeKrNzfw0acvoarqZR3VV+2VrhkEBAqzjPzs2AQfWm6nav7/1lblsL97BqukcPeVzVjMJnZ2B3j6wgSb6rTXsa1lglsXF16GowuCgIKMiISqquzvmaHUoePiVJx3rC1gf7eP86NBNtW52Nfj45dHhqnPtyECB3p8vGFpIeUuCwuLbOxq97K1OR+9TsBkkBAEUZvOmbcV5U7uWFLI8y1TIIq8Z33VvOAzOC0GwskMBQ4Tty0uYHg2TtuYH1WQWFhkpzbfxvH5cvj337AAVPjU8x3U5pn5+i2N8wNEo+xs91KWbeHkYIBjA7OUZZuZCiWocpn5yNWVGCSJu1cU8ZUdvfRNzyEK8OfWABaDSJZRhycSZ0m+kQ3uOE6LAZ1q4tBgmpVlTkqcJvqmo7zS7SOtKIjAiD/G7o5pKlwWJoIJrHqBTb5ezloK+IV7Ndsd9RijuZezXEVR2dft40ivn2VlWVwcC1PsNPPp66q1bK7Xx5MXPCiAVS8xHU5wfGAWTzhJtzfKhpocPrihnNm5FI+dnuD0cJACh4lIIsMfTo0DWibnsurZsiCPleVO3rSskK/s7GPUH0cUVI6MxPBEM9y8uFhz9qrmwGciKR54sYdr6l2sLHNwuH+WtJxCEgQO9QcIxTPcvCifa2SVZ1s8mA0S8ZTCt/cOsKg4C4NOZHouzaJiO5sb3AzOxrk0Geb8eBjQYAYfe64DoySSyCjkO0zMpTL0dA3RNhkjai3iW2OHya1bT1dC5sW2IIvydCxwZnBk6dh+KQSofOGGWgqyjPz53CSRRIY3LitkPBDnUK8PRIHxYAK7QUJWVJaVZJFrNzAwE+OJ85MoisKmWhc5Fm39zqTXIP46UWBBgY3Hz0xQ6DCxNF/i2KTAx6+qYu280MX9z3dSm2thPJDgkVPjnB0OIIgSRp3E1oX5bL/k5eZF+bxpWRGheJpv7x0gnuHy1Pt/Zk1FTga6JdKKgpBMk2s3EopnyLcbKcsxM55IE0nKZP3nl/qHmMvl4s477+TOO+/8i99z7bXX4vF4/revf+Mb3+DWW2/9D3+msLCQ0dFRXC4XgiB8AnhBEIQFqqqGX6vn/p/Z6yao/qNMlmU6OzsRBIGVK1dy+vTpyxnKP8IkSbpc/n3VPnpVNbsuefj10WG+c/sCPrSxmvufbeehY8Pct7YYXzCEomoaou9cW4ZZL/Hhq6r49p4+Hj4xwn3rKwjF04TjGSQEzo8ECMaKcVoMCILAu9aWc34kiCeaoibv3w9cGXTawJBRL/HNPT18/eYmCp0mkmmZ3R1eVpQ7efeV5Xzm2Us83BJmYV0GvSTwm8MD2HQwl5I4NhTh+gVm3rqqlC9v72J3h5ebFubzcruHvZ3TXFXnZl+3l1yrHm9EISNnyDYbCCcyWPV6xqMZin/0bRzVN3LwaCfZ5UWoqibM/bkbark4FuKHB4ZpGQ+ztNTBrYsL+fruPg73z9JUYOPEQIAqt5XTw0F6vFGtZwS8YWkh29s9JDLQNRVhSYkDSRA41j/LaCDJJ66tZk2Fk9bufg71yqwoc/LA9bXsaPfgiSR5+5oSBmbmePTUGKmMzOqKfDqnInjCSXa2e7m63s0H1pcTSWbY3TnNsxemyMgKk+E4fzg+xrrqHM6PBumfnuOdV5Sxqc5F63iYly956JvWZNpaPElaPAIGXZpURqHILlKij1JtE5iNas56fXUONblW2ibC7O6cIS1rXOXNddmsKM7jrZEAhxD5Q9BJejZBod3AVbVuKt0Wbl1UwG+OjXB6JAiqQLnLTDytcMOCPGRFZcAXpy7PitOko3UiwtHBoLbfiUjHZJhv7uknlpIZ8MVYU+nkQxvK0UsiT56f5IVWL26bnlRa4eeHR9CJIxh0mtj3B9eXsazUwYvnhzkyMscP5olUy0qz+MQ1lQRiaba1etjTNcOOjmnMOpFPXFNFY76NPV0+dnVM8+CuflChocDKA5trCMQ17vLebp+2PwsoqsrsXJr6PCv7umfIt5v4+FWVRFMZur1RXmzzYjZIKLLCsxemUHGiWJ2IgsCDBVeScz6K36YHQcDttLG3L4I37OXMVIamXAMHu6YYC8ucHQ0hCvDMxSktMEoiNzXnMRFKcH40xKryLN69tgynRU/beJhv7u7DZZY40jdLOJHmI1dVoSgqe7qmWVuVzYc3VdIyFuaxM+McG5dBEDk3EmIilGA2mmI8kESvE/nOK4OY9SKCKPGBDWVsqnXz8MkxJAG2Nudj1IlMhpJMhFJ8YR4x+ddYaY4ZUWdAkVPIgNtqQJ2X8btjcQE/PTnNcCBFce7f6fD+RlPntwT+Gtu3b9/ffH2j0XgZKauq6nlBEAaAOuDc33yx/6K9roKqIAh/9Rv+11g8Hqe1tZXi4mJKS0uBfy3PvpbDSf/WJEkimUz+u6/ZTDq+cFMDH378HDsuebl1cQF3rSjmT2fG6ZrwE04J3NCcz96uaR45Ncr711eyqiJ7vic5RV2BjT3t0yDAXQ16nh9I8929fXx1ayN6nch0OMnsXIock44/nhilOtdG3byu6N6OaVQVPr25lp8dGuCLL3fx1Zsb6JgIE0lkuHVxIXk2Ix+/upKv7R7ge6/04BDSzMxlePCmOp46P8ljp8dYVZHNopIslpQ4ePbiFFfX57K5IZc9HV6+eGMde7qmKXSamZ5LgQqRpIwqCHjCKRQEHqy+Dr2gcjamR+qepT7PRp8vRiwts7IiG6d5lL2dMywtdbCo2E6Vy8y2linO55jR6yU+cW0VX93Rw6+ODPPd25sw6EQO9PpIKQIrKpy80Oal2xPl3tUlPHlukqpcCytKbPR0d7NvBGQEtizI46lz4zzXMoUkwCOntDKbMp8pPn5m/NWqG0ZR0/oc9sfItxs5NxxELwrcd2Ul3Z4oR/v97OuZQVGgyKnJFU4GEwTjafpn5lhfncM715YxPBvjmQtTdHsiZJl0zMRkXhyUYTCAoqo49KBPz+HWG7myysmliTAuq54so8DengBH9TlsrK2hZTyCSIJr612cHQnx6W2dbKpzaSX0ySjFDjO1eRZODgbn+5w6gvEMjYV2PrW5hiyTjiN9s/zs4DBVLjMrK5xMBRNcmowQiKcRBYFTw0HOjARxmHQE4hnq86186tpKrAYdp4aCPHR8lGhSWzP589kppsJJanL0XJySCCdkcsx6Lo6H+cgzndyyMJ+SbK1i4jDpiKVkfrR/iPU1OVxT76JlPERkOobdJNHjneOjz3awsdaFN5IgkZFZV6WV18+Phnjk9DiKoiJKIguLzXijCardVh49PY7ZIPGtWxrJtRuY7R3m0wensAgq12c8zIpGehQz4aRMlknP8cEQSflVpy7QOZOic8YPKuhEuLLcTjgNFye039VL7dPznw2NenZu5BLLy7PomIxQ7DTw9a11HB0I8MdTE3x5ew8LCu2kMnD7Yg0wsqQ0iypXNe9/shN3lpFOT4Rjg/5XN9BYVe5kXXUOOy55mJlLs7HGxVwqw8FeH+uqs8m26FFVeOr8FMVO07/DEf5nVpptQtAZEOQUKBpxSgBmoinq8m2o8hQDswnW/Z3+7r9q/6ikZmZmhpycHCRJQhCEKqAWGPyHPNhfsNdVUH0t7S/Jtf2jg+q/HVT6t3Z9Ux5XNxXx1IVRVlc6kdKaokcoKfCxa2vYUOvCpJN4sW2KZaVOVlVk89ZVJfR4I/zi4CCyKvDJzTXYouN8aGMZP9w3wO9PjPD+DZU8fnYMnSjw5Zsb+PbuPr61q4fv3LGAbIuBPZ3TLC9zsqzcyVe3NvKV7V184cVODDqRKrflsvh3XZ6VW+uMPN89B6rCdY1umksc5NgMfPyZdn5/YphPbq7l3jWlfPLZdp5vmeSOJUXs7ZxmT4eXleXZXBoPYhA0PUtVJ5Bt1lEYmWEUCzOSCZecQNAJvNUww6qrmvnI0+0c6fNz+5ICrq7P5YU2DzORJLl2I3cuK+I7ewfwRtK8eVUJBVkmPrC+kgd39fLo6XHuXV3CcxenaCyw88B1NRzp9/PIqXE+/1IPqqpSnG3iV/s6iGHi/MQcRp3I13b1Xe6TXlnjRgCO9c+yZUEudy0vonMyyk8PDWLWCRQ5jOxo9/LSJQ9GSZtI/fCmCjbUuri2wc2CQhu/ODxCfpaetCzzx1OjgIaFM+o0Sa5tLVMM++N0eyJc25DLfVeWkZIVvr23n/bJCE0FNmRF5chYjENjHhRZwW4UuL1KR5kdDO56/nR2kl1dM6Cq3NScz72rS3j7GpXnLk6yvd3DgV4f2UYdn9pcTUm2mXesyfDIqTEO9s4C0DUV5V2PtmA36YgkMuRY9ayrzqHQYWImkiKUyHBdQy5bm3MZ8sXY0TFN33QMURDo8c7xvj+3U+k2MxlKoqLylRvriKUyHOidZVurd57PC/euKubGBXm0TYR5rsWjBUJVoTTbwhdvqEFRVV66NM2+7hn29/jQSSJ3Ly/itsX5tE1G2N0xw8uXvCgq5NkNLC3JYnVlNjcuyOMLL/cQScjU5llpHQ9zcjAIaINHqyuyGQnEMRtE/jiUIiHpeNB3kkqzSjSa4F/yNlKXa+Fr82X9x89O8GKbl4/PZ807Ory82DbNO1cXcnIwQLtHw1yKKKCqNOVZ+djVWua9v9fHK10+UBU+eGUZZoOO6xpzKcwy8r19g4wGEqwsd14+TACcHg4hCwIf2VRBbZ6VjKLykac7KHNZ+chVlUxHknR45njTskIEQWBn+zSJtIJOEPnt8VF80RSD/jjfuKUBvfTXDxZV5FiQ9AaUBKiCts6kojGPl5VmYRIyDPlfG+zg32KvleTctm3b+PCHP8zMzAw33XQTS5YsYc+ePRw5coQvfelLr/r3Z4H3q6rq/7sf8G+w/3FB9T+Ta/t7hMr/GvtL1xcEgS9vbWTLwAyfea6dcEqhwmVmPJDg1KCf9TU5vGVlCW3jQX55eIiaPBs5Fj2ry7Pp8kRxmPUsL3PS2zXBuqochpbEeKF1CptB4sSAn7tWFFOeY+HzW+p54IUOvrGrhy0L8gklMmxdqJ1wy1wWHrylic+/0Ik3nOSqFe7LbcpYLEa1KUGWSSIYky8PBBU5zbxxWTF/PjPKmeEAqyqy2Vjr5qVLXtonwsgqnBjSBjxAJddqIKWkkFNxzA4nAzoHW6baeKlwKV/3n+bT2WvoMmdxq8NEfZ6VQz0+bltcwHWNubzY6mF35zT3ri5lWUkWBhEyAlzXqNWoFpVkccvCAl6+5CWcSBNMyHxiZck8xs+Noqja/q4o0joR1SY+1SiSoOmUxtMq50eD3Lu6hMlggv3dPpxmCU8wwfdfGaTXG8Gupvhm104KM1Fm1l/Ng6lSRoMa7ednBwc50DVNnsPMod5ZGgusfPGmOgySxK4OL384MUaB3UCFy8pUKM7FsRDy/FDSvp4Zjg7MIqgqiYzK2iqtElGWY+KRU+Ps7vDithkIxdL8vj1BRZZIgaWfHq9MjkmH26ZnZ7uX1vEQ71hTRoFORpXBKScJIfDpZy+xvi6XHKtGsmousvPxa6oZno1xoMfHicEARr1AIiXz+NkJUOdnyoDW8TCBWJrZuTTDs3GuqXfx9tXF9Hqj7OuZ5dRICFVVkYDt7V421bq4vjGXzqkoiiojqAKPzvdktzTlYpofOsu1GZkIJvjkti5uXJDHkpIsjvb7EQUNjPDE+UlaJ0Jc15BLIq2puawodzARTPDLo6P87vgoOkkkrcCXbqylLk87hPzh5Cj7emYptBtoGQ9xZiQIaP3fGguMCib0AT8vuZuI6Ex87soKJEFgMpRgR8c0G2pc/H/svXeYG/W5t3/PjHpZSavtvTf3XrAxYBvTeygJCaknPQFCOmmQAqQTQkgCCQmh2wYMNrZxw717197ee5NWq1VvM/P7Y9Y+yXlJTqg/3vec57q4WK+tmZE0832+T7s/y0pdtI2GePnMOC6zjkcPj2g6sgYJi0HCE4ojqCqtY2HuebmJy2pdtI2E0CNjNYk89Hov37FUUJFlY1Z+GguKHBzo8dPtjeANJc6pxrzeMUGmRaQiU3ue2sZCTISTfGiJVlvd1e5FECDHYeT+7Z2c6JsEQWBPxwRWk0QoJlOUbuGKmdlvah0qdZtBZwABBFVle+sEBlHAE4xroBiTSpc38qaO+U7YO8X9vfbaa7n22mv/j99ff/31XH/99Wf/OP9tn+gt2PvKqb7d9G8qleLMmTOYTCYWLlz4hjuid9up/tdGpb+3dBN8oMbM4/VxziuzcvuaCl4+PcoTRwbZ2znBqsoM7lhTwVc3NPHQrk6unJPL344NUp5hoXsiymMH+zjfLaCoCh9cVEDneEiDCFgMXD1Ha2AocJn42sWV3PNKK3851E9RuplZf8cHLXCZyXUYCU9HUqUZVkotSUZHR2kJGgnEYhS7Lfz5UD/lGVaqcmxcMyeXg90T/GFfLw6Tjj5fBFmBAX+Uy2dmsb15nCybgWA8hSeUnEYCKowGYqiChF2JowoCh2IGlifH2BKvIBhLcWF1Jr/b10fbWIiaHBtLSl3saPFyw7w89nVOkFQFBAS2t3i4fp5Gh/rQ4ny6vWEOdk9SnGGjPNPCS/UjPHN8kFg8jpKMI6jTipqihCrpEYxGJiMpmkaCmHQCTxwZREVFBBwmiaGpKJ5gAlWVmRQkvlx+GQviHsb6JAYsST4zfpwZapB9ukx2U8mZUSMg0OEJ860XWzDqJNrHwuQ6jHzivEIKXNr4w8BkjItrM1la6mTQF+PF0yP4IylMBpGDPZMc7JlEEkBWVGbn2bm6RMVscdE8pef5UyP0TKUwSwLX1VrJN8ToyjSztSfJj7Z2ADAj6uG78WY8GNgkZLGrQyChgCgIxFMKz5wYIhyXOdw9SbHbzBcvKNMEHfb3cahrksVFDsoyLPRPxqgfnJoWSBDY2e6jxxclw6pJjhU4jHx4cR5NIyH2d/k5OdCDrGi6sJ+Ym05ZdhonxlK8dHqUX+7qBUFldXUGn1xeRI83zIb6UZ47OYI8PWbz/curyHMY2dk+wStnxvjV7l4QYFWlm08sK8SoE2kYCvCbPT0E4gqiAD/b0c15ZemkW3XsaJ1gRUU6X1hVQkpWOdI7ye/29aPXiwwkBH6VuRBUAQWw6yWeOj6MwyTRNBICVSAUT3Hn+kYG/fFp0WAVh1nHVDRJLAXRRIIsq56bF+binx47+8NRLzoUPlijoyTDziMng9y3rYsfXV0NgjautajQQfNYiO9vbufuSyqRVZWOiRiXlP8nvm9/1yRGvciCQgeyorKjxYtZJ/Dg7l7MehFEkY8szueSukxe7/Dxh4MDfGVNOZL45tKlGTYDBoORlKqCqlLgMDLkj9Pp0YD+GSboHA++q/0lb2ShUAiL5c2TnP5vsv9nVGpCoRBHjx4lOzub2traf5pieC8i1TdK/05OTnLixAk+s6aOheXZnBkNEYyluGp2DtXZVh7d18tEKKGl8JYVcWowwH3bOsh3mbnnyjqun5vLjpZxGr0KiqwgSQLzipwgaPWZpPyf55ydn8bls7KJJmWshn9sbOjxhmkfj3DDvHxK3VYe2NbB5sYx8ksr2dkfZ2aenXuvrCPDZuL+7R34Qgl0Oq37eDIm862XmvGF45xXlk48pTKvyMnNCwsY9Mf58JJC9JKIJAgIokgikcQoCQytWEuOQeFA3XkUL68jpai8cqydOjeY9QI727wAXDkrm0gixctnRnnq2DAz89JYXOJk46lhxgNanVoSBSoyrSAI9PtifOTPJ/jLwW5i4RCk4ucib6dJRJWTqLEQySkv9V1DpJJxUtPfjVUnUOjUUqDeUIJil4n7hA5+MPQ6S+JjHDJk0WlyY0tE8DiyiFgdlJlUplQdhQ4jn11ZzGUzspBVldaxEIqqMByIce+rHXz6ydNsaxnHbtIRiCY52R9gS9MYgWiKz68q4Ynb5vHz62eQ7zAiKypmvcjp4SD3Hw6xtTfFoR4f8aTM7Lw0zAaJR0/42TpqpqYknyy7tg+WFJlmcwa/M1eQFCVqIxpBqdBlZl1dFjpRYF/nBAe7fShA30SUuzY2cdvjpzjYNYnLosOoE4mnFPomIsSTMh9dUsD3Lq/ixvk5BKNJjvROIssK4YSmw7m83K2NgKgqDpNEMJ7iN0d8PHnKg9UgoSgaWtFt0bOrbYKvvtDMaCBORaYVFXCYdYQTMj/Y0s7G+lHqcmyY9BI6SetA39vh47PPnOHxw4M8c3yYaFLlq2vKuHN1KdVZVrY1j/PXI4PoRIEMq57hqRiCAFubxjHqBH52XR13L7dx76XluCw6rAaJ2hwboViS+sEA3lACnQ7GgtMzmoLA1bOzcFuN+KNarViT/lPxhpNsPDWGSSeRklV0yMiqgiU9h4JMBx+bbSGRSvKjza08trcHQRT4+PJCvnNJBbFEku9ubmVj/SiSKLIgR4taE7LK4Z5Jlpa6NInATa0Eo9qozSeXF5Bu0VM0jZSUFZX1p0aYk+/goqo3P60gCAKZLjtnF4jr5uZi0gu0j4dpGg6SYRGZCsc1EYz30P5fRxTC+yxSfas2NjZGV1cXs2bN+m/xV/9/pH8HBwcZGBhg/vz5mM1mfnyNiase8vP7fb18/eIKvnhhGV9Zf4aH9nTz3curKcuwoBMhnlL42LIiLEaJmxfm0zwS5KX2MEvroqRZVdafHKYiw0qvL8yDu7r45iXVnN1L9E5EsRp1tIyG+PPBfj62vAhBEHixfhSTQeSyGRnUmQNsFM283BGjwdtJLAWfXFFCmlnPNy6p4psvNHHftnbuvaqWU/1TqAqoispHlxZxXrmbnokIf9zfy0+vncmOVg8b60e4anYOG+tHEIxm5CkvitHAydEoF9Tk8Eqjh7tm15DX20rjhMJ50Qi1Ttjf4eWyMiPFWW5m5Np4sX4UVRD55HnFmPQid6wP8PDeHr57eQ2BaIpXm8dxWYz4wjHkVAIlJU/rWgqaiogoMXl2sRC0axYSEdRYiHBYRDDaUA0mInGtrm3WCfT5InyTKtIzCwjozNjVJJeONdFhcPJCegkbTUUoioJJkSl2mfBHkoyHEgz4YszJt3Pn6nJ84SR/PNhHy2iI6kw7aSaJgckoR3r9WmQsiPz29V6ePTFMJJEimlC4ZX42FXo/CUsmO7qCvN7hA1TKM8zctrSQXIeRV86MseHUCIe6J5EkgQ8tzueig6+wIWBmu7OS3aYC1DRw6GBhkYOCdAvhhCbkPbfAwQcX5TM6FedvxwbxBhPU5Nq0udmxEBPdCe0zE2DTmVGqs2xMRlJ4w0lWlLmYW5DG0T4/W86M8vKZMVBVZuTa+fq6CvqGx3nl9CgnBiMc6uvFpBP5j/MKWFrq4mjvFBsbRvnN630oikpdro1vratgPJRgY8Mom86M80LDGEadyJcuKGZJiYt2T4RXGsd4tXkMVRWozLJiMYjMyLVTnWXTVGGSMoVOMy81jPPi6THSjDoCCZnbFheQYTMwgUrDcIjJaIpvrC1nfpGDWErhKxuaKXZbuP+aWoKxFHesb6LAaWJX24QWDaNlDNLNErcuLiIlqzx5bIg/7O8j3Qj3XV3FX48M8tihAUour2ZBbRl32gP8eHsXoxGF1UUGRvs6cKSl8dUL8njowCj7OibIdVnObfCP9fmJJBXGgnHueqEFARWzScdDN86kcSTI4FSML5xfzLA/zuamMXxRmV+tKXvLkWSJ20afTj9Nc0qxvCydXR0T3PdaJxcV6EBJ0e2N4P4XdKZ32v7Xqb7H9mZvHlVV6ezsJBAI/Ntybe9F+vdspHoWh5hMJv+BGlKWYeX2NVXcv/kM+7t8rKxw85GlRfxhfx9/OtDHrnYv6VYjiZTMYwf6+Ol1MzHqRe5YU8HtzzTw6909pNtMyKrKnWsqODUwxaMH+nixYYTr5uXSMR7izHCAjywtxhdJ8MqZERIphWvm5nKge4J1tRn0draRn5/Pd2e4+fXOLl7v8JJhhGybVoMuTrfw5dXl3L+1na9vbGZgMszKinT6JsI8cXiAeUVOPr2yhO+90soL9SN8akUxP9jchqIo5NiNjAdjpFCIRqMkVMh3mJAVhRMDU6ysdPPsiRHM6ZXcdJ6bky+2cHQwzOzQFFm6GI2yyuwCK/lOE4Ig8LFlRfxuXy/rTw4zHkwQS0EsEQZZRlVlBFVGRUAQJARR6+xE1P0dwDyFqKogSZrUViJGSk4g6TV9znBC4Oyt55WMGpNVkUna07hEF0UfGuSovZCcVJhMi57m8TAHeqfO1SRPDwW4c0MTSVkhGJdZWuriipnZOM06frqjC4Mk8KkVpaSZdDSPBtnWNE4smUIQRJ4+OUZemoECV4iGwSkyrXrmFKZxqGuSuzY2s7zMSbbNhKKomA0CiRSsPznCVM0qLji2jcnQEIfSinDrZFS9gZdOj4Gm4IkoQMPgFKcHAyiqRuGvyrJSlmFBUbT6XppR4kOLC0kpKmeGpjje5yepgCCo2kxuhoWr5+QwMhVjyB/DadHTPBLkM0/WsyBbwul0kRr1km0zEE7K/Ob1fvZ0+LikLhOLXtQiVIuO1rEw39zUyg3z8ri4JoNT/VOkFBVZUXlwTx+rKkJcMiODUCyFKIosKEijbTzMvVs6yXcZSSRVgjGZH1xeSVmmlclIgg31I2xvnkAQBf5yZJDXWr1kGFI0TkRZXOxkfpEDgBcbxvCEUnz/ilLCCZkHdnQRTiiEUzEEFQQVZBQEVWUyqvB6+wQ5ThOBSByToBBPqhh1ErdfWMbXXmjlwdd7uf+aGupy7diMEsGYQmVxDjXlLgKBIKkpP8uyFF6KCIwGE/xwfwLnyTNMRZOgCvRNRFhbk3GOJmXSCfzt6CAmSeCxA/1EZU2LdU1tFvMLHW95LarMsrDPYEZNxhAlkQKXCVUFu1HH64MpVEGixxth0bQAxXth4XD4//n07/vKqb4ZSyQSnDlzhrS0tDcl1/ZeRaqJRIL6+vp/ikP82PJitjWP8djBAWbkpXFJXRavNo3x8plR8p1m7r2qloHJGD/Y3Mbjh/r49PmlZNgM3Fhn5fEzIfonE3z8vCJyHCYuTTPSMhrkyaP9VDl0vLS7ETsmLvG1YbxwFUadyIZTI5zonwQVqo0ByssrsVptqKrKWDCGxSDiS8h875VWvnmJhrdbUpLOjHwHZwYmqc608OULy+mfjPC1jU08sreHr11cyeqaTF46PcrSMhcXVLp5+cwYn1pRxCP7+tGZbKTCk4h6I0NTUTJseg53T/Kx5UU8e2KEfZ0+rpubS0WGhQP9UdbNqeHkCa0zudsT5nj9adwOG3MyHKwsd/HciWEUVVv8mK69i4KAIoiaZqVOIJFUNQcp6VDkJLKcRBIlREmP22okmkwhq5qcmaIqmkg0ZylTIiZJoMoO/oSRF6RS7e9MYBIU3A4LztxMhkcDCMDldVnU5drp9UV4rWWcqWgKvShxuGeSw92TTE9uUOQy0T4ewmaQ2N3mRScJfPv8QkKTHnySm1eaPBzunURQoSrbyuUzs7l1cQHPHR/i1aZxFLSO2DsuKsNikNhwapTNnRNsci5FEASKHUaWVrixGHS82jzOeCDBskIrJXaFNk+MM14FnSRQ6rYQSchsaxwnqXVxERPgxYYR8p1m2saCiILAR5flI8sqR3om+duxoXMdvtfPy+X6uTkcbO5jR1eAfYNJVMVDYbqZu9aU4bYZ2d48zgv1ozywowdRULlpXi5XzsrkaO8UGxrG+fXuHlRVxWU1cP/llYiiwKbT4+xq87CjzYMoityyMJdrZueQkFX2dU3wxJEhwgkZs17iUI8fu0mHTtJYxIXpZu5aXUrDUJCTA1M0DEZREDja5+dTT53GadJUoZwWPb/f189IIA6qVttV0OqrKip6Aa6YlYNBJ7KxYYRT/T5WVzq5Zk4O39zUxoO7e7j3ymo+v6qYe17t5NkTml5xKC5T5DLyl8NDzMi1k+1y4nI5ebKlg3SLzEdnWzgxGGQyJuOPwNwCO7evLuO5k6NIIuSmGfjS802MBeKYDRLnlbsY8sfo8Ma486Iy3o7NL3TwZ0lCTapEEwqZNgMC8OElhfx2Tw+CKNMxXWN9rywUCv1vpPp+tEAgQGNjIxUVFW9aru29iFQTiQTHjh2jqqqKzMw3nq6WRIH7r5vJ1Q8d4Ld7unGZ9Qz4YuhEAZNeIs2sZ47VyNWzc3jp9CjzihwsLknHZZGQABmVTPt02kYQ+OyqUnq9Ie5/tZWwzsqtnfuwvXYIZXiAD936YVwWPX/Y34dOFAmZszFbtA68Pe1e2sdCfOGCUsZHBnmxM8q3Xmrmu5dVc6R3kqahKXIdRto9EY73T7K4ROtS/uvhfna2evjo0iLqB/z8Znc3d19aQcPgFOtPjrK4xMnhbhWCPpLBCTY3iqyry2Zbiwe7UUdNloU97V6unZPDZbOyeHBPHz/Z2kEkCZ+/oIxf7eqhJe7iuqx0xr0TBP2TKKmklkYVJa0ZChUUFUHSSDaxpIwAyKoAqoKiKOj0JhYWu2gdDVKeZeXaOTl8e1Mrl9Rl0jQcZCgQn061q6goxGVo9GsRliCImHRQnmUjzaSjbSxEc6dX66hE4GC3D28oTtdEhFBC4Y6LyjmvIp2+iSg/e60TTyjJrHw7saTMwW4f4fg0vhJ4+MAQswpd9Pv8+CMJ1lRnYNCJ7G7zcqB7kposKwP+KAadSHW2ldbREN/a1MbMXCsWg/bY2k06ilxmvOEkz54YZvrCEBA4MhDiMFqN0CBCsUOPjRihlEpKVanNsnLFrGyGA3EOdk9yfLqDFlQOd/lYWupibmEaXd4wTrMOSYQNp0Z4rXmMOdkGJhMikihSnWOlYzzMnRuaWVmRTvV0h67FoM1FPntqlO6JGB+Yl8OFlek8eXwEvSjgjyT4xc5ubpyfx83zc+n0hOj2RtEJ8NTxEbo8Ua6ZnUWPN0o0qXD5jCx8kSSvNI7y8plRLEYdSVnlu5eWkOs0k+s0k1QUGoZD3DAnG5tRx8BklGN9WurdbdGTYTMQjqemhc1VApEkqgBWvcCPrq4l32Fid9sEckpBJ6h4Qwmy7EY+ubyQX+/p5dWmcS6fmc3aGjdbGsdRBYG1NRlcMzubr73Qyq/XH+e+lk140jJoyF/LTQsLqS0wU+A0cWJC4oxvlNWFEi0trexoSWA3Sjy4pw9JULGbdfz2ppkM+WN8a1M7H1teRFH6G0u7/btWm2NDkPSgqASiSWbk2jQJO1Xluko9z7anOHIOgfjeWCQS+V+n+n6zs3Jtc+bMeUut2e+2U/V4PIRCIZYvX/7fXl9ZhpW71tVw76bTCKkpbliQR0m6hZ/v6OK540N8cHEhH1xcwJmhAL/d00P2lSb+0hAkzSxhN+n57e4eyj9gxW0zYtZLfC0/yZd9EqKicMlkJ1isiC+8QOqmm+kd0aKA0kwrjx4c4LVWLzctzOevhweozrZyQVUGjfER5lxRwX1b27lzfSPRRIqlpU6+cEEp33u5lV/t7ObH1xi5anYupwb8PHawj5ocGzctyOPhvX3cuaGZSFxGBbx9UwiSAAYLajJENDDFqQETKVnleL+fC6sz+N3+ftrHwiwrTeeP+/rpnIjx6fNLWVGeTuNwgFfOjBNJKuxuHSOZTJ1zqLKiaKnd6YhVg6ULWmpX1ebgFFVFrzOTa9fzzXWVPHdymPUnR/jQonxWV2fwWouX715exS92dmEx6Mh3mjjWHyAvzch4UANWlGcYCScUmoanNMVOUWJWvp0PzMtjOBDneJ+fY33+6VSryF+PDHK4d5IeTxhvOMkXLihjZUU6nlCC77zciiTA5RUWQnGZwaiOfV2TKLKCzSSRaTeyrjaD6+fl8tPXOmkd0yKIqkxtnOKLF5TxQv0I21rGSama47QadNjNOlIKjIfipFv0rKnJQCeK7G73MhqIk+80kZNmYiwYo3M8jqJq4yyt42EG93ZjkCR8UZnqLDM3LyqgbSzMoW4ffz0yiKJqHb4fXVrIomIH2463sbMvwd7+KKoK55W5+OiyIgQBbRyqaZw97V7STDq+ua6KQpeJLY3jvNQwyrH+KQRgTr6d2y8q5WjPJOvrR3lgRzd6SdsI3X5hCTPz0tjSNM7WJg9H+vwIwPkV6Xx0mQZvmQgl+NmObjq92iztAzu6uagqg/IMM8+eGKYmXZqe+RQ50jvJ7s5JbltayOUzs3mxfpQj/VMYRIVEajpDoSpEknDXxhYWFtg40T/FLGOUxaUuHm0KsrV5nEvrstjb4eO5k5oy0tn7RyeJ3Lo4D5NOx+eFPh6QC/lz7iL0qSR6FVbbkyiqCQF4vcNHZZaVRXVlPPR6L4lUHFlNMTtTR4MnxeUVFrpGJvnz8XEybEY+veKfS7v9u5ZpNyJIWk3VG06QaTcgTFOV6jL0zAga8cf/t1Hpnbb3Vffvv0rhKopCS0sL4+PjLFq06C3POr1bTlVVVdra2hgbG8Nisfzb1/fhJYUsrcxGr9ezujqT88rdXFTtZuOpYZpHAuglkTvWlJNIpvjWi00EEzJfOC+fr11cRVKR+cWOTuTpGu5IVEGVJBRR5A+lK1FEERSV+tNNHOiPsbQ0nZ9eN5OvrK0kllJ5YHsn/kicy2fmIE5/9rU5dm5bWkQkKQMKF9dmYtbr+Ma6KixGPfdt6yAY0zRhdaLEN15o5pG9fQhoWqprajNZUeEGBC6fmY3e7tIcYTxM79AYsqywtXGM5WXpGCWRnW1e9nX6iMkqoihQNE0nqsmxIcsptp8ZJBmLgZJCFERAoNghaZP/ioIqCAiCxh1GEEGSUEWJApeFz6wsYTSYpMMT5pK6LPQ6gc2NY9y8MB+9pP38xQtKGZmK4TDrubjGzfBUjPMr0qnIMNPpiTDkj8I0H9luEGgcCvDb13vp90W0mqRJz1fXVPCp84qoy7VquMhwEgWBB3f3cNtfT/HF587gjySpTpeQJJF5lQUMT8UxSwI3L8ynKtPGs8eH+MxTp/n6C820j4VZXe3mpgW5eEMJfrK9ky8/18i2Zg9GncRN8/O5dXE+OQ4jJ/o0dKOAwGQkyfMnR3jq2BAjUwn0kjYuMhGJMxqIY9SJ3LIgj7svreS6ubmogoQ3kgRU2jxRHtndycjEFJGEjCDAnII0dJLAg3t6+NxT9RweVRgOpih2mVla6uJI7xSfe+Y0zx4fZsgf07qPnWaiCYXvvtzKU8eGWFrmoiTDggCIqJwZCvDMiRHmFbu4+7Jq0kw64ikVRVHZ3uJhaDLC9XOyuKQuA0FVkUTY2+njh6+20zwapH4wQPeENsr1pQtLyLDqeOb4EPe+2kFKEcjUK3Q19zE1PsHjhwcpcZlZVeHmtRYPTx8fAgQSSUWrMasK55e5uP3CYs4vSeNwXwAlHuPzRzdwxdMPMd8l8vTxETyhOB9fVoCsqDx+eJAHd/Vg1AmkZIWDXX4Azju9j6un2tiSVsqrrkqWTPWTMdqPoij0BWSGpjT1o/tf62R/pw+TQeKBa+oYjUnYDBKbO0L84LV++ifj3FJrIOjzkEgk3taaZNSJmI0GVFQG/VEsegmbQdS6nyWtAWzUHyGSePeCjP9qoVDoHdOufr/a/xWRajwep6GhgczMTGpqat7WXNW74VSTySQNDQ04nU7mzZvHoUOH/u3XiqLAT6+fxZUPBfj1rm5+eFUNHz+vhObRIL/e1cXPb5hFps1AVpqJ/skoC/IslKQbcTlNfGZlCb/a1c0zx4e4YV4ej43rKYqOc/5gI09Wr8ISDXFDqoeGiJWo7OfGBfkgwHnl6bgtBr71YhMGncQvd3bRMBhgnk2l3xflTwd7yU8zogI/3trBV9dWsLDYxTcuqeTul1r48dZWit0WIikZWVaozLTw5YvKuH9bB8f7prj/2jr8kQSvtXgpzLDTEzZDIgLxMIqcoGUwwVdfaKIiw8yedg+72ycoy7IxOhXjwT1dyIqANxBBlbUFHxUEUa81I6EyMCWf/fAQBQGLXiScnJ5vFg3okLljaTpOpwWjTmJ3u5dPryhhVYWb1zu8fHBRAVfPyeHZE8NcPTuHq2fn8OLpUT6ypJBCp4mdbV4EUcRq0CErKilF5aLqDK6Ylc2J/imeOz7E5jNjiKLE0jIHToueJSVOeibCpBT4yJICSt1Wen0RXj49SjSpYNfDydEkx0eSKA1+RFTW1Waxri6TDyzIo34gwC92duINJxGBYExmcYmV+67N4KevddHpiSIIAtGUwnMnh7EbdUQSWipzWZmD0gwLLcMhGoYD2E06lpc5ATg1EGDIH0EQQFbg2ZPD05gOFb0k8uHFhSwodlA/OMW2pjFe7wkBKllmgVnp8KmF+exoHGJ7b4p2r1aemJmXxjVzcrhticLfjg3yWqsmzFCXa+X2i8pAFVhfP8z2Fg9bGseQJIGPLy9iaYmL508Ns6PVw642D6KgjV/94Ioq+iZibKwf4btbOilwmhjyx1he5uKjS/LY3eFjc5OHH7zSgQqUZVi4dXE+OlHkvLJ0fr+/l13tk2SZYN+oyr6xSVA1NaBESuXjT55mOp8/vRHTUvDXzs7m5oV5mpbpkS50qRSqJPJQxWq+17uDz7Xt5Av5a3js4CDfuLicNTUZbGnyoJfg7ksqeOb4ME8fH2ZxiROj1cpHRk5yypxNn8mNMxVFMRhRVZX9fRF0Oon1J4dJKSoIAp9cVsgfD/QzFogjCloE3zwaZkZeGh88r4KJiQkaGxtRFOWc0HdaWtqbXvvy0u1MjSiMBzQHnWk34gnGEXMkTbdVCdLni1Kb895Ej+FwmJycfx+3+H+jve+d6uTkJM3NzdTU1Lwj6jLvtFMNhUKcPn2a8vJysrPfHPXkrOU6TNxz9Qxuf/okz50c5paFBdyxuoJvvdjCQ7u7iCZlBidjVGXZqB8J0zoWZpnLxfmVGTQNB9lwcpgBXwRPOMW9a2qZuekk0dHTbCycR4d7GQPtUywvc1OSoXXdpWSV3+/rId2q5yfX1LGlcYzNjaPsU1X0Z5oxSgLfu6JGQ/ttaeP+7R3cubqcpWXprK7JYEujBoy/tC6TuKywq9XLiD/O7ReV882XWnhkbw93ri7nGy+2Eoim0JmsJBIaEk1VUqjhSQYGQwzqjCAZECSJzpEEiqIwFZrWmlEUBFFEEgVkVVuIcmwSnnASWRVQ5SSCTo8AhJIKiHrsFgPRuIxekLGY9HhHh6hyCezvmOC6Ogfr6jJ4rdXDzlYPl8/SNEifOjbI9XNz2dKkofVMeol8p5mRQIKrZmezri6LP+7v5+ljg5zs97Og2MlIIEFhuoWyDAtHe6c43D2JxSgRSSosLUvnouoM0kx6jvZN4o+muLZSz2VzitANjPDNU2F8KYECp4VtrV52tntZWOSgeVoC72trK+jyhNnZ6uFY3xSKqiAKAsvKXFw+M5tANMX6+hG6vRGMkohBJ3Kk18/hHr8mtydAKJ5iW7OX6aozpW4LF1Zl4LTo2NbioXk4iEkvkEgpPHlskJfOjGAz6hkLJqnJtjAjN41D3ZM8dSbI+qYA6vSRrq5zMhGDrc0etrd4mJlvp2U4SJpRojLbRsNggC8928hls7Kpy7Gxt8OHCCQVledODCMrKh9bVkSp28KjB/qJKzI2o0TfRIyL6zK5qNrNA691cXoogDA9hhJOKFw5K5uqTAs/2t6NokLPRIRvvdjKlbOzMekldrVNcnGtm8+9/DuGwgkOl8znCWcdJYkpakoLiOgM7O/2IwpoTg0VURB4sWGUg90+itMt1IfgswMH0dssPJg+nydz53Ob5zQfuDKXJ44MsrtjgvrBAKgq2XYTNdk2Pr6siG+81MqzJ4b51I03Ij3ye+KKQJqcYHNmHce69Zj7R+ibjCNIOmYV2hmcjKGq8KdD/YQTMm6LnnuuqOKxQ4OIosiPr67Bbjdjt9spKSkhmUwyOTnJ8PAwra2t2Gw23G436enpGAz//ShMSaaNVkEgnFSYCGs14j5vGFGykuswoSopeiYi75lT/d+a6ntsf78L+3u5trPzne+EvZNO9c3Mx/53dtnMHPYtLGbj0S5m5dmZmefgqlnZbKwfQRQFvnhROYuLXdzxXAN/PDxGXXEODrOOT5xXzJmhAEd6JrmwJou6mSUMuj/GPL+fXMHN7/f3k1AFBicjbGoYZWGxk4NdE/T5Inzj4koy7UZuW1bEygo333zhDMFogiKXiWhSJtNm4AdX1PDDV9v4+Y5OStwWuj1hXCYd/miSwnQLF1Sm0+UJ89DrPfz8+jo+sqSAxw7283rHBN9YV8HdL7Ug6M2I0/OiqjpNOlIUxFQSVU4hShKKMO1MBR2CpHXzmnQQTcoIkrZ4jAaTnJ19USXt1pVVAUGnozDdRkJWiMRSJBSBkbiRedVVXGWe5IdbuzjYMUapJUGZU+LVxlEurXVz5axs/nZsiKbRTtJMBlLxFHMLHNy1tpzf7OnhmeNDWAw67lhdysJOB7/b20fLeJgCp4UfXV2L1aAjmkrxzPFhNjd6kERB6/ztmcRqkAgnZLItIpLZRdtLe9nhE/DYsvl+92vUXb2a/ovm8vzJYQ50+QCFsnQL4bjMVbNzqMyy8sud3SBIGCSRQ91+DvVoCjOCIOA26yjLtJJUVA0XqKgsLHZQkWmlZTTEqYEp9DqRbLsRbyjBnw/1nxPpzncauWxmNrPz0jjUM8nGU8MMR6OIgkAgKmM36fnm6gL+uLeb0xOgqgp6UcQfSbIkQ2aew8CGzhQn+/0ICKwod3PL4gLCcZlnjw+x8dQICpBp1fO9y2uIpVSePDLAXw4P8PzJESIJmdIMMx9cmM+m02M8frifV86Mkuc00TgcZGWFG7dNz9bGcQ73+FlS4uTMSJA0k57vXFJOy2iIV5o8PPR6LyoCTouedSU2hFAIo9HEbmsBGXKc+7peJXX+x/hCQwxUSKlaLb7Ubebm+bn4YylerB/jSI+PXDNc5O9CFzXSqnPygrOaOTlWLqnLZHuzh9/v68esF7hqVhabGsc53OtnWamLdXUZbG32suqqKsY+/EXGG8PcVSqQKirgQH+QHk8IQZT40gUlBOMpTg0EUYE0oxalX1jl5oEd3fRNxvjWJVUUuv5xndPr9WRlZZGVlYWqqoRCoXNRrKqquFwuMjIysNvtbxjFFjjNIOpQVZWnjg+TbddEChC0koaoyPS8h7jC/wk11feVUz1rZ+XaABYtWvSOium+E05VVVW6urrw+/3/9nzsv2N3X1bNib5JfrW7h5vm57G9dVzDk4kCBU4zFoPEJxdm8LP9Y/x8Rwffu7yGiXCCYCyFJAq0jAapb2knzaSjtrYWy0QUWYWZuWlEkykeP9TH44f6UVSZfIcZ87ROpAr87egAKQWunZfL7lYvd61v5MNLC7liVg7r6rJoGw3ROR5iTbWbTy4v4oHXOvnTwX7KMsx8+aJyvvFCMw/t6eHbl1RyZijIk0eHKM+wcl5FOq+1ehF0ZtRUAp0kaHVQUaQ2x0qXN0pS0VK8iihORyiixsGNJhEkHQUuE1PxFOGEtijKKkiCoHUATzvUQY8fOR5BUAVMaXb2d/mYV+RgdoGTdIuO1oCOqxZXc7l+jN/sG+CZfS0cHEmiAi6zngdvnMmrTeM8fWKEQ90+Pnd+CfGkwp8O9qMTNXWPpAIZViMjU3G+90obty0pwGk1sKvNS2WWlbsvraTfF+N4v59XzoxhECGqiKw/7QGhAMWmki5HaXMVkLdhE9bvzKF9PESaSeKCyiyO90/xu319PLKvF1lVSTfruXlRPtXZNhqHgzxxZAAQtXqzonFpg3GNmSvAdMQ6OQ1zEEi3GChKN5Nu0dMwFMBh1FGTY6N3IsKj+/vP3ct2k47bLyonGEuxo3Wcxw8P8LiiIoiwtMTFdfNy2NrkYW/HBAf6VKxGHcG4zIIcAylZZlf7BPs6J7m4LoNgLIUAZFh0+CJJ7tnSwY0L8rhrbQU/3d7F6eEAAgIpWbvvvnNZFSf7p3hwTzf1QwEcJh0ryl0sKHJyxczsaYSnxkNfWeHGYjKwdkYOS8vS+eoLLUxGkkxFknxlay+lNdcgyDKDBge3+JrZ5yzlmYYEgQQocgpRFMi2G/jAvFxmFzgYD8QJxAZxmUTGoil+dsFH+erpTXx87CSNznx+I5ZyW7+fyYh2n5xfns4tC/OoHwry5LEhFhQ5uXF+Hkd6/Dyyr5+UolLotrJwdS2iILC43M1nnzlNVaYJu0nHr3d1gSBwXpmLAz2awPn6+lEEYEaeg1sW/mutVEEQsNvt/xDF+nw+BgcHCQaD56JYt9t9bk0qcJkQdQYEZPZ3TXJ+hYukohJKChgkkQyLRM9E9B1Zv/4d+59QUxX+G9buO6fD9m+YoigEAgEaGhrIy8ujsLDwHedSRiIR2tramDdv3lt6/Vm+sMVioaqq6g2v7+DBgyxfvvwtHb91NMgNvztILBqmItPKp1cW88C2TgRR4IHrZhCd8nG4P8RfTvpYU5NJ00iAUELhE8sL+O0eTTXl+1fNwmHW8Y2NTfhiMg/eNAebUcfgZITvvdxCKC4DKklZW1DTTHqGJiNcVa7jI6vnEozJPPx6N8d6/bitOiZCCUrSjaQUldGpGN+5VOvs/OoLzQgC/PTaOg50+fjD/n4+vryARcVOvrK+iWhS05HMtOkZ9U6ihHxovXEqqiAiSRKlGRa6PGFUnQFR0uO0aGoqMppzNRiNZFgNeCNJ6nLtNAxqizKCgKjX9GTlZBwhEUFOJUFRMDjcGCSRR2+dg1En8pfDA7x6ZpQ/fmgOZoPIZ546TTCWxGLUsbTQxo7OADdX65hX4OCRUyHGwyl+ecNMbEYdD2zvoH44CILExXXZfGJZIUf7/Pz1yKCGvRMFjEY9D1xbR7Zd+4y+tuE0nnCKn183gyyHhe6jZ/jWqQi5qRA2NUmLIR1BVTDoRBQV7r60irpcO6Dyl8ODvHJmDJNO4/eq0xVQZZp65LYaMOtFvKEEsZRCrsPEjBwbwViKU4NTKAosLnGSZTfQ4QnTMhokpaiICDgtEotL0qnLsbHp9Cjd3igGnUAipWI1iKyuycShS/Fcg5ekjFZ7VAXOr3Rz7dwcWkdD/PFAH4mUik6Ci2uzuHZONl5/kMcPD9Pu0+p2c3PNfPGCUgaDMk8cGaDLE0WSVFIy3Lggj5w0I88eH2YslKA6S4vOh6ZirKxw0zYWYjyYoDrbyurqDJ48NoSiqMzIs3OsbwpBEFhV6WZgMkKXN8rdl1aRbTdwoMvHrjNDDMU0otZZgQdFUVAVBQQw6bQGt6SiCaknFZVEMslPrqrhSK+fJ44OsrY6g08sL6TLE+bbL7eBqlLkNOKy6GgZC/HL6+sYnorzw62d3Lwwj2vn5HCib4oHdvYgSDq+dnEVC4s1YMPutgkeOTDAwkIbxwY1etfcAk1MHlRWVbjp9IaJpGDDpxaSk2Z8gxXh37O/j2InJjSFovT0dDpCer7014OgKtTmOmjzRFBV+NwiJ6tmlXHf9k5iBicbP73oLZ/7zdgnPvEJvvOd7zBr1qz34nTvHdT47+x9FalGo1FOnjxJXV0dLpfrXTmHTqd7y5FqJBKhoaGB4uJi8vL++a5SEIS3LHFUk2Pn+1fP5JsbGpiRZ6c808ZdF2tp1J+91snnFmewqMDCcFRkW/MYBr3Et9aWIkwNc/uqfB7aP8JX1p+mNMNGty/G19ZVYTNqX/PLp0eZjKT47uXV1ObYOTUwxVPHBuj3RRGBE2MpqronWVqWzi2LCmgfC+MJxrAbRb50QQlOi5G7X27l/u2d/PjqGu5cXcZ3Xm7jD/v6uP2iUo50a/Jrzx0fJppUUIG5+XbuvqyaT//tFKOhSRDAZTXgCydIKSJdnjCCpEcQRUySylQkOf0oKAiSjmRKZjQQpyzTyunBoJb+FSUknY6qTCutYwFIJZAVGVGAHKeJ0WgUxWjhVP8US8tcrKp08/LpMQ72+Fha4iIpq6RUgS9dWM6sfDtnxpo46BFZPSedD9Sq/PKQj/s2N3LnqnzynUbqh8OAQLpFhygKLCtLZ0GRg7s3tdI1ESMVl/n6C80sKHTiC4Ton0px+4WlZDksyIrC7/tVLHKCH/bvxGmUGA6n+HHlOvpSWlr8ni3tVGdbMUgi9YMBZuencefqMgRR4OHXezjS46ck3UxNjpWRqTgtoyFkVSHLbkRWVfZ0TJyL9gUBDvT4sOolokkFgyRy8/wcHBYDJ/un2NXqZWvTOAICi0scfHx5ISNTCbY2jfPS6TFUVcVmlPjyRSWUZlh5+cwYrzWPs7vDCyoUOE38x8pi9nVO8FqLVp+uzrbR7U/isuhwW3Q0jEa5Y2MLq/IlVuSZGZiEREq7tvqBKW5ZVMCvbpzJk0cHeaVRO2dllo2rZ+eQ7zSxq83LU8cGeej1XvSSwOdXlbKywo0nGGf9qRF2tnpQ0DCGABl2I7UueEEWKXQYWZyhY/tglEAkjqqqiJKOZaVOvnRBKYqqcrw/wO/29hJLyawsc+G26rliZhZT0RSbzoyS5zRqG5rpDuHr5+VQlmHhzg3NPH54iK+sLmVxsZMX6kc5ryydfKcJSQAFgXBCyxwkUwpPHR9CEuDYQBgBFaNO+34FAS6sdNM8qm3gHr559ttyqGfXnDeKYk0T4wg6I0oiwi2zHfz1lELnRIytnWEWVqfIc5rY1R/ShOvfA7B+OBx+26Wy97u97yLVcDj8bxXg36qlUilOnDjBkiVL3tTrzuqzzpw5E4fjX6PDjhw5woIFC96WZut3X27h6YOdfHlVEedXutnT5uHB3T2sKE3jvEITj52aYjKSQlbhqnI9N6+oxWQ2M+CLcP/2doamEuh0umksnY1QPMXezgkuqcvik+eVIIqw4eQwTx4dYE1NBsUuMy+cHMAXU3Bb9UxFk1gNIlfPzGJjwyiCqnL3ZVVYDRLfeqkFi0Hivmtq2dbs4aljg8wvSKNlNEQ4mcKql/jWJZWcGQ7y3MkRvnxhKS6Lnu+sP0Eq7MNq1DMn38GhnklNSFmQcNkM+KOp6XlTQRtanyYlaXVUAVUQEfQ68p1WPnVeMfduaUNNxjQoRCpBlpRgpS7Exng6Rns6C4vsfGVNOagqd25owiRp0UrreARZVVlXl8nHlxWxv8vHr3Z186ULSjm/Uhu/+P2BPmZkmmj2JlmQowdRx8mRBOdXpfPpFcWagsiBfm5ZVEiJ28z+Lh/HeiaIpLRISScyvdgKdE9E+HClheteX4844eXF6pX82VDKrYvyqc2xcbzPz77OCcaDCY3dPE1iEtAE0y0GEZtBIppSCERT6CWBTJsRnQjjoQSJlEJ1tpVlpemoqGxtGmcsGEcn/meatSLDAgJ0eSLkpBlwmHW0jobRiQJzC9Po94YYj8iUppsZnooRl2FJiYPV1Rk8c2yInumNl6rCBdUZ3DA/j1AsxQOvdeIJJZAElWvm5HLt3Fy6vWH+dmSQTo/2ObvNIrdUSowmdOzuTzIZk8m1GxkPxnHbjCwodLC300c4KbOs1Em23cjLZ0axm/TIskooITMzz87lM7PY1uw5t/Ho8kQIJ2WcJoloUiahiiiygpJKoCopECSQ9OgkCZtRYF1tJpfVZbKxYYzNjeNUZVtpHw2zrMzJF1YVIwoCP9vRzYn+KVRV4bxSJ70TESKJFD+7rpYdrV6ePjHCXatLKXVbuHNjC+UZmgi6P6aS5zDR6Y1Rl59G30SUYFzBbhRREQglFcw6gVjqbNsXgMCnVpRoHdPvkqmqytx7thObmuCzy7Jxi1F+fFQTp3CZdSwtdbGtO8a2O1Zq9dd32a688kqef/75fwrFeYftfyNVQRDeVYcKb76mqqoqvb29eDyeN9Rn/VfneDtO9e5Lq+kaD/HbvX1k2AxcUJ3J8FSM9adGONgXwGkx8qVl6Wxu8bOpK4nRMcGNC/PwR5N4QklKM2zMLnDQPhZkR+u4RhsSBLY2jbGrTaMaTYQT1ORY+eCiApwWAwXCBOu7BRqHg6iKzLx8O6uqMlhQ4uKezW38YHM737u8iq+sqeAHm9u4d0s7NqOEqsKJfj+Li52UZVp45sQI3d4o183NoX5gij/u7+MXN8xg3bxSXj2aJBwP0TwcAFEEASxGEV84hiBokaA6PY8qiCL26fqdKgiIeiNZdiMP3TSbH73ahiqnKHEa6J5IAiLxUJTigWNI5Rcip1Kc7J8ilpAxGSTOr3Tz5JFBVOAz55fQMhJkT5uXWxbmc16Zi5cazDx9bIhlZS7W1GSwu91DoyfGnAInX1lXRsA/xfr6UXa0T9A4MMlUUlMDunZuDoosYwqN0OsxMJmAWxbmMx6I0+EJ0zSizZA+0RHhxdKrKFlopnEkwKwcG1fMytIIV/EUL50eY3a+nRsX5DPkj/FiwwhjwQS1OTay7AZaR0OEYkmy7EYqMi2MTMXp80URBZU0g462sTBtY+FpjKNAbY6Na+Zkk24zsq1pnN3tXlIy6CQtG3LlzGz0Oo01e6xPAzOUuc18ckUxuQ4N2rCpYZRDPZMYRJGPLClgRYWbFxtG2d7s4fV2LwadREJWuHp2NqNTMTbWj/Jaq4e1NZma0xAgzajDH5PZNKDjqplubs9P8vBhL8OBGIIg4DSJLCpxcuPCXF6oH2XzmTFSqkq6xcDnVpVQk23jtRYvL54e4b5tnTCd/v3keUUIwK+2t3BkIKwxoJUUakpGlEQEQYekN3LNnGxqcuzsaPWyoX6Mbc0ThOIp1la7uW1JHpubPDx9YpRESuHLF5ZiNUooqoJeVLlxQQ6RhMLdL7fzx4MDfOH8Yg50T/LogQEeuLaGa2Zn8dTxEQx6Pd+6pJKqLCsb6kc1Gb14ghyHlWA0TjipYjPoCMa1DdHMPDtnhkKsnZHDly4sfctrxL9jgiCQk55Gr3+CsGDhopmluE43kGkV8UcVtjR7kAwWTrQPkz2v6B3rD/ln9j8hUn3fOdX30zlkWaapqQmdTvdP9VnfyN6JZiiDTuS3t8zlpj/G+cn2Tm6/oPRcl56iClQ7BQpsEvdeN48/7O/j+VODvN7pwRdJkuew8oMra0gz6enxRvjOyy3kO018bHkxk+Ekezu9NA4FkERoHQnx8SdOUZVpxROI4IvKXDUrC4NOU7S564Um7ryonHuurOW7m5r53ittzMi1IwjQMRbGahS4pDaDXW0eJEng+nl5tI6GePr4IItLnHzxwjLu2tDEw6/38u1LKxnyRzndOchkPIIkSaiySiguY5REkoqKrAqIkjTtUCUCMRlBp8dmMRFJKCQVbZC9fnAK5CR5jjS6vUEEQWXKaCPkSCczEWZccZJQDBzt83N+pZuZuXYUIDfNyOrqDErSLbzeOcHuNi+XzczmI0sK+cGWdjY3jjErL43eiRhGnUTfRAR/VCErI4PPrMlgUd8k97/WiSxIqIkIh+qbsZCgLWZnIBDhztXlLC/TIsYfbmnHohf57uVVDPvjnBqY4mDXBLICZ4ZDfPjxetKtOsaDCWwGiRyHiX2dExzr9eOPpShxW0ikNMRhUlbRS+ALJzgciqMCJp1IZZaNfKdZk/QbC5NjM5LnNNE2FuK+7RrMPyGruG16bp6fR9t4mH2dXna3eXFZdExGZcpcemYVprO7bYJvvdTK7Hw7dpOeuKzgthiIJmT+emSQbm+ED8zLxaqXeP6UprSj14mgwqfPL+G6YIJH9vaysWEUVI2E9MkVRdQPBHju5DC/2z+EMB2B3zw/FzkZZ1ubn3te7SDDBFFZRBVgTp6DHm+YH23tpCTdRFmGlZSspU8dZh17OyY41D2JoMpEEklIJTgrNypIEkgSgs5Ipk3PomIXVTk2FhQ72d40xh8ODICqkGYxoNPruGp2DiadyJ+PDHP7+iYmwgnW1rg52O3jpzt6uPeKKm6an8tTx4epy/HxhVUlfPOlVu7Z0oEvksSgl5BVGAlosok3zs+lYSiAKOkZC8RR0bq1Q/EUJp2IKMLpoRDzilz8+Krq9yTlOiPPQV+/jv5JrSEpwyKRkAV+/oGZPHqgn729ITo92mggcK7ZyWazveNrcjKZ/LcCk/+b7X3lVN9PFo1GaWhoID8/n8LCwjf12r9Xqnk75rTo+dNtC/jgo0e5b3snkihww2w3XcM+Dg3HEUwpPpev8rlVpYgibG/RoAVjgRgPbOsgzaLnVL8fvSRy3bw8HCYdraMhmoaDLC9L58sXldHni7KtaYxd7V4URcaqE3CadVw6I4fFJen8cmcX33uljbpcO+GETDiR4njfBKvK0xmYjDLoj3Hl7GzSrXqePj7EqYEp/mNFMXesb+IvhzWB5Y8sKeAPBwfY2erFZTYgWZ2gKohyklyHnuFgioQsIkg6EKeRg6pKMKkiGU1cPCMHo05iS+MYgWiSO9c3IcspnEaR8WACQVHRy0niokSr0Y1PZyYVnEDQG3mxwcj5lW42N44jigJJWUu8VWRZqc22s7lxjHUzMpmVn8bCIgfrTw6zpWkcp9XE7avL+cnWDn60tYN7r6zWNim+KKogsajYRcNggEZvlGKHjoGgnyK7RIEuRDCk5/RYkoahIJ9YXkRlpo3KTBtNI0EQ4K615aQUldbREHvaJxAEAZ1O5HCPn3AihayAy6InKSuMBxPIqsrMPDtmvUTjUJC4LFObZSXNrKPPF+XM0FmxaU2zdP50V+pDr/cwPBUHVLyhJEf7/Fw7N5fzK908sK0DX0Tr1LWZjSwtcfGBBXlsPDnMS6fHkBWVnDQTX1lTRqbdyKbTo2xpHGVflw9BhVn5ady6uIAtjWO80jjGtuZxcp0men1R8p0m7EaJvZ0+Tg5McXFdJnU5Ngb9MSRBQFZUtrV4WVOTyXcur+HxwwM0j4QAGaMEFjXCbfOc9IcFtrX46PFGEQSVurw0arJs7Gn34g3FUGJRQOtSVgVNNEEQtfr8wiIHXd4w33q5lbU1mVw+M5PnTg2Tk2agPMPKhlMjDE7G+MIFJVwyM5e93X46xkIUOA18eGEuiwrTeGBHDz/f0cXX1pbRMhbi8UMDfGB+Lg6zngF/HKdFz/cvqeSp40M8dqCflxpGCMYVYrKAIGrcZRBQ1bNrgUoiBXkuKw/dNBOT/p2bavhXtrzUyasnTOe6fDOsOk6PJTDoRD63qoRTw00EMbNgQRWJRAKfz0d/fz+hUIi0tDTcbjcul+tdj2L/X7H3VU0VNPWZ/+aa3rb9d925Pp+PlpaWt9ww1djYSGFh4X9be/13rd8X4ZY/HsUfDHBdKczMsdIWd/Dk0UHSzDqy7SZax8LML3aytiabxpEApwenGPDHpoEAWk0SQFFVdAKUuC1UZdtRVZUdreNk2vSsLbOxv8tHT0Al0ypxw7w8DvdNcaJvElWRqc6ysrTUxd+ODLKkxMlHlhbwlfVNVGRZ+ca6Sr66sQlZhV9eP5NNp0d55sQw37msill5dr73Sitt42EUVeSWRfnsbvcy7JlETWqLvirqABFBlBB0OlTJwPxiN1+8oJQ0s57PPl1PvsNE40gQRVFRkzFURUYCUqkkM4LDNFqzsScjBEU9qpyCVAzJmcd3rqjjR9s6WVjk4Hifn+9dVsXM/DSO9k7ywGtd3HlRKcvL0hmZivGlZ88g6HTcd80MyjOtNI8E+eGr7eQ6jXx8WRH3vtrBsopMPjrfTVNnP70pB1uavURTGmbRqhfJNMNQUMaiF/nE4mwq8930+WLct72L6+bm8MFFBSQVhe+/3EqvL8IPr9Jg7vduaaN1LMyqynTCcZnj/f5pMR4VVAFlGsMAAg6TDrNeQ865LXoNaRjWhMX7JqLIqopB1LRXzyt3s615nG3NHkLTIzgWPXxqWR6TCZGX6keZiqUoz7QwMhVDVQXmFzmoHwgQSaZYUZ7O0lIXfz40wEQkgahqnbUX1WRw7dxcBn0xfrW7m3BcRhJhdU0GV83OIRhL8eTRIRqHA4BAvsvIJ5cXg6qyuXGcEwNTKCoIqMzJT2NFeTr1QwGO9U6SVAWUaXWcgjQdINA/lUROJkFJcVY4WJIkVFFElPTTq5XKVbOy+fCSQqKpFM8fH+HlxjFEAfSiwP3X1pLnMLG5cZy/HhmgJN1MntPE/i4f8wscnOz3M7/Qzp2ry9jf6eOR/f0sKLCxotzJw3sHiSsKLrOeykwbxwf8zMy1c8P8XDrHw7zQMEokJSJKogYsmTZRALtRTyihUJ3r4I8fmoPL8t45KH8kyfL792BIBfnrR+byxP4OXm4P8edb52A1SnxvcztWVxZ/++g/TkSoqkogEGBiYgKfz4coiufoTm8lilVVlZUrV1JfX/+eZCX535rq+8P6+/sZHh5mwYIFmEymt3SMdypSPWtF6RZ+e00JX9nUxbOdEUbjCT69OgsBeOLIIL5ICLNBR7rFQCSZwmaUpmHqZr6xrhKARw/0cXowwOxcO8UZFnomwrzWPE5SVRBVhTK3g1kl2Vy5oIT9HeP8/sAgv93Xix6VS8ptDIZkmkYiXDkrmw8tzueJo0PMK3JwyyIN9nC8389HlxXxo62dbG0e48rZ2exq8/D44QG+f1ml1nGrCJRlmLl2Tg5z06I8cUah0SNTmmGhNtfGvg4fiiCwuNjJng4f6VYDToue00MBfOEUdkMSSRBIqSkkVKwmHf5wHAGVBbNLaO2OEpAMFIsx/EYLk1MxhEiAR/b1YTVI/MeKEppHzrC73cvM/DQWFjnJSzOysX6EZaUuTg8FNHYw4jmR87pcO1+/uIL7t3dw75Z2zEYjV1dZGBoaYvHsGtI8MV444+GKmdkUppvp8oQ53ucnIWup6l/sGwVhBEXWxlAah/z8xBdlYDLKeDBOvsPEL3Z0MhpMIE/zj/d0+KbVZQRqcm2MBxOMBWIUOI1cUOEmGE/xescEowGNUjUR1rRFV1W4KXVb6JuI4jLriCRk/nJkiPbxMNfOyWUinGBPhw8RhUhS4FB/hA8uymdNTSY/myYZqcCiIge3LMznk+cV8VLDCK80jvF65wRmvchdq8upyLTyQv0oO9s87Gz1IkyjIj+/qpj28TB72id4rdVLgdPEyFQch1lPZZaN5pEg92xpJ8tm0OavBQGXWUIviZweCXF6JIRJJ5Ka1nO1myRiSYVef/IsrVKjaIlGJBGuHzjJEsVHWXyKV425PFq0HFRNDHxpqYvKLBsfXlpA32SE+gGNhtQ0EiLPYeKKWdnkpBn42Y5uujxh1tRk8NnzS9jW7OHRA338YHMHX1tXzrWhBBtPjXB8IIDTrMcm6PBHkuSk6fnwojw2nBrl+5s1fKIg6RAEgZSi/KfTmO7IDsRlyrLSeOzWOTjM723E57TokXQSsaiMoqq4zQICAqOBOOWZFgocJo6Pn812/KcfEgQBh8OBw+GgrKyMRCLBxMQEfX19hMPhc1Fsenr6m+ofeY8c6v9v9j/SqQqC8H/cQIqi0NzcjKIobxs48U5Sm1RVpb29HTUS4YXPr+CHW1p58dQwx/5WTyihUJBuZm1NFm1jIQ52TUyPG2jvcdgf49svtRBNyKQUlaosKwtLXOQ5TfR6w8iqwopSJya9xMHuSQ50TzKvII3hqRixhExumoHRQJxoCm6okJgKizy4p5tvri6kLsfKXw4N8IsbZrCzbZwnDg/y6xtnMjffzoZTI1xUncGHlxbysx3d3LG+mXBCZkWZi4M9fjYdamJJWQbfvrKEH2xup28yxrLSdEJxha9cXM7SEhcZNiPrT41SmWWlfTyESS/SNxnDbTXgmYwiKwp2owF/KIbNKOEszsPtGWJsKkJuUTbWSBJ/KISSCDMejPHBRYWkW/WsrHCzq83Lx5alsJl0XDs3l4f39rKrzcsTRweZXeDAH0nyu329/CJrBg6znjkFDtbVZbHpjAc1mWJXm4dbV9YiSBKPHewj227k1iUFGCSRuhwbu9snWFOTwQcW5DHsj/HMiSE6xsNUZZhIJJN0TcWYTKg4jDocZh2TkRSqKrCw2IFOFDnWN4lOhKSqcno4iKqoiILKsD/Ok8eHQZXRiyI3zs/jkrpMTgxMsaN5nGdPDKOqKuWZVu5aU47JoKXMXzk9xoEunwYkyJH4+PnV7OkO8lLDKHf0NeG26JkIJzUmtFXPtmYPtz/fyIXVbiIJBVmBbJsBfyTFL3d2c36lm8tmZBGIJTnY40dQFYIJONwzyWUzs1ldncHPd3QxMKlRmnSiQIHTxOUzM9nR5uVQ1+S5ezSeVMi2G9BLIqNTcWIpBb0kkJKZnqfWOsBFnYRwlsqF5mD3ZFSg83ehyjJPFiyiIjLB9Vcv4bFDfdy9qZWbFuZh0ImcHgxyw/xcerxh/ri/n5bREP+xoog+X0yTqTPp2NM+QVG6mUtnZOM063lwdxeff+YMSVnFrJdITjOgr5qdzYAvyitN3umsAajTGRZUFUVQEaYjeYMooMlSSKybmcuPr6p+z1K+/9VcVhPjQZFDPX7cJhEVTYS+PNNCgcvErv4A3lCCTPs/r3caDAZyc3PJzc39hyi2v78fURTP1WKtVusbOk5Zlt9RkM/71d53TvWsw3s3TRTFf+jOjcfj1NfXk52dTXFx8dveSb1TkWoqlaKhoYG0tDTmzp2LIAj84PIq6mwxnmlP0T0ewB9J0eWNUOA00TEWIppUWVbqYkZeGp3jIQ71TKKoKjl2AyOBGI+fxdWpKpWZFi6oymBugVYf+9mOLk4OaLD325YUcMWsbJ48NsSLDaOk2TL5zhUlfP3FFn63f4hry0Tax2X+cqCbWxfl88Nt3WxrHudDiwv42gvNvFivaV4KqspULMV3L6uizKWnfcTP1t4Uly7JQi9JfPmiMu5a38QzJ4aoyraxpERLt9+4II+O8RCPH+xHEATSrTrGUjKeYAxF1mZZh6e0mpo4neUxTKuyeAJxZuan0TzkRySGmoxRl6d1HK6pyWRbs4e9nRNcNjOblRXpPH9iiD8f6kfU6fnc+aWEEzLfeLGFB3d38a1LqgnEkuxonaDcbcAkCbzaFefEeDOlbguD/jhfW1uBQdIWqj8fGsCkk/jg4gIcZj2Dk1Hax8NcPzeXWxblM+yP8bUXWqh06/nILDObWqZoDijYDQInB6Y0eTsg12lhZCqKoKqsqnYzMzeN/d0+jvdNYdJLpGSF508Oc7Dbx5wCB8OBOGa9REWWhZbRELevb+KyGVnMyLOzTefBoKgoqsopj8y2dq22Ojvfzo+2duAJJdBJItl2A9fMzeXKWTk8cWyQ7S1eUFWqs+3ctaYcgBdPj7C9xcOO1nFEQWR1TQa3LMzntRYP25o9/Ghrhya7J4l8aFEBOWlGXu/08UrjGC+e1uhBjulRDlEQaBoJ0uGJnHOyAEkFrdtb0HRtrQaJhdkSccHM0YEg183JIX9iiH1ngjyTu4BnFRmjnOAzgVNYbQu478pq/nxkiKePD6EqKvOLnNy8MB+AjadGePbEMGeGpvBHU6yqTOe2pYU8/Hovjx8e5FD3JOeVp2My6PCHEwiiQJHLxNKydI71+Xn6+DDa46OCKHJW4F5VFW0UTNUicIOkaSanVB0fWpjPNy6pfE+akv6Z1ebaGB8Tee7EMJ+da0IERqYzHQVOEyg+uryRf+lU/97+axQbj8fx+Xz09vYSDodxOBznarFn19lwOPy2aUpf/epXefnllzEYDJSXl/PnP/8Zp9MJwE9+8hMee+wxJEniwQcfZN26dW/rXG/V3ndO9b2wswAInU7H1NQUjY2N7xiwH96ZSPUsaKKkpITc3NxzvxdFkZkZEq+uW8yhbh8vNYyyt93D/p4gqqp1PzaNhhj0xxjyR7EZJC6bm8v8Im3m7tEDfZh0InU5GiP2x1s7KUo3oZcEujxh5hWkMR6I87djQ7isBj60KJ9ESuGVxnEK0y188cIy7t3SjlfK4OLqBFtbJpjtTFDuENhwcph5eRZK3RZeOD2KgEBtro2W0TBNg5MIvgCfWF7I/bsGeKlhlBsX5JNpMzIjL40jfVMUppvPFUEE4AsXlPHFZ08TTSqMTKU07dRUnEyrHr0kMjwV03i1cZmpaJLRQBwBgSF/jIvrshAkiVQKpESUbU3jzMi1U5phoSrLqulkzshCJ4nMLXKwvWWC1VVuMmxGMoBPLi/id/v6+OuRAaaiCZIqfGhuBrMrCjg1MMWzx4e0OVtR4pnjgxzv96MoUD8UYF1tJhOhBN5QnIf39pJtMzAnP40jPZM8drAfWVGQEfj+3klS0ylfg06AhIBB0MAUfb4Igqot2TvafLzW4gVUyjIsfPmCUtJtBg53+3j2xDCvnBlDEuGDC/O5anYOnlCcZ08M82LDKBvrR3GYJD4710h1RRnPnRrjhfpRXmsZJymrGHUSd6wu53ifn431o+xo87KiPJ2GAT+WaSfdNBzii8+dYW1tJiXpFnSiiKgTEETY2eql3xdldU0Gcwsd7Gn3YtZLxFMKTx0fpthlxG7SIyBg1gmUZVrxRRJsm34/6rQ2rDgtOq+iRbZmg0Q4oUWx2cYUAdlIw0iQ8yvS+dDiAlDyuGjoT9zjh1O2PGKSgefmXs41oSjJsTGWpauc6BeJJhV6fRE6x8NUZtm4fp4Wvf7l8CCo2tyyqsLXL65gR5uXPx/sp2WsH4MocNXsXCRJYG+7lyeODmlycWeF4AVJmzadVr9RVBVJmB6pVlWSioDZZOI/aiU+epFWR+b/R6c6J9/BvkY9o8E4ewcEMqzSdBMbFLrMqIpMpyfC0tK3Bt0xGo3notizZLyzqWJFUXj11VdZtmwZFovlbb2PtWvX8pOf/ASdTsfXv/51fvKTn3D//ffT3NzMM888Q1NTE8PDw6xZs4bu7m5JVdX3Ttdu2v5HOtWzTm9oaIj+/n7mzZv3tr/sNzr+W7WJiQlaW1vfEDRxNgoWBIHl5W6Wl7tRVU2yrXUsSL8vyrA/pjWcSAH8kQTPN3h4vn4MFBmrXuSDC3O5oCoDEFh/apiN9SMossLMPDtfW1tBSlG5f1sHD+7uQS8J3La0kH5fhEf397K8LB27SceG+hHKMiyIosgz7QpZNhPBRIQ7NrZrF6qqLC5J466LK7nn5Sa2NHu45AMzcNktLO8J8mLDKBdUZaAoKqcGA2Tbjexpn2BdbRal02o6TrMOq0EikkjANAxdlWWK0u1k2AwMT0XPdVie6JsiJSs4rQZ8wRjheAq9JKIgoCYinBwMnEv5r5uRxW92d9MwGKA628bx3ikMeh1nhqZIyDIGSWJ1TSYDk1FeaRwDQeTyunTmVmpd4AuKnDQMBemdjLOuLouRqRgn+qfwR5OAwLZmL9taPNMfg3bO725uA1VzHEadOC1jJ5Bh02Mz6uidiCAAer2OZFIm3Qjn5elwWM283BEhHNeyDb0TEe7Y0MSyUheCIDARTlKbbSUhKzx5bIgDXRN8fHkR2XZt3tuig1Bc4eU+gewChS9dWEq23cBzJ4enzy+RZtJx++oyrpqdw4N7utl8ZhxJEvjAPA3mMB5KsOHkMJsbx1FUlTSTxBfXlFObraW6N58Z45G9fSBAUbqZmxfkU5BuYm+Hj81nRun1aZuflKxlF6alE9BLEll2A7GkgieUQJQERFVAUZme6RSJxxMMpHRMRkOgwt4OH6F4iuvn5dGz7ErqD/RzS7kZ0ZHGs6fG6A1Ncsfqcl490k9KgZvqLGzvinL3phaurEvn/OpsNpwapsClKc281uZlT/sEC4sd9PoiJGWVmmw7gViSV5rGUM/6zbNQEkHrTIdpEKKoNSIZBJF0qx4Q8ERkyrIdfGWhicJ067n14OyaIIqa/u9bIa69VVta6uC3koRLr2dnb5wsm5ZFAS1zYNdDpyf8jpxLFEWcTue5CHJqaoojR47w61//mqamJj796U9z2WWXsXr16jcN17/44ov/8z0tXcr69esBeOmll7j55psxGo2UlpZSUVFBd3f3YuDf1+F8h+x91/2bSqXeFRHxv7eGhgZAW/Bmzpz5tiANb2QDAwOoqkpRUdFbeu3w8DBz5sx5w0YpWZY5duwYS5cu/beOp6oqw1MxzgwF2N3m4UCnl/FAFJdJpCrTwrE+P+kWHdVZVvZ1+ajJsvL1dZXoJZEfbG6jzxvitmVF7Gzz0jUeQlBTVGdZaR8Pk2bSPjd/NEWew0QwniQpw92XlLPhxCAtnjifmwmBhMBjLSrXz8vl5kUFeEJxbn+uiQXFTmJJmZbxCD++upZ7trTjtBi475o6dKJAjzfM115oQVG0KBVVRUnGMegEStOttI8FUVUFAbDoBWQ5xcJiF/s6PBgliCUV5HgUq04lmZbHj6+dSU2OnaSs8tmnT1PkNFKRZWNjwzifOK+Yxw728YH5edy0QEsVBkIhPvd8C1FFx6pKN59eUYJJLzLkj3HnhmZW12byHyuKAXixfpS/HR3iM+cXk5dmpMsb4a9HBpmVb2dNTea52upFVRlcOSubu19uJRpPoQgCqgpGSZMlS8nK36UJVRQF9KLKxcUiK4ttqEY7+wdivNrsQVGgPMvCdy6rwmbQcaDbxxOHB5iIpBBUmJdr4IZaC15c/PXoAN5QkhK3md6JKHMK0lhR4Wb9yWHGggnm5NnIdpjY3uyhxG1BFAV6vBEybAaumJlF03CQY/1T5DtNeIJx4tMUp/JMK7vbJkCAqmwbXeNhwgkZSRBQ0VCdF1a5qciw0u+PsafdSziuYDNJpBl1BOIpgnEZh0nP4hInI4E4TSMhLq5OZ4kjSF5xOd/b2k1KVvjiBWU0jQR4rcXD1DS4f0aOje9erjG428dC/GJnN75wAhX47MoSLqrJIBhL8siebo4OBBFUFYNO4JsX5FCel4EvBk8dG9Sk86bHkgS0oFJjJjPtPQUQ/m45nI5QRVTSrVrjlDecQhV1fHZVGStcQVxOB6WlGtxBUTTsoaIo/7C+nXWu77aDjSZlFv5oF04hQqkdjo8lUVWBD8zLYWWFmz8e6MOQlslTH5//rl3DqVOnePTRR/nUpz7Fli1b2LVrF7fddhuf+9zn3tLxrrzySm666SZuvfVWvvCFL7B06VJuvfVWQGMM/+lPf/qAqqrr38n38O/Y/7hINZFIMDk5SWZmJnV1de9KJ5okScTj8Tf1GkVRaG1tJZlMsnDhwn9a0H+z9VpBEMhzmMixG1hdlY5KNQe6JvnD/l4O90xgkeDLF5ZSnW1jYYmL3+zp5vub2/j+5dV8cFE+P3y1nd/v7SHDpmNJiZbCvCEbTrncbG2b4IFra/nOy+0Uu82srSnkni2dDE0luGlREd/Y1EZr2MoNCwo46uvj5dMj1JiD5GWmc8XMDDY2eEAQ+diyQopcFj61vJif7uxm85kxrp6Tw893dmspt+kUoSonyU4z4g0l6fZqu2pBFECBUEKmJtOKOu2IYsn49OclEUkmkBIxXjgxxDcvr0EvCVw2M4unjg7SMhbi/KpMLp2RRft4iI31IywtdeHUJXnpeA9xRWJ5qYv9nT7ax0N8cFEBezsmMOjEc853Mpxk/akRFpU4WVOTiawo/PnwAE6zjrvWVCArCo8d6CPdomcinOD29U2ggtumxxdOoYkbaG0vi4qdXDU7m/rBAC/Wj5BmFFBVgVd7ZNqDCdaWRTg9MAkIVGbo6Z6IcsfzjXx8WRGLix3s7ZjAFwmAAK0TCYbkXC6oSmduoYPvb26jyxNBp9OExleUp7OiPJ1Xzozx7Ikh6odC5DqMfPmiUgqcZk4NBHj6+CB/PjQAgsC8gjQ+t6oEg05kV6uXjfWjtI6FEdBGcFaUp/OZFcX87egQB7p96CUJWVZ4rdXLa0wgT2vDWk069DqBsWCclKqlewPxFDtavSioiILIvk4fnU4TgaYuJiNJvn+FxquelZ/G2tos7ljfSCiapGcizPF+vwZ5yLbx0aVab4AAtI2HOK/Chd2k5651VXzzpRY6xiPEZYFHjkxQ5fKhKDInRhUsBoHKLDu93ij+WApBVZEErU6uqmdHmeDclIZwVupAwBtOIIgi+e407r+mFp2/H5fLRXFx8T88t6CtDXq9/pxzVVX1XBSrqiqSJL0rUaxZL5GRZsUzNsUXZpvIyXTxyhmNMrWhfow0k0QqMvWuMoBDoRA2m42VK1eycuVKgDcMoNasWcPo6Oj/8fsf/ehHXH311ed+1ul0fOhDHwL4Z30473lQCP/DnGowGOT06dM4HA5ycnLetdbuN+v4kskk9fX1uN1uamtr/+V1vZXZsLMPrCBotavzK92srEhnW/M4925u455XO/mP8wpZVenGZqjkJ9s6uP35MwSjKQwSyAIsKHTwyZbX+HK0mCcPjPLttq1sW/xRdrR5WVubwctnxvjQonxK3SY2nRnlP+pEqjMMHBpK8uGVNm47r5y7NrbQFLKQ7VYo1gUBbQ5xRYkNVVVZUupicbGD504MkW7VMTIVRURFAVRFRlEUitPNOC162kaD6CQoclno8YbOjVwc7JxAVBKoCBh1EEtOp+qSMY4NTBHoHSCtpJCLazN59vgwKRVuXlgAwMeWFXJmKMBPt7dza7XE3hGoytL4wc3TCi2/3NmNqqpU5aRxos9PdpqRV86MkVIULq3Lot8XYVuzhx5vlPMr0nl0fx9H+/xEU1pEHU2GEQRNgmw0qI0D6UWBpKLhBU/0T3F6KEgyJVORaea7l1WhEwV2t0/w/Ikhfnskgk4U+OzKQma5RZoHJtjQGuKXu3qw6QWCSYUryozMLnSxsSXIw3t72d3uxaLX0TMRYW1tJr5wgqeODbKrzcstC/NoGQkgK1CbY6XbE+auDc2cX+lmSYmLQDSFUSdR4DJSPxjgs0+dZmGxi6loklA8xYIiJ5k2A0d7/ZwamDqn91rgMrOuNpMch4nX273s754k02ZiXqGDZEqenlHVkH1lGVb6JsKcHg5Rlm6k2Cyjtzk5NRjQ0sLAL3d2c/nMbNbVZWgC5wmZO1aX8dLpUR7Y3sXlM7O4dEY2fzzQR6HTzPwiB5vOjNE2GuTzq0o5PRygyxPhY8uKsBgl9rR52TcYOlcnTckC9YNB7fkVNGepKNom4OwzJysqoggOk8SMXDsui4EzwwEGgzLnV2dz/9U1dLc14c7M/G+BMX8fnZ6NYmVZPrduyLL8jkexswrS2On1MhBQWFLmZHOjl0+vLCISl3mlaYxQPMngZIyi9HeHAfxGWqpvFDzs2LHjXx7nL3/5C6+88go7d+48990UFBQwMDBw7t8MDg4CDL/ti34L9r5zqu+WoxsdHaW7u5s5c+YwMjLyrqaY30xNNRTS8GAVFRVkZWW9o9fxXx3qf51Bu2RGNguLXdy5vpGH9g3gDSVIM+nRiQL+cIJ8p4EfXlnDhlOjbG4c5aIjTdziHOMX1evo0KVxvq+TXa0CP7mmhlcax9ja7OHi6nR+f2gYv5TLjYts3Lu1g31dPi6qymBJiZPtrT6unjuTfafDiGKMlArbG/qodsik2dO4YaaThoEpHtzdA4qKTi+CoJCcXmz8kRROix5UDYAwp8BBz0QEVJmO8RAOA5itZgYmI6RSKoKqgCiRjAaRBIlHXvLwtS8X4o8kNWctiIRiKbJsBuwmHR+ancbDR3w81SExFVP52jpNfnBGrp2fXTeDLz17hqm4wqA/ysN7e6c/Z437f8+rbdqfFRVBFNnbNYHVIBFLKSwodFCZZeXZE0OAwGgwAdONOSlZ4byydBYUOznQNcHJPj+SoNLpifCrnd18dFkRy0rT2do0TiCmfZ9PnxjBubKEVfNrWDgjyV0bmhgPJTFIKgZRocBh4DuXlLOr089fDg2QVFRm56fx8eWFGHQSp/qn+POhfn6xsxuAq+fk8OElhUxGEhrft2mcXa0TGA0CnzqviAuqMhiZirPpzCi7Wj3IgFkvkmHTs7DYyeqaDB7Y3oUvkqTAacIXSfKng/2AViMVRVAUlaZhbXRDVqE6x0ZOmklDVw6HqHQbuaZMYMHMOg73BdjR5uXi6Qaobc3jPHF0gPWnhokmZG5bms+KCjdLS108cWSQzWfG2dHiQQV+cEU5BXqFed31PORP55ubWgCBeQUOLpuZhSSKWPQSbeNhTDqRWFJBOYu4P1cyVXAaBfLSDASSAsNTcbLsBmbkpjEZTXJqMEAspWI0mvjhNTO5amYmp0+fJjc3918qWL2R/X0UC7xrUew1s7PZ1TjIibEIqxaYARV/JMl1c3OpzbFy99Z+WsdC75pTjUQib5v7u3XrVu6//35ef/31f+iDueqqq/jgBz/InXfeyfDwMB0dHQBH394VvzV73znVd9pUVaWjo4NgMHhOUPztyL/9O/bvOlWPx0N7ezuzZ89+xyHTf7/r/VcPYYbNwKMfnsvdm1p5+kQfyElqsyyUuF282jTOvs5JblqYx9HmYX5XdzE/b3iO54oW80zFCu44s5ldyyo42T/F0hIXu1rH+fxsHQ6jyL7eCF9fl0ex08TmM2NcWOXm+nm5HOlt4cHdPZwcDPKxZYXs7ZxgR3+SKz8wk3gkhN/vx6pXiYW1Cb9EQtPDtOkF4ip0ecO49SqqqhANhCnwJ2F6djGViHPZ3AL2tHsRBQFZVbElooT0RhxKgkAkwCF3LtGEzDMnhjHqdEiSwBNHBvjOpZX09fdRbIebF+bx1LFhMp0WyjL+cwRgf9cE46EkX15dzorydEam4vxkWzvBmMytSwrQiwIvnxljeCrO1y8ux2018K1NLZRnWLAbJZ45NgiCiF6EJCoS2vhYTpoRQYDjfX5O9vm5sEob89jR4mFD/Qh3PN+Iw6InGE3w7UuqsJt0/Ob1Hn68tYO1tZmMBWJ4w0kuLtbRG9bzQmeM9sA4Fxd6ONSjObBSt4nG4SB3rG/itqWF1ObYMBskRFHApBd56fQYA74oH1iQT12One3NHhwWiZSi8vDePjY3jrOgyEn9QAC9TuKamdmMBuLs7ZjktdYJZEVjE59f4WZRiQuXRccf9vfR74uzrNRFrsPAWDDBiT4/CgJ5DgPBeIrDPb5z+r4d3hg/9QpktjQzEUlQmm7mE+cVo5cElpa62Nw4xp8O9CMATcNBVpS7cVr0fGx5EWPBOMf6/Fj0IhP+CKUP38eCgQF+YMvgzgUfIqHXUz8U5MOPn0ISBcJxGUEUtI2XoKE5owntOhYXO3GadPT6IrSMhkgpCiIivnCKQV8It91EShXIc6fx8C1zKHcbqa+vp7CwkJycnLf97P53UWwqlTr3b96Mg72oyo07zcZENIJZL5FlM9Dr05qVCtMtCIpM61iIi2vfHQWZcDj8thtCv/CFLxCPx1m7di2gNSs98sgjzJgxgxtvvJG6ujp0Oh2//e1vufTSS9/zzl/4f9ypJpNJTp8+jd1uZ/78+ecitXcSzvBG9t+lf1VVpa+vj/HxcRYtWvSOKvOcbYY42yH870T+Bknkvmtq0Ymw8VgvC4sdXDk9mvHXIwPUZNv4eJmB+zqz2Zo7ixv7j/KL6ksYLSpnZq6NV5vG+fA8Fwe6VSaN2aytldlwapTxYJzLZ+Xw8N5ezgwFmZ2fRlWWhZMDARYUO7l8ZjYlbgvf39LO5sZxrp+Xx2vdEXwRBVCRRBFZUUFNEYyrFDkMDE3F8ca18YWITk/roUbIrdPSxKJAOJnCG06gMVfBqiQJCSaSogYzV6NBfrmrm5MDAW5YkI/DrOdPBwd48VALS0qc5OXnsfNAPzq9Js5+96YWPreqlGy7kaeODlGRZWVFeTrC9IzlyFSCL19UxsqKdPZ2TNDri/KJ5UUUuS384OVWogmZLm+ETk8EabrrM6loUVFq2qFmOYzUDwYJxhIYJY3iZDFIXD03l+Xl6Xz9hWa84QTFLhOF6SZcFgP3X1PLk0cH2dw4jqqqrCqQ+PDKakwmM1uaxnny6AAtXlAUuHF2OvPcCi2jMtv6U/z0tS4MOi2C/MbFFdTlpvFq0zibTo/wzZdaUFWVQpeZe6+sQa8T2d/pY9PpETbUjwBQm2Mn12Hi8lnZzCv088i+fmxmCbfFwP4uH693akQoBIEMq4GUouANJzg1MIUkCXzmvGIWFDnpn4xyz5Z2ihx6bqwxYcvIoWk4xIsNI6RklZ6JKD/c0sb18/LIc5rYcGqYfKeJVZXprD85wl0bG/n8+WVEkjIn+qdYVZlO70SUe3f0cJWlkpvccX5bfjGiKPKzE8/Qd8c3OTkU4GDX5DSbV0AQtD5sfySFgPbMdnnCZNtNdE9E0YsiX1hVitUg0TAwwf4uP91TMjOzzHxvbQ45phQnTzZRWlr6jmea4J9HsWf//2Y6ikVR5OPnFfPTzX46PWFKMyz0TkTOrQH5Dj0to6F3/D2ctVAo9LbRrZ2dnf/077797W/z7W9/+20d/52w951TfafSv2fTqmVlZf/H7lGSJJLJ5Dtynjeyf+W0FUWhqakJQRDelPLNv2NvxaGeNVEQuPfKWqJJhb+d6CPdZuTzF5Ry14Zmfrm7iweumcW8rv08VbSUhw49Rl7xJM/WXsT1M7L42Y4uxnwBitJN7Gyf4OtrK7SZxxYvH1iQy1PHBtncOIbDrKN/Uhs4L8vUqCsz89JYVOTkhYYRqrOtPH1kAKYXN1nRItIil4m+iSgT4TjZiQDD+jQK4n5GjHYaXIUoigqyNs5ysMtHPKU5ZZtRwi+kgSoQl4xa00ksxLHeSdIsRq6YlYMOhS0Ng7zSHWfNgixGpuLsbJvg4rocZuWn8ci+Xu7a0ESuw4gvmuLmxYVMRVPEUzJPHh2gMstCTpqB3e1eHj3QR5pJYlvzGI8d7EdVVdJMOvKcZtrGQtp4j6IgiJqyy0cW53P1nFxODUzxk61tVGWaUVWV3+7p5UDXBJ87v4TnTgwTjKVYU53B/s4JvrqxSRNXz7Mz7Zsx6gQODKsUdUxx5Swzl8/MomM8xIHuSQTg9HicpVUlXFFuYn61n+++2oM/ptXstjcO4zJLXDs3B6NO4E+HBtBLAkP+ON/e1MLa2kzcVgO+cJJMm4E5BQ4ahgI8vLdXu99UlSybkVuXFDIr344nGOdHWztIyioLi50kZJXBySgjUzGU6fvswT29Wie3qo0XVabryMwroMBl4pnjw4iCyA+uqKDLG2FL4zj3vNqOSSegKir3XFFNvtPMomInv9rVw4+3dSAKAtVZFj57fimKovLkK6d4mflsz5lJXGfkPwYOkRvx0xhNcKDThzLdbGQXZVaLfrbiJs2k13jF8RStoyEahgPnaqqPHexjVr6DSCJFUJa4YWEhX19djM/r4dSpU+h0unNsXJfL9a5Sg/5rFPv3/53rm/gXDvbG+bn8YU8njx8eYl5hGod7pwgnUlgNOkrTTTQO+s8d5522cDhMQUHBO37c95u970ZqFEV52w7vbFp11qxZpKWl/R9/Pzo6SiQSoazs3REHDofDdHR0MHfu3H/4fTwep6GhgaysrLdFbnojQYC341D/3hIphU88cYpT3SN895JyZAV+8Eobq2u0UZA7NzRyfq6R2pJMfrt/gA/VGdnaHSczzcyKcjePHuznvqtr2VA/Qvt4iEdumcPG+hHWnxrFapQw6nTkOAwM+BM8fMsszHodw1Mx7ni+EUEUSCa1zUhphpUebxg1lWBmno2m4QCqCi41wQQ66kKjxAQdPRY3SipBXbaF5pEQArLWtKQq5NkNDAdi2iKhKEiCtpzKlgxmV+TzzTUldHZ2IqRl8aMdA8yYjhBPDoV46ObZuMxaivL5k0NsbvRon+m5j1VTHxHOihWo2neQZtZT5DLROhaiPNPKrDwN26hJqWjR29kxIKNOZEaujebRINlWPT+8qhqjTmRbs4e/HhlEFCCekrlxfh4fmJ/HwGSMX+zsZHAyxtyCNOoHAyzP0/HRVTX86fAQh3smqcm2UpRuYXuLh+vm5ZJtN/LEkUEiCZm1tRm0j4cYmIzxpQvL6BwLsrXFSyKlkGsVGA7D3DwLd6yt4kT/FNuax2kfC6GgYjXo+OCiAs6vcGPWC/xuXx+72yewGzXGsDwNs1cUFaNe4oJKNzPz7RgkkUf2aTOs37i4Er0o0uEJ8dcjg6RkGbtBwheVQdCipYSs8PFlhVw6IxuAZErh7lda6RwPIQkC183N5fr5ueglTY3pKxuaiSZlil0m7lhTQaHTDKEQG37xJE/nztfwgdOTpupZ2IQokR0PUBQYp95dgiSnWJFrpLiyAIMk8NfDg9iMEl++sIyxYIyGgQAHenykkPjChZV8flXJOQpbZWUlLpeLyclJvF4vk5OTGI1GMjMzycjIeMv88Ddrfz+yc/bns85RkqRzDlZVVR5+5TAPHw8wO0OgfjDEty+pYHZ+GtuaPTxe7+e1O1aS53jnr/uee+5h5cqVXHXVVe/4sf+J/S9Q/+2aqqp0d3fj8/n+ZVr13U7/vtHxz3YeV1dXk5GR8bbPoSjKPzwo/6wh6c2aQSfym5tnc/Mf4/xsRw93X1rBohIHO1q9dHu1WszOoRhHxocQBZUNbTGqcuycHgxw84J8jDqR7a0e1tVmcqx/imO/eAybL4hSdj6yovL9K6oIxVN8a1M7W5s8XDs3lzyHiUy7kRF/9NxgvYqKqiigqnR5w+SmGUnIKp6QiiDL9JndVARHUVVw6xXKLQmaAVURQFDIsBoIxpNk2Ax4A3HMRh2pZBJZUZGjAVpG7LS3t1NZUYHFauXjy0X+sL8XVRC4YX4BLrN275wVSTdIIt+4pJJIQqZlNMiWpnHm5jtYUeHWRMVPj3DDvDxuWpDHd15uQS+JhOMpNpwa+U/0poYPwmaUuGJmNhPhJDvbPMipFAX5tmmfK3DJDE3P9uHXewEw6UVUFQpdJn5ydR33b+/g5MAUNr3Ah8+vwWUzcefqcvZ2TvD7fb00j4apybFx4/w8JFFkUbGTvx0dYmuzBxBYXe1mYZGDZWXpXD03j1/v7qZ+cApBEJkKx3j5UCPzC+ysLLbQMR4i3WxAQOCxg/38+WC/xiuOplhS4uL2i8pQVJWdbdpGQKeXcFp07Gzz8lqrF0VRESSBvDQj25o9FDhNbG0eQ1UUvnJeJguqi5mKpXh0fx+HeyYBlT8d7KdxOMj18/JoGJyi2xPh+rm5+CJJNpwa4XDPJJ88r4i/HdWoSB9bVsjGUyN8bUMT187NpTbXzvqCBcxIBvjE8HH2l87j5ZSbpCqeG4XxGmxMuC3IKrgTUQ6OCez0DGi3n6risug41O1jRp6dUFJGFvR89eIqPrG86JwsZHV19TkVq7PcW9A21V6vl6amJlKpFOnp6WRmZuJwON7ViQP4xzTx36eIz65HqqqyNM+AcmE1j+zWGuv2d/mYmWenMtMCsoczw8F3xam+E5jC/xvs/xmnmkqlaGxsxGg0smDBgn+ZVpUkiVQq9a5dy3+tqY6NjdHV1cWcOXPeNEHkjUySpHNOVVXVc+/lnUolO816fvfBudz0xyN844UWZFVFURV6PFpdtHk0iCqnKHIa6ZmIc2ZwClWFH23rIMNqYG+7l9psK7ZUlN/a6oi7TLhiAaKqBVsySm6Wi3kFNl46PcK6ukzWnxxhxBfWhuunAey9nhAoSQpdZvono8zMNZGSwROK47bo8MVF2lz5qIkEhdlplBVnYB/sYyquIioqWcYUzWGFOXl29oXiqCrTs6AqopIkGQlSH8pmzvRDvrY2k431w4yHFfomIgRiSdJMelpHg+zr9HH9vDzmFDiIJWX+dmSA3DQTX19XQSSh8MTRAUrSLVRlW/n+Zk2+TRAFfJGkxoed7iZFFEkziTx00ywsBh1PHR1EVRWWljo51OOn09PMXWvLcZr1PHVskHyniVyHib8eHqJ3IsqnVxQzGU7QMR7CZZYIJ1W+8VIbX7yglDkFaThMOlKKisMk0T4e4tubWviPFSWUuC2EEykEQaDQaWJX+wT1g4H/j73zDo+rvLb+75zp0mjUe7UsS3KTXDEG45jebYOpuYE00vORAgklNwkkNyQkgdwkJCShhJACAdNtwISOweAqybKKrd41vddTvj9GMwjHgItkSTdez3OfG8Ce82p0zlnvu/faa7G+vgCtRqBx0MfiknQWFFn4V5uNJw5E2dTtISzJFKQIXFenJz8nE2tUz8M7RuhxBBFFgfd6XHzub3soSjPS4wqRYdJx6/lzKMtMYcAd5IfP7ScmK5xSmYUzFGNPv5vX9sdQAZNW5K1BmYguXq5/t9vF6qosPnVSMVta7by4z8r2nhYAlpZZ+OTyYhAETqvK5t43u7n9+bhz17fPmMXKymxOq8rmz+/08fjuIRQVzEY9q0+t4wVbFf9qszFmyYtJq+ErSi/vWKPsyprFjzpeoN7WhauwlBsXXk4wKrG6KpsBd4gXW2xxRy2Njh9cNJdPLi8hGAzS1NTE3LlzP7Q/mJqaSmpqKuXl5UiShNPpZGhoiNbWVsxmMzk5OeTk5ExqPmmiTKzVaj9wirVareh0Or58ajGlGXrueLGDt3oDNAy1UJNrRFRldvV6OHcSxEqHGqn5v4hpV/5VVZVoNHpEfyfhk1tWVkZxcfHH/nmPx8PAwADz588/2mV+JBKuRytWrKCrqwuXy0V9ff2EPUQ7duygrq4u+cAc6+n0w/BOp5MvPLyTSkv8NPCD59qpztEzxxzjqU6J61aWsrl5FFGEnFQ9TYNejFqRYCxRFlVRVYFPD7zLUnc/36y/kk+WaLjk/CV02YPc9HQrcwsttAx6QFXQiCCrIlUmmQNBQJHJStXh8EepzkvFHYwx6gtTZDEQkRQc/jBaVSI7RcdlSwr509Y+orICssRFC3LZ1DzKaSUGtg2GiUlKckBcAyiiiDa7hPs/tZisVANNAx5+9GInJ83KYnefB4NWYPWcLHb1eYjJ8N1z5qACT+weYs+Ah9OrcxCAHb0ePOHYmOdrnLSzUvScOmaqEB96jA/UK6rK1UuLOGVWBiEJbn56H2uqMvnq6gpaR3z86tVufOEohRYDI94oP1s/l5JMI0/sGeafOwepyDbhD0UIxFR+uWEBEUnh7le66HeFWDMnm+29LvLSjPz4ohp2DXh5aFsf7mB8YzLgDvOZk0u5cGE+zYM+/rlrkLZRP4qiUmgx8P0La8i3GOI9yR0DPNs0mqiakp2iY16engFXmG6PzLlVZs6Zl0evV2VrZzyIXFFAFOO/86xUPf6IhCDAF06t4NTKLBDgjhfaaR7yc0aVBVnUsaffgyckoagqmSYdN51bxZy8+Eu3adDDT144kOxrXrQwnw2LizDpRX7zWjdbO5wgQFaKjk8uL+a0qmwGPWG+/1wbEUlBUeKdeTUeKAyqQIpew/KKTNoHXIwEZbIiPgoiXjLCftoKqvAhcv2aSlZWZqICv32ti609Pm46t5rPrCwjEAjQ1NTE/PnzD9lW+jioqorP58Nut+NwOACSBHs0+aRHCrvdnhwrTGzIg5EYb3S42NrpYlefm1FPkN9/cgmrqrIm/Pqf/exnuf322yftvXsITEn5d8aTasInd/78+UmvyY+D3++nq6uLurq6o1zlR0NVVd555x3MZjN6vZ6ampoJFSTt2rWL2tpa9Hr9pBFqAo/sGOT2Z5s4b46FNDHKP/e6+OSSApqGfHTbQ/zX8iL+9HYfn1pexN93DnHx/Fz2DHqJxWT++6U/8M2TP8t51la+0PMWt885n+6iKn5/zWJ0GpGbntpHpz0EihKPOpPVpCOcKkURxogo/uPFvz8tChLx/lskGCLXosPui7K6Kos3DjjjJueyzKISM40DHk4qz8Afkdg37EVR4p21ynSBbo8KpjTy8gr4wyfruPnpVgKSwG+uWMiwJ8wTe4bZ1u1EUviAw4yixg3wUeNl2VBUZnaemZPK0nmzw4kvIrG2rpC/7+gn0WsV4qZPH3jCDXIMvVHLb69ciFmfsHuMceszrYz4wiwrTefmc+ckf7fbu53c9XIXkqLyqRXFXLIoPgsZlWQefKefl9tsCAJ8+4zZrJwdfyEGIzJ3vnSA5hE/OhHWLixgbV0BZqOWhn4PP93SgUknEojKiAgsLU+nKN3As01W5heauf6MWTQO+Hi328XuPg8y8YD7Odkmyi0gx6K8MRCPCPzOOVU4QzJNgz427x0hnNhYCfFwcJNOwBtRuHhuJteeWomAQK8zxPeebUUkbgkYlVXmFZj5xJxs/ra9n1SDlu+eM5tNTVZe228nzShSlplK87CXK5YUsagknfvf7qPLESArRYc/IqPTily1rJjmQS/bulwIWi1aQUCKV94xakVCkky2Dqo8wwTR0JGSQ1AREAUAFbM+Hsk35JP4xplz+PLqCnw+H83NzSxcuHDCTlvRaBS73Y7dbk8mu+Tk5JCdnT3hYiebzUZ3dzeLFi36QFvs4JEddyhGhkk3KfaJV1xxBX/4wx8+4DQ1yTjRUz0SJMZSRkdHWbZsGQbD4UUWweT3VCORCMFgkLKyso91VjlSqKqKVqulp6eHwsLCo9oxHwmuXl5MrzPIQ2/t59wSOLkig3/uHuXra8q5980edve7qcpNYXOzlRXlGfyrzcGVywp5aNsg7gvWcWprO6/mVnNN68tcYnTz/ajCq+0OFEWl0xZAleN905gcf+tlxEI4NUY0isyFei/PSpnxl6KioAogakSQVCKRKIKg4g5KqMRPjPEMS5WoItA85CMjRY/NH6U0w4iiCmOnSfDERDKNKq6gD6fPws0bm+jxKlx/xmx0GoGyLBOfXF7Mjj431XmprK0vwBWI8df3+slK1XLzufG+8A82tbOsPIObz53DP3cNMuSNkJ2i4+87BxizYEcdS1/RqArf2vc8ZUqQR0uW83ZGGZGwzJN7hrl6eXzGNRSTcQdj5KTq2d3v5devdvHVT1QgotLT148CWFK0PLJjEElW2bC4CJ1GxB2MohEFLCYdd7/axZmDXq5eXsx+q5+WET9LStJJ1Ys83TjC8/usLC/LYHuvi9JME7dfXIMvLPFyq41/tdl4r8eFRhDITNHRaQtxUkXGWHi5ypk1OaTqtTQPeXmpK4gytmFQFYVH3t5PXqqGt4dkRAF+un4uuWkG2kf9PLZzkB5nPFv1uRYX7/XvZUlJOu/2ujHqRO5YW0uqQcur7XaebRrhd292oxEFrjmphNIME1/5RAVnz83hFy910jTkxazXUGAxUJmbws8umctjOwfZ2DCMoghEZYUH3u5HBUSNBogT6smz4haS97zWTY5Zx0/Xz8WkW8bG3cM07xrkmpOKWFKWQYfNz0stdjqdYa5dWcGXTivH6/Wyb98+6urqJrQnqNfrKSoqoqioCEVR8Hg8ydOkXq9PnmKPdbYzQaiLFy/+t2rZwb3YPL3+qEd2Pg7/KeXfaUeqh3PqkmWZlpYWBEFg+fLlR/yLnsyeaiJKzmAwTAqhyrJMdXU1TqeTvr4+/H4/GRkZ5OXlkZmZOeGeoZIkcU6en+45OWzZb+OqRan0OII8+E4/Fy8s4ImGYdbX5fN04whGrUBIihNDqkHk+cxqLjq7iDf2xXj5v/4f51xwErOfaePBbX1IsRjI0phJ+djvXBXxag0IsowiitgjCoImPlqTSAcRBQFVkUBRWFCYRtuoH0FVCcbiBuuVuWbaR/xIioBGEBjxhBl2B+MneuIk4ArJLC/PYHuPG8lrpUMoJM1oQOsdoKfHhyU9g3veGkYrCvy/0yvwhWUefLsXFZVTZ2fz5gEHzzePohXj/d/r/7mXIU8YQRTRaeKlRhGSGaEq8AW5l09Y2/Fm5rLXUshC9wAlQoRn9wq0j/i54ezZ/OHNHrSiyB1r5/JWp5O/bx9kxNPGGUUSm7ol5uSmcusFc/jz2/08tns4mbKzq9/LZ1eWcUZNDht3x6PgtnY6kFWVypxUvnP2bPRaDRuWhPjHjgHe6nKCKpBh0tE24mdRiYVFpek83zxKUbqR2bmp7Orz8lanC0GIeykvLk3n/Hl5VGSnsG/Yx09ePEBmio7lZRl02QPsGAkSk6X4xsgg8uS7B5hbkMpwAPpcYS6cl82lS0rY1edha6eTF1psqCiUZZrYN+zn1NmZrKnO4dV2O76wRKpBw+/f7OWVdjtXLi2mdcSHMxhjUYkFhz/Kb17v4a/bBynNNLJ30Ee+xch58/J4t9tF62jcCjJFryEYi5s5ZKfo+fWrXQQiEhaDhh9v3o+kqHTZg1RkmyhIN2DQihg0Il3OEBcsLObmc6vweDy0tbWxaNEiTKbJcRoCkuM4mZmZzJkzh1AohN1up729nUgkQlZWFjk5OWRkZBzRM/5RhPph6/g444mjdXb6TyHVaVf+BT7SjD4cDtPY2EhhYSGlpaVHVfqUZZmdO3eyYsWKY1nmv2FoaIje3l7q6+tpaGj4t7GXY8GHCZIURcHtdmO1WnG5XKSmpibl/Mfaww2HwzQ1NVFaWkp2Xj7f+OdeXm3uBzk2FnQuY9KKRCWFmnwzbSM+FhSkst8W5BNzsnm5zcE9Vy3gly93EZIUzqzN4eFtfaiKjKAoces6xkZSEEiVIvgFLYIIellC1WgxGXWEYjIROV561Igi4WgUFIW6IjM5Zj2v7o+HaS/PEMixDfGitiA+DyrE51xVFTSCgF6jEpbiquLTZmfxVqcTUQDBlIE2PRedKFCda8IViDDkV7DowR9TiR+m30+PUdW4vaBRpyHTpMETlkEQ+eaZldz7Rk8yQQXiuaC5Zh13x/aif+op/rDwAl7IreF/3/sLZXNKeePyL/D717vRauIuP186rZyzauMika37R/j9m/3IKmg1InddNp8CiyEegdbh4I9v9RKOyVTlpnLH+rloxu6L9lEft21qJyqrpBt1XLK4gLNrc4jJKrc+04ovIrOyMpMdPS48IQmzXkNIUshK1fPTdXNJN+mIyTJ3vdzFrj4PaUYtvnD83jPqBCKSgsWo4+trZrGwOI1ITOW2TW30u8OsmZONMxCldcRPKKagqCpmHZxWZuTU2dmU5mfz05e66LIHObMmm+ZhH4PuMGmGuMm+NxTjlvOqmVdg5tV2O0/sGcEZiKIoCgtL0vneeVVoNSI7e938bfsAg+5IPJIt8ZwAKXotxelGDtiD8cAFFTRivMRclmEiN01PICqzf9Qfz+gdd88LosDSihweuGYRQZ+X9vZ2Fi1adNxGYw4FWZZxOp3Y7XbcbjcpKSnk5uaSnZ39kRW6IyXUj8N444nxvJE45X4cyZ522mns2bNn0nvH43Cip5pANBo9ZOqAy+WipaWFuXPnkpV19I10VVXZtm3bhJFewgrR7/cnBUSHmiU92s8+3PlTVVXx+/1YrVYcDgcajYbc3Fxyc3OPeJedKHnV1tYmxwaiksKX/tHItvYhCs0CI+4wgiAgKSoQnwVNiwTw602cKtl4W5/PhQsLyDBp+euOIVRJAkWKO94KxGc3ETDptYRi8ZEgrRQlJmiYpZPolfVkpuhxBeMG+XlpeqzeMIIsjSWeEDeGcMZnUZfbDlDv6OGBmjNJi/jx6oyAiqqozM410+cMoBUhFFMxaVRCkgCigCCKiCYL6Vm5mPQiNn+MFJ1IToqWUV+UmKKwpkTHqZWZPN8ZYveAf8xFKZtHdw6ysWGET59cxvPNI9h8UYx6LZGYQkmWkX5nmByzFrNWpLSnjbctZZwz2MhX+t8h9uMfQ1ERe4e83L55P6gyN51bzbKyDPw+Hz09Pez2pfFMsw2NAJ87pYzz5uWBAP6wxLc27iMQlYnJKrNzTHzulHJm56bw4+f3s98a5POnlPB2p4u9wz5MWgGDVkMgKvPDC2uoyTcjKQovt9p56L1+YpKCKAqk6bUsKk0jEFHY3e9hfX0B/3VSMe7X3+HVVxp4rHR5PAlI0IzNlgpoBYGQpHDtitK4t64g8OyePh7eaaU4w4BOFOhxhgEVnagiKQJX12Vw7sIiTEYTjYNefvNaN55IDJ0gcnpNfCa6KN3AIzsHeXLPCEadSFhSsBg1LCvLwB6I0jToY3FpOitnZfFej4td/Z64C9fYzLBJr+XKpcXoNQL3vdPHObU5fHFVBcGoxC1PtxKISNx5yVzMBi0NAx5++3oPWZYUnvzSSchBLx0dHSxatOiIWkuTDVVVkyM7drsdRVHIzs4mJycHi8WSfD9MNKEejKMxnli1ahUNDQ0nSPV4LWQ8DkWqAwMDDAwMUF9fPyFlmIkiPUmSaGpqwmw2M2fO+8KSifj8YzV0CIfD2Gw2bDYbkiSRnZ1NXl7exyoNbTYbnZ2d1NXV/Vs/R1ZU/vfVLu5/s4MsvYQ3GEMjQDQmocoykkaDRpZRBKgSo3QqRmRFBkUi0WdM1WsIxOLSHY1GiBsHqKAqEgIC+RYDI95IvF+HgE4rEIupCCIIcnzWNMOoxReRkMfmP1VVQVDhLGsr2zIrMUcDjKRkoCgqaQaBk2dl8UqbjXNqc9jSaosTuEZEQcOsbCPdrgiizoSQko6o0cG4yC8BQBDi5uuKwJx0kWWlZkKqlmf2uUgz6fCG45mvaSY9/rCEeex0pxHizkKSrLJnwIMkSaxOV7ju9CrMefG5xj+81cMrbTaK0g0MeSJcWZ9DjclPZlE5Nz1zgKq8FPQakT39HpaXZ/DFVeXc/3YvO/s8/GTdXIbdEf66vQ9nIEZhupFhT4SvfaKC06vj89D7rX7ufrULqy+CSNx9aHV1DtV5Zu58Ke5+9P0Lqhl0R9jZ5+a9bifhWPxkXpppYGGKSt5rW3is8hTSYyHu2P531KVL2HPxJ/n7ewPYA9H4KV6AVL1IXqqObmeYuQWp/ODCWrSiyIg3wi/+1UGPI4CIiEaEuTla6rJUdtsF2hwxrl5WhM0f5Y39DmKKOraJinJ6dTZfXFVO06CX1/Y72NHrJjY+okxVUQUBURBZUpZOjz1IMKZw1bJi7P4oL+yzotdAVW68H9rnCuMJRTllViZzC9PITNHx9x2DhGQNj163lBQ5cEhRz3RELBbD4XBgt9vx+XxYLBb0ej1Op5MlS5ZM6thOAocynoAPZsWqqsppp52WzLI+TjhBqgmMJ9XxOaMLFiyYMFXcRJBeYpSnvLz835Iptm3bxooVK466xzlRDkkJJB4+q9VKIBAgMzOT3NzcD/RhVVWlr68Pu91OXV3dRz6Qb3c6uX1TK302D0osCigokhx/gABUBSkaQRSFZIFXFQUqs0yM+qJxUhU0cZ8bVUVU5XipVhC5fEkRj4+luSR6kmkGDZ5gFBEFs16DPyKhjvVZBVR0cgxRihHWGUmJhQmIuqSBhCjAufNyebHFzrIyC7v6PChjMmNRFNFrBGKKgCpoEQwGBI0eUdRg0Gk5Z14u5ZkmnmwcZsQnYTJoCUXj/SUVFQQthWkiokbLiE8i+TjFFTzcsb6WOblmmod83LaplXn5qbSNBshO1XHDmbNBgFueauHChflctayYX7zYSuNIiDOrs7D5o+y3Brn7sgXkmHU832zlb9sH0IoC4ZjMNStKWFdXCEBIkrn3jR62djoRBYFl5elctDCf+YVpbNw9zD93DXHJ4gJSdVpeP2BnwB1OjqtcsKCAM2tyKMsy8sI+Gw++08eSsnRq8uNOVm1DHiJq/B4pDLlY5BlgrnuIl8+8jOZBH18/PT4n2zzo48W9Q7Raw2MmCwJVeSmsKM+gwx7kvW4Xnz65lPpiC6+223n9gB1/REYF6vL0nFcGGSlGRJOZB3e7aR8NIIoCFqPIilnZzC1IZcs+G+3WABcvzGdeoYXGAQ9bWm3kW4wsLrXwVocTXyShGI+X7QGK0o2YDVqcwSh2f5Q0Y7yaEBvb0el0ev786SWUGiP09fWxaNGi40JIE4mEeLOvrw+9Xo9OpyM7O5vc3FxSUlKO2wnxw8rEp512Gvv27TsuaxjDCVJNIBaLoSgK0WiUhoYGcnNzqaiomNCb4lhJ1el00tra+qGjPNu3b2fJkiVotUeuBZtIh6RDQVEUXC4XNpsNl8uVHEh3Op0AzJ0797A2A1FZ4dEdg/xtez+DTj+SMpZ7Gg2hRgIgy/HMr7GTJKIIiAgaLQadBhCJKQo5qTrs3hCM2f4ZtCKhsRxUk05DSIqfAlVZRiuqVOen0TrsR9HqEZExCirhmIQsy2ODNwKKImEQVaKKiqpAqkFDRFIQURE1WiKJGQtBQNDqQdCg0+lZVp7B9j4fVy0vYleflwP2EHqdFlkVuO7UCs6Zl0s4KvPAtj5e7/Bw0zlV+ENhfvfW4Jinb1wSKwgqZ9fm8IVTy1FVle8+1YIvFOXXly+g1xnkV6904QpGyTHrCcVkfn3ZAjwOKz6fjx3uFJ5pGkVVFa5aXsLli9/fsO3u93DHlv2oqsC8glQ+s7Kc2TkpDHvD3PRUC/kWI0tK03mp1YYvIpFp0uIKSpxcmckNZ85GEOJl/O8/10aXPUiuWY/VHwUVTHqRUFRhVnYK3zhjFsUZRkZ9UX6wsZFYKMyZ1jb6U7NpSSsgpDGgqCpZKVpWzs6mrsiCx+3m/h12yrJMfO7UcpqHvLzX7aLbEUJRVXJSdaxfVMiplVmkGDT86uVOtve44ydSXxSdRmR5mRlvMEKzNcLyAg0LCs3sc6g0DvkJx5S4eYcokqrXEJUVIjEVQRTGsoLjlZR5hRbOrs1ld7+brZ0uvnPWbFbMymR3v5ufvnCAVXOyuH7NLFRV5bev97C128Md6+dzcj4MDg6yaNGio3pupxoHl3zD4XDyFBsMBsnMzCQnJ2fS/YnHI0Gsv//97/nLX/5CS0vLcbnuGE6QagKxWAyXy0VzczPV1dXk5k68u8exkGqiFP1RAoadO3eycOHCI+7HTIZD0sddz+12s2/fPhRFwWw2J/uwRyLOUJ1O+NS19DuC3Dv/XF4oriOsyMSCPlCiqIqAKMaTYwx6DQaNBm9URhQ0iBoRRVVRpRjlWSZ6nCEEQUArEJ8vFECORuP8LGjif0erIyvNSEyB2nwzHaMeHEEZUY6hAaKqil7yI8kKMVmOl26l+O1s0GmJqiAIGrRaPWl6DVevrOCPbw9w/ZoKNu+zMuSNsaQ0nXe63aiCiE6n46KFBVy4IJ8hT5jbnj/AuQsKmJNn5nevd6EiIo7lo+anaXEEYqwu1jA7L42oaOCh7cNcv6aC1VXxkq83HOMHz7XR5wqztCydq2u0qIrKrFmzCEZlvvJoE+GoQopBw1dXV7CiIhNZVfnh2N9ZV1/A5uZRvCGZkyrSGXCF8IQlfnHpfPLSDEQkmWcaR3hs92A8jQWB6vwUTirPoGUkwO5+N/9vzSw+MScbuz/KM40jvLDPGjepGEt8t5i0hMe8mL8xuI1Tm98CFX694EJeK5zH0hILMtA67CMixQ3yjVqB9fXxdJ3idCN/fref55vjc6++sESfM4QoQHqKDlcgxjUnl7C2roBeR5CXWm280mYnpqik6EQ+MSeTmkwtejnEQ41eXFFYVZFOVpqJAU+YnX0eMlP0XLG0iIik8tC2fhaXpnNmTQ7be928ccBJWaaR4gwj/ohEy7Afgy4eI5eTqsfmi/J6h4svrq7kilojIyMj1NfX/58g1IOR2Egfb39iVVV54IEHePHFF3nyySePt+DrBKkm0NvbS3d394TPhY3H0ZRnFUVJStwXLlz4kbu9PXv2UFNTc0QzZomSyWQbOoxHKBSiqamJiooK8vPzCYVCyT6sLMvk5OSQm5t7eI4voRDirl2gKChLl/JKf5DbN7cz4vCgBN2AQp5Zj9UXt++bV5CKUadlV58HhHg0mzI2FycggEaTLB+JogaLyYA/plBXlMa3zqzkzQMO/rytn28sNVKSl8VNLw6xri6fK5YWcc2fd6MTVXw+P0oshBYFWRVA1CJotAhaLcUhL2cNNPJw7Vnco+/gdv08RDEeV9ZiDSCIOs6bn8d58/N5smGYtzpcIAiIgoBWr0cnCgSiStLPNxBRyLMYGPGEEYSxR0eND8emaRVuOjWT/JwsLOnpSApc/9heIpKMPyyzMN/Ad86fS4pey/1b+3ipzcqNZ1Xx+J4huu1BVs3OoijdyGO7hvj6mkrWVGcTiMo81zTC043DRGWVskwjly0pYnl5BqGYws1Pt6Cq8K0zK2ke9PFej4tuR3y+NFWnYcWsTOYVmtFrRH73Zg/lmUZ+cEE1zpDErl43G/cMERzLHQVIEVRSNSq2mMjZc3O57tQyRAHeaOzkj7tcGDRa0lO0cUUuY73zqMyq2Zlcv2YWgijQbQ9w9yvdDHnic6tpBg2nzs7mlMoMNu+18l6Pm5WzM4lKCk2DPmKygqIqCIjMyTWRaxLwBiO0OFVMOpGa/FTsgRi9zlD8xIoICeN8ICtVh1EnYvNFickyeWkGwpIS74Ejcs6CQm44OR2nw0FdXd1xO8FNJI5GlDRe7DSZ/sQPP/wwTz75JM8888ykjiR9CE6YPySg1+tZvnz5pO4Yx/vnHg5isRiNjY1kZmZSW1v7sTfex2WqjsdE908PFx6Ph5aWFubNm5f0MTWZTJSVlVFWVkYsFsNut9Pd3U0gEEg+eB86K2cyoaxalfzHM2tTObM2l3e6HNy2aT/do25s0SCqEENUVdpG/KiCiFYEGRUEHaJeC7KEoNGTbk7BE4xg0MA9Vy4gM8XA955txRGIkqrXUF8U37D0R1JYVVJEVa6bxkEPn1xeTE2+GX84xqdWlPKHt3pQpPgJVhW1XJgv0NIxjCgI6HVaQOW//QW4DTEEQSCqiJwyO4dt3W5CMYXiDCPXr5nF5UsKufuVLnrcEiUZRorSjezodSOKIoFIvHc74g2TYdTw3xdUk5Wq58F3enm704EvpnLPLj+fXABpAwO8MwKOQIzPzNfjU9N4qsXN955t48qlxWxpsXLegjxOqshkSVk6TzYM88TuYWKyQllWCifPygDipFWRnYKkqMwvTMMRiPK/r3Zh0onoNXGnpB9cWE1tfhq1+WlYTFr+tLWPBYVmLCYtu/vdvH7AgaIqaASBDJOeV9rtlGaaeOuAA1mGH6+tJTfNQMuwl2cbR+l2BBAF+FebjXe6HBSkiPR6Y6Qb9fxkXS3ZqQZsvgh/eKuHxgEvggBbO120jwZYXp5Opy3AiDfM508po9Bi5I0DDl5ps/HCvlEQYEGRhfPn5TEnL5V+V5gfPd9ORBKZnZtKICLRNBLBH4s/V6IA/U4/7pCKQavh7JpMKvMsPLpriJik8ItL55GZoufuVzoZ8UT4/gU11Jek0+sI8t/PtVOWk8ZXF6ficjqT1n0zDVarlZ6eniNW+R7Kn3hwcHBC/YkfeeQRHnvsMZ577rmpINQpw7Q8qUqSNKmORxC3+ps/f/5hlSMCgQCNjY3Mnj2b/Pz8w/r85uZmSktLPzaUN0Gosiwn5ejHA6Ojo/T09FBXV3dYN7yiKDidTmw2G263m7S0tOSs3OFuflzBKD/d0sGrbVZ8gTCo7286RI2G+SVZKKpKy4CD5aVpbO9xsnp2Bm8ccPLV0yo4oyaHl9ts/HFrDz84pwI8wzy8X0CrEfnp+nk8tmuQx3cN8eA19Ty/z8rG3UPccOYsHnq3H1lSqCkw8263iyKDgiMgEdbox1yWVFKlCBefVMGzLU7mFqZxy3nVbNw9xKN7RlhXV8g1K0rY1efmZy/3sGFJKZ88qYTfvtbFmx1OEOLqRoNOJCbJ3LG2lqo8M0PuMN96fC+nVWVy2uwsfvdGN76IzDUnlfKPHf2Um1WuqNah1WoZlUw8tMdJMKaSotdw79X1pOrHEkdUlZufaqHHGUJRId2k5aIF+SwutfDDTfspSDfwP2vnohGgecjHg+/00ecKIwrx+czaPDPZqTre6nRRV5zGLefNQSuKDHlCfO+ZNhQ1Tsq9zhCjvkhSbFWSYWB5eSa1+WZaR3w8s3eU8+blcUl9Ac1DXl5pHqDFLqGMzfDmW4wsKrHgCcZ4t8fFufPyuHJpEbv63Lzd6WJPvxsVsBi1nFmTyymVWeSadfzPCwfodgQpyzQx4A4hK6DTxEPUdRqBy5cVs7AwjUBU5q5XOknR6/jxxTWYDVp+/Hw7HfYQV9Zlo8ZCvNwVxBZSmZtvwmSI+xWP+iIUpukpyUxBr42HyxsMBn5xTh5GJcyCBQv+owj1ozBR/sQbN27kgQceYPPmzVNp+HCi/JvA8SDVRBbix5WXE64mH5bN+mFobW2loKAgOeN5KEy2IOnDrtnT04PL5WLhwoVH9TCqqorX68Vms2G325P9mdzc3MPuISuqijcUd1SKSirpJm18jjIice5v3sHu9HD6nAy+fFo53326lZik8qvL5xOOyVz3twYWZot885y5bG518ejOQf74yTrs/ii3PtvKp08qpmHAS0O/B2HsFlZUhQyTDk8oRoZRQ/lID01Z5Xyt9SVsuhQ2zl7Ffdcu5l+tNv65e5g7L5nHrOwUHtzWxwutDk6uyKTVGiDbYuaOdXN5cZ+Nv77Xn0yhic9RyszJSeHU2VksKU3nT1t76LIF+M3lC8hIiV/7V6900jzsR1Dhx+eVUVuWRywWw+Px8GzjMJs747O/S0vMfOG0WeSYDTy/b5QH3+nn62sqyEszsnH3IE2DPlRVQasR+cIp5XyiOgedRuC1/XZ+90YPFy7IZ1l5Bnv63ezodTPkjofDa0QozzJRnpXCrj4PsqryPxfXUpZlIior/GjzftpH/CwtS8cTluiyB5GVeL80zaDltKos5haY8TuGebQtilYj8vVPzGLIE6ah30PDgJfY2Ml3UbGFZeWZzC0w84e3ejhgC3BGdQ4Of5S9Q15kVUUcyzm9cmkx6+oLiMkqm/eOsnHPECrCWBC7mrxnBFGDyJifsgrKWDc78fpUVBWdKGAxaFAUGXdExWIQyU8zEFVg0B1B1Gr55bkFFKcozJ8//3jOTU4YrFYrvb29k65SPhp/4meffZZ77rmHzZs3f+yhYpJxglQTkGV5UqPZAPbu3UtFRQVpaWmH/O+J8ZKRkZGjGgBvb29PDmV/2OcnCPV47ZIVRaG1tRVRFCfU5D8YDCb7sKqqJvuwqampR/XC2tw8yg3/3MMXTsrj7Hl5bO1w8L+vdnPTOVVUpET5w9uDdHgU7v/UIka9Eb75RAvXrSxh9ZxsPv+3BiRJQa+BaExh5awM6kvS+f2bPfxsfS1/2z5AKKbwwyVpfGbLCFeNNLA8R8u39fV86bQKTqnM5GuP7qUqz8x/X1CDqqo81TDMIzuHUXVGSjNNBKIy7qCEII5RdiIvVYnnrsaUMetFVeG/lhdzyaLC5M+2r9/OD1/oAUFlUYmFb54xG7NBSyAicf3jzeSl6VmYb+LZZgeCACcX63hvKEZtvpnvX1ibfEv8aWsvL7XaMGgFIpKKUScyOzuFVquf2dmp/GhtDTpRxO6PcsszrQjAJ5cXM+SJ0DrsoXU0MJYsAwatQEVWCq5QDKsvyqdXlHDRwnwEQeCZxhH+un2A8kwT6SYt7aP+MVGSgkYUOW12FidXxonziT0jbNo7ytKydPLS4v7Fo54wiYTcUyqzWFdfQGVOCu2jfn62pYNQTCYebi5g0ouUZBjptIfIMeu57cJq0k06Gge8/P6NHmRUzqjOwaDTst/qp3nIx8JiC5+Yk82oN8Lje4Y5dXYW31gzix5HiO8/10ppppGvn5SFz+fhH80B2j0abjjZwsmlKcybN2/GEupEn1APB+P9iR0OR9KfOCMjI3ngeOGFF/jlL3/J5s2bj8mgZ4JwglQTOB6k2tLSQmFh4SFPkoqi0NLSgqqqzJ8//6jIp6Ojg7S0tEOWi6dCkBSLxWhqaiInJ4eysrJJu25iZ2uz2QiFQh/owx7uNVVV5VN/3k3HwCi/uXw+Oo3A//vnXiw6levqUwiY8vifFw/w7TMqOaUyi28+3oxBIxCRFQacIXSiyu+vWshdL3cSkxW+c/ZsvvTIXj53cgnuUIxn947y8KcX871n20jVa7ntohquf6yZ7FQDt11Uw7N7R3j4vSFuOW8OS8sy2Nbl5O43+plbnEmKTkNYUmgdCSAKArISt7nTiyo/uHAOcwsstI/6+eGmNqSYRIpBy/Wnz2ZZeQZuj5v/eaEDe0Rkw+Ii/rZ9gByznhvOms1bHU427R3lZ5fMZXZOKsPeMA9v62d7rxsEgZWFWlaV6CgvyMYa1fOTl7o4rSqbL6+uoGnQy9udzngfVImn6KToNZRlGRlwRYhKMv9vzSyWlWcgKyo/er6dLnuQL64qRyMKHLAG2NrpwBtOjCRBRoqODJOWHkeIBYVp3HxeFRoUtja08nCLFBdGZaXQ6wgRlZWx2WSV2TkpfHJ5MbWFaXiCMX64uR1XIEZhuoFBTxhVBbNeQ1iSMepEbjm3mtJME02DXp7bO0LbiJ/4fHI8iajAYoiXpAWBS+sLqchJocse4PFdw9QUmDmjJoceR4gtLTaMOpFCiwFvWMLqC4MgkKrXYtSJRGMyfkngmgWprCmKV4YSSvfs7OwZM5N6vE6oh4OEP/Ff//pXHnvsMebNm0dbWxuvvfYahYWFH/8Bk48TQqXjiQ9Lqpmo2dhDCZWmSpCUCFaurKwkLy9vUq81Pnkj4Vk6PDxMW1sbFosl+RL7KJWlIAjcdO4crvyjnWeaRrhySSEnF2nZtD8EGUXU5aSSk6Ln1XY7p1RmUZRu4L0eNxkGkfX1+TzdOIzdH2VBsYUndw9h0Mb9d9tG/ZxWlY2sjNJpD7KgyMKWFjtRKW6U/8SeEeyBKOfPz+eVdjt/fKuXH15o4P53+qnKz+C2C2tQVPh/j+5FhTFCVRFROXV2Fql6LZKs8Nd3+9EJKt+7qIaH3+3nzi0HWL8gC4Pko8cHn11ZzIUL8pmTb+bulzu5+akWVAHW1OQyOyfejii0GDl5dhY7+jyUZpl4bzjM9lGJykwrg54YZj2cWy4SDgaoK0rj6YZh9BqRb51bSSAi0zbq460DDoIxFVGAu17piptkaASiMpxamUmaUUdpppGmQS/+iMK1JxWzsDidA9YAbx6w0271gwrNIz4++9cG8gwqjoiAIAr88MIa5uSZiUoK//tqFzt63WSm6OhxBvmfFw+giVvuohEFrltVxunVOQSjEo/sGOJfrXEzfUlR+eGmdmrzUwGB9tEAS8rSuXZFKT2OEI0DHt7qdCLJcRJ8dNcgQDzCTxRoGw3ExW5jW//0ND16rYgvIqERRVbOysRs0NJhC9AZlLmw2szl9dlUVVUBJHuHfX19iKKYrLAcT6OEI8F0IlSIixpLS0u59dZbWbx4MT//+c9ZuXIl5513HtXV1Vx44YVcccUVx5yyM9MwLUn1eNzQCfXvePh8PpqamiZkNvZg0p4qQnW73UmTismOiTsY472HVVXF4/Fgs9no6urCaDQm/9uhrODqii1cUF/CpoYeZus8rK7M4NWeCJv2jvKtM2azpjqbJ/eM8NvXu+OnOVXhooWFnFWbyzNNI+zsc1NfnM7G3cPsG/ZRm59G85CXz51ShqrC/lE/CwvT2NRspW3Uz5ISCxv3DPH1Rxvjmb6xGIKg4cYn9oHOxNfWVKAVRX79WheOoATE+4woMiaDhjfa7by+30FJpokBZ5BvnD6LBUUW/mftXO7e0sZTzU6MOg1F6QbOnRu/t2rzzfzy0nl8a2MzrpDEAaufxkEv9cUWPGGJh97ppybfzI8ursXmi/Lafhubm60EJRVRErh7m4vSNC/2QJR+n8pVi7JZVGRGo9Wyd9BHWFL52ifKmZNrpscRZOOeYYY8ISwGLW93uni7Mx70rahQZDEQiMrYAxFkVeGANcDiknS+sKqctiE3b7YOsteuIqkKIgL//Ww7FdlGYrJKvyvMufNyue6UMsKSzL9a7fxj5wCqEs/I/eObvTz8bh9ZKfHT6py8FL577hz6nSF29rp5bb+dQERGFAQOjAZ4dOcg+WkGdva6SdVp+HHXJopbG9iXWcbPll9GSpqZz62qIEWv4f63e7EHotx4VhWpeg1/equXiKTw6RUlzC1I44AtwJY2ByeXpvDFk3KoqpqdfPYsFgsWi4XKykoikQh2u50DBw4QDoc/YJQwHURM041Qx2Pr1q385Cc/YfPmzRQWFqKqKi0tLWzevHnStTHTEdOSVI8HDo5/s1qtdHR0UF9fPyFqtfEn1akQJAEMDw/T39/P4sWLpzRlA+IbpYyMDDIyMpgzZw6BQACbzZb0Ak0Q7Hjh2FdOLWbL3iHetop8s76Ys2oVNu0dxXZSmJp8MzIKb+y3s3ZhHrv6XezqdXPJokLm5pvZ3uPmsiVFGHUiDQMe5hWaeavTSSgqU2gx0DYS4Jx5eWiAzXtH2N3nRpai8XAAIZ5/qigKQUVFq4h894l9aDXxsqWKgCJLiKicOy+Pz51SSjAq87+vdNMw4MFs1DAnLxVUFevwIBvmaAmrqewd8pOfpickKaRp4i/q/dYAnrDMmTW5NA/7+PHz+6nNj/eigzGZL6+uQCMKFKQbyEszEo7JrF2YT26agdZhP3uHvPhCKqIGHm1w8OReJzpRJRCDObkpRCSVUV+Yd7qdDHnCfHZlKRcuKCAQlfjd6z1s73FRmZ2CgMqzTaNxURJxo/xUg5b3Om3EvHb6gxpMOoGbz60iLCm0DPt4td2GKyghCgJbWqzs7HWTk6anYzRARoqOH1wwhzSjlj39Xh7bPUS/K278sN8W5L+faWVugZn9o/E0m0sXFVKUYaBl2E/DgJd3e1wwloH7s9yTyLbM44ClEBWRef4RXt9vpmnQQ1gCQVD5yQsHSGx0REHgz9v6x+47kdo8I7ecXkj17MoPvT8NBgPFxcUUFxcjy3LScWz//v2kpqYmFbBT4QU8nQn1vffe46abbuK5555LlnwFQWD+/PnMnz9/ilc3NfiPJtUE0XV3d+NwOFi2bNmEPTQajYZIJDJlCt+uri58Pt9RWyVONhJzchUVFUSjUWw2W/KUkJ2djdlsxtbTw2VLCvnnez1c6ghywYI8NjWP8suXu+i1B9EAZoPAZ958BIs3hb+VrcDW2c+KWZn8eVsfo94I9cVp7OnzcOGCvHgpc8hHbUE8LzQe7g07uh2gSJgNGvxRmYI0IyOeEIgiGkFFifgJRfwg6hA0IggaTAYtBRYjW/aN0usIsKg0g6ZBD/PyU+h3hfj+s618rj6VgjQdmUXldLy9j9m5qfS7Qnz3yRZuOLOS8uwUHnq3j+IMI186rRxFVXm5zc4Tu4dxhWMYtSLPNY2yqNRCZoqOB7f1saAojWtWlKIRBapyUtnZ52JBkZkNi4sY9obZ1uVi37APk06g2x7kgK0PYIxsYNNeK3v6PLhCMXodIc6oyeYLp5aj04jxcaW3eii0GCjLTKF5yMPbXXLSgGNObioNA14qc1LiDk4hiauWFbG4NIP9o37e6nDQNhwAVBz+KDc80UJphhFXKIY7JHH50iJWV2XRNOhlW5eL1/Y7UVUVjSjQNupDrxUoTDfwdqdEfpqRS+oL8bzwEh1+2JM7C0mjxSTHaCKV8ICbiAx1RWnML0xjd7+X/VY/58/L4+TKLPqcQf6+Y5CsFD0/ObeU6tmzDvve1Gg0SRJNpMKM3wAezYjJ0WI6E+quXbv41re+xTPPPENJSclUL2faYFoKlVRVJRqNTuo1BgcHiUQi+P1+tFottbW1E1rmGR0dxev1Jvuyx9PMuqWlBZ1OR3V19bTsDX0UZFmmt7eX3t5e9Ho92tR0vv68ldJUhUUl6fxz9xDRmMIps9KpL07jD1v7uO2dv1IScHDd2d/kk51b+cS3P82Xnu7giiUFZKfq+f2bvfxifQ0/e6mD2vw0FpVa+N2bvaiyjKrE1adaESRZRa+Jn24isoIgarEYtXgjMowFp4taHSadjqgq8PU1swhEZP7y7gCyCidZFL69rp6hoMQPN7UiiFp+vHYu/9w1zO4+L7+6fD6eUIy7X+3C6Y8xr9BM87Cf/z6/mkWl8dGDQETmWxubEUWoyTPTMEjmpPIAAGE/SURBVOAlEH1/Uza/MI2yLBOpeg3PN49i1InccGYV+RYDzUNefv1qF0vLMrjxnNmIwO/f6Ob1Aw7mZevINap4YiIHXBK+iIw45pSkEwWyUnSM+iKUZpr47jlVGJUwezv6+Gt7PAB+cUk6I94IvY4QKnGSTtFrmFtgpizThM0f5Z0uJwsK0/jSaeX0ucJs73axtTOeOCPGjaWYlZNCZoqexgEP6SYdly0pxOqL0NjvpdsZjPdLBYHqvBRqCyykt+1lk9dIWG/gutaXMYWDPLLkQgY0qcwtMJNu0tJpC2LzRynKMFCRlYqiKjQN+UkzaPntugrqqism7P48eMRkMv10pzOhNjY28uUvf5knnngi2aOehjih/k1e9DiQan9/P52dnVRWVlJWVjahn62qKg6Hg/b2dsrKysjNzT0uD0U0GqWpqYn8/HxKS0sn/XqTgUTJur6+Hr1ej9vt5oGt3dz37gjEIlRkaulxxVhfn88Vs1P46t+bmO0d4Qe7/sn3l1/FUEoWv1+Vwe3DZuz+CD++uJYv/qOJDfX52AMxdvS4Kc8ysnfIi6BIY6cwAVFQSdFp8MekeNlR1MRnUOU4+UiygqgzoNFqyTUbkBQVT1iOe/LLEioCc92D3Db4Bu3XfZ5IWi6/3WZHUiEQjRvjX7Y4Xh7zhyX+8FYv73Q5Meg0fGpFKWdU52DSa/jjW7280mbjJ+vmMicvFUlR+fWrXWzrcjG/KI1gVGLIEyYcjStuxeRrQxg7jarkmg2kGTX4wjI2X4SafDNn1uaQa9bxQvMI23t9nFossqzQiBcTO4Yi7B30xeNtE7aKeoGIEo9z++KpFZxaGVfJ3/mvDhoGvJxUnoFJp6HbEaDPGR6bF42nBhVlGJMCIbNBwzdPn4VGFGga8vFKmw1nIBpX94oiC4otLChKo2XYz64+DzX5qczKSaHDGqDbERorx4/pLNR4MpCqCmg0YDHqiMkqgYhEVoqOjBQdwaiM1RfFbNTzh8sqWVpTMWn3qqIouN3uZDBFQieQk5NzzBmso6Oj0zYtZ9++fXz+85/n8ccfp6amZqqX81E4QarJi04yqXo8HhoaGsjIyKC+vn5CP3u8Q1JiftNut6PRaMjLyztio/rDRSAQYO/evVRVVX3obOx0RsKUwu12s3Dhwg+UrGOywsW/fxeXw86tp+Xwl502Drglbj8lg8Y/PcUjtadz75v30mvO5WdLLuPmGg3+ghLueaOb2y6IOyM5/FFq8lN4db8DVZZBjsVD0gU1mSxj1IrxsHRFRhC18cqFGu8xImjQ6vUUWAyM+GKYDRqCEYVUOcrP3/gTHbkV3L1oHRd0beeTi3IwXnUVHdYAtzzbCqKW/7m4htqC92eif/lyJzv7PJRlptDlCGDQaqjOTWXvsI+zx8K0BQG2dbm46+VOLq7L59MnlyIrKne93MnOXjefP6WUwnQj7aN+nmgYJlWvZUVFBqGYTOuIH5svgkEnEI2p75sjKComnYba/FQqsozYPAHe6QtSlSFw3Un5jPii7Or38e6wRFRSkr6/InF3o4issqwsg09UZ5GTque5plHe6XJxWlUWS8rS6XeG2NbtYsgTjsfyjZFhcYYBX0TCG5I4Z24ui0osNA352NnrxuaPoCJg0oksLLJQmZOCIxjjtXYH6SlazkqNEO4f5HVDAX5Bz5rqHOYWmHmzw0HzoI/TquLzr5ICv/hXBzFV5LfrK1kxt/z43cCQLBOPDw/Pzc0lLS3tiCpG05lQ29ra+MxnPsMjjzwyE3qmJ0g1edFJJNWRkRG6urqoqKjA4/Ewd+7cCfvsj+qfHmxUnxDmTIQoyul0sn//fubPn/+hZhbTGYmgAlVVP7QMv7PXzafuf48L56Ry1txcvrWxmdWzzKx/52m+m/sJVvc18tWG5/jyBTeSWZLH9y+o5sv/aGJ+YRqzsk08snMQVZFRJQkUOfm4ZaTocQdjIArx/67IccN9tHHCFeIx5WaTkbAMnzm5hKcaRyhMN3LZ4kJ+8mInV3a+zSc73uL+Oafz3JxT+JZmkFM/fwl/eLOHVw84yUjR44/IfHFVOWuqs2kY8PKTFzu4enkxGxYXsX/Uz7/arLy23xFPlBEgVa8lO1XLgDtCmkHDmupsUvRaGgc8tIz4WTkrgwVFFuyBKJv3jmLSiVy+pJgMk5aX22w0DHg4e24unzu5lLCkcOeWDtpG/SwrTyfNqKXTNv6ECUXpBkpTVcxijPdGVWQEvrumjPxsC62jAf65YwBrIEaKTkMwGld0qmPG9XqtQH6agaxUPcPuEPZAjMWlFj61vAR3SOK1djtvd4/N0AoCGgRq8lPJNuvZ0eNGEAVOr84hEJFpt/oY9kSI1w+E93NRgUS+7vsOSyTnahHiVvopJh2/WT+HU+ZNbPXpSJHIL7bZbPj9ftLT08nNzSUrK+sjy8TTmVAPHDjANddcw9/+9jfq6uqmejmHgxOkOh6RSGRCP09VVTo6OvB6vdTX1xMIBOjv72fBggUT9vmHK0hK9GWsVmtSmJOXl4fFYjniHujg4CBDQ0PU1dUdc8lpKiDLMk1NTWRkZHzsXPCPnm/nH1v3872zK3ivx8UrbXZ+vr6WV1/ezYteAz+U9/FexUI298M3VhWwzxbl5TY7iiyBKoOioCIACihq0hHJoIFwJIaqKmg0WrQ6PZIKIioyGtDquGxRAa0jAay+COvrC3jw3X5uPXcObz+7lW26PO564w8Uh318f8XVdOXP5uoVZfz5vQEuWVTERQvz+dXLnewb8bOoJJ0+dwiTTscvN8xDN6YC/t0b3byx38FXP1FBVFLpsAd4u8NBRFJI1WsJy0pcmTtGTAmoYzFt47+2uBfvWLYr8Z9DAepLLKyuyqEqL4WXW208t3eU+mILtQVmGnvtdDqjRGVAAItBS0W6hiydRLtbZTSo8vmVJZy3oACHP8rPthygxxFixawMslINDLpD7BvyEZXlD6zPpBMJxeKZuV9fMwtFUdnT7+H1Aw484RgCAnqNyJy8FHLNRvYOenGGYiwtzWB+kZm+xzfxVslCUqQIn2x7g3LPCI+d92kaghoumJ/P6jlZjHgi/HlbP2FF4Hcb5rBqign1YIx3InI6nUknooOrVtOZUHt6erj66qv585//zJIlS6Z6OYeLE6Q6HtFolI9Z22FDkiT27t1LSkpKUrzj9/vp7OyckPJvIgP1aARJsizjcDiwWuMB1RkZGeTl5X3sfFxikxAMBlmwYMGMjKyKRCI0NTVRUlJyWA4soZjMFfftYGjUzs1nz+IX/+og3aTn5nNmc8MT+6jMSeXmc2fz7ceb8YajBKNKnEgVmcQBx2LS4A3LpBtE3OEYqiRj1KjxAGxRQNQYEAUVrSIT1egRtTpqCiwMukJ8ZXUFv3y5k8+uLGHzXitpJh1fWZTGf2/pZ769jx/u/Cf2az/Pt0PleGIKZVlp/PySeWg1Aoqisrl5lL/vGCSmQE2BmfPm5bOwKI32UT93vdzFZUsKuWpZMaoKv3q1k3e7XHzv/GrqSyxs7XDw61e7WVqezldOq6DHGeSul7swagW+cfosjHoN923tpcMW4Ly5edSXWOh1hnhu7wjBiES+xYAzECMqK8T7rwrFGSbW1uWTo3qJyCr3NwSQVIWzavKw+SMcsAYY9cbdjBJCI5OOMZtCgcqcFIozTURjMnv6PSCobFhUxKISC6O+CH99dwCbP4JOKxCT4s5TJRkGAhEZd0hidVU2i0os7LcFaOz3MOSNm/mL7zeK436/gpj8/amqAoKIXiuQZzaQZtQy4o0QVQV+fUk1p82b/lqCYDCYFDvFYjGys7MRRRG73T4t1fr9/f1ceeWV/OlPf+Kkk06a6uUcCU6Q6nhMFKmGQiEaGhooKyujuLj4A/++tbX1mHZdE23okBA+WK1WXC4XZrOZvLy8f0uCkWWZffv2YTKZqKqqmnEKX3i/B1xdXX1EHqH9rhBX37+DaMBDVU7cFL4kQ09ZVgpvd7nIMGpxhWLxvilq/P9UAZBBFVCR0QpxkhNFkGUVRY7EHYAEA6oYt3hH1IBGT07Ex01nV3Hz21bOm5tLtz2I1Rdhw+JC7nu7n6vn6tGYs/nbrlFuOruKuhIL332qhSG/TIHFyI/XziUzJX7qOGD1871n25mVk4onJOEIRuNrUhWMOg2LStMxG7T0OkN02ALMyUulON3IiDdM+6gfg0Yg26zHH5ZxhWIIwljurPr+iVWnEbCYtOg1AjZ/fGTo0kVFfGJONkadyE9fPMD+0QDl2SZcgSje8PulcL1W5OplxZw8KxOLUcvPX+qgadDHhiWFFKYb6XeGeLnVRiAqk5ca/578MfBFVdTE6RiIe/mqaDUCp1dnc+miQgIRmRf2jfL6AQfyWNlWEARKMw2AQK8zRJ5ZzyWLijAbNbxxwMGuPi+1ESfr9rxEVKvjsepVDFryWJZvIifHgtUXYe+gFzQ67r28llVzi//tfpnukCSJrq4uhoeH0el0HzCsnw7kOjQ0xOWXX84999zDqaeeOtXLOVKcINXxmAhSdblcybzQgz1+o9EojY2NLF++/Kg+e7IdkhIRTFarNWlenSgRt7W1UVRU9IFNwkyC2+2mra2NBQsWHFVPudse5It/30O/1Y0qx1DU+OkLVUWRpfipRlEQRBFFUUGVEbV6lFgEGDO/V+X46UsQkoHWxP8RVdBhFAW+0byZXy66hIuyZcI5ebzWbufLp5Vxz5s9rKtJ461uP2mpRn66fi43PdVKSIr74TYMBrj6pBKeahjBZNDw9U9UUp2fynefbCGGwN0bFmDUiRwYDfCrVzpxhWJU5qQSism4glH8EQWdRiBVLwICnlAMnQjV+WnIisp+mx+jRuS0OdnIssrbnU5iiszKWVlkmLS0jQQ4YPMjCvENRcI3TAQU4ML5eVy1rIierg46AnoeaXShxOmZmBz/0xpRQFJUavPTmFtgRivCy+0O/GGJ8xfkUZKZQuOAm3c6XRT77Xx634ukl+bz0qnreaXHj0krYtAIuEMyKirpRi2esESe2cA3Tp9FRFJoHPLyWrsdT1j6QMmY+K8PURQw6QTSIgE8skhUZ6A+S0t9dSFGnYanGobxS/DbDXM5tbboKO7EqUei5Lt48WI0Gk0y/cnhcKDVapNq4qmw+hsZGeGyyy7j7rvvZs2aNcf9+hOAE6Q6HrFY7LBDvg+FwcHBZH/iUHmhsiyzY8cOTj755CP+7KmwHAwEAgwMDDA4OIjJZKKoqIi8vLwZF/47OjpKb28vdXV1x6SCDkQk7n87ntQy6AygIhCTYmhEDTqtBp0I2WnGsZ6liiiIDDkDBKMS0UgIJRoCQI6GETV6VBGQJUStgdKIj0FzNuf37KQ9o5jO9Hxq81I5YA2SZRLRCQqOkMS1K8q4751+vnlGJWaDlp+82AFaHV84tYJz5+fS6whx9yudDHki8ei3sMJN58xhWUUGqqryp629/KvVxnfOrmLFrEwaB7zcsWU/NXmpfOP0Snb2evjzu31YDFpWzc6iwxagZdiPXgtZKXE1rS8sIYgqFoMWs0GLPyLhC8uUZBr41PISascyZB96t5+oFM89lRQFvQizsox0u6JYTDpuv6iGbLOeTmuAP7zZS78nTIHFQExWcAViyKo6NsLzfktCGbMsBBJ7GlAVsowC58wycFJFNgHRxP3bBul1hpIiI4G4yYM7GCUkqZw3N48za3PwhSUefKePfneEkyoyKM8y4QxG2dblIhhVyEzREYrJhMdCylOMen53xTxWVk8L8/YjxnhCPdSpNBwOJ8MpIpFIMvUqPT190q0TrVYrGzZs4M477+Sss86a1GtNIk6Q6ngcLamqqkp7ezuhUOjfRjMO/nPbtm3jlFNOOeLPnwrLQYfDwYEDB5I/k81mw2q1EovFyMnJIS8v77g4vBwLent7cTgc1NXVTVlpS1VVdvS62dnrZnuPm6ZeO6kmPWajjj5nEGPYgyngx2qyoKoqOiGuhBUEFaNGQyAa78+qikKKTgOiiqKImPQavBEZVdTyuVPKuWBBPLggKin85rVutnW7Ecf63tlmPSIqNn+MwnQjRRlGnP4YXQ4/GkFEEFRkRUBR5TEFbPx3qqJi1IhU5qbgCsQY8UXIMGpZXJpOICrTOOglEpMxaUVCUvzZEcfmV9NNGm45dw5FaTq27Gyjwamh2RoC4mrcOfmpzC1IY9+wjwO2AJ85uZQLF+TjCUvc9lwbVn+Ub55ZSXmmiY17hnltv4NTDUH+69k/EjSmcv/C82nLLCEv4MSVlkVMUdEAAgoSIqdXpHDFsmKG/CpvHHCwtcuFrBwssFIRECmw6KnITiXVILKj100oqvDpk0s5bU4W3lCMH27eT1gW+ONVCzip6t9ToGYCPo5QD0YinMJms+HxeCY1YcfhcLBhwwZuv/12zj///An9bFmWWbZsGcXFxWzatAmn08mVV15JT08PFRUVPPbYYx+ZQX2EOEGq43E0pCpJEo2NjVgslsPqNb7zzjtHRKoJQRJwXE22+/v7GR0dpa6u7t9sFCVJSiqJA4EAWVlZ5OXlHVHU2mRDVVX279+PJEnMnTt3WhiUHwrvdrv4/F92oJeCnG7y87rfiFYr8MlFOfxxu4OaTIGQIjLslynNNNFhC5Jm1OIJy+RbjFx/xiyeaxxle7+Xc+blc9WyIrpsQX62pZMFJRlsWFRAuzXArj43bSMBUgwacs16opLCiC+MXiOytCwDb1iiechHZoqW+iILTUNenEGJNIMGUQBPSPrAg6kV42VbrShwUnkGp8zOoshi4IFt/bQO+9COlXINWpF5WQLleRk81+KiJNPIVcuK2D8aoGnQQ6c9hALxUqwqIIoqkhLfPOrHPkNFQFEURFFAi0qW301Qp8evT+G8zu185sDrKI89ysttdv62fYBITEIYUyKbdAJpOrAGITtVy/WnV1KYbmR3v4eH3x0gqigsKLSgqCojvghWTwRFiG8MGBMqaTUCJoOOP169kCWzJjdxabJwpIR6MBKtoYTYaSITdlwuFxs2bODWW29l7dq1R/05H4a7776bnTt34vV62bRpE9/97nfJysri5ptv5mc/+xkul4s777xzoi53glTH40hJNRgM0tDQQGVlJQUFBYf1dw6XVKcqYSZBRtFolHnz5n2swldRFJxOJ1arFY/Hg8ViIS8v72Nn4yYTsizT3NyM2WymsrJy2hD9h+HVdjvfeHQPsy1w5bIifvLCAQpTRSpSJN4YgTXlJrYPRsg164jKCqGYyuJSC691uPj0ilLOn5/HP3YMsrnZGg9VEAQyU0187pRS8tIMNA16+fv2QSqyTVy2uIi9Qz5earNi0IiUZJjoc4UIxaRkmVVVFVRVIMOkIUWvxeoLIwgCNXlmMlK0tAz7cYdi5KTqsBi1DLrDRMbGbzSCwBnVOXxxVSn7Bt1s2tNLk00lpqoYNCLr6gtYPSeb/DQD977Vw+v7nZw3P4+afDPDnjBbWqz4wjLLKzLIMxvYN+Kny+6nNt9MfYkFrz/M1r2DeDQGRFQQRURVpSDdyKgvSppRy/fOryLNoKVp0MemvaP0ukLvm0IQ3xBIiopOo2F1VRZVeWZUVeWRHUMIosq3z5xNql5L87CXf+4cRtRq+cs19Swsy57K2+SoMTIywsDAAIsWLZqwak0iYcdutxMKhY46Ycfr9bJhwwa+/e1vs2HDhglZ23gMDAzw6U9/mu9973vcfffdbNq0iZqaGl5//XUKCwsZHh5mzZo1tLe3T9QlT5DqeEiSdNixQQ6HIyl8SU9PP+xrHA6pThWhSpJEc3MzaWlpR0VGiai1hNApJSWFvLw8cnJyjtsMXEIMNtNEVS/ss3LjYw3MyRA4vSqDe7cOUJ1jIDvNyDtdLk4u1vPuYJTydA2jQRWLUUNRegp7hnwsL8/g0kWFvNpm55X9TkSNlve3hsLYiEhcFJTsQ6JiNuiIShIxBXJSdIgagVFvJOlIBIny6PvRiKoaT9QRgbJME/OL0nAFY2zrcpKii79MAzGFikwjK3IkwoZMnmm2U55lwqAROWAPAmDQikQkldqCNE6elYGqwubmEZwBidPmZJNh0rJnwEufM8is7BQWlaRj1Iu80mrDEYhxo2mERYFh2uYs5plQGnv63WM/rYBeI1CVl4o3LDHgDnNGdTafWlHKgCvEa/vtvLbfgV4jYNKq+KICSkLJjIBGgHyLgXSTll5HGJNBx5+vXURN0YSVB48rJoNQD0YiYcdut+NyuQ47Ycfv93PZZZfxla98hauvvnpS1nbZZZdxyy234PP5+OUvf8mmTZvIyMjA7XYn/0xmZiYul2uiLnmCVMfjcEm1r6+P4eFhFi1adMTmBx9HqlNFqOFwmKamJkpLSw9rfvPjkEjasFqt2O32pKpwsiwT4f1g9Jlqm/jCvlG+s7GJLPzUFqbzRqebnFQdKuDwxyhK1zPkiWLWqYTGbtP8NB0jXgkJAVHUUpqVwpm1OciKyivtdoY9UYoyjFRkp7Bv2IcnFE/GUVQVf0RCEEUENe5ShCBg0oqkGbU4gzFkRaXAoicrRUevM0RgLMJuZWUGsgK7+tz0u8KoikJWio6vrK5gQXEaL+0d4ummUTxRUFWBuhILt5xbhU6jweoLc++bPTQP+0k16AhFJRTi94uKkFTkigLJ8rFeFAhGFRQSYzQCaUYtc3JTicoKzUN+Fpek8eXVFXTYAuzuc/Nmh4OIrCZFTVkpekw6gUFPhAKLgWtXlDIrJ4X9oz5++3oPqTqB5fkiikbPgB/228Nkper5++eWMivv8DfN0wnHg1APxviEHbvdDhw6YScQCHDllVfymc98hmuvvXZS1rJp0yaef/55fv/73/P666+fINWpwMeRqqIotLW1IUkS8+fPP6ry5rZt21ixYsUhSyRTJUjy+Xw0NzdTW1s7kQ37D+BQlol5eXkfyDI9Fng8HlpaWqYkGH2i4HK5eGxrC/fuCREK+uPjOgKIY+5FYz78KErcnICEs5EoYDZoMGpFPBGVmDL2CAni2NCKijr29/Wa+P+Q1PipTCOKhKS4I1Hyz479dYMm3leMSGNkJsavqSKgF+I91eIMA4tL09nW5cQekFhWaOLMwZ08ramgHTNmg4ZARCY3zcDqqizs/ihvdLi4YEE+n10Zz4T9nxf202kPcu2KEqrzzWzcM8zuPk9y3jQiyfz8pU684Rj/tbwYURQ4YA3wXo8rHjQOGLQCcwvTKM2Mn+y9oRhfPK2CnFQDnXY/r+13MOAOE/8JxtsOqug1IvUl6ZRlGgmGI7za4SXXrOUHq7OoLS+cFGHOZGMqCPVQGJ+w89577/H2229z3nnn8fe//52rr76a6667btKufcstt/DXv/4VrVZLOBzG6/Vy6aWXsmPHjhPl3+MFWZY/ECI+HomyYnZ2NrNmzTpqwtu+fTuLFy/+t4d0qgRJNpuNzs5O6urqjttc2kRaJkJcit/d3U1dXd2MG/dJIPEz1NfX0+mM8aW/78Hl8XBGVSbvdrsIxWTmFqaxd9BLik6DUafBHoyRohXRagV8YSXuiyuIYwQpohEhpqhjatc4aaoQD0QXBFRFRRXi5VKtKOCPyAiCgE4DmgSZApkmHcvL06ktSKNz2M2Wdke8miKIpJu0fPHUchaXpfPY1gNsanERE0V0wDd2bOTkz2/g3YIaXmq1sXfIlzyNWowatFoRdyCGrPKB37tyCFMHUYiXm2sK0ijLNLK100n7iJ+19fnMzkmlZcTP7j4XVn8svtEASjJNlGWZGHSH6HWGWV2VxRdWleELSzyxZ4RX2m1km3WUZqRg9UUY8kRAEKjMNvLQZ5djFOTk/KZGo5nS+c0jwXQh1IMRiUR44YUXuOeeexgYGGDJkiVcdNFFXHjhhRNSHfsojD+pfuc73yE7OzspVHI6nfz85z+fqEudINXx+DBS9fv9ybJiXt6xqf927drF/PnzP1AClWV5SgRJ/f392Gw26urqpmwnfrSWiQn09/djtVqn9Gc4VgwMDCSV1omfwRWM8r1nWnm1ZYjyNBFZVehzRjBqBVRUwtKYx26CfARhbKZTQBQhJqmIoog6pmSV1fg/C2On3cQzaNAKyIqKrEKKTkN5dgrBiEyfK4RWIzArKwVJUelzhYiNVVEKfA5+8PqDhFLM/P6Ma+kSUllZlkpBdyvPUYSgKkg6A2XOIT7V+y71v7uDjbuHeGzPCAuLLMwtMGPzR9nZ6yYQVVhSZqEs08S2bhcj3ghnVOdwUkUGXfYgTzUOoxMFFpdYcAZj9DhCBGMyqgpGrcDCLD0LGt9G9Hr5R/Vq0sxGLl1chDMY44A1QPOQh5gyZnmIgKCqmPQagjGZ4gwjFy7IozwrlX1DXh7ZPUJ1joGHP38y6akfbFGMn9+MRqNkZWWRm5tLenr6tBLCTVdChfhm+pprruGss87i+uuv58CBA2zatInNmzdzyy23TOps6nhSdTgcXHHFFfT19VFWVsbjjz9+RA5rH4MTpDoehyJVm83G/v37qaurm5A0loaGBubMmUNqauqU9U8VRWH//v3Isjytxk0O1zIR3vchDofDzJ8/f9r8DEcCVVXp7u7G5/Md0ktZVVU2N1v5+Zb9jLp88eg44qdLETDoNERjylhMXEJEJAAimrFbSU44N4lx8VHcKnHMFnFcSVlQheRnJNoPWSk65hWmkZtm4LV2G4FQlArnIB3pRViiAb6662lOGmzhV9/5X97oCSIIMH+kk1u3Psze/Cr+suh8RtJyMeo0hGSVWdkprKzMwhmI8laHg1BMJS9NT0RWcQXiCVECcRGTRiPgD8VINWhZPSeL2vw09FqRB9/pwxeWOHtuDuFgmNaGToYsuWMlcZXKoIPZS+eSnaLnlXYbrqDEF1aVUZZlos8Z5rm9Iwy6w5gNGsJRBWlMtIWo5eRSE/deezImw0dvzhIbQZvNhtfrxWKxJOc3p9IPezoTaiwW47Of/SwrV67kxhtvnFYbkQnGCVIdD0VRiMVi8UWMZW3abDYWLVr0kSq2I8HevXspLy8nLS1tyhS+e/fuPayElqmEqqpJ+zS73Y7BYEhmw2q1Wvbt24fRaJyxPsSqqtLW1oaqqsydO/cjf4ZgVOaPW3t4avcQVk8AgyBj1II7JAHxk6sgxE+hogiSoiAIGhBAJ8Y7pRH5/ZOpRoBgNJ6YYzGIFGeYCEYV+txhjFqBNXNyyDbr6bIF2NXvJhxTEAWVs8xRPvePX2AzpPGbZRvoyCphjqOPvtxSFEAnqERjClc3vcQlrW+ARsNP1l7P7pRCdFoNsbHTdZzHVAozjJj1GnqccUXwybMyyTEbaBrw0mUPYjaIWAxaRn1RZFUdC0QXmJOXwqKSDGrtPWx/bQ9bKlcwx9FHjbWbnuxiOormEJITwW3xE3ieRY/TH8UbkVlTncOliwrINOn4zevd7BwIsKrMxO+uPRm97sjIKKF4T5SJDQZDskw8WYK8Q2E6E6okSXzhC19g4cKFfO9735uRz+sR4ASpjkeCVBVFYd++fYiiOOEnuZaWFgoKCrBYLMddkBQKhWhqaqKiooL8/JnlCpNQFI6OjhIMBsnKyqK6unpG9lCPdo5WUhS2dbnY0mLjldZRXIF4KDeqgl5UiUjx06dGEDBoRRRVJTzmcpRq0JCi1eIORZGUONEUWHQEIhJWvxTPHU2EgwsCxRkG3CEJbyjGybMyUBTY3uvGHAvwycZ/kRNyc8/yDXgMqeg1Il//RAULiy088OR23g4ayPU7Mem19BsyqS9NZ0Ghmfd63BywBdGJkGoQCUQUonJcNqTXiqTo44HtUUmlPMvEWTW5lGWZaBz08GTDCLlmPXMLzPQ6g/Q44z+7qkJ6xM/ywVZmuYbwmNJ4uu4sMkw6LltSSFRWOTAa4L0eJ+Exy8QEREFAFbVcMT+NH2xYPiGnzGAwmNwIyrJ81KHhR4Lh4WEGBwenJaHKssxXvvIVZs2axY9+9KP/64QKJ0j1g1AUBb/fT0NDAwUFBZSVlU34TdDe3k5qair5+fnHtWSZUMfOmzfviOZqpxMSm4LS0lIURZlxlokQL4M1NTWRl5dHaenRR4apqsqgO0zriJ9Oe4BBdzgehYZKVFIIRiQ84RjS2IktIelVx+ZO4ypfFVQVnQjLyyzMy9Hi8fp5ayDCcEBBVeOCppWVWVy4IA+9RuTel9vp9EiogkqmFOa8JaW81uVm1CexZk426+rz2dnr5rHdw0SVsRGZZGkajBqR3DQDzkCUYEyhJMPI4pJ0nMEYO/vcRCSFrFQdwYic3BAoikqqQcOysgxqC8yY9Br+vn0AdzDGUnsXUiRKZ0YRnhRLcq7WpIsbW5iNOlqGvWhEga+unkVdSRotQz7uf6cPV0Tgy0stfP3CZZPyLB4cGp6RkZEMDZ+o601nQlUUheuvv57c3Fx++tOfzsgWzVHgBKmORyAQ4L333qOmpmZS5hxlWcbtdtPV1UUsFvvAWMlkksHo6Cg9PT0zWh3r9XrZt2/fv20KYrFYUkAyXS0TE4hEIjQ2NlJeXn7cKgWqqsZdmKIKgaiENyThDsWw+aMMucM0DHgYdIfosvlRZAVRlZEUmbqCFFYVa9g9GKDBFhdGpehEgjEFo0ZBI4oEYipVOSZOr8ll/6iftzocyHHPQURBQCOIxMYcyoQxf+H4muJrE8ZZASqoY5aJ6SwotKDXCjy+ewhHIEp9SToi0GHz4w3LKGNq4bIsI/NyTBR3trA1ZGC/IYulpRbqSzIY8oTZO+hhwB0Zu378mnqNgCiKaLQ6bliZydWnLzou90lCL2Cz2XC5XJhMpmSZ+GhbS9OdUG+44QZMJhN33333fwqhwglS/SBkWcbr9U64ZP5QgqQEGVitVkKh0DGPlXzYdXt7e3E6nSxcuHDGqmPtdjsdHR0fO/ZzsAH4dLBMTOBos1yPF/wRiZ29bv7VauPVNivuQASTVmFJcRpD7hCdjnDSeMGsg1NmZ6Mg8F6PE29Ijvv3jkXcCcTjZAVBBAFMOg0peg2+sERMVslPM7Ck1IJWI/JutwtbIEquWY/FqGXYEyEYU1BUBQHIMRuYX5hGRVYKe4e87On3UpplpDbfTL8rRJctSERWkq2UvDQ9pZkm3CGJLnuAmjwzX/tEBZ6wxJsdDl7d78RsMvKjNVmcs2LhlGy8DmWQkDBGOVwf3elOqLfeeiuSJHHPPff8JxEqnCDVgy6sqkSj0Qn/TEVRkGU5rro8xANz8FjJRJy2FEWhtbUVURSpqamZsTf24OAgw8PDhzT2/yioqpo8GTgcDlJTU5Mng+O9uUicshcsWDAhCvLJhqyobO9x8dzeUV5sHiYYjmLUqNSkK+SmpdDpitLtiiZdmFBBpwUxHCEmalA1WvRSlIKAk9SyIoZiGtwhCQ2g04pExsq6CWtAs0FLVW4KxelG2q1+Om0hZuUYqS9OZ8Adpn3Ehzccz0gVBZGKbCOVOanEZJVtXU5SDVouWpiPrKh02oM0DLiJSAk3JRWNGHdg8kRU5uSmcOtp2ayo/2hx2PFEYm7bZrMlfXRzc3PJyMg45HM7PDzM0NAQ9fX105JQb7vtNtxuN3/6059m7HvnGHCCVA9GJBKZsM86Goekgw3q09PTk6etw71BE327nJycSekLHw+oqkpnZyeBQOCQ4yZH+ll+vz/pSXw8LBMTSMTn1dfXz8jSezAqs6V5iCe3d7PPLhGKjoVOqApmvYhFL6IqEvaQiqSCLhbFKEfx6lNA1MRrvYKAVhAozjSQazbQ5whhC8TtE5eVp+MOSrSO+LD6ook/TqpepCo33hZpHvKRotdw/vw8opLCAWuAtlE/MVlNlnUL041kp+rptAWQFJX/OqmEhYVmuuwhntk7wpBPYlWFmW+uzGFezZxp+0wkfHRtNhtut/vf4tYShLpo0aIpr74cDFVV+clPfsLAwAB//vOfp936jhNOkOrBmChSnQjLwcRpy2q14nQ6k3ObOTk5H3rDJvxvKysrj9moYqqgKAotLS3odDqqq6sn/AWYsEy0Wq2oqpoUOk2UZWICIyMj9Pf3U19fP2EjWccbifuppqaGVEs6ewe9vNPppNsRYtAdot8ZxBWIjJlJqKDI5Hvt+PQmgoYUio1QPzsXBWga8DLoDicNKVTAYtSg12qw+6PkpOq5dHEhAtAy4mNHj4tQTEk6QuWYdRSkGel1BvFHFc6pzaG+JJ1eR4DtvR66HSFUVUUU41aEaQYRnUaDJypw1YI0/mtxDrNnz57Kr/OIkIhbS1RbJElCVVXq6+sxm81TvbwPQFVVfvGLX7B//34efvjhaXeCPo44QaoHIxqN8jHr+1hMhkNS4gFLGNSbTKZ/S4Bxu920trbOaP9bSZJoamoiOzub8vLySb9eNBpNehKHw+FkRuSx9rb7+vqw2+1TGo5+rPD7/ezdu/dj76dARKLXGaR/6y76HnqMA1ml7MqbzYg5M25MoaqIqoysxM0ezqrNIcdsoG3Uz+v77YRjSjJFRwXSjVoCMQlUOGduHkvL0ulzhXirw0mXPYg6No+rqpCVqkMAnAGJ2bkmvriqnIik8naXk1fbHYhaPd88KY3Ta3KpqKg4Lt/bZGB4eJj+/n7y8/NxOBzEYjGys7PJycmZclcnVVX5zW9+w+7du/nHP/4xY7UbE4QTpHowjoVUj6dDUqKcabPZ0Ol0GAwGfD4fixYtOq5D5xOJRFLO8VTHjocsy8nels/nS/a2jiQj8v+C0xO8P4JVV1d3ZCd4qxVx3z7UggIic2rotAVo7HPQ0GNj54Afe1CJ91RVBVVR0GtUTq3MZGlZJjFF4dnGUbodwXGW93EzKJ1GICKpzMpJ4eplxWhFgV19Hl5utxGKymPPW5xoUw0awhJU5mfw1TotCyvyj2l8aapxqJKvJEk4HA7sdvuUujqpqsof/vAH3nzzTR5//PEZW5GZQJwg1YNxtKQ61aHiiV6hKIrJUZ2Z1MPz+Xzs27eP2tpaMjIypno5KIqS7G25XC7S0tKSQqcPe2klxGFarXZSytbHC06nk/37909KHzgYCrOrY5CdnTYaR6OMhKDfE0WRFRRFRkClIsvE2bU5FKYbaRjw8MI+K5KiohUFYmNztxpRQJJVjHoNl9QXcEplJkPuMH/fMcCAT+ac+YVcUR6hoqRwRuXqHozD6aFOlauTqqo88MADbNmyhSeffPKIYzAPhXA4zOrVq4lEIkiSxGWXXcbtt9/Obbfdxn333Udubi4Ad9xxBxdccMExX28ScIJUD0bCUelIMFWRbYfqPUYiEaxWK1ardVIi1iYDiZf4woULp+U6E5aJCaGT0WhMlt4TO3NZlmlqaiIzM5Py8vIZS6hWq5Wenh7q6+sn5CX5UUictvqHrfGZ0rCeVpdKjzOCKxAZO83KGLUCq6uyWVyWjijA37cP0ucMYdLFT6+KGh/hMWk1yKKOG86qYr7OSklJyaSnn0wmhoaGkrnNR3L6PNjVKdHSmGhzlL/85S889dRTPPPMMxO2+UqMG5nNZmKxGKtWreLXv/41L774ImazmRtvvHFCrjOJOEGqB+NISXWqCDUajdLU1ER+/oeXtmKxWFKQk+gX5uXlTapl2pFieHiYgYEB6urqJv0lPlEYH74uCALZ2dlYrVZKS0spKiqa6uUdNRK/i0WLFh33vth4cwSHw4EPEyMxE3ttMTrsQQ6M+sadZhWqc1M5oyaHqtxUXt1v58VWBxlpKfz28vkoti7KyspmnBXneBwtoR6Mg81RjqalcSj84x//4B//+AebNm2atCi8YDDIqlWruPfee3nhhRdOkOpHXfT/CqkmBElwfDNQE0YCVVVVh+38lDgVWK1W/H7/lDsPJQILPB4PCxcunLHye4/HQ1NTEzqdDlEUJ+1UMNlICKvq6+un/HeRGIFKnLY0Gg3mjGxGJSMtoyHe7XbTNuLFE4yiqgopei0rq/K47fxKutv3MWvWrGSZcCZiogj1YIzfuDidTlJSUo7K1Wnjxo088MADbN68eVJUyLIss3TpUjo6Ovja177GnXfeyW233cZDDz2ExWJh2bJl3HXXXWRmZk74tScAJ0j1YEiShCzLH/lnpqp/Cu+XSufPn3/URgKKoiQJ1uv1kp6eTn5+/jHvXo/k+m1tbQiCMKONKRLq2Llz55KRkfEBl6xgMJh0yZpqdeZHYXz83MKFC6fl7yIcDicV2gmv55ycHOxRDQ0DXhaVWig2a2lsbKSqqors7OypXvJRY7II9WAc7OokCEJyQ/hRLZhnnnmG3/3ud2zevHnSPcTdbjeXXHIJv/3tb5PkLwgC3//+9xkeHubBBx+c1OsfJU6Q6sH4OFKdSkIdGhpicHBwQkulB2eYpqWlJTNMJ+OhninRcx8Ht9tNW1sbCxYsOORuPWGZOH7jklBnThfiSojcJEli3rx5M+J3cbBJfWZmJunp6fT09FBTUzMtLSAPF8eLUA+FSCSSLBOHw+EPhLAn7tfnn3+eu+66i82bNx+37/n2228nNTX1A2Xfnp4eLrroIpqbm4/LGo4QJ0j1YHwUqU6lwnei3IU+7jrjBTmHmoU9FiQM5UtLS2e0gMRms9HV1UV9ff1hqSsPNvFITU1NblymaqZPVdWkyG3OnOnrMPRRUBSF0dFR2tvb0Wq1yY1LTk7OjJsNnkpCPRjjPbR//vOfJ6sYr7zyClu2bJmUsJEEEiOCGRkZhEIhzjnnHG666SaWLl2afGf86le/4r333uPRRx+dtHUcA06Q6sH4MFKdKkGSLMvs27cPk8l0XAO5E+Wh0dFR7HY7Op2OvLw88vLyjmoWbbobyh8uEl7E9fX1R0WI4y0Tx3+vubm5x02opSgKe/fuJS0tjVmzZs1IQoX376l58+aRlpb2Afeh42lFeayYToR6MGRZ5v777+eBBx5Ar9eTm5vLxRdfzMUXXzwp5ixNTU18+tOfTupVrrjiCn7wgx9wzTXX0NDQgCAIVFRU8Mc//nG6bsxPkOrBkGUZSZI+8O9UVU3+u+NZuotEIjQ1NVFYWEhJSclxu+6hEAwGk2YTgiAkCfZwXlgul4v29vYPLZXOBEyWsCoUCiW/V1VVk0QwWaNFsizT2NhIbm7ujDZESPSzPyykIGFFabPZJnWs5FgxnQkV4K233uLWW29l06ZNFBYW0tfXx6ZNm3jhhRd4/PHHp/2GZQpwglQPxnhSncr+qd/vp7m5mTlz5kw74UVCOHI4s7AjIyP09fVRV1c3Yx9AVVVpb29HlmXmzp07aRurhGWi1WolEolM+AhULBajoaFhxs9v+nw+mpubWbhw4WFt0g6VuftRKTDHC4ODg4yMjExbQn333Xe54YYbeO6556Z8Uz+DcIJUD0aCVKeSUBPJJjPhZJeIrRodHU0SQX5+PqmpqfT19eF0Ome0/62iKDQ3N5OSksLs2bOP231w8AhUZmZmcgTqaIgg0c+uqKiYsUEL8L59Yn19/VHNRyZSoBIpMAmnrOzs7ON6jw4ODjI6OjotRpgOhZ07d3L99dfzzDPPHBcP7v9DOEGqB0NRFKLR6JQR6sDAACMjI0ecHzodIElScqTE6XSi1+upra0lMzNzWpXcDhcJc/9EhN5UIWGZaLVak0RwJArtUChEY2PjjO9nJxTXE2WfmBDmJfqwx6u/Pd0JtaGhga985Ss8+eSTMyrVZ5rgBKkejJGREUwmExqN5rgrfA8cOEAkEmHevHnT8mE7HMiynDzZpaenY7PZ8Hq9ZGRkkJeXd9xmYY8V0WiUhoYGysrKKCgomOrlJHE4lonjkeg9zps3b9LnCicTifnsyQyMSNj72Ww2FEX5QH97ot4D051Qm5ubue6669i4cSPV1dVTvZyZiBOkejDuvfdefve737FixQrWr1/P6tWrJ33sQZblpBqzsrJyRp7qIE5EjY2NFBcXf8Cu7+BZWIvFkgxen44vlkSG6HTsZx+MhGWizWZDo9EkicBkMuH1etm3b99h9x6nK+x2O52dnSxatOi4KaQTbQ2bzUYoFPpA+f1on8/pTqitra189rOf5dFHH2XevHlTvZyZihOkeijEYrFklNHWrVtZsmQJ69ev5/TTT5/whzoRdzbTZzcPl4gSiRqJk1ZiZnO6zBYmRDAzMZN2vIAsEokQi8WYP38+2dnZM3ajljD4X7Ro0ZS1Q8bPbXo8nqMySJnuhLp//36uvfZa/va3v1FXVzfVy5nJOEGqHwdZltm6dSsbN27ktddeY+HChaxfv56zzjrrmPs6iRd4ou84U+HxeJLh6EdinXjwzKZer0/2tKbiBZooMdbV1U2aSfjxgM1mo7Ozk6KiIlwuF6FQKOn1PJ0tEw/GyMgI/f39U2Lw/2E4VMxaYlP4YRvu6U6o3d3dXH311Tz00EMsWbJkqpcz03GCVI8EiqLw7rvvsnHjRl5++WWqq6u55JJLOPvss4+4vJZ48c30F7jVaqW7u/uw3YU+CgkvUpvNhiiKSYI9HqM4o6Oj9Pb2HpfIs8nEoZJmDmWZmCi/T9f+dmJ+s76+flpUMD4M4/1zDzVnPN0Jta+vjyuvvJL77ruPk046aaqX838BJ0j1aKEoCrt37+bxxx9ny5YtVFRUsHbtWi644IKPLRv29fVhs9moq6ubNjvwo8Fk/hzhcDjZK0yIRvLy8iZlA9Lf34/Vap32L/CPw8DAAFar9SNHmBRFSZbfx1smTpfyO7z/c0xXIvowJOaME/65er2eWCzG0qVLp813Ox6Dg4Ncfvnl/O53v+PUU0+d6uX8X8EJUp0IJGzfNm7cyPPPP09BQQFr167loosu+kBZNxqN0t7ejiAIzJs3b9qeEj4O45XK8+fPn/SfY7wpQjQaTZoiHKs7TsJTORgMsmDBghn7+4B4Cc/r9R5R0sx0sEw8GH19fTgcDurq6mYUoR6Mvr4+hoaGMJvN+Hy+aSfOGxkZ4bLLLuPuu+9mzZo1U72c/0s4QaoTDVVVaW1tZePGjWzatImMjAzWrVvHqlWr+MpXvsK1117LNddcM2P6WgdjqryIExg/Czs+Xs1isRzRWhLxc6IoUlNTM2N/H6qq0tHRkRzFOpaNwfiRkkQpc7KqA4fC0WwMpiMOPmmPF+c5nc6PHYOabFitVjZs2MCdd97JWWedddyv/38cJ0h1MpF44d1///3cd999LFmyhIsuuoi1a9eSn58/417ksViMxsZGCgoKpoVtmSzLSdchn8932K5DiVnamW4on9jATcbG4ODqQGLzMlGWieOhqipdXV0Eg8HjUvmYTBxO6ToxBjU+x/R4bV4cDgeXXnopP/rRjzj//PMn5DPD4TCrV68mEokgSRKXXXYZt99+O06nkyuvvJKenh4qKip47LHHZrQg8zBxglQnGzt27OC6667jD3/4AwUFBTzxxBM8/fTTiKLIxRdfzPr16ykqKpr2L/ZQKERTUxOVlZXk5uZO9XL+DQe7DiXKbQfnl063jcHRImGfmJqaOumzzRNtmTgeiY1nNBqdMZmuH4aj6QVHIpFkdSASiRx15eVw4HK52LBhA7feeitr166dsM9NJFqZzWZisRirVq3i17/+NU8++SRZWVncfPPN/OxnP8PlcnHnnXdO2HWnKU6Q6mTCarWydu1aHnnkEWbNmpX896qqMjg4yBNPPMFTTz1FNBrl4osvZt26dZSXl0+7F0vCRGCmuPIcahY2Pz8fs9lMc3Mzs2bNmtH+t7Is09TURHZ29nG3TzxWy8TxSAQVqKpKbW3ttLvvjwQDAwNJ0d7R9kwPrrwk8mEnog/r8XjYsGEDN9xwAxs2bDimz/ooBINBVq1axb333su1117L66+/TmFhIcPDw6xZs4b29vZJu/Y0wQlSnWxIkvSRyj9VVRkdHeXJJ5/kySefxOv1cuGFF7J+/fop6VkeDLvdTkdHx4T5rR5vqKqKz+djcHCQoaEhLBYLxcXF5ObmzkjldeKkXVRU9AHXqqnAwTObJpMpOVLycd9tonSt1WpnbEh6ApOhVk64kNlsNpxOJykpKYf93R4Mn8/H5Zdfzle/+lWuuuqqCVnfwZBlmaVLl9LR0cHXvvY17rzzTjIyMnC73ck/k5mZicvlmpTrTyOcINXpBrvdztNPP80TTzyBzWbj/PPPZ926dcydO/e4v3hmsrn/eCSSTRYuXIggCB+w9Uvkws6E2dSEH3F5eTn5+flTvZx/g9/vT5YyE5aJh8rcVRSFlpYWjEbjcU3+mQxMxAn145Aoryb6sKIoJr/bj9voBgIBrrjiCj772c9y7bXXTsr6xsPtdnPJJZfw29/+llWrVp0g1eN10ROkenhwuVw8++yzPPHEE/T393POOedwySWXTPr4x/hRk/nz50+LEYCjxUedtBNB1lardUrUrkeCcDhMQ0PDjPAjhn/P3E2IcUwmE/v27UuKxGYy+vv7sdvtx338J/Hd2mw2YrEY2dnZ5Obm/lsfNhQKceWVV3L11Vfz+c9//rit7/bbbyc1NZX77rvvRPn3eF30BKkeObxeL5s2beKJJ56go6ODs846i3Xr1rFkyZIJJdjEKUKn01FdXT2jTxFDQ0MMDg5SX1//sSftaDSK1WrFarUSi8U+ELw+1d9BIBBg79691NbWkpGRMaVrORrEYrEkwbpcLtLS0qiqqppRlokHY6oI9WAcLCJra2sjIyOD008/nc997nOsW7eOL3/5y5P6PdtsNnQ6HRkZGYRCIc455xxuuukm3njjDbKzs5NCJafTyc9//vNJW8c0wQlSnYnw+/288MILbNy4kZaWFk4//XTWrVvHSSeddEwPeCwWo6mpidzc3CnND50I9PT04HK5juqlF4vFkrOwoVBoUhWZH4eEP/SCBQuOyFd5ukGWZRobG8nOzsZkMiUjAWeCZeLBmC6EejAUReGNN97gr3/9K2+++SYFBQXceOONXHDBBZO6GWtqauLTn/40siyjKApXXHEFP/jBD3A4HFxxxRX09fVRVlbG448/PqPzfA8TJ0h1piMcDrNlyxY2btzInj17OO2001i3bh2nnHLKEVmjhcNhGhsbqaiomJb9usNFwu0pMaJxrC/qD5uFPR7B64lQ7pnuDy1JUnKMqbi4OPnvDxbjmM1mcnNzp5Vl4sFIEGp9ff203ATEYjE++9nPcvLJJ3P22Wfz7LPP8sILL2CxWLj//vtn/GZ5BuAEqf5fQiQS4ZVXXmHjxo1s376dk08+mfXr13Paaad9pGIwcRqaO3fujCwvJpAoXRsMhklRTiuKkjSm93g8k3rKSmSITkRQwVQioVYuLi7+yGjDhEo7YU6v1+uTatfpIiKb7oQqSRLXXXcd9fX13HrrrR+4//v6+sjPz5823+XHQVVVBEFI/v8ZhBOk+n8VsViMN954g40bN7J161aWLl3K+vXrWbNmzQcerM7OTmw2GwsXLkwma8xESJKUnN0sLy+f9OupqpoMXk+cshLWc8daEkwk5kxlhuhEIBaL0dDQQFlZ2RFXP4LBYFKlDUy5iGy6E6osy3zlK1+hsrKS22+/faYRURIJEu3q6qKysnKql3M0OEGq/wmQJCmZCfv6669TV1fH+vXr6ejo4KmnnuKll16aMTvYQyEajdLY2EhJScmUBL0nTlmJkQeTyZQk2COdKRwYGEhGhU3XEujhIDH+M2vWrGN24BrvOpQIVMjNzZ0Uy8RDYboTqqIoXH/99eTl5XHHHXdMyzUeDhKE+uKLL3LjjTfy5JNPUl1dPdXLOlKcINX/NMiyzLZt2/jud7/L4OAgy5cv55JLLuGcc86ZkSfVUChEY2MjVVVV5OTkTPVyAD6Q/KLVag87+aWnpwe3283ChQunlQDmSBGJRGhoaKCqqmrCx38SgQo2mw2/309WVha5ubkTYpl4KMwEQr3hhhtISUnhrrvumpZrPBLs3LmTq6++mr/97W+sWLECu90+bZ7rw8QJUv1PQywW46tf/Sp6vZ7//d//paGhgccff5yXXnqJWbNmsXbtWs4///yPzYSdDkj0gqezfWIoFEqO6giCcMih/cRccCgUmvGG8ol52pqamkk3Tz+4x22xWMjNzT0qy8RDYSYQ6i233IKiKPz2t7+dlms8HIzvm77wwgvs3LmTyy+/nBdeeIE///nPVFRUcP/9988Ua9ETpPqfhjvuuAOdTseNN974gdKZoig0NTUlM2GLiopYu3YtF1544bRMlnC5XLS3t8+oXnCijGm1WpEkKSnEGRgYAJjREXTwftVgKuZpD2WZeLQleIgTaiLXdTqSlaIo/PCHP8Tr9fLHP/5xWq7xcKAoCqIo0tjYiMvlory8nMsuu4zc3FwuvfRSzjrrLG666Sb+67/+i/Xr10/1cg8HJ0j1Pw2Jm/ijoKoqLS0tbNy4kc2bN5OZmcm6deu46KKLpkUpxmq10t3dPaOVsbFYDKvVSmdnJ6qqUlRURH5+/nHrE040AoEATU1NzJ8/f8qrHAfb+iXsKHNzcw/rfunr68PpdE5bQlVVlf/5n/9haGiIBx98cEa3CgDee+89vvrVr3LLLbdw2WWX4XA40Ov1pKWl0dXVxSWXXMKDDz7I0qVLp3qph4MTpHoCH43E3OfGjRt57rnnMJlMrFu3josvvnhKMmETQp66uroZaYifgCzL7N27l4yMDEpLS5NmE4k+YSJabSYQrN/vZ+/evSxcuBCz2TzVy/k3hMPhpJJYluVkheBQa50JhPrzn/+cAwcO8PDDD89IMdv4cm8wGOSqq67iuuuuY+3atciynNwk/Otf/+Lb3/42t91226Qm60wwTpDq0eI/MZhXVVW6u7t54okneOaZZxBFkbVr17Ju3bpJz4RNXNvn87FgwYIZvTv/MDMEOPQsbH5+PpmZmdPyJZ+IBayrq5sRZfiEZaLNZvs3t6z+/v5pT6i//vWv2b17N4888siM3FTKssyXvvQl7rrrLtLT0/H5fKxfv57bb7+dVatWEYvF0Ol0yRJ+Y2MjK1eunOplHwlOkOrR4j89mFdVVQYGBpKZsJIkcfHFF7N27doJz4RVVZW2tjZUVZ2StJ6JRGL853BmNxOOQ+M9c482u3Qy4PF4aG1tnbGOT+PdshwOB6IoUltb+2/B9tMBqqpy7733snXrVh577LEZPb+8c+dOampqGBgYYO7cufz85z/H7/fzmc98hsrKSlpaWrjhhhu4//77k5vOGWQCcYJUJwL/6cG8qqoyMjKSzIT1+/1ceOGFrFu37pidjWRZprm5GbPZTGVl5Ux5sA6JhBXk7Nmzj7g3raoqXq83SQDHKsQ5ViSEYjM1Z3c8+vr6cDgcyTK8y+VKmnlkZ2dPeYlVVVUeeOABtmzZwpNPPjljZ8rHl3bvuecebrvtNl599VW0Wi0PP/wwDQ0NrFmzhocffpibb775uETVTQJOkOqx4EQw76Fhs9mSmbAOhyOZCVtbW3tEpJgw+M/Ly6O0tHQSVzz5CAaDNDU1TcioyXghTiIhJJELezxOMA6HIxmnN1OFYgkcqoc63swjIZpJCJ2m4oT40EMP8cwzz/D000/P2A1MglAjkUhyU/CnP/2JX/3qVzz22GOUlpby+uuvMzw8THl5ORdccAEwo06oCZwg1YnAiWDeD4fT6Uxmwg4ODiYzYT9uHjMSidDY2DhtA7mPBJOdNDPe0k8QhCTBTgbh2Ww2urq6WLx48YwuQcLhi5IO/n4PNyB8IvD3v/+dRx99lOeee25CS+z9/f1ce+21jIyMIIoiX/ziF/nGN77Bbbfdxn333Zd0wbrjjjuSBHe0SEwcjIyMcP3112M0Glm9ejWf+tSnePzxx/npT3/Kn/70J1atWvWBvzcDCRVOkOrE4T88mPew4PF42LRpE08++SSdnZ3JTNjFixd/4KXmdDrZv38/1dXVMz4qKpE0c7zmaQ8OBx+fC3ussFqt9PT0sHjx4hkpkhmPo1X5jp81jsViyfB1s9k84QTw+OOP8+CDD7J58+YJV1UPDw8zPDzMkiVL8Pl8LF26lKeffprHHnsMs9nMjTfeOKHX83q9fOlLX2LlypUYDAZ6e3vxer385je/4YEHHuCWW26hoaGB4uLimUik43GCVI8WJ4J5jw1+v5/nn3+eJ554gtbW1mQmbCQS4dvf/jZbtmyZFjOxxwKHw8GBAwdYtGjRlJRJo9FoclQnHA6Tk5NDfn7+URHAyMgI/f39LFq0aMYTam9vbzJr91gESQnLRKvVSiAQmNBRqKeffprf//73bN68+bi4ha1bt46vf/3rvP322xNOqpIk8Z3vfId3332Xbdu2AbB3715+85vfcOmll3L++efT1NREXV3dhF1zCnGCVI8WJ4J5Jw6hUIgtW7Zwzz330NjYyNq1a7niiitYuXLllItEjhaJU910SZqRJAmHw8Ho6CiBQCA5SpKenv6xBDA0NMTw8PCMN/mHOKEm/JUnUuF7KMvERCzgkSq1N2/ezN13383zzz9/XMbxenp6WL16Nc3Nzdx999089NBDWCwWli1bxl133XVUa5AkKXmvxGIxnnzySb797W/zrW99K0nYP/jBD5AkiTvuuCNZIj4cc5ppjhOkegLTA48++ij33HMPjz76KI2NjWzcuJEdO3awcuVK1q9fz6pVq2bMCWloaIihoSHq6+un5ZplWU4SgNfrJSMjIxm8fvALrb+/H5vNRn19/bQY4zkWTBahHoyEZWIiFvBIlNovvfQSP/3pT9m8efNxqdT4/X4+8YlP8L3vfY9LL72U0dFRcnJyEASB73//+wwPD/Pggw8e0WcGAgFSU1OJRqPs2rULnU7H0qVL2bJlC/fddx81NTXccsstbNiwgXXr1vH/27vTqKbONA7g/yCUWhCqVRRBiyJ6VETcRqoDdrQMjo0mgtbRqeDeTotaHZ1DYdzqqZU5FLAqtlPcOjhqCatEplYLLlgdsYIEd5AREAx7NpAsdz70cAdwQ0hyE/L8PiImTxDvP++b+z7Pxx9/bKBXxwkKVcK99PR07Nu3D4mJiW0+O1Kr1cjOzoZIJEJOTg4mTpzIzoQ1hdXf07Qcz/Dy8jKLEHreWdiysjK9bJOaAmMFanvtWyZaW1uzHZ3afySQlZWFLVu24OTJk0ZpHq9Wq8Hn8xEQEID169c/8eclJSXg8/mQSCQdfsywsDD4+PiAz+dj6tSpGDFiBM6dO4cPP/wQ06ZNg0wmw7JlyzBs2DCEhIRg2bJl+nxJpoBClXCvsbERPXr0eG5QtsyETUxMxNmzZzF27FgIhULMmDHDJI51MAyD4uJiKJVKeHp6mmUItT4L+/DhQ/B4PAwbNgxOTk5mve3LVaA+TWNjY5vZsJmZmQgKCoJMJkN4eDgyMjKMMhOYYRiEhISgT58+iI2NZb9eUVHBPn9MTAwuX76MY8eOdegx165dyzaE2bNnD27cuIG4uDjcvn0b3377LXr37o2IiAicPn0aR44cgZeXF9atW2eIl8clClViflpmwiYlJeH06dMYOXIkBAIBZzNhGYbBnTt3oNVqzb7jU+sxdG+++SaqqqpQXV3N+VnNzjKlQG1PLpfj8OHDSE5Oxp07d7Bo0SIEBwdj4sSJBq/1woUL8PX1bfNz2bFjB44ePYq8vDzweDy4ubnhm2++6VDIr1u3Dnv37kV1dTUcHBywb98+XLx4Ef/4xz/Qs2dP3Lt3DwEBARCLxRg6dCjOnz+PL7/8El988QXGjh1r0NdqZBSq3Y1Wq8XEiRPh4uKCjIyMbt2LGPh1+zI3NxcikQg//PAD3N3d2ZmwhjgT+rTnv3nzJl555ZUud4/iWsvwBI1G88Sbg9ZnNa2srNijOqawS/AsJSUlaGhoMMlAbZGbm4s1a9bg6NGjuH37NlJSUnDt2jXMmzcPmzdv5rq8DvnLX/6CoqIi2NnZoba2FgkJCSgvL0dcXBzmz5+P3/72t7C1tcW7776LyMhIeHp64vHjx6itrTXKqtzIKFS7m+joaOTm5kImkyEjIwN//etfLaIXMfBrwLXc5JSZmQkXFxd2Jqwh5nvqdDoUFBTAwcEBQ4YM0fvjGxPDMOx56hfNdW19Flan07EBa0r9f80hUPPy8vDhhx8iJSUF7u7u7NfVajVKSkrg4eHBYXUdU1BQgPj4eOzatQsAEBwcjIqKChw/fhzHjx/HtWvXUFVVBY1GA0dHRyQkJHBcscFRqHYnZWVlCAkJQUREBKKjo5GRkYERI0ZYZDMKhmFQWFjIzoTt06cPhEIh+Hw+3njjjS4/vkajwfXr19GvXz+zb6HIMAxu3rwJGxubl15tNzc3swHb3Nxs0GYIHWUOgSqRSLBixQqIRCIMHz6c63I6TavVgsfjtfk5L1++HOXl5UhISIBMJkNBQQGampqwYMECAB2b6WzGKFS7k3nz5uHTTz+FXC5HVFQUMjIyqBcx/v+Zp0gkQkZGBl577TV2JqyTk9NLX/zVajXy8vLg6upq9ttXOp0ON27cQM+ePbs8sKB1MwSVStVmrJqxAtYcAvXmzZtYunQpjh07hlGjRnFdjt60bpj/0UcfobCwEEeOHIGrqyv7Pd08UAGOQrVb/0S5kpGRAScnJ0yYMIHrUkwOj8fDiBEjEBERwd480dTUhPfffx+zZs3Cvn378PDhQ7zgzR6AX9vU/fLLL3Bzc+sWgSqRSGBnZwd3d/cuB5+1tTUGDBgALy8vTJo0CY6OjigtLcWlS5dw69Yt1NbWQqfT6an6J5lDoN65cwdLly7FkSNHzDZQn/X/pEePHtBqtQCAuLg4uLm5Yc+ePW2+x1T/XcwdrVQN4NNPP8U///lPWFtbo6mpCTKZDIGBgbhy5YpFbv92xNNmwvL5fMydOxeDBg16ImQaGxuRn5+vl0kzXNNqtSgoKEDv3r3x5ptvGvS5dDod6urqIJVKUV9fz3Yb0ufc0pKSEshkMpM+znT//n0sXLgQhw8fxrhx47gup1Nar0Zba70CtYDV6PPQ9m93lJ2dzW7/bty4kXoRd0D7mbBKpZKdCevu7o6rV69i//79iIyMhIODA9fldolWq0V+fj6cnJzabM0ZQ+tuQzU1NbCzs0P//v27NLfUHAL1wYMHWLBgAeLj4zFp0iSuy+my0NBQtlPU6tWr8eqrr7YJ09YTZsx02kxnUah2R61DlXoRd45UKmVnwpaVlUGhUCAyMhKzZ8826wuERqNBfn4+nJ2dMXDgQE5rYRgGCoUCjx49QnV1NV599VX2LGxH2zuaQ6CWl5dj/vz52Lt3L6ZOncp1OZ3SOjBTUlKwf/9+LFmyBFlZWVCr1YiOjoa9vf0zV7IWhEKVkOfJysrCJ598gvfffx9nz57Fw4cPERAQgLlz52LUqFEmeyF/mpYbrAYNGoQBAwZwXc4TWg9e79GjBzsXtmWodXv379+HXC436UCtrKxEUFAQYmNjMW3aNK7L6ZTWK83s7GxcunQJb7/9Nnx8fHDnzh3ExcVBoVAgKirKIEfXzAyFKiHPcuLECXz++edITU1lQ6ihoQEnTpxAcnIyiouL4e/vD4FAAG9vb5O9sAO/Hn3Jy8uDm5ubUfrKdlVLOz+pVAqGYZ44C2sOgSqVShEUFITIyEi88847XJfTKa0DddOmTcjKysKDBw8QFBSEHTt2oGfPnigqKsKOHTvg5uaGTZs2cVwx5yhUCXkWsViMKVOmPPOmJLlczs6EvXXrFqZPnw6BQIBJkyaZ1IW+ubkZ165dg7u7u1nOqG1uboZUKmUHg1tbW4PH45n05Jzq6moEBQVh+/btmDlzJtfldNmhQ4dw+vRpJCQk4Pr169i4cSNmzpyJVatWwc7ODuXl5XB2djap33uOUKgSog8tM2FFIhHy8vLg5+cHoVCIt956i9MLf1NTE/Lz8+Hh4dEtPku/d+8eampqYGtri6amJk7Owr5IXV0dAgMD8be//Q2zZ8/mupxOab1CVSqVWLt2LTIzM3H16lUMGDAAV69eRXh4OHx8fBAWFoaePXs+8fcsFIUqMSw3Nzf06tULPXr0gLW1NXJzc7t9P+LHjx/jxx9/hEgkQm5uLt566y3MnTsXU6dONep81e50BAh4cstXq9WipqYGUqkUcrkcvXv3ZufCcnVhb2hoQFBQEDZs2IDAwEBOatCn48ePY8GCBVAoFFi3bh1UKhUiIyPh6uqKq1ev4uuvv0ZcXJxJzg3mCIUqMSw3Nzfk5ua22Xa0pH7EarUaWVlZEIlEuHjxIiZNmgShUIhp06YZdNqLSqXC9evXMXLkSDg6OhrseYzlRZ+htpyFffToERoaGuDo6AgnJyf06dPHaFuScrkc8+bNQ2hoKNuSTx9KS0sRHByMyspKWFlZYdWqVVi7dq1B35y2rDgnT54MGxsbXLhwAQ0NDfj8889RVlaG7du3w93dnf0+WqGyKFSJYT0tVC21H7FGo8H58+eRmJiIc+fOwdvbG0KhENOnT9frtBelUonr16/D09PTKJN6DO1lb0piGIYdvF5bWwt7e3s4OTmhb9++BtuKVyqVeO+997Bs2TIsXrxYr49dUVGBiooKjB8/HnK5HBMmTEBqaioOHTpksDencrmc/d3h8/moq6tDTk4OFAoFwsLCAABfffUVAOqS1A6FKjGsIUOGsNtxH3zwAVatWkX9iPFrA4aLFy8iKSkJZ86cwahRo9iZsF2Z9iKXyyGRSDBmzBjY29vrsWJu3L9/HwqFAqNHj+7UxZthGMjlckil0k6fhX2RxsZGvPfee1i0aBGWL1+ul8d8HoFAgNDQUISGhurtzWlpaSkYhsHgwYNx5swZXLt2DXPnzmWn58yaNQt1dXX4+eefoVKpoNPpusXvlwFQqBLDevjwIQYOHAipVAp/f3/s3r0bc+bMsfhQbU2n0+HKlSsQiUQ4deoU3N3dIRAIMHPmzJdaacpkMhQWFsLLy4uTYe36VlxcDKVS2elAfRqFQoGqqipUVVXB2tqaDdhnnYV9kaamJixatAhCoRAffPCBwbdAS0pK4OfnB4lEgsGDB+vl/5FMJsMXX3wBAFizZg3KysoQFRWFqVOn4g9/+AM8PDxQWVkJDw8PBAQEQCQSAaCbkp6BQpUYz9atW2Fvb49vv/3WIrd/O6JlJmxiYiIyMzPh6uoKgUCAWbNmPfdgfX19PW7duoWxY8eyd2KaM0MEanuNjY1sswmGYdhmEx39+T1+/BiLFy+Gv78/1qxZY/CAUSgUmDZtGiIiIhAYGKjXHZ9///vfyM7Ohq2tLT755BOUlZXh73//O8aNGwc+n4/CwkLcvHkTISEhcHFx0dMr6pYoVInhKJVK6HQ69OrVC0qlEv7+/ti8eTPOnDlD/Yg7gGEYSCQSdiZs3759IRQK8e6777aZCSuRSKBQKODt7a3Xz2a5YoxAbe/x48dsswmNRtNmLuzTqNVqLFmyBFOmTMGGDRsMHqhqtRp8Ph8BAQFYv349gK7fm1BaWoqKigr85je/AQDk5+fju+++g52dHUJDQ1FTU4O4uDiUl5ejsLAQJ0+ehLu7O7UifD4KVWI4xcXFmDt3LoBfb9JZtGgRIiIiqB9xJzAMg9u3b7MzYXv16oU5c+bA1tYW0dHROH36dLf4GXIRqO2p1Wp2LmxjY2ObgLWysoJGo8Hy5cvh7e2N8PBwgwcqwzAICQlBnz59EBsby369K8MyZDIZXFxcoFQqER4eDltbWyxduhQ3btxAXl4eFAoF1qxZAwcHBzQ0NECj0cDZ2dnSJ9B0BIUqIeaGYRgUFxdj+/btEIvF8PLyYifqDBgwwGw/52oJVE9PT5N5DVqtFtXV1bh79y4++ugj+Pj4QCaTYezYsfjss8+MUueFCxfg6+vbZk7sjh07MHny5C69OU1MTMSSJUuwePFijB8/Ht999x1Gjx6N0tJSvPLKKxgwYADCwsLg5uYGgD5D7SAKVULMUXJyMqKjo5GWlgaFQoHk5GSkpKRAp9OxM2FdXV3N5iJoioHankKhwMcff4yioiJoNBpMmTIFgYGBePvttw165tiQ0tLSsHjxYmRlZWHMmDG4efMmdu3ahbt37yI/Px+FhYUYNGgQ12WaEwpVQsxNQUEBVq9ejfT09DazXRmGQUVFBTsTVqVSgc/nQyAQYOjQoSYbVkVFRVCpVCYdqDqdDuvXr4e9vT2ioqLAMAwuXLiAlJQUyGQyHDhwgOsSOy0pKQlLly7FwYMHERQUBLVaDRsbG5SVlcHV1ZVWqC+HQpUQc8MwDJqbm194DEQqlSIlJQVJSUmoq6vDrFmzIBQKMXz4cJO5SBYVFaGxsRGjR482mZra0+l0bRoedMfPFMViMYKDgxETE4Pg4GAAtN3bSRSqxDLV19djxYoVkEgk4PF4OHDgAEaMGNFtexLX1NQgLS0NSUlJqKysZGfCjhw5krOQMJdA3bx5MxQKBb7++utuGagtMjMzMX/+fNy7dw/9+/c32X8TE0ehSixTSEgIfH19sWLFCjQ3N0OlUmHHjh0W0ZO4vr6enQl7//59+Pv7QygUYuzYsUYLDXMIVIZhsH37dlRWVmL//v0WcYykrq6u27yR5AiFKrE8LXdvFhcXt7mgW2JPYrlcDrFYjKSkJNy+fRvTp0+HUCjExIkTDRaw5hKokZGRKCoqwuHDh2Ftbc11SUZBW75dRqFKLE9eXh5WrVqFUaNGIT8/HxMmTMCuXbvg4uJi0e0TGxsbkZmZiaSkJFy/fp2dCevj46O3VZq5BGpsbCzy8vLwr3/9i8aakZdBoUosT25uLnx8fJCTk4PJkydj7dq1cHBwwO7duy06VFtrmQmbmJiI3NxcTJ06lZ0J29lVm7kEalxcHHJycvD999+b7VEZwhkKVWJ5Kisr4ePjg5KSEgDA+fPnsXPnTty7d8/itn87orm5uc1M2MmTJ0MoFMLPz6/DoWMugRofH49Tp04hOTm50032iUWjUCWWydfXF/Hx8RgxYgS2bt0KpVIJANST+AU0Gg3OnTuHxMREnD9/HuPGjYNQKMTvfve7p/Ydbun+ZOqBCgCHDh1CWloa0tLSukUPZcIJClVimfLy8tg7f4cOHYqDBw9Cp9NRT+KXoNVqkZOTg6SkJPz0008YPXo0BAIB/P398dprr0Gn02H37t3w8/ODt7e3SQdqQkICjh8/jhMnTnRpni2xeBSqhJCu0+l0+M9//gORSIQff/wR7u7usLKyQlNTE44ePWrSx1G+//57HDx4EGKxmAZvk67iJFS77+lponetL3InT56Eh4cHHjx4wGFF5GmsrKzg4+ODqKgo/PLLL3jjjTdQVFSE8vJy/OlPf8LRo0fR0NDAdZlPSE1Nxf79+5Genk6BSsyWZRz4Inp15swZrF69GqdOncLgwYO5Loc8A8Mw2LZtG7RaLa5evQoejweJRILExETMnj0b/fr1g0AgAJ/P53xrXSwWY8+ePRCLxXB0dOS0FkK6glaq5KWcP38eK1euhFgshru7O9flkOe4d+8eqqqqEB8fDysrK/B4PIwZMwafffYZrly5gujoaFRVVSEoKAhCoRAHDhyAVCrFCz4S0rtTp04hKioKJ06c0HsHoWXLlsHJyQmenp7s17Zu3QoXFxd4e3vD29sbJ0+e1OtzEstGn6mSDrOxsUGvXr2QnZ0NLy8vrsshesIwDIqKiiASiZCeng5bW1vMnj3bKDNhf/rpJ2zbtg1isRhOTk56f/xz587B3t4ewcHBkEgkAH4NVXt7e2zYsEHvz0dMCn2mSkybjY0NpkyZgv3793NdilHcvn2bXc14e3vDwcEBsbGxqK2thb+/Pzw8PODv72/2TSl4PB6GDRuGsLAw5OTk4ODBgwCAJUuWYObMmdi9ezdKS0v1voI9d+4ctmzZghMnThgkUAHAz8+P861tYllopUo6zN7eHlKpFO+88w74fD7Cw8O5LslotFotXFxccPnyZezdu9cimv23zIRNSkpCcnIympqa2JmwQ4YM6dIK9ueff8aGDRuQkZEBFxcXPVb9pJKSEvD5/DYr1UOHDsHBwQETJ07El19+SY3ruyc6UkNMm729PRQKBWpra+Hr64v169dj+fLlXJdlFKdOncK2bduQk5Njkc3+GYZhZ8ImJyejvr4es2bNgkAgeOmZsFeuXMHatWuRnp5ulBvd2ofqo0eP0LdvX/B4PGzatAkVFRVmPdicPBOFKjFtLaEKAKWlpfDz80NsbCwEAgHHlRnesmXLMH78eISGhuL111+3+L7ENTU1SE1NRXJyMiorKzFz5kx2JuzzAvbatWv485//jNTUVAwdOtQotbYP1Y7+GTF7FKqEmKLm5mYMHDgQhYWF6N+/P4VqO/X19UhPT0dycjL++9//sjNhvby82oysKygowMqVKyESiTB8+HCj1dc+OCsqKuDs7AwAiImJweXLl3Hs2DGj1UOMhpNQpXOqhLxAZmYmxo8fj/79+wMA+vfvz16YKyoqDHaTjbl4/fXXERwcjODgYMhkMojFYsTExODOnTvsTFg7OzusXLkSx44dM2qgLly4ENnZ2aiuroarqyu2bduG7Oxs5OXlgcfjwc3NDd98843R6iHdH61UCXmBP/7xjwgICMDSpUsBABs3bqRm/x2gUqmQmZkJkUiEH374AWfPnsWYMWO4LotYDtr+JcTUqFQqDBo0CMXFxWynn5qaGmr2/5LUajUNGCfGRqFKCCGE6Ak1fyCEEELMGYUqIYQQoicUqoQQQoieUKgSQgghekKhSogZiomJwejRo+Hp6YmFCxeiqamp2zX6J8QcUagSYmbKy8vx1VdfITc3FxKJBFqtFseOHcPOnTsxY8YM3L17FzNmzMDOnTu5LpUQi0OhSogZ0mg0aGxshEajgUqlwsCBA5GWloaQkBAAQEhICFJTU7ktkhALRKFKiJlxcXHBhg0bMHjwYDg7O8PR0RG///3v8ejRI7anrbOzM6RSKceVEmJ5KFQJMTN1dXVIS0vD/fv38fDhQyiVSiQkJHBdFiEEFKqEmJ3Tp09jyJAh6NevH2xsbBAYGIiLFy+yjf4BUKN/QjhCoUqImRk8eDAuXboElUoFhmFw5swZjBw5EnPmzMHhw4cBAIcPH7aIObeEmBrq/UuIGdqyZQuOHz8Oa2trjBs3DvHx8VAoFNTon5D/o4b6hBBCiJ5QQ31CCCHEnFGoEkIIIXpCoUoIIYToCYUqIYQQoicUqoQQQoieUKgSQgghekKhSgghhOgJhSohhBCiJxSqhBBCiJ5QqBJCCCF6QqFKCCGE6AmFKiGEEKIn1i/4c04aEhNCCCHmiFaqhBBCiJ5QqBJCCCF6QqFKCCGE6AmFKiGEEKInFKqEEEKInlCoEkIIIXryP8hM8pvaZg3FAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "z = coherence_vals\n",
    "x = KS\n",
    "y = ntopics_arr\n",
    "#x_grid = numpy.linspace(0, 132651, 1000*len(x))\n",
    "x_grid = numpy.linspace(kmin, kmax, 2000)\n",
    "#y_grid = numpy.linspace(0, 132651, 1000*len(y))\n",
    "y_grid = numpy.linspace(ntopics_min, ntopics_max, 2000)\n",
    "B1, B2 = numpy.meshgrid(x_grid, y_grid, indexing='xy')\n",
    "Z = numpy.zeros((x.size, z.size))\n",
    "\n",
    "\n",
    "spline = sp.interpolate.Rbf(x,y,z,function='thin_plate',smooth=5, episilon=5)\n",
    "\n",
    "Z = spline(B1,B2)\n",
    "fig = plt.figure(figsize=(10,6))\n",
    "ax = axes3d.Axes3D(fig)\n",
    "ax.plot_wireframe(B1, B2, Z)\n",
    "ax.plot_surface(B1, B2, Z,alpha=0.2)\n",
    "ax.scatter3D(x,y,z, c='r')\n",
    "ax.set_xlabel(\"K\")\n",
    "ax.set_ylabel(\"Num Topics\")\n",
    "ax.set_zlabel(\"Coherence\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "point = unravel_index(Z.argmin(), Z.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "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": [
    "plt.figure()\n",
    "y, z = zip(*sorted(zip(y, z)))\n",
    "plt.plot(y,z)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,5))\n",
    "plt.contourf(x_grid, y_grid, Z)\n",
    "plt.xlabel(\"K\")\n",
    "plt.ylabel(\"Number of Topics\")\n",
    "plt.colorbar();\n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}