40_perf_summary.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "02c142ce-e8e1-4bd5-bc82-6c2b1649783c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Modify the file 00_setup.py to define input/output file paths on your system\n",
    "# The information in 00_setup.py will be used across notebooks\n",
    "from importlib.machinery import SourceFileLoader\n",
    "setup = SourceFileLoader(\"setup\", \"./00_setup.py\").load_module()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73fba715-9dc4-42ea-b4d3-3e9fadf4adbe",
   "metadata": {},
   "source": [
    "## Performance Summaries\n",
    "Get metrics from different runs and combine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ebd8e9d9-30c0-4579-9bb0-29c64f97b404",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import importlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7463ea3b-0e94-47d2-9f96-64c697bf56b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a184dd8d-ef74-4aac-9a56-1016cd5c8884",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as ticker"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "322cf9b1-5a3b-4673-b58b-6ffabbc754ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sba_nn.sba_nn import sbnn_plot\n",
    "sbnn_plot.plot_defaults(style='seaborn-colorblind')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c427500-5775-4afc-8c99-8290885fa39c",
   "metadata": {},
   "source": [
    "## File Dictionaries"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ead8da18-941d-4955-b9df-c4f9cb326fcc",
   "metadata": {},
   "source": [
    "##### Initial embedding tests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "10bae3a6-1f84-45da-973e-5fc733cfdf70",
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_1_tests = {\n",
    "    'base': Path(setup.temp_path).joinpath('02_'),\n",
    "    'naics':  Path(setup.temp_path).joinpath('03_'),\n",
    "    'naics+hier':  Path(setup.temp_path).joinpath('04_'),\n",
    "    'naics data rand':  Path(setup.temp_path).joinpath('05_'),\n",
    "    'naics+hier data rand':  Path(setup.temp_path).joinpath('06_'),\n",
    "    'naics cust test': Path(setup.temp_path).joinpath('10_'),\n",
    "    'naics data gen': Path(setup.temp_path).joinpath('11_'),\n",
    "    'naics+hier data gen': Path(setup.temp_path).joinpath('12_')\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d3b691b-9403-429b-8196-82d24713c869",
   "metadata": {},
   "source": [
    "##### Loops over injection rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "34521235-afbf-4602-8fa0-e75dd391ccf4",
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_2_rate_loop = {\n",
    "    'irloop naics+hier data gen': Path(setup.temp_path).joinpath('13_'),\n",
    "    'irloop naics data gen':  Path(setup.temp_path).joinpath('14_'),  \n",
    "    'ds irloop naics+hier data gen': Path(setup.temp_path).joinpath('26_'),\n",
    "    'ds irloop naics data gen':  Path(setup.temp_path).joinpath('27_'),  \n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b7dd98f-262c-4130-a3a6-29c84bc8cf62",
   "metadata": {},
   "source": [
    "##### Downsample tests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "87196dc1-d5cc-4712-92aa-bab2f36accd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_3_ds_tests = {\n",
    "    'ds naics': Path(setup.temp_path).joinpath('20_'),\n",
    "    'ds naics data rand':  Path(setup.temp_path).joinpath('21_'),\n",
    "    'ds naics data gen':  Path(setup.temp_path).joinpath('22_'),\n",
    "    'ds naics+hier':  Path(setup.temp_path).joinpath('23_'),\n",
    "    'ds naics+hier data rand':  Path(setup.temp_path).joinpath('24_'),\n",
    "    'ds naics+hier data gen': Path(setup.temp_path).joinpath('25_')\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "921ff3e7-bf77-4e39-860c-963f4c7d9747",
   "metadata": {},
   "source": [
    "##### Loops over downsample fractions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "904f2d1f-c104-4b2a-b02b-6ad1cb81f12e",
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_4_ds_loop = {\n",
    "    'dsloop naics': Path(setup.temp_path).joinpath('30_'),\n",
    "    'dsloop naics+hier':  Path(setup.temp_path).joinpath('31_'),\n",
    "    'dsloop naics data rand':  Path(setup.temp_path).joinpath('32_'),\n",
    "    'dsloop naics+hier data rand':  Path(setup.temp_path).joinpath('33_'),\n",
    "    'dsloop naics data gen':  Path(setup.temp_path).joinpath('34_'),\n",
    "    'dsloop naics+hier data gen': Path(setup.temp_path).joinpath('35_')\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b703f0ed-4f7b-4e87-b66e-aac95b6fd5d4",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea74a6ac-9a4b-4d2b-9a48-4e58c375682c",
   "metadata": {},
   "source": [
    "##### Test metrics (regular, downsampled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "80b66444-1629-4ec8-b593-149d389984cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_dict_list = [dict_1_tests, dict_3_ds_tests]\n",
    "test_dict_all = {k: v for d in test_dict_list for k, v in d.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0373bb3f-6c55-4794-91af-ef16241c615e",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_dict_all_metrics= {k:v.parent / (v.name + 'REPORT_metrics.csv') for k, v in test_dict_all.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d7b014b1-1e28-47a8-bc08-e61f945b4630",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_set_metrics = pd.concat([pd.read_csv(v) for k, v in test_dict_all_metrics.items()],\n",
    "                        keys = test_dict_all_metrics.keys()) \\\n",
    "    .reset_index(level=0) \\\n",
    "    .drop(columns=['Unnamed: 0'], errors='ignore') \\\n",
    "    .rename(columns={'level_0':'model'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "76096108-bdef-4602-942f-887b58d2394e",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_set_metrics.to_csv(Path(setup.temp_path).joinpath('40_REPORT_metrics_nonloop.csv'), index=False)\n",
    "test_set_metrics[(test_set_metrics['dset'] == 'test') &\n",
    "                (~(test_set_metrics['dset_naics_holdout'].isna()))] \\\n",
    "    .to_csv(Path(setup.temp_path).joinpath('40_REPORT_metrics_nonloop_test.csv'), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e1b32349-0892-4be6-8a0a-b9809d601cdd",
   "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>model</th>\n",
       "      <th>dset</th>\n",
       "      <th>accuracy_score</th>\n",
       "      <th>f1_score</th>\n",
       "      <th>precision_score</th>\n",
       "      <th>recall_score</th>\n",
       "      <th>average_precision_score</th>\n",
       "      <th>roc_auc_score</th>\n",
       "      <th>dset_naics_holdout</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>naics+hier data gen</td>\n",
       "      <td>test</td>\n",
       "      <td>0.667965</td>\n",
       "      <td>0.454397</td>\n",
       "      <td>0.337824</td>\n",
       "      <td>0.693812</td>\n",
       "      <td>0.385357</td>\n",
       "      <td>0.739736</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>naics+hier data rand</td>\n",
       "      <td>test</td>\n",
       "      <td>0.666348</td>\n",
       "      <td>0.456108</td>\n",
       "      <td>0.337785</td>\n",
       "      <td>0.702016</td>\n",
       "      <td>0.381281</td>\n",
       "      <td>0.738753</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>naics+hier</td>\n",
       "      <td>test</td>\n",
       "      <td>0.548938</td>\n",
       "      <td>0.430542</td>\n",
       "      <td>0.287638</td>\n",
       "      <td>0.855634</td>\n",
       "      <td>0.363896</td>\n",
       "      <td>0.727930</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>base</td>\n",
       "      <td>test</td>\n",
       "      <td>0.621576</td>\n",
       "      <td>0.437015</td>\n",
       "      <td>0.310590</td>\n",
       "      <td>0.737017</td>\n",
       "      <td>0.353046</td>\n",
       "      <td>0.721018</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>naics data rand</td>\n",
       "      <td>test</td>\n",
       "      <td>0.659355</td>\n",
       "      <td>0.433432</td>\n",
       "      <td>0.324159</td>\n",
       "      <td>0.653836</td>\n",
       "      <td>0.352953</td>\n",
       "      <td>0.718785</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>naics data gen</td>\n",
       "      <td>test</td>\n",
       "      <td>0.607956</td>\n",
       "      <td>0.437327</td>\n",
       "      <td>0.306259</td>\n",
       "      <td>0.764511</td>\n",
       "      <td>0.352904</td>\n",
       "      <td>0.720349</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>naics cust test</td>\n",
       "      <td>test</td>\n",
       "      <td>0.537649</td>\n",
       "      <td>0.424242</td>\n",
       "      <td>0.282138</td>\n",
       "      <td>0.854761</td>\n",
       "      <td>0.349626</td>\n",
       "      <td>0.718901</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>naics</td>\n",
       "      <td>test</td>\n",
       "      <td>0.540746</td>\n",
       "      <td>0.426159</td>\n",
       "      <td>0.283730</td>\n",
       "      <td>0.855721</td>\n",
       "      <td>0.346411</td>\n",
       "      <td>0.718357</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  model  dset  accuracy_score  f1_score  precision_score  \\\n",
       "4   naics+hier data gen  test        0.667965  0.454397         0.337824   \n",
       "4  naics+hier data rand  test        0.666348  0.456108         0.337785   \n",
       "4            naics+hier  test        0.548938  0.430542         0.287638   \n",
       "4                  base  test        0.621576  0.437015         0.310590   \n",
       "4       naics data rand  test        0.659355  0.433432         0.324159   \n",
       "4        naics data gen  test        0.607956  0.437327         0.306259   \n",
       "4       naics cust test  test        0.537649  0.424242         0.282138   \n",
       "4                 naics  test        0.540746  0.426159         0.283730   \n",
       "\n",
       "   recall_score  average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "4      0.693812                 0.385357       0.739736                 1.0  \n",
       "4      0.702016                 0.381281       0.738753                 1.0  \n",
       "4      0.855634                 0.363896       0.727930                 1.0  \n",
       "4      0.737017                 0.353046       0.721018                 1.0  \n",
       "4      0.653836                 0.352953       0.718785                 1.0  \n",
       "4      0.764511                 0.352904       0.720349                 1.0  \n",
       "4      0.854761                 0.349626       0.718901                 1.0  \n",
       "4      0.855721                 0.346411       0.718357                 1.0  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_set_metrics.to_csv(Path(setup.temp_path).joinpath('40_REPORT_metrics_nonloop.csv'), index=False)\n",
    "test_set_metrics[(test_set_metrics['dset'] == 'test') &\n",
    "                (test_set_metrics['dset_naics_holdout'] == 1) &\n",
    "                (~test_set_metrics['model'].str.startswith('ds '))] \\\n",
    "    .sort_values('average_precision_score', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "14840b3e-e1db-4937-a535-ab3121c5d1b8",
   "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>model</th>\n",
       "      <th>dset</th>\n",
       "      <th>accuracy_score</th>\n",
       "      <th>f1_score</th>\n",
       "      <th>precision_score</th>\n",
       "      <th>recall_score</th>\n",
       "      <th>average_precision_score</th>\n",
       "      <th>roc_auc_score</th>\n",
       "      <th>dset_naics_holdout</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>naics data gen</td>\n",
       "      <td>test</td>\n",
       "      <td>0.645615</td>\n",
       "      <td>0.447497</td>\n",
       "      <td>0.327483</td>\n",
       "      <td>0.706361</td>\n",
       "      <td>0.380296</td>\n",
       "      <td>0.727207</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>naics+hier</td>\n",
       "      <td>test</td>\n",
       "      <td>0.663918</td>\n",
       "      <td>0.446207</td>\n",
       "      <td>0.335390</td>\n",
       "      <td>0.666392</td>\n",
       "      <td>0.380095</td>\n",
       "      <td>0.726571</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>naics+hier data gen</td>\n",
       "      <td>test</td>\n",
       "      <td>0.657880</td>\n",
       "      <td>0.448441</td>\n",
       "      <td>0.333442</td>\n",
       "      <td>0.684522</td>\n",
       "      <td>0.379818</td>\n",
       "      <td>0.726954</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>naics+hier data rand</td>\n",
       "      <td>test</td>\n",
       "      <td>0.667152</td>\n",
       "      <td>0.446197</td>\n",
       "      <td>0.337033</td>\n",
       "      <td>0.659954</td>\n",
       "      <td>0.379570</td>\n",
       "      <td>0.726183</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>naics</td>\n",
       "      <td>test</td>\n",
       "      <td>0.658246</td>\n",
       "      <td>0.447250</td>\n",
       "      <td>0.333081</td>\n",
       "      <td>0.680505</td>\n",
       "      <td>0.379322</td>\n",
       "      <td>0.726497</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>naics data rand</td>\n",
       "      <td>test</td>\n",
       "      <td>0.665686</td>\n",
       "      <td>0.445767</td>\n",
       "      <td>0.336089</td>\n",
       "      <td>0.661705</td>\n",
       "      <td>0.378818</td>\n",
       "      <td>0.726329</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>naics cust test</td>\n",
       "      <td>test</td>\n",
       "      <td>0.652406</td>\n",
       "      <td>0.448357</td>\n",
       "      <td>0.330866</td>\n",
       "      <td>0.695236</td>\n",
       "      <td>0.378482</td>\n",
       "      <td>0.726526</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>base</td>\n",
       "      <td>test</td>\n",
       "      <td>0.600835</td>\n",
       "      <td>0.425687</td>\n",
       "      <td>0.300766</td>\n",
       "      <td>0.728097</td>\n",
       "      <td>0.338248</td>\n",
       "      <td>0.697015</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  model  dset  accuracy_score  f1_score  precision_score  \\\n",
       "3        naics data gen  test        0.645615  0.447497         0.327483   \n",
       "3            naics+hier  test        0.663918  0.446207         0.335390   \n",
       "3   naics+hier data gen  test        0.657880  0.448441         0.333442   \n",
       "3  naics+hier data rand  test        0.667152  0.446197         0.337033   \n",
       "3                 naics  test        0.658246  0.447250         0.333081   \n",
       "3       naics data rand  test        0.665686  0.445767         0.336089   \n",
       "3       naics cust test  test        0.652406  0.448357         0.330866   \n",
       "3                  base  test        0.600835  0.425687         0.300766   \n",
       "\n",
       "   recall_score  average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "3      0.706361                 0.380296       0.727207                 0.0  \n",
       "3      0.666392                 0.380095       0.726571                 0.0  \n",
       "3      0.684522                 0.379818       0.726954                 0.0  \n",
       "3      0.659954                 0.379570       0.726183                 0.0  \n",
       "3      0.680505                 0.379322       0.726497                 0.0  \n",
       "3      0.661705                 0.378818       0.726329                 0.0  \n",
       "3      0.695236                 0.378482       0.726526                 0.0  \n",
       "3      0.728097                 0.338248       0.697015                 0.0  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_set_metrics.to_csv(Path(setup.temp_path).joinpath('40_REPORT_metrics_nonloop.csv'), index=False)\n",
    "test_set_metrics[(test_set_metrics['dset'] == 'test') &\n",
    "                (test_set_metrics['dset_naics_holdout'] == 0) &\n",
    "                (~test_set_metrics['model'].str.startswith('ds '))] \\\n",
    "    .sort_values('average_precision_score', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "26fb4cce-a3ca-42f8-8b0c-94d5099ff098",
   "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>model</th>\n",
       "      <th>dset</th>\n",
       "      <th>accuracy_score</th>\n",
       "      <th>f1_score</th>\n",
       "      <th>precision_score</th>\n",
       "      <th>recall_score</th>\n",
       "      <th>average_precision_score</th>\n",
       "      <th>roc_auc_score</th>\n",
       "      <th>dset_naics_holdout</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ds naics+hier data gen</td>\n",
       "      <td>test</td>\n",
       "      <td>0.656572</td>\n",
       "      <td>0.448060</td>\n",
       "      <td>0.329591</td>\n",
       "      <td>0.699485</td>\n",
       "      <td>0.365265</td>\n",
       "      <td>0.727972</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ds naics+hier data rand</td>\n",
       "      <td>test</td>\n",
       "      <td>0.645440</td>\n",
       "      <td>0.449289</td>\n",
       "      <td>0.325351</td>\n",
       "      <td>0.725757</td>\n",
       "      <td>0.363327</td>\n",
       "      <td>0.726720</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ds naics+hier</td>\n",
       "      <td>test</td>\n",
       "      <td>0.596180</td>\n",
       "      <td>0.442513</td>\n",
       "      <td>0.305231</td>\n",
       "      <td>0.804224</td>\n",
       "      <td>0.348779</td>\n",
       "      <td>0.719503</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ds naics data gen</td>\n",
       "      <td>test</td>\n",
       "      <td>0.615227</td>\n",
       "      <td>0.430971</td>\n",
       "      <td>0.305529</td>\n",
       "      <td>0.731169</td>\n",
       "      <td>0.339131</td>\n",
       "      <td>0.710802</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ds naics data rand</td>\n",
       "      <td>test</td>\n",
       "      <td>0.587188</td>\n",
       "      <td>0.430768</td>\n",
       "      <td>0.296997</td>\n",
       "      <td>0.783800</td>\n",
       "      <td>0.332818</td>\n",
       "      <td>0.706708</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ds naics</td>\n",
       "      <td>test</td>\n",
       "      <td>0.525769</td>\n",
       "      <td>0.420679</td>\n",
       "      <td>0.278023</td>\n",
       "      <td>0.864013</td>\n",
       "      <td>0.328941</td>\n",
       "      <td>0.706302</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     model  dset  accuracy_score  f1_score  precision_score  \\\n",
       "4   ds naics+hier data gen  test        0.656572  0.448060         0.329591   \n",
       "4  ds naics+hier data rand  test        0.645440  0.449289         0.325351   \n",
       "4            ds naics+hier  test        0.596180  0.442513         0.305231   \n",
       "4        ds naics data gen  test        0.615227  0.430971         0.305529   \n",
       "4       ds naics data rand  test        0.587188  0.430768         0.296997   \n",
       "4                 ds naics  test        0.525769  0.420679         0.278023   \n",
       "\n",
       "   recall_score  average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "4      0.699485                 0.365265       0.727972                 1.0  \n",
       "4      0.725757                 0.363327       0.726720                 1.0  \n",
       "4      0.804224                 0.348779       0.719503                 1.0  \n",
       "4      0.731169                 0.339131       0.710802                 1.0  \n",
       "4      0.783800                 0.332818       0.706708                 1.0  \n",
       "4      0.864013                 0.328941       0.706302                 1.0  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_set_metrics.to_csv(Path(setup.temp_path).joinpath('40_REPORT_metrics_nonloop.csv'), index=False)\n",
    "test_set_metrics[(test_set_metrics['dset'] == 'test') &\n",
    "                (test_set_metrics['dset_naics_holdout'] == 1) &\n",
    "                (test_set_metrics['model'].str.startswith('ds '))] \\\n",
    "    .sort_values('average_precision_score', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aea845cc-2496-4a5f-bb6e-1938983e7a78",
   "metadata": {},
   "source": [
    "## Downsample Loop\n",
    "Plot.   2 rows - NAICS only, NAICS+hier.  2 columns - test and holdout.   3 lines per plot (unmodified data, fixed rand, generator rand).  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "98621c37-224c-4a05-b6b6-3cb89b28ee94",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics_dict = {k:v.parent / (v.name + 'REPORT_loop_metrics.csv') for k, v in dict_4_ds_loop.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "54ef5722-8523-46e8-9ffe-4c86dc7ed480",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics = pd.concat([pd.read_csv(v) for k, v in ds_metrics_dict.items()],\n",
    "                        keys = ds_metrics_dict.keys()) \\\n",
    "    .reset_index(level=0) \\\n",
    "    .drop(columns=['Unnamed: 0'], errors='ignore') \\\n",
    "    .rename(columns={'level_0':'model'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c6b66bb3-c7df-432c-9228-63ab73cf08b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "model\n",
       "dsloop naics                   45\n",
       "dsloop naics+hier              45\n",
       "dsloop naics data rand         45\n",
       "dsloop naics+hier data rand    45\n",
       "dsloop naics data gen          45\n",
       "dsloop naics+hier data gen     45\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds_metrics['model'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f23c515a-0f49-4a40-8a50-9748a3f5394c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature - data treatment\n",
    "cond = [ds_metrics['model'].str.contains('data rand'),\n",
    "        ds_metrics['model'].str.contains('data gen')]\n",
    "choice = ['fixed randomization', 'shuffle randomization']\n",
    "ds_metrics['data_treatment'] = np.select(cond, choice, 'unmodified data')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "cdda121e-23d9-4999-97d1-f203a7f43ed7",
   "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>data_treatment</th>\n",
       "      <th>fixed randomization</th>\n",
       "      <th>shuffle randomization</th>\n",
       "      <th>unmodified data</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>dsloop naics</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics data gen</th>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics data rand</th>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics+hier</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics+hier data gen</th>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics+hier data rand</th>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "data_treatment               fixed randomization  shuffle randomization  \\\n",
       "model                                                                     \n",
       "dsloop naics                                   0                      0   \n",
       "dsloop naics data gen                          0                     45   \n",
       "dsloop naics data rand                        45                      0   \n",
       "dsloop naics+hier                              0                      0   \n",
       "dsloop naics+hier data gen                     0                     45   \n",
       "dsloop naics+hier data rand                   45                      0   \n",
       "\n",
       "data_treatment               unmodified data  \n",
       "model                                         \n",
       "dsloop naics                              45  \n",
       "dsloop naics data gen                      0  \n",
       "dsloop naics data rand                     0  \n",
       "dsloop naics+hier                         45  \n",
       "dsloop naics+hier data gen                 0  \n",
       "dsloop naics+hier data rand                0  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(ds_metrics['model'], ds_metrics['data_treatment'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "d1f42c17-fc94-4025-9df8-268635e499b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature - NAICS only vs NAICS+hieararchy\n",
    "ds_metrics['hier_flag'] = np.where(ds_metrics['model'].str.contains('hier'), 1, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "cbc1fcd4-647e-44e0-87d5-fd0f8218ef31",
   "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>hier_flag</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>dsloop naics</th>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics data gen</th>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics data rand</th>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics+hier</th>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics+hier data gen</th>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dsloop naics+hier data rand</th>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "hier_flag                     0   1\n",
       "model                              \n",
       "dsloop naics                 45   0\n",
       "dsloop naics data gen        45   0\n",
       "dsloop naics data rand       45   0\n",
       "dsloop naics+hier             0  45\n",
       "dsloop naics+hier data gen    0  45\n",
       "dsloop naics+hier data rand   0  45"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(ds_metrics['model'], ds_metrics['hier_flag'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "07612ffe-fa9c-487f-b85c-9fa763d7855c",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics.to_csv(Path(setup.temp_path).joinpath('40_REPORT_metrics_loop_downsample.csv'), index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f251b557-39b8-41b4-bf84-5b12741751f5",
   "metadata": {},
   "source": [
    "##### Plot\n",
    "I didn't record the sizes for all tests but they should all be the same (or very similar)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0d46aa3f-31c4-4e1c-a48b-fb57641fa4cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics_test = ds_metrics[~ds_metrics['dset_naics_holdout'].isna()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "8ae60fbd-4500-4266-8bf9-aaaebd98121e",
   "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>model</th>\n",
       "      <th>average_precision_score</th>\n",
       "      <th>dset_naics_holdout</th>\n",
       "      <th>hier_flag</th>\n",
       "      <th>data_treatment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>dsloop naics</td>\n",
       "      <td>0.377062</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>unmodified data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>dsloop naics data rand</td>\n",
       "      <td>0.378691</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>fixed randomization</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>dsloop naics data gen</td>\n",
       "      <td>0.380609</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>shuffle randomization</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>dsloop naics+hier</td>\n",
       "      <td>0.380598</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>unmodified data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>dsloop naics+hier data rand</td>\n",
       "      <td>0.378425</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>fixed randomization</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>dsloop naics+hier data gen</td>\n",
       "      <td>0.381015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>shuffle randomization</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>dsloop naics</td>\n",
       "      <td>0.348513</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>unmodified data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>dsloop naics data rand</td>\n",
       "      <td>0.348434</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>fixed randomization</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>dsloop naics data gen</td>\n",
       "      <td>0.353664</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>shuffle randomization</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>dsloop naics+hier</td>\n",
       "      <td>0.373376</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>unmodified data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>dsloop naics+hier data rand</td>\n",
       "      <td>0.382847</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>fixed randomization</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>dsloop naics+hier data gen</td>\n",
       "      <td>0.384249</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>shuffle randomization</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          model  average_precision_score  dset_naics_holdout  \\\n",
       "38                 dsloop naics                 0.377062                 0.0   \n",
       "38       dsloop naics data rand                 0.378691                 0.0   \n",
       "38        dsloop naics data gen                 0.380609                 0.0   \n",
       "38            dsloop naics+hier                 0.380598                 0.0   \n",
       "38  dsloop naics+hier data rand                 0.378425                 0.0   \n",
       "38   dsloop naics+hier data gen                 0.381015                 0.0   \n",
       "39                 dsloop naics                 0.348513                 1.0   \n",
       "39       dsloop naics data rand                 0.348434                 1.0   \n",
       "39        dsloop naics data gen                 0.353664                 1.0   \n",
       "39            dsloop naics+hier                 0.373376                 1.0   \n",
       "39  dsloop naics+hier data rand                 0.382847                 1.0   \n",
       "39   dsloop naics+hier data gen                 0.384249                 1.0   \n",
       "\n",
       "    hier_flag         data_treatment  \n",
       "38          0        unmodified data  \n",
       "38          0    fixed randomization  \n",
       "38          0  shuffle randomization  \n",
       "38          1        unmodified data  \n",
       "38          1    fixed randomization  \n",
       "38          1  shuffle randomization  \n",
       "39          0        unmodified data  \n",
       "39          0    fixed randomization  \n",
       "39          0  shuffle randomization  \n",
       "39          1        unmodified data  \n",
       "39          1    fixed randomization  \n",
       "39          1  shuffle randomization  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds_metrics_test[ds_metrics_test['rate'] == 0.95][['model', 'average_precision_score', 'dset_naics_holdout',\n",
    "                                                  'hier_flag', 'data_treatment']] \\\n",
    "    .sort_values(['dset_naics_holdout', 'hier_flag'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "5283ec73-5564-450e-b550-4fc2467566cb",
   "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>model</th>\n",
       "      <th>dsloop naics</th>\n",
       "      <th>dsloop naics data gen</th>\n",
       "      <th>dsloop naics data rand</th>\n",
       "      <th>dsloop naics+hier</th>\n",
       "      <th>dsloop naics+hier data gen</th>\n",
       "      <th>dsloop naics+hier data rand</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rate</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>0.01</th>\n",
       "      <td>4468.0</td>\n",
       "      <td>4468.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4468.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.05</th>\n",
       "      <td>22342.0</td>\n",
       "      <td>22342.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22342.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.10</th>\n",
       "      <td>44684.0</td>\n",
       "      <td>44684.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>44684.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.20</th>\n",
       "      <td>89369.0</td>\n",
       "      <td>89369.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>89369.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.40</th>\n",
       "      <td>178739.0</td>\n",
       "      <td>178739.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>178739.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.70</th>\n",
       "      <td>312793.0</td>\n",
       "      <td>312793.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>312793.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.90</th>\n",
       "      <td>402163.0</td>\n",
       "      <td>402163.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>402163.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.95</th>\n",
       "      <td>424505.0</td>\n",
       "      <td>424505.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>424505.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.99</th>\n",
       "      <td>442379.0</td>\n",
       "      <td>442379.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>442379.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "model  dsloop naics  dsloop naics data gen  dsloop naics data rand  \\\n",
       "rate                                                                 \n",
       "0.01         4468.0                 4468.0                     NaN   \n",
       "0.05        22342.0                22342.0                     NaN   \n",
       "0.10        44684.0                44684.0                     NaN   \n",
       "0.20        89369.0                89369.0                     NaN   \n",
       "0.40       178739.0               178739.0                     NaN   \n",
       "0.70       312793.0               312793.0                     NaN   \n",
       "0.90       402163.0               402163.0                     NaN   \n",
       "0.95       424505.0               424505.0                     NaN   \n",
       "0.99       442379.0               442379.0                     NaN   \n",
       "\n",
       "model  dsloop naics+hier  dsloop naics+hier data gen  \\\n",
       "rate                                                   \n",
       "0.01                 NaN                      4468.0   \n",
       "0.05                 NaN                     22342.0   \n",
       "0.10                 NaN                     44684.0   \n",
       "0.20                 NaN                     89369.0   \n",
       "0.40                 NaN                    178739.0   \n",
       "0.70                 NaN                    312793.0   \n",
       "0.90                 NaN                    402163.0   \n",
       "0.95                 NaN                    424505.0   \n",
       "0.99                 NaN                    442379.0   \n",
       "\n",
       "model  dsloop naics+hier data rand  \n",
       "rate                                \n",
       "0.01                           NaN  \n",
       "0.05                           NaN  \n",
       "0.10                           NaN  \n",
       "0.20                           NaN  \n",
       "0.40                           NaN  \n",
       "0.70                           NaN  \n",
       "0.90                           NaN  \n",
       "0.95                           NaN  \n",
       "0.99                           NaN  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds_metrics_test[['rate', 'train_size', 'model']].drop_duplicates() \\\n",
    "    .pivot(index='rate', columns='model', values='train_size')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "79a2048a-b1e0-4d0e-ba2d-e2e54e912095",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_sizes = ds_metrics_test[~ds_metrics_test['train_size'].isna()] \\\n",
    "    [['rate', 'train_size']] \\\n",
    "    .drop_duplicates('rate')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "d93e8973-72a9-4f9b-8740-a1935595d699",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics_test = ds_metrics_test.drop(columns='train_size') \\\n",
    "    .merge(ds_sizes, on='rate')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "1ac52220-9b2f-414b-8245-a76e42751e78",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics_test_pivot = ds_metrics_test \\\n",
    "    .pivot(index=['train_size', 'hier_flag', 'dset_naics_holdout'], \n",
    "           columns='data_treatment', values='average_precision_score') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "71d995e9-9e60-4116-9fec-3220cd40b9b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics_test_pivot = ds_metrics_test_pivot[['unmodified data', 'fixed randomization', 'shuffle randomization']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "d4706f3f-4f15-4df7-b21d-5982195990ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_metrics_test_pivot.reset_index(level=[1,2], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "171c81c2-2227-4aa2-94b5-c943897e2ead",
   "metadata": {},
   "outputs": [],
   "source": [
    "groups = ds_metrics_test_pivot.groupby(['hier_flag', 'dset_naics_holdout'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "cf0ea7fe-dcdc-4dee-b1f1-96ae0634ea94",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x600 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(2, 2, sharex=True, sharey=True, figsize=(11,6))\n",
    "for n, g in groups:\n",
    "    g[['unmodified data', 'fixed randomization', 'shuffle randomization']].plot(ax=ax[int(n[0]), int(n[1])])\n",
    "    ax[int(n[0]), int(n[1])].legend().set_visible(False)\n",
    "    ax[int(n[0]), int(n[1])].set_xlabel(None)\n",
    "    \n",
    "ax[0, 0].set_ylim(0.32, 0.39)\n",
    "ax[0, 0].margins(tight=True)\n",
    "#ax[0, 0].set_ymargin(0)\n",
    "ax[0,0].set_title('test dataset')\n",
    "ax[0,1].set_title('holdout dataset')\n",
    "ax[0,0].set_ylabel('NAICS only'\n",
    "                  )\n",
    "ax[1,0].set_ylabel('NAICS+hierarchy')\n",
    "fig.supxlabel('training case count')\n",
    "fig.supylabel('PR-AUC')\n",
    "ax[0,0].legend(title=None).set_visible(True)\n",
    "plt.subplots_adjust(wspace=0.05, hspace=0.17, bottom=0.1, left=0.12)\n",
    "ticks_x = ticker.FuncFormatter(lambda x, pos: '{0:g} k'.format(x/1000))\n",
    "ax[0,0].xaxis.set_major_formatter(ticks_x)\n",
    "fig.savefig(Path(setup.temp_path).joinpath('40_PLOT_downsample_comparison.png'),\n",
    "        bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8c87e96-8d0d-40ee-ad9f-0b8a597b6d53",
   "metadata": {},
   "source": [
    "##### Count max results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "0d04300b-523f-42e4-8f3f-56aaa7534a9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "winners_df = ds_metrics_test \\\n",
    "    .sort_values(['hier_flag', 'dset_naics_holdout', 'rate', 'average_precision_score'],\n",
    "                ascending=[True, True, True, False]) \\\n",
    "    .groupby(['hier_flag', 'dset_naics_holdout', 'rate']) \\\n",
    "    .head(1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "497df71d-8f0e-46d3-823f-7cb88a2cde7a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "data_treatment\n",
       "shuffle randomization    24\n",
       "unmodified data           7\n",
       "fixed randomization       5\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_winners = winners_df['data_treatment'].value_counts()\n",
    "count_winners"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "35ad2b77-ea6d-44fa-ad7b-ab3591418302",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "data_treatment\n",
       "shuffle randomization    25\n",
       "unmodified data           6\n",
       "fixed randomization       5\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quick roc score\n",
    "ds_metrics_test \\\n",
    "    .sort_values(['hier_flag', 'dset_naics_holdout', 'rate', 'roc_auc_score'],\n",
    "                ascending=[True, True, True, False]) \\\n",
    "    .groupby(['hier_flag', 'dset_naics_holdout', 'rate']) \\\n",
    "    .head(1) \\\n",
    "    ['data_treatment'] \\\n",
    "    .value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "14188765-9c6a-4497-b45c-07efbf176ba2",
   "metadata": {},
   "outputs": [],
   "source": [
    "count_winners.to_frame() \\\n",
    "    .to_csv(Path(setup.temp_path).joinpath('40_REPORT_downsample_winners.csv'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "b586ac63-e867-415d-af70-03de13443262",
   "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>rate</th>\n",
       "      <th>0.01</th>\n",
       "      <th>0.05</th>\n",
       "      <th>0.10</th>\n",
       "      <th>0.20</th>\n",
       "      <th>0.40</th>\n",
       "      <th>0.70</th>\n",
       "      <th>0.90</th>\n",
       "      <th>0.95</th>\n",
       "      <th>0.99</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>data_treatment</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>fixed randomization</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shuffle randomization</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unmodified data</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "rate                   0.01  0.05  0.10  0.20  0.40  0.70  0.90  0.95  0.99\n",
       "data_treatment                                                             \n",
       "fixed randomization       0     1     1     1     0     1     1     0     0\n",
       "shuffle randomization     2     1     3     3     4     2     2     4     3\n",
       "unmodified data           2     2     0     0     0     1     1     0     1"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(winners_df['data_treatment'], winners_df['rate'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "3109c73d-b936-4c1d-b2fb-bfcd50759f38",
   "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>dset_naics_holdout</th>\n",
       "      <th>0.0</th>\n",
       "      <th>1.0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>data_treatment</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>fixed randomization</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shuffle randomization</th>\n",
       "      <td>12</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unmodified data</th>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "dset_naics_holdout     0.0  1.0\n",
       "data_treatment                 \n",
       "fixed randomization      1    4\n",
       "shuffle randomization   12   12\n",
       "unmodified data          5    2"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_xtab = pd.crosstab(winners_df['data_treatment'], winners_df['dset_naics_holdout'])\n",
    "test_xtab.to_csv(Path(setup.temp_path).joinpath('40_REPORT_downsample_winners_by_ho.csv'))\n",
    "test_xtab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "ba40c945-da55-4647-8e2c-c0e0d2b70366",
   "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>hier_flag</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>data_treatment</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>fixed randomization</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shuffle randomization</th>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unmodified data</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "hier_flag               0   1\n",
       "data_treatment               \n",
       "fixed randomization     4   1\n",
       "shuffle randomization  10  14\n",
       "unmodified data         4   3"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(winners_df['data_treatment'], winners_df['hier_flag'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf7342b7-1331-493b-ba4d-9936a31d4e4e",
   "metadata": {},
   "source": [
    "## Injection Rate Loop\n",
    "Make plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "896753d1-56a3-40dc-9f62-b7e06e3487c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "inj_loop_metrics_dict = {k:v.parent / (v.name + 'REPORT_loop_metrics.csv') for k, v in dict_2_rate_loop.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "acda7ddc-2c35-400c-b9a9-4c11bffba2b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "inj_metrics = pd.concat([pd.read_csv(v) for k, v in inj_loop_metrics_dict.items()],\n",
    "                        keys = inj_loop_metrics_dict.keys()) \\\n",
    "    .reset_index(level=0) \\\n",
    "    .drop(columns=['Unnamed: 0'], errors='ignore') \\\n",
    "    .rename(columns={'level_0':'model'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "137c4503-e410-4ff9-98f6-2479147a7693",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "model\n",
       "irloop naics+hier data gen       70\n",
       "irloop naics data gen            70\n",
       "ds irloop naics+hier data gen    60\n",
       "ds irloop naics data gen         60\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inj_metrics['model'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "dce1ff35-4414-4806-b0e7-15172deb565d",
   "metadata": {},
   "outputs": [],
   "source": [
    "inj_metrics.to_csv(Path(setup.temp_path).joinpath('40_REPORT_metrics_loop_inj_rate.csv'), index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d73333cf-d3dd-4230-a463-6023f0daa20c",
   "metadata": {},
   "source": [
    "##### Comparison plot - regular"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "77d2d4aa-4c07-4a77-a2dd-64a3146eb3d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Non-downsampled tests, also test cases only\n",
    "inj_metrics_reg = inj_metrics[(~inj_metrics['model'].str.startswith('ds ')) &\n",
    "                             (~inj_metrics['dset_naics_holdout'].isna())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "1e6bded7-e7fc-4068-a325-ebde3577608e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "model\n",
       "irloop naics+hier data gen    28\n",
       "irloop naics data gen         28\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inj_metrics_reg['model'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "dfbd2a44-a1a3-48f1-bbfa-18e1ed4e15cf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'sba_nn.sba_nn.sbnn_plot' from '/Users/valeriecarey/Documents/projects/2024_05_blog_nn_rand/code/sba_nn/sba_nn/sbnn_plot.py'>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "importlib.reload(sbnn_plot)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "62e7e997-e606-485c-a085-ae6e39a94cdc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2, figsize=(12,4), sharey=True, sharex=True)\n",
    "sbnn_plot.plot_metric_dependence(inj_metrics_reg[inj_metrics_reg['model'] == 'irloop naics data gen'],\n",
    "                                 xformatter = ticker.PercentFormatter(1),\n",
    "                                 ylabel='PR-AUC',\n",
    "                                 title='NAICS only',\n",
    "                                 xlabel='injection rate', ax=ax[0],\n",
    "                                marker='.')\n",
    "sbnn_plot.plot_metric_dependence(inj_metrics_reg[inj_metrics_reg['model'] == 'irloop naics+hier data gen'],\n",
    "                                 xformatter = ticker.PercentFormatter(1),\n",
    "                                 ylabel='PR-AUC',\n",
    "                                 title='NAICS+Hierarchy',\n",
    "                                 xlabel='injection rate', ax=ax[1],\n",
    "                                marker='.')\n",
    "ax[0].legend().set_visible(False)\n",
    "plt.subplots_adjust(hspace=0.001, wspace=0.05, bottom=0.1)\n",
    "fig.savefig(Path(setup.temp_path).joinpath('40_PLOT_injection_rate_comparison.png'),\n",
    "        bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "382f9e15-c2d6-4cb2-8853-eb1642c03331",
   "metadata": {},
   "source": [
    "##### Comparison plot - downsampled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "2a7e04ec-1d77-40c5-b6f5-169a021d69b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Non-downsampled tests, also test cases only\n",
    "inj_metrics_ds = inj_metrics[(inj_metrics['model'].str.startswith('ds ')) &\n",
    "                             (~inj_metrics['dset_naics_holdout'].isna())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "7d332f0f-0e82-4657-988d-3e1a2dae9edd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "model\n",
       "ds irloop naics+hier data gen    24\n",
       "ds irloop naics data gen         24\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inj_metrics_ds['model'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "2f49324a-4883-4238-868a-d11706465bff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABBoAAAGrCAYAAABucvQoAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAA9hAAAPYQGoP6dpAADjGklEQVR4nOzdd3hT5dsH8O/J6N6Dlu7SQlll770UwcEQUFEExa2oqK/i3ooLBXHATwQXKqAgIENk71mmrJaWDuje6UzzvH+kCU3TlTZt0vb7ua5epDnrTptyzrlzP/cjCSEEiIiIiIiIiIjMQGbpAIiIiIiIiIio5WCigYiIiIiIiIjMhokGIiIiIiIiIjIbJhqIiIiIiIiIyGyYaCAiIiIiIiIis2GigYiIiIiIiIjMhokGIiIiIiIiIjIbJhqIiIiIiIiIyGyYaCAiIiIiIiIis2GigYiILGLFihWQJMnoy87ODj4+PujYsSOmTp2Kjz76CJcvX7Z0uGTl4uLi9O+hXbt2WSyOESNGQJIkzJo1q17bh4SEGP1NyGQyuLm5ITg4GEOGDMEzzzyDNWvWoKSkxLzBExERmQkTDUREZFWKi4uRmpqKixcvYs2aNZg3bx4iIiJw8803M+FArZIQAjk5OYiPj8f+/fuxaNEiTJ06FYGBgViwYAGEEI127F27dukTHnFxcY12HCIialmYaCAiIovbtGkT8vLykJeXh5ycHFy9ehUHDhzAggUL0LNnTwghsG3bNvTs2RPr16+3dLhEjW7IkCH6v4m8vDykpKTgwoULWLVqFR555BE4ODggNTUVzz//PMaOHYuioiJLh0xERKTHRAMREVmcvb09nJyc4OTkBBcXFwQFBWHgwIGYO3cuTpw4gZ9++gkODg5QqVSYPn06Tp06ZemQiRqVXC7X/004OTmhTZs2iIiIwNSpU7FkyRLExsZi9OjRAIBt27bhoYcesnDERERENzDRQEREVu++++7DihUrAAAqlQovvPCCZQMisrA2bdrg77//Rq9evQAAv/zyC/bv32/hqIiIiLSYaCAiomZh6tSpuOmmmwAA//77b41VDZs3b8aUKVPg7+8PW1tbeHh4YNCgQfj0009RUFBQ5Ta6JnyvvfZalcsHDx6sH6u+ceNGo+UlJSVwcHCAJElYsmSJwTLddrpkyQ8//IAhQ4bA3d0dDg4O6N69Oz7++OMam/upVCp8/PHHGDRoEDw8PKBUKuHt7Y3OnTtj6tSp+N///geVSmW0XUxMDL744guMHTsW/v7+sLGxgZOTEzp27IjHHnsM58+fr/aYlRsslpaW4rPPPkOvXr3g4uICLy8v3HTTTdixY4fBdidPnsR9992H4OBg2NraIjg4GHPnzkV2dnadjlNYWIj3338f3bp1g7OzM9zc3DBs2DD88ssv1cZaVzk5Ofjwww8xaNAgeHl5wdbWFn5+fpg8eTL++eefWrdPSkrCE088gZCQENjZ2cHPzw933nmnRW7ybW1t8eWXX+q//+yzz4zWEULgyJEjeO211zBo0CB4enpCqVTC3d0dffv2xZtvvomMjIwq9y9JEkaOHKn/PjQ01KhRZcW+DUVFRdi0aRMef/xxdOvWDS4uLlAqlWjTpg3GjBmD//3vf2xgSUTUWggiIiILWL58uQAgAIidO3fWaZvffvtNv80nn3xitLy4uFhMnz5dv05VXyEhIeL8+fNG286aNUsAEIMGDTJalp+fL5RKpX4fc+fONVpnz549+uWXLl0yWKZ7/rvvvhNTpkypNrabb75ZlJWVGe372rVron379jW+LgDi6NGjBttlZ2fXuo1SqRQ//vhjlT/v2NhY/Xrr168X/fr1q3IfMplM/PTTT0IIIX788UdhY2NT5XrdunUT+fn5NR5n7dq1olu3btXGe/fddwu1Wl3jPqp7P+3evVt4eXnV+PN45JFHqty/EEIcPHhQuLq6VrmdXC4XS5cuFcOHDxcAxMyZM6vcR22Cg4MFADF8+PA6b9OxY0cBQLi4uBi9f9atW1fre8DX11dERUUZ7be27QCI2NhY/frPPvtsresPGDBAZGVl1etnQ0REzQcrGoiIqNkYPHiw/vHBgweNlj/zzDNYuXIlAGDcuHHYu3cv0tPTcfHiRbz11luwsbFBXFwcbr75ZqNP13Wf3B49etSoMmDfvn0oLS2FUqkEAOzcudPo2Lrn/P390b59+yrj/+CDD7B27Vq89NJLOHPmDDIzM3Hy5ElMmjQJAPDPP//gu+++M9ruxRdfxOXLlyGXy/H666/j5MmTSE1NRWpqKqKiovD111/rp1WsrF+/fvjoo4+wY8cOnD9/Hunp6bh8+TLWrVuH0aNHo7S0FA899BDOnDlTZcw6zz77LM6dO4ePP/4Y0dHRSE9Px7Zt2xAREQGNRoOnnnoK27Ztw+zZszF06FDs2LEDaWlpuHLlCubNmwcAOH36ND788MMaj/Pcc8/h/PnzeO2113DhwgWkp6dj9+7dGDVqFADgt99+w5tvvlnjPqpy8uRJjB07Funp6YiMjMTPP/+M2NhYZGZm4tSpU3j22WchSRKWLl2Kd955x2j7tLQ03H777cjJyYGjoyM++eQTXLlyBampqdi0aRMiIyPx1FNPITo62uTYGkr3d5Gbm4tz584ZLFMoFLjjjjuwZMkS7Nu3DzExMUhPT8fZs2exZMkSdOjQAcnJyZg8ebJRQ8m8vDxs2rRJ//25c+cMGlTm5eUhODhYv9zV1RWzZ8/Gb7/9hqNHjyIhIQEpKSk4fvw43njjDbi7u+PQoUN4/PHHG/GnQUREVsHSmQ4iImqd6lPRIIQQdnZ2AoAYOHCgwfMnT57U7+/OO+8UGo3GaNs//vhDv87zzz9vsCw+Pl6/bPPmzQbLXnzxRQFAzJ49WygUCiFJksjIyDBYZ8SIEQKAuO+++4yOiwqf6P78889Gy8vKykSPHj0EANG/f3+j5R4eHtVWUjTUXXfdJQCI+++/32hZxSoBuVwudu/ebbTOxYsXhSRJAoBQKBTi1ltvrbIi4J577hEARNu2bWs8DgCxYsUKo3VKS0vFyJEj9VUYSUlJ1e6jqveT7uc7cuRIUVRUVOXP4uuvvxYAhI2Njbh27ZrBsjlz5ggAQpIksW3bNqNtc3NzRefOnfUxNGVFw/z58/XH3bp1q0nHy8vLE2FhYQKA+P77742W79y5s8rqhfo4ffq0kMvlQpIkERMT06B9ERGRdWNFAxERNStubm4AgKysLIPndZUACoUCX375ZZWf7k+ePBm33HILAOD777+HRqPRLwsMDERYWBgA44oF3fd33HEH+vTpAyGEwTpFRUX6CouKY9orGzhwIO69916j52UyGe6//34AQFRUFNRqtcFy3fd+fn7V7ru+dMfdtm1bjetNmzYNw4YNM3q+Q4cO+oaEarUan3/+OeRyudF699xzDwDg+vXrSEhIqPY4ffv2xcyZM42eVygUWLhwIQCgtLQUP/30U43xVrRr1y6cPHlS3z/D1ta2yvUee+wxtGvXDiUlJVi9erX++bKyMvz4448AgIkTJ2LMmDFG2zo7O2P+/Pl1jsmcdH8TgPHfRW2cnJwwefJkALW/BxoqMjISvXr1ghAC//77b6Mei4iILIuJBiIialaEEABglEjYu3cvAGDo0KFo27ZttdvfddddALQ3ZGfPnjVYNmLECACGiYbc3FycOHECcrkcw4cP15fwV1zn4MGDKC4uBlBzomHcuHHVLouIiACgbSpZ+WaxR48eAIBPPvkEf//9N8rKyqrdT1V27NiBmTNnomPHjnB2doZMJtM387v11lsBaBMAeXl59Ypdl6Bp165dtcNGwsPD9Y+vX79e7b50N71ViYyM1O/flOaLupvasLAwtG3bFvn5+VV+qVQqdO/eHYB2CI3OmTNnkJOTU2t848aNg729fZ3jMhfd3wRg/HcBaBNAP/zwA+644w4EBQXpm5bqvj755BMAwMWLFxscS2ZmJj766COMGDECPj4+sLGxMTiW7udqjmMREZH1Ulg6ACIiIlPobvg8PDwMnr969SoAoHPnzjVu36VLF/3juLg4dOvWTf/9yJEjsWzZMpw4cQK5ublwcXHBnj17UFZWhj59+sDV1RUjR47EBx98YDDTgi7pEBwcjNDQ0GqPXVNFgoODg/5x5Zkx5s+fjxEjRiA1NRW33XYbPD09MWzYMAwZMgSjR4/W3xxXVlZWhtmzZ+OHH36o9rgV5eTkwNnZ2eTYdTfXdVkHAAoLC6tdr1OnTjXG2LlzZ1y+fNlgtoPaXLhwAQAQHR1d7eurLC0tTf+44rFqik+hUKBDhw41zojSGHR/E4Dx30VqairGjRuHEydOmLSf+jh06BDuuOMOg59dYx2LiIisGysaiIio2YiPj9c3rKt8U6v7NL62G8mKyyt/gq+rRigrK8Pu3bsBQJ9Q0C0bPHgwbGxscP78eSQnJwO4kWioqZoBQJVDCqpS8RNqQDvk4tChQ5g4cSKUSiUyMjKwdu1aPP/88+jRowe6du2Kv//+22g/n376qT7JcPvtt+OPP/7QN1jUNfOruF3lIRumxl7f11eRk5NTjdvqltdUfVFZfW5qKzZGzM/PNzp+dWpb3hgqVgdU/ruYOXMmTpw4AYVCgaeffhrbtm1DbGwsMjIy9O8BXbPOmn7/tcnNzcXEiRORlpYGb29vfPjhhzh48CCSkpKQnZ2tP5aucWVDjkVERNaPiQYiohYgJCQEkiRhxYoVlg6lUR04cED/eNCgQQbLdAmEijeFVam4vHJSws/PDx06dABwI3mg+1c3ZMLe3h4DBgzQLysoKMCRI0cA1J5oaIiePXti7dq1yMrKwvbt2/Hee+9h2LBhkCQJ586dw2233YY//vjDYJuvvvoKgHa4yPr16zF58mRERETA09MTTk5OcHJy0g/5sBZ1/f3VtTIBuHHzr+sPUJevXbt2GW1vSnxNSfd34e7ublBxceXKFWzZsgUA8OWXX2LhwoUYM2YMQkJC4OHhoX8PVK6gqY81a9YgJSUFMpkMO3fuxLx58zBgwAD4+fnB1dVVfyxTEkTU+uzatUs/zIaM6WYXeuuttywdClGtmGggaqXeeust/cncwcEB165dq3bduLg4/boVL76rs3nzZv36QUFBBg33qjNr1ixIkoSQkJBa1z19+jReeukl9OvXTz8G2NXVFZ07d8bMmTPx119/obS0tMpt8/PzsXDhQowaNUq/rYeHBzp16oSxY8fi7bffxo4dO0weA09NY9myZfrHN910k8Ey3Xun8vR+lVXsy1DV+02XLNi5c6d+6kOlUokhQ4YYrbNjxw7s378fJSUlBs83JkdHR4waNQqvvvoqdu/ejZMnT8LT0xMADKZlzMzM1Ddd1DVirEpt01o2tfPnz9e4/L///gNQ9e+uOu3atQMAxMbG1un/o8oqHqum+NRqNS5fvmzy/hti//79+oqGMWPGGNygnTx5Uv+4sd8DumN169bNYHhSRSUlJbh06VKDj1VZY53PeC6rO93PqS43wKb8jIio+WKigYhQWFiIt99+22z7q3gzmJCQYLZO5nl5ebj33nvRo0cPfPzxxzh69CjS0tLg6OiI4uJinD9/Hj/++CMmTpyITp064fDhwwbbnz59Gp07d8azzz6LnTt3IjU1FXK5HEIIXLx4Ef/88w/eeustjB49usau+GQZq1ev1jf1u+WWW4xuZoYOHQoA2LdvH1JSUmrcD6D99Ldr165Gy3UNIU+dOoU///wTQgj07dvX4FPtig0hdRUPYWFhCAwMrOerq79u3brpG1zqehEAMKhUqO5mo6ysDCtXrmzcAE20du3aapedOXNGfyOvK8Gvi7FjxwLQNgD9559/TI4pMjISrq6uAIA///yz2vU2b95sluqAuiouLsbTTz+t//755583Wq5T3XsgPj4ee/bsqfYYSqWy1n1UPFZN66xevdpgSEpjMOf5jOeypufg4ICIiAh9c1wiar6YaCAiANqp/szxSVNaWhrWr18PuVyOxx9/HIDhxVp9ZWVlYeDAgfqborvvvhu7d+9GUVERsrKyUFRUhKSkJHz33Xfo1q0bYmJi9NMNAtoLu/HjxyMhIQFeXl5YuHAhUlNTUVhYiKysLOTl5WHPnj148cUXa5yxgCzjl19+waxZswBoy9h1XfIrmj17NgDt1IdPP/10lX0A/vrrL2zatEm/vkxmfBrUJRqEEHjvvfcAGFcqDBgwAPb29oiJicGvv/5a5TrmolKpar1ZiImJAQB9ZQMAtGnTRp8c+euvv6rc7t1337W67v9HjhzRTyVZkVqtxjPPPANAe/M7Y8aMOu/zpptu0jf9fOKJJ5CUlFTj+ikpKQYzf8jlcv00oOvWrcP27duNtsnPz9f3OmgKusaguiaPs2bNQv/+/Q3W0VVyAFW/B0pLS/HQQw/VmBzw8vLSP66pUkB3rPPnz1d5LklKSsJLL71U7fbmZI7zWWs9l+mG4ZnSbNWc+vXrhwsXLhgkTYmoeWKigaiVCwwMRLdu3aBWq/HKK680eH8//fQTSktLcdNNN+HFF1+EJEn466+/kJGR0aD93nvvvTh37hwUCgV+//13/Prrrxg2bBhsbGz06/j5+WH27Nk4efIkvv76a9jZ2emX/fbbb/qbiw0bNuDpp5+Gt7e3frmjoyOGDh2Kjz76CPHx8QgICGhQvGSawsJC/RSDubm5SEhIwKFDh/DFF1+gd+/euO+++1BQUABHR0f8/vvvVVYidO/eHY899hgAYNWqVbj99tuxf/9+ZGZm4vLly3j33Xf1n/wHBgbi1VdfrTIWX19f/Th33UwWlZMINjY2+h4Rugvyxko0pKWloV27dpgwYQKWL1+OM2fOID09HcnJyThw4ADuv/9+bN26FYBhebxcLsedd94JAPjxxx/xzDPP4OzZs8jIyMDRo0cxa9YsvP3227XO0tHUQkND8dBDD+GNN97ApUuXkJGRgT179mDs2LH66pEXX3yxxhkuKpMkCT/88AMcHBwQGxuLHj164MMPP8SpU6eQmZmJ9PR0nD17Fj/++COmTZuG4OBgffJG5/XXX4eXlxeEEJgwYQI+++wzxMXFIT09HVu2bMGwYcMQHR0Nf39/s/wcysrKDKbeTEtLw6VLl/DHH3/gscceQ2hoqEGFz5IlS4z20adPH30C4JlnnsGiRYsQExODtLQ0bNmyBcOHD8e2bdtqfA+Eh4fDzc0NgHZ61StXrqCkpARqtdqgoeOdd94JuVwOtVqNW2+9FWvXrsX169eRmJiIFStWYMCAAcjKykJwcLBZfj5VMef5jOcyIqIGEkTUKr355psCgAgODhZ///23ACAAiMOHDxutGxsbq1++c+fOGvfbuXNnAUD8+uuvQgghhg8fLgCIL774osbtZs6cqY+nsk2bNumP/84779T5NZaVlekfP/HEEwKAaNOmTZ23r4/CwkLx+eefi4EDBwo3Nzdha2srgoKCxIwZM0RUVFS12wUHBwsAYvny5aK4uFh8/PHHolu3bsLBwUG4uLiIkSNHis2bN9dpe53NmzcLAEIul4ukpKQa4x4yZIgAIGbOnGniK66/5cuX63+vtX1JkiTGjh0rLl++XOM+i4uLxfTp02vcV0hIiDh//nyN+9G9XwAIW1tbUVhYaLTO+++/b7Dfa9euVbs/3ToVfz+V7dy5U79ebGys/vmKf381fY0ZM0bk5+cb7DM5OVmEhoZWu82wYcMM/r4qHrfysWv629f9/Q4fPrzadWraV8Vla9euFZGRkdXGfNdddwm1Wm3S/nUOHz4sgoKC6vTzPHnypNH2Bw8eFK6urlWuL5PJxJIlS/T/59X3b0n3t1yXrzZt2ojPP/9caDSaave3e/duYW9vX+0+XnjhBYPzQVV0y6v6qvie+fjjj6tdz87OTqxZs6bBP5+a4jPn+ay1nst077/K/xfURve63nzzzVrXrelnVPH/weoUFxeLr776SowYMUJ4enoKpVIpfHx8xB133CE2bdpUa4w7d+4UKSkpYu7cuaJ9+/b6vw8dlUolVq5cKWbMmCG6d+8uvLy8hI2NjWjbtq2YMGFCjcfQndd0r23Hjh1iwoQJwtfXV8hkMqP3fX5+vvjss8/EsGHD9K/F399fDBs2THz66aciOTnZYH3d+/DNN98UGo1GLF26VPTr1084OzsLJycnMWDAAPHTTz8ZxXX+/Pka/y4quu+++2r9/5yoLphoIGqlKl9Y6k5eI0eONFq3rhdmBw8eFACEi4uL/sbs+++/FwBEZGRkjfHUdOExfvx4AUC4uroa3UjVle7iTKlUCpVKVa991CYxMVF07dpV/7NSKpUGNyUymUwsWrSoym11F3dffvml6N+/v357Jycn/faSJIlly5bVuH3FG1mNRqO/yXz33XerjbviBcj+/fsb9DMwRXWJBhsbG+Ht7S0iIiLEnXfeKebPn19rgqGyTZs2icmTJws/Pz+hVCqFm5ubGDhwoPj444/r9PtfvXq1Pp7qLrYOHDigXyciIqLG/enWq0+ioaysTOzfv1+8/fbbYvTo0SIsLEw4OjoKGxsbERAQIO644w7x+++/V3uzmZ6eLubOnStCQ0OFUqkUHh4eYuDAgeLLL78UarW62uMKYZlEw86dO4VKpRJvvfWW6NKli3B0dBQuLi5iyJAhVV5AmxprYWGh+Pbbb8W4ceNE27ZthY2NjbCzsxNBQUFi3Lhx4qOPPhLR0dHVbp+YmCgef/xxERwcLGxsbISvr6+YOHGi2LNnjxBCNEqiQZIk4ezsLAICAsSgQYPEnDlzxJo1a0RJSUmd9nnmzBlx9913izZt2gilUil8fX3FbbfdJjZu3CiEMD4fVKbRaMSSJUvE4MGDhZubm5DJZNW+ZzZs2CBGjRolXFxchK2trQgJCREPPvigOHPmjFl+PlUx9/msNZ/LrD3REBcXJ7p06WLwt1E5+ffYY4/VGOP//vc/4ePjo0+AOTs7Gxyv4rlJt38HBweDYzz//PNVHqNiouGLL74QkiTpf+dKpdLgfX/8+HERGBhocI3g4eEhbG1t9c99/vnnBvvXvbdfe+01MWHCBAFAKBQK4eLiYhDfG2+8YRSbbtvZs2dXGbsQQmRmZgo7OzsBQPzyyy/VrkdUF0w0ELVSlS/MdBdWAIw+Oa/rBfxDDz1kdBLLzc3Vn6CPHDlS7bbVXXiUlpbqb7anTJli8uvUWbFihf413H333SIjI6Pe+6qKWq3WJwhcXV3Fzz//LIqLi4UQQsTExIjbbrtNf9FS1achuos7d3d34e/vL9atW6e/ibhw4YIYMGCAACCcnJxEdnZ2tdtXvpGdP3++ALSf4ld3I/rcc88JAKJr164N/CkQ1Y8pVVNElZn7fNaaz2XWnGjIz88XHTt2FADEiBEjxK5du0RRUZEQQojs7GyxYMEC/c+4qsoT3X6dnJxERESE2L59u75a5OLFi/r11q1bJ1544QWxb98+g2TOtWvXxNtvvy2USqUAIP766y+jY+gSDXZ2dkIul4tZs2aJ+Ph4IYT2OkGXxIyPjxdeXl4CgAgMDBS//fab/lgajUacO3dOvPXWW+Lnn3822L8uWeDu7i5cXV3FihUrREFBgRBCiISEBHH77bfrkxaXLl0y2Pa3334TAISjo6PIzc01il0IIRYtWiQACE9PT/3Plqi+mGggaqWq+gRr0qRJAoDo0aOHwU1pXS7M8vPz9Z8K6D7Z07n33ntr/JRBiOovPKKjo/XHfv/9901+nTpFRUUG1QY2NjZi5MiR4qWXXhKrVq3SXwjUl+4EDkBs3brVaHlpaak+EVHVDb3u4s7W1rbKsv7U1FT9pwyVLzwqbl850ZCamipsbGwEALFlyxaj7YqKivQXO9VVWxA1NiYaqCHMeT5r7eeyhiYaHB0dhY+PT41funOZqYmGd955RwDayqnqqnn+/PNPAUB4eXmJ0tLSKmN0cXERCQkJJr2+ij755BMBQIwePdpoWcVqiMmTJ1e7D93wBE9PT5N+Z7pEAwCxY8cOo+VFRUXCz89PABDvvfeewbKSkhLRpk0bAUB8++23Ve5fN2ztueeeq3NMRNVhM0gi0vvggw8gl8tx8uRJfSf9ulq1ahXy8vIQGhqKIUOGGCybOXMmAODXX39FYWGhSfut2HjLw8PDpG0rsrW1xY4dO3DXXXdBkiSUlJRg586d+OijjzBt2jQEBQWhc+fO+OKLLwymhKur33//HQAwcOBA3HzzzUbLFQoF3nzzTQDA2bNnq523fsqUKejYsaPR897e3hg4cCAA7dRmdeXt7a1vCLh06VKj5WvXrkV6ejrs7e1N6uJPRGTN6ns+a+3nsoZSqVRISUmp8au+U5zqZv147rnnDKZdrWjixIlwcXFBeno6jh8/XuU6M2bMaFCTzFtvvRUAcPDgwRpnbHn55ZerfF6lUumvGebNm1evaZEHDx5cZQNiW1tb/XS+la8VlEqlfnamqq4HDh06pL82eeSRR0yOiagyJhqISK9jx4544IEHAGg7rJeWltZ52++//x6A9gQuSZLBstGjR8Pf3x85OTlYs2aN+QI2kbe3N3777TfExsZi4cKFmDZtGsLCwvTxnj9/HnPnzsXAgQNN7ix+7NgxAMCYMWOqXWfkyJGQy+UG61dWeXq6inRd9jMzM02KTTcTw4YNG5CSkmKw7H//+x8AYNq0afrO8kREzV19z2et4Vw2efJk+Pr6Vvmlm0q3b9++VS7v27dvjbG9+eabENqK6Wq/dAkbUyQlJelnAZo9e3a18bdt2xb5+fkAbswaVNngwYNrPV5KSgrefPNNDBw4EJ6enlAoFJAkCZIk6WdpKSgoMJgGtyJ7e3v06tWrymXHjh3Tvx9vv/32WmOpSn2vFR555BHIZDKcOHFCPzWuju56YPjw4YiIiKhXXEQVMdFARAbeeust2Nvb48qVK/j222/rtM3Fixexb98+ANDPNV+RTCbTf1quu4irK09PT/1jU2+wqxMcHIynn34av//+O6Kjo5GRkYFffvlFP2ViVFQUHn30UZP2mZqaCgA1Tm1nZ2enn5Net35lzs7O1W6vUCgAwKQEEAAMGzYMnTt3RmlpKZYvX65/Pjo6Wj9doKmvl4jI2pl6Pmst57LMzMxqqw00Gg0AID09vcrlaWlpZondVNeuXdM/ri62yq+hoKCgyn21adOmxmMdPHgQHTt2xDvvvINDhw4hMzMT9vb2aNOmDXx8fPTncUBbnVAVT09PyGRV32YlJyfrH9d3utf6XiuEhIToKx4qVjXk5ubqqyx4PUDmwkQDERnw9/fHnDlzAADvvfee/pOBmlS84AoPD9dn/St+zZ8/HwCwe/duo/npaxIcHAwnJycA2oumxuDu7o7p06fj8OHD6NSpEwDtkAJzXQxaA11Vw3fffQchhMHjrl276odlEBG1FKaez1rLuWzXrl3VVhvobnxjY2OrXB4XF9cosdem4hCF8+fP11o1IYTArFmzqtyXrrKwKmq1Gvfccw+ys7PRo0cPbNq0Cbm5ucjLy0NKSgqSk5Nx6NAh/fq686kpx6hcKdPUHn/8cQDAypUr9YkS3WNPT09MnjzZkuFRC8JEAxEZmTdvHtzd3ZGamorPPvusxnXVajV+/PHHOu9bCGHSJ0EKhQLDhg0DAGzbtq3aTw/MwcHBAffddx8AQKPR4PLly3XeVvcJSWJiYrXrFBUV6ctYa/tExdzuv/9+ODg4ICYmBjt27EBpaSlWrFgBgJ9ekOWFhITobw5GjBhh6XCoBanr+YznMuvm6+urf1zdkAhzOHjwIK5evQq5XI6NGzdi3LhxRtUDFSsS6qOpXkt1xo8fj8DAQOTl5eG3334DcGPYxKxZs2Bra9vkMVHLxEQDERlxd3fHvHnzAACfffZZjaWSf//9N5KTk6FUKpGYmIi8vLxqv3QXeT/88EONDZQqe/LJJwEAOTk5WLBgQZ2305VPmkL3iRMAk062ffr0AQBs37692nV27doFtVoNALWOczU3V1dX3HPPPQC05ZK6fg329vb6C1IiopamrucznsusW0hIiH5o4oYNGxrtOLoeFd7e3tUOhfz3338bdIw+ffrAxsYGQOO+lurI5XJ9s8elS5ca9GtgE0gyJyYaiKhKc+bMQUBAAPLy8vDuu+9Wu56uC7SuSZaTk1O1X3fffTdkMhmSkpKwdevWOscyfvx4/UwO77zzTp2acC1dulSfoQeAI0eO1DoUQq1W45dffgEAODo6mtQM6e677wag/TTkn3/+qXLf77zzDgCga9eu+jG0TUk3fGLdunX4+OOPAbAJJBG1fHU5n/FcZv0efvhhANrfVW3DT+o79NHV1RUA9P0eKktMTMSiRYvqtW8dBwcH/TXD/Pnz9cmNpjR79mwoFAocOXIEc+fOBaBtAtmhQ4cmj4VaLiYaiKhK9vb2eOuttwBUn3G/fv06Nm3aBEB7w1obPz8/fbdn3UVdXa1cuRKdOnWCWq3GtGnTcO+992Lv3r0GzY6uX7+OH374Ab1798ajjz5qMP3YqlWrEBwcjAcffBAbN2406MRdUFCAzZs3Y+TIkThy5AgA7RhGe3v7Osd355136rtAT5s2DStXrtTHFhsbizvvvBMHDx4EAP1NflPr06cPevfujZKSEhw+fBgAh00QUctX2/mM57Lm4fnnn0dkZCSKioowcuRILF682OD1Z2dnY/Pmzbj//vsxdOjQeh1jyJAhcHR0hBAC06ZNw6VLlwBoe0Rs3boVI0aMMEuPhffffx9eXl7IyMjA4MGDsWrVKv3vWQiBs2fP4v/+7//w008/NfhYVWnbti0mTJgAANizZw8AXg+Q+THRQETVmjVrFjp27Fjtcl3ZqFKpxMSJE+u0z6lTpwLQXuyZ0r3a09MThw4d0l8Erly5EsOGDYOtrS08PDxgb28PPz8/zJo1CydOnECnTp0MLjSUSiXy8/OxfPly3H777fDy8oKjoyPc3Nzg6OiI8ePH67uNz5gxAx988EGdYwO0pYh//PEHunTpgpycHNx7771wcnKCu7s72rVrh/Xr10Mmk2HhwoUYN26cSfs2J10TKABsAklErUZN5zOey5oHJycnbNmyBQMGDEBOTg7mzJkDb29vuLu7w9XVFe7u7hg/fjx++uknlJSU1OsYrq6u+PTTTwFob8AjIiLg7OwMJycn3HLLLcjJyTGYvam+AgICsHXrVvj7+yMhIQF33XUXnJ2d4eXlBQcHB0RGRuLTTz81eaptU1S8HmATSGoMTDQQUbXkcnmNFym6RlhjxoyBu7t7nfY5ZcoUyGQylJaWmpypd3Fxwe+//46oqCi88MIL6NOnD7y8vJCXlwelUolOnTph5syZ2LhxI86cOYPevXvrt/3ggw9w6NAhvP3227jlllv0zefy8/Ph6uqK7t2749FHH8W+ffvw448/QqlUmhQboO1wfuzYMSxYsAADBgyAvb09CgoKEBgYiBkzZuD48eN4+umnTd6vOU2ZMkX/aQw/vSCi1qKm8xnPZc2Hn58f9u3bh19//RV33HEH2rZti4KCApSUlCAkJAS33347vvjiC/2n9PXx2GOP4e+//8aIESPg5OQEtVqtn8Hk1KlTiIyMNMtr6dWrF86fP4/58+djwIABcHZ2Rl5eHry9vTFixAgsWLAA06dPN8uxqjJq1Ch4eHgAYBNIahySqG5eFiIianH++OMPTJkyBfb29rh27Rr7MxAREbVCx48f1zeyvnjxIvszkNmxooGIqBX58ssvAQD33HMPkwxEREStlO56YNSoUUwyUKNgooGIqJVYunQpdu/eDZlMhueee87S4RAREZEFbNq0CT///DMA4IUXXrBwNNRSKSwdABERNZ5Dhw7h7rvvRk5ODrKzswEATzzxBLp06WLZwIiIiKjJJCYmYsiQISgoKNA3ML3tttss2qCaWjYmGoiIWrCioiJcvXoVcrkc7dq1w8yZM/HKK69YOiwiIiJqQmq1GlevXoUkSQgICMCUKVPw7rvvWjosasHYDJKIiIiIiIiIzIY9GoiIiIiIiIjIbDh0wkI0Gg2uXbsGZ2dn/Zz2RERErZUQAnl5efDz84NMZr7PQXi+JSIiuqGxzreVMdFgIdeuXUNgYKClwyAiIrIqCQkJCAgIMNv+eL4lIiIyZu7zbWVMNFiIs7MzAO0v2MXFxcLREBERWVZubi4CAwP150dz4fmWiIjohsY631bGRIOF6Mo3XVxceOFDRERUztzDG3i+JSIiMtbYwwnZDJKIiIiIiIiIzIaJBiIiIiIiIiIyGyYaiIiIiIiIiMhsmGggIiIiIiIiIrNhooGIiIiIiIiIzIaJBiIiIiIiIiIyGyYaiIiIiIiIiMhsmGggIiIiIiIiIrNhooGIiIioBUpUZWPn9WgkqrItHQoREbUyCksHQERERETmtezSYTxyYA00QkAmSVg6aApmd+hv6bCIiKiVYEUDERERUQuSqMrGw/u1SQYA0AiBRw+sYWUDERE1GSYaiIiIiFqQy7npEBAGz5UJgYOpcZYJiIiIWh0mGoiIiIhakDBnjyqfv3/Pb3j9xBbklRY1cURERNTaMNFARERE1IJklxgmEmSShPYuXijSqPHeqX/R/o+PsPTiIag1ZRaKkIiIWjo2gyQiIiJqQfYkXwEADPdth7d63IxwFy/4O7hiXfxZvHj0b0TnpePRA2uw6L+9+LTv7bgloKOFIyYiopaGFQ1ERERELcieFG2i4Wa/CIxoG44ARzdIkoRJwZE4N+kFfNFvAjxsHXAuOwXjtn2HsVuX4kzmdQtHTURELQkTDUREREQthBACu8srGob5tjNabiNX4JkuQxF95zw832U4lDI5/rl2CT3WL8DD+1fjekFuU4dMREQtkNUnGlavXo0RI0bA3d0djo6O6N69Oz7++GOUlpaatJ8DBw7giSeewMCBA+Hv7w87Ozs4Ojqic+fOmDNnDuLi4qrcbsSIEZAkqdavBx980AyvloiIiKj+LuWmIbUoH3ZyBfp6BVa7nrutAz7tdzsuTH4RU0O6QSMEvrt0GO3/mI93Tv4DVWlxE0ZNREQtjVX3aHj22WexcOFCKBQKjBo1Ck5OTtixYwdeeuklbNiwAf/88w/s7e3rtK9Nmzbhm2++QVBQECIiIuDj44OcnBycOHECixcvxvLly7Fx40aMGDHCYLtbbrkFISEhVe6zpKQEv/76KwBg5MiRDXmpRERERA2m688wwDsYtvLaL/PaOXti1cj7cSAlDs8f3YBDaVfxZtQ/WHLxEN7vNQ4zwnpDLrP6z6WIiMjKSEIIUftqTW/dunWYNGkSnJycsHv3bvTq1QsAkJ6ejlGjRuHMmTN4/vnn8emnn9Zpf+fPn4e9vb1R0qCkpAQvvvgiFi5ciICAAMTFxUEul9dpn6tWrcJdd90FV1dXXL9+vc5JDwDIzc2Fq6srcnJy4OLiUuftiIiIWqLGOi+2tvPtjD0r8XPMCbzR4ya83XOsSdsKIbA67hTmHduE2PxMAEAPDz982vd2jPZr3xjhEhFRE2uq86LVpqg/+OADAMC8efP0SQYA8PLywtdffw0AWLx4MXJycuq0v06dOlVZmWBjY4NPPvkEdnZ2SExMxH///VfnGJctWwYAuOeee0xKMhARERE1Bl1FwzAf4/4MtZEkCdNCe+D85BfxSZ/b4Gpjh5OZ1zBm6xLctm0ZzmenmDtcIiJqoawy0ZCUlISjR48CAKZPn260fMiQIQgMDERxcTE2bdrU4ONJkgRZeVmgra1tnbZJSEjAv//+CwCYPXt2g2MgIiIiaoir+ZmIV2VDIckwwDuo3vuxlSvwQuQIRN/5MuZ0GgKFJMPfiecRue4zPHHwD6QW5pkxaiIiaomsMtEQFRUFAPDw8EBoaGiV6/Tp08dg3foqKyvD22+/jYKCAnTu3Bnh4eF12m7FihXQaDTo1q2bPhYiIiIiS9mTHAsA6OMVCEdl3T44qYmXnSMWDZiIc5P+DxOCuqBMaPDNhYMI/2M+5p/egUK1aY25iYio9bDKZpCxsdoTZVBQ9dn4wMBAg3XrKj4+Hm+88QYAIDMzE1FRUUhMTER4eDhWrVqlr2yoiRACK1asAMBqBiIiIrIOu5NjAADDfKr+kKa+Orh6Y93oB7A7OQbPH9mA4xmJePn4Jnxz4QA+6D0e97TrAZlklZ9dERGRhVhloiEvT1uS5+joWO06Tk5OALTNLEyRmZmJH374weC5Xr164fvvv0eXLl3qtI9du3bhypUrsLW1xX333VenbYqLi1FcfGOqKFPjJiIiotq15vPtnpTy/gy+pvdnqIvhvmE4cvvTWHklCq8c34x4VTbu27MSC//bi8/63o6hjXRcIiJqflpd+rlHjx4QQkCj0SAxMRGrVq1CQUEBevfujUWLFtVpH7omkBMmTICHh0edtvnwww/h6uqq/9JVZBAREZH5tNbz7fWCXFzOTYcECUPMXNFQkUyS4b6w3rg4+SW832scnBS2OJqegGGbv8bk7StwOSet0Y5NRETNh1UmGpydnQEAKpWq2nXy8/MBoN5TckiSBH9/f0ydOhUHDx6Ej48P5s6di1OnTtW4XU5ODv78808Apg2bePnll5GTk6P/SkhIqFfcREREVL3Wer7dW17N0MPDD642jT8Tlr1CiVe6j0b0lHl4NGIAZJKEtfFn0XntJ3jm0DpkFFV/DUdERC2fVSYadNNQ1nRxoFtW1ZSVpnJzc8OkSZOg0Wiwfv36Gtf99ddfUVhYiKCgIIwZM6bOx7C1tYWLi4vBFxEREZlXaz3f6qe1bOLhCz72zvh20BScmfg8xgd0hFposOj8PoT/MR+fnd2F4jJ1k8ZDRETWwSoTDT179gQAZGRkVNvs8dixYwC0/RXMQdcPIjU1tcb1vv/+ewDAAw88UKfGkURERESNbU+K9nppmI9l+iR0dvPF3zc9hG1jH0E397bILinEC0c3ovPaT7A69hSEEBaJi4iILMMq75QDAgLQt29fAMDKlSuNlu/btw8JCQmwtbXF+PHjzXLMHTt2AAA6dOhQ7Tpnz57F0aNHIUkSHnjgAbMcl4iIiKghMosLcCbrOgBgqG/j9WeoizF+HXDijrn4fsg0tLV3wZW8DEzb9RMG/70YB1PjLBobERE1HatMNADAK6+8AgCYP38+Tpw4oX8+IyMDTzzxBADgqaeegqurq37Z2rVr0bFjR4wePdpofx9++CHS0owbFGVlZWHOnDk4duwYXF1dMW3atGpj0jWBHDNmDIKDg+v3woiIiIjMaG/5sIlOrm3gbedk4WgAuUyGB9r3w+U7X8JbPW6Gg0KJg2lXMejvxbhr50+IzcuwdIhERNTIrHJ6SwCYOHEinn76aSxatAgDBgzA6NGj4ejoiO3btyM7OxuDBw/Gu+++a7BNTk4OLl68iKKiIqP9vfLKK3j99dcRGRmJsLAwKBQKJCUlISoqCiqVCq6urli9ejV8fHyqjKe0tBQ///wzANOaQBIRERE1Jt20lsN9wywciSFHpS3e7HkzHo7oj9dPbMXyy0exKu4U1sWfxdOdh+DVbmPgZtv4jSuJiKjpWW1FAwAsXLgQv//+OwYOHIgDBw5g06ZNCAgIwPz587Fjxw7Y29f95LR48WJMmTIFBQUF2L59O/744w+cO3cOkZGRePPNN3Hx4kXcdNNN1W6/fv16pKenw8PDAxMnTjTDqyMiIiJqOF2ioakbQdaVn4Mrlg2ZhqgJczHGrz1KNGX49OxuhP/xIb78bx9KNWWWDpGIqFEkqrKx83o0ElXZlg6lyUmC3XksIjc3F66ursjJyWk1HbGJiIiq01jnxZZ+vs0rLYLbL69DIwQSpr2GAEc3S4dUIyEENidewP8d24j/slMAAO1dvPBxn9swIagLJEmycIREROax7NJhPHJgDTRCQCZJWDpoCmZ36G/psJrsvGjVFQ1EREREVL0DqVehEQLtnD2tPskAAJIkYXxgJ5ya8By+HXgn2tg54XJuOibtWIGRW77BsfTqpzYnImouElXZeGS/NskAABoh8OiBNa2qsoGJBiIiIqJmak95I8hhPpadbcJUCpkcj3YciMt3zsMr3UbDTq7A7uQr6LthIWbsWYmE/GxLh0hEVG/ns1OhgeHAgTIhEJ2bbqGImh4TDURERETN1O7kGADW25+hNi42dni/9zhcmjwP94X1AgD8HHMCHf6cj1eOb0JuiXGDbyIia/d3wn9VPl+q0TRxJJbDRAMRERFRM1SoLsWR8qEGw3yaZ6JBJ9DJDT8Nm46jtz+DYT7tUFSmxoend6D9H/Px7YUDULNhJBE1E4fTrmLxhQMAABkM+87csf17fHfpMFpDm0QmGoiIiIiaocNpV1GqKYO/gyvaOXtaOhyz6OMViF3jHse6UbPQwcUbqUX5ePzgn+j+1wJsSjjfKi7Oiaj5KlCX4P49v6FMaHBPu564Ou1V7LzlMZya8BzGB3REUZkaD+9fjfv2rEReacuu2GKigYiIiKgZqjitZUuarUGSJEwI7oqzk17Aov4T4WnrgP+yU3Drv8tw8z9LcSrzmqVDJCKq0kvH/sal3DT4ObjgqwGTEODohhFtw9HNww8bxjyIj/rcCrkkw8orUeizfmGL/v+MiQYiIiKiZmhPciyA5tcIsq6UMjnmdB6C6Dtfxv91HQEbmRz/XruMnn99jgf3/Y5rBTmWDpGISG9b0iUsPr8fALB8yF1wt3UwWC6TZHgxciR2j3scAQ6uuJSbhv4bF2HpxUMtslqLiQYiIiKiZqakTI0DqXEAmm8jyLpys7XHx31vw4XJL+Ku0B4QEFh++Sja/zEfb0Vthaq02NIhElErl1VcgAf2/Q4AeLLjINzsH1HtuoN9QnFywnO4NaATisvUePTAGkzf/UuLa37LRAMRERFRM3MiIwmFZaXwsnVEJ1cfS4fTJEKdPfHbiPtw8NY5GOgdjAJ1Kd4+uQ3t//gI3186grJW1M2diKzLU4fWIqkgB+1dvPBRn1trXd/TzhHrxzyAT/rcBrkkw2+xJ9Fnwxc4mZHUBNE2DSYaiIiIiJoZ3bSWQ31CW1R/hroY0CYY+299CqtGzECokweuF+Zi9v5V6LX+c2xLumTp8IiolVkVexIrr0RBJkn4adg9cFTa1mk7mSTDC5EjsGfcEwh0dMPl3HQM+PtLfHvhQIsYSsFEAxEREVEzU7ERZGskSRKmhnbH+ckv4rO+t8PNxh6ns67j5n+WYvw/3+FcVrKlQySiVuBaQQ4eP/gnAOCVbqPR3zvY5H0M8glB1B1zcVugdijF4wf/xD0tYCgFEw1EREREzUiZRoN9KXEAgOG+YZYNxsJs5Qo813U4ou+ch2c6D4VCkmFz0gV0++szPHZgDVIK8ywdIhG1UEIIPLRvNTKLC9DL0x+vdx9T73152jli/egH8Wnf26CQZPg99iR6r/8CWxIvYOf1aCSqss0XeBNhooGIiIioGTmddR25pUVwUdqhm3tbS4djFTztHPFF/wn4b9L/YXJwJDRCYMnFQwhfMx8fnNqOQnWppUMkohZm6cVD2Jx0AbZyBX4adg9s5IoG7U+SJDzfdQT2jn8SQY5uiM5Lx7ht32HUlm8RvPp9LLt02EyRNw0mGoiIiIiakT3J2mETQ3xCIJfxUq6i9q7e+GPUTOwZ9wT6egUiX12MV09sRoc/5uOn6OPQCDaMJKKGi85Nx3NH1wMAPuw9Hp3dfM227wFtgrHxptkGz2mEwKMH1jSrygaenYiIiIiaEX1/Bp/W2Z+hLob6tsOh2+bgl2HTEeTohsSCHNy/91f027BI30iTiKg+yjQa3L/nVxSoSzHCNwzPdB5i9mOkF6mMjysEonPTzX6sxsJEAxEREVEzIYTQVzS01kaQdSWTZJge1gsXJr+ED3uPh7PSFsczEjFi8zeYuH05LuWkWTpEImqGPj67EwfTrsJZaYsVQ++CTDL/LXV7Fy/IKs0oJJckhLt4mf1YjYWJBiIiIqJm4nxOCtKLVbCXK9HbM8DS4TQL9gol5nUbheg75+GJjoMgl2T4K/4cuqz9BE8fWlflJ4fUeBJV2c22uR3RyYwkvBn1DwBgUf+JCHbyaJTjBDi6YemgKZCXJxvkkoQlg6YgwNGtUY7XGBrWsYKIiIiImoyummFQm5AGNx5rbdrYO+OrgZPxVKfBePHYRmxMOI8vz+/DjzHH8Gq30ZjTaQjsFEpLh9miLbt0GI8cWAONEJBJEpYOmoLZHfpbOiyiOilSl2LGnl9RqinDxKCumBnep1GPN7tDf4z1j0B0bjrCXbyaVZIBYKKBiIiIqNnYkxILABjmG2rhSJqvTm4+2DBmNnZcu4znj27AycxrePHY3/j6wgHM730rpoV2h1SpZLm10wgNVOoS5JUWI7ekGHmlRcgtLUZe+VduaZHhvyW677Xr5pUWI6u4EMlFeRX2KfDQ/tXYnRyDbu5+aOfsiTAXT7Rz9oCz0s6Cr5aoam9EbcXZ7GS0sXPC0sFTmuT/iQBHt2aXYNBhooGIiIioGTDoz8BGkA02yq89jt3+LH6KOY5XT2xGXH4W7t79Mz7/bw8W9L0Dg3xCLB1ig+iSA9rEgHEyoPrEgPE6+aUlEBCNEudPMScAnDB4ztvOEe2cPbXJB2dt8iGs/Hs/B5dGGRNPVJM9yTH49OxuAMD/Bk+Ft52ThSOyfkw0EBERETUDsfmZSCrIgVImR3/vYEuH0yLIZTLMat8XU0O6YcG5PfjozE4cTovH4E2LMSWkG+b3Ho8wFy8kqrJxOTcd7Ru5fFkjNMgvLamxSiBPbVxVUFUSIa+02OzxySUZnJW2cFbawkVpC2elXfm/tnCxsYOzovxf/Tp2+nUL1KW4/d/voamQsJBBwpMdByGtWIUreRm4kpeJ9GIV0oq0X4fT4o1isJUrEOrkgXbOHhUSEdp/Q5094KCwMfvrptYtr7QIM/f+BgGBB9v3wx1BXSwdUrPARAMRERFRM6CrZujnFQh79hIwK0elLV7vcRMe6tAfb0RtxfeXj2BN3Gn8FX8OI3zbYfu1aGhQdV8BXXLAODFQhyEGBusWI1/dOMkBl8o3/tUkAyp+X9X69nJlg8rFlw6egkcPrEGZEPrmdpV7NOSUFCI2LxMxeRn65IPu8dX8LBSXqXEhJxUXclKrPEZbe5cqkxDtnD3gY+/MYTFksrmH1yMuPwshTu74vN8dlg6n2ZCEEI1TB0U1ys3NhaurK3JycuDi4mLpcIiIiCyqsc6LLel8+8De37Ai+hhe7jYKH/Qeb+lwWrQzmdfxwtEN+OfapSqXhzp5oKhMjdzSIqjUJWY/vkKS3UgGVKgSqFxFUFUyoHKiwE6usKqb60RVdr2b26k1ZUhQZRskHyr+m1NSVOP2Dgol2jl53ugH4eShfxzi5AFbNlilStbHn8OE7cshQcLOcY9huG+YpUNqsKY6L/KviYiIiKgZ0DeCZH+GRhfp0RZbxz6Cj07vwLzjm4yWx+ZnGj2nSw7UmgwwGl5QaQiC0g62VpYcMKeGNLdTyOQIdfZEqLMnRqO90fLM4gJt0iE3A1fyy6shctNxJT8TCapsFKhLcTY7GWezk422lSAhwNG1mmoIT3jaOrTY3wlVLa0oHw/vXw0AeK7LsBaRZGhKTDQQERERWblEVTau5GVAJkkY3MybFDYn94b1wisnNkMjDPsKrBoxA2EungZVBy05OdBceNg6wMPWAX28Ao2WlZSpcTU/C1fyy5MPlaoiVOoSJKiykaDKxu7yYUoVuSjtKiUhbjwOcnKHUiZvipdITUQIgUf3r0FqUT66uPngvV63WDqkZoeJBiIiIiIrt7e8mqGXpz+n/mtCAY5uWDrIuK/AnaHdLB0amchGrkB7V2+0d/UG/CMMlgkhkFaUb5B8uDEsQ9uENbe0CCczr+Fk5jWjfcslGYIc3cpnyrgxQ4YuEeFma99UL5PM5KeY41gbfxZKmRw/DZsOO/bFMRkTDURERERWjtNaWs7sDv0x1j+i3n0FyPpJkoQ29s5oY++MAW2MZ3QpVJciLj/TIPkQk6etiriSl4GiMjVi8zMRm5+J7deN9+9uY1/eE8KwP0SYiycCHNwgl3G6TmsSn5+FOYfWAQDe6nEzenr6WzagZoqJBiIiIiIrtyeFiQZLakhfAWr+7BVKdHLzQSc3H6NlGqFBcmGeQT+IG30iMpFSmIeskkIcS0/EsfREo+2VMjmCHd21yYdK1RDtnD1qrGBqqmlXWxON0GDW3t+QW1qEAd7BeDFyhKVDaraYaCAiIiKyYmlF+fgvOwUAMMQn1MLREFFFMkkGPwdX+Dm4Vvn3mV9ajNi8TIMZMnRDNGLzM1GqKUN0Xjqi89Kr3L+3nSPCnL2M+kMcSUvAS8f/hkZUPe0qmS5RlY0PT+/AzuQYOCiU+HHo3VCw90a9MdFAREREZMX2Jmv7M3R184WnnaOFoyEiUzgpbRHp0RaRHm2NlpVpNEgqyDFIPlRMSGQUFyCtSIW0IhUOpV2t9hgaIfDI/jUY5tNO24OCTLbs0mE8sn8NNNA2fp0UFMmfZQMx0UBERERkxXTDJob7ctgEUUsil8kQ5OSOICd3jDDOQyCnpFDfB6JiNcTZrGRcL8w1WFcDgS7rPsFI33CMD+iEcQEd0d7FizOh1MGF7BQ8vH8NBG7MLvNbbBTm9xnPISkNwEQDERERkRXTN4JkooGoVXG1sUdPT3+jZoSJqmwEr37fYNpVACjVaPDPtUv459olPHvkL7Rz9sQ4/wiMD+iEEW3D4KCwacrwrVahuhQHUuOw83o0dlyPxuG0eIMkAwCUCYHo3HQmGhqAiQYiIiIiK5VTUqifTm8oG0ESEaqedvXbgVMw2CcEmxLPY3PiRexJuYIreRn46sIBfHXhAGzlCozwDcM4/44YH9CxVQ0LKClT43BaPHYmaxMLB1OvokRTVuM2cklCuItXE0XYMjHRQERERGSl9qfEQUCgvYsX2jq4WDocIrIS1U272snNB893HYG80iLsuBaNzUkXsDnxAuJV2diadBFbky7i2SN/IczZE+MCOmKcf8cWV+2g1pThREYSdpRXLOxPjUWButRgHT8HF4xqG45RbcMx0jcc269fNkjcLBk0hdUMDcREAxEREZGV4rSWRFSdmqZddVbaYUJwV0wI7gohBM7npGBz4gVsSryAvSmxiMnLwOLz+7H4/H7Y6aodyhMPza3aQSM0OJ15HTuuR2NncjR2J19BXmmxwTredo4Y6VueWGgbbtS/YrZz1Ykbqj8mGoiIiIisFPszEFFDSZKEzm6+6Ozma1TtsCnxAhJU2diSdBFbki7iGWirHbQNJSMwwjcc9gqlpV+CAV3iZMe1aOxMjsGu5BhkFhcYrONmY48RvmEY2TYMo9qGo4ubb62NMWtK3JDpmGggIiIiskKq0mIcTU8AwIoGIjKfytUO/2Xrqh3OY19qHGLyMvDl+X348vw+g2qH8QGdLNK3QAiBmLwMffPGnckxSCnMM1jHSWGLYb6h+qqF7h5+kMtkTR4r3cBEAxEREZEVOpQWD7XQIMjRDSHOHpYOh4haIEmS0MXdF13cffFCpLbaYfu1aGxOvIDNSZWqHQ7/hXBnr/KkQ0cM9w1rtGqH+Pwsg8RCgirbYLmdXIHBbULLh0KEoY9XIJQyeaPEQvXDRAMRERGRFdL3Z+CwCSJqIs5KO0wM7oqJ5dUO57KTtUmH8t4O0XnpBtUOI33Dtb0dAjo2qNohuSAXO5NjtImF69GIycswWK6UyTHAO0jfY2GAdzBs5byVtWb87RARERFZIX1/Bg6bICILkCQJXd3boqt7W/xf5EjklhRh+/XL+sRDYkGOdlaLpAvAYaC9ixfG+WuTDhWrHRJV2bicm472FZosZhSpsCs5Rl+1cD4n1eDYMklCX69A/VCIQW2C4ai0beofATUAEw1EREREVqa4TI1DaVcBsKKBiKyDi40dJgVHYlJwJIQQOJuVrJ8+c19KLC7npuNy7j4sOr8P9nIlRrYNg5vSDr/FnoIGAhIkjPELR2qRCqcyrxnsW4KEHh5++uaNQ33awcXGzkKvlMyBiQYiIiIiK3MsPQFFZWq0sXNCB5fmNdUcEbV8kiQh0qMtIj3a4sUK1Q6byqsdkgpysCnxgsE2AgLbrl3Wf9/ZzQej2morFob7hsHD1qGpXwY1IiYaiIiIiKxMxf4MtU3JRkRkaVVVOyw+vw9LLx02Wvf17mPwRMdB8HVwsUCk1FQ45wcRERGRldld3p9hOPszEFEzo6t2eL3HTZBVSpTKJQmPRAxgkqEVYKKBiIiIyIqoNWXYnxIHgP0ZiKj5CnB0w9JBUyAvTzbIJQlLBk3RN4Sklo1DJ4iIiIisyMnMa8hXF8PNxh5d3X0tHQ4RUb3N7tAfY/0jEJ2bjvAKs05Qy8dEAxEREZEV0U1rOdQnFDKJxadE1LwFOLoxwdAK8exFREREZEX0jSDZn4GIiJopq080rF69GiNGjIC7uzscHR3RvXt3fPzxxygtLTVpPwcOHMATTzyBgQMHwt/fH3Z2dnB0dETnzp0xZ84cxMXF1bqPPXv24O6770ZAQABsbW3h5eWF3r17Y+7cuSbHQ0RERFSZRmiwNyUWAPszEBFR82XViYZnn30W06ZNw/79+9GvXz/ccsstiI+Px0svvYRRo0ahsLCwzvvatGkTvvnmG1y7dg0RERGYNGkShg8fjszMTCxevBhdu3bFrl27qtxWCIFnn30Ww4cPx59//onQ0FDceeed6NWrF1JTU/HFF1+guLjYTK+aiIiIWqtzWSnILC6Ao8IGPT39LR0OERFRvVhtj4Z169Zh4cKFcHJywu7du9GrVy8AQHp6OkaNGoV9+/bh9ddfx6efflqn/d1777146KGHEBISYvB8SUkJXnzxRSxcuBAzZsxAXFwc5HK5wTpvvfUWFi5ciEGDBmHlypUIDg42WH706FHY2dnV/8USERER4cawicFtQqCUyWtZm4iIyDpZbUXDBx98AACYN2+ePskAAF5eXvj6668BAIsXL0ZOTk6d9tepUyejJAMA2NjY4JNPPoGdnR0SExPx33//GSy/ePEiPvjgA/j4+ODvv/82SjIAQN++faFQWG3OhoiIiMwoUZWNndejkajKNvu+dY0gOWyCiIiaM6tMNCQlJeHo0aMAgOnTpxstHzJkCAIDA1FcXIxNmzY1+HiSJEEm0/4obG1tDZZ98803UKvVePjhh+Hm5tbgYxEREVHztfDcHgSteg+jtnyL4NXvY9mlw2bbtxCCjSCJiKhFsMpEQ1RUFADAw8MDoaGhVa7Tp08fg3Xrq6ysDG+//TYKCgrQuXNnhIeHGyzfunUrAGDYsGHIzs7GkiVL8OSTT2LOnDlYsmQJ0tPTG3R8IiIiah4SVdl47sgGiPLvNULg0QNrzFbZEJ2bjuTCPNjKFejrFWiWfRIREVmCVdb7x8Zquy0HBQVVu05gYKDBunUVHx+PN954AwCQmZmJqKgoJCYmIjw8HKtWrdJXNgDa/g0XL17UH+e+++5Damqqwf5eeOEF/O9//8Pdd99d43GLi4sNGkbm5uaaFDcRERHVrjHPt5dz06HRpxm0yoRAdG66WeaI11Uz9PcKgp1C2eD9ERERWYpVVjTk5eUBABwdHatdx8nJCYDpFxCZmZn44Ycf8MMPP2DDhg1ITExEr169sGbNGnTp0sVoXSG0FxRPPfUUfH19sWvXLuTm5uLChQuYNWsW8vPzcd9992Hv3r01HvfDDz+Eq6ur/kuXKCEiIiLzaczzbXsXL8ggGTwnlySEu3iZZf/sz0BERC2FVSYaGlOPHj0ghIBGo0FiYiJWrVqFgoIC9O7dG4sWLTJYV5dkAAB7e3v8+++/GD58OJydnREREYHly5dj3LhxKCsrw1tvvVXjcV9++WXk5OTovxISEhrj5REREbVqjXm+DXB0w9s9x+q/l0sSlgyaYpZqBgDYzf4MRETUQlhlosHZ2RkAoFKpql0nPz8fAODi4lKvY0iSBH9/f0ydOhUHDx6Ej48P5s6di1OnThnFAQCTJ0+Gt7e30X6eeOIJAMDevXtRUlJS7fFsbW3h4uJi8EVERETm1djn22mh3QEAjgol4qa+itkd+ptlv1fzM3E1PwsKSYZBbYxnuCIiImpOrDLRoJuGsqZPIXTLqpqy0lRubm6YNGkSNBoN1q9fr3/eyclJn1xo167qTxd0z5eWlrIxJBERUQsnk6Tyf2Vmq2QAgL0p2p5Tvb0C4Ki0rWVtIiIi62aViYaePXsCADIyMqpt9njs2DEAQK9evcxyTF0/iMrNHnv37g0A1SYRKj6v6xtBRERELZMu0aARopY1TaPvz8BhE0RE1AJYZaIhICAAffv2BQCsXLnSaPm+ffuQkJAAW1tbjB8/3izH3LFjBwCgQ4cOBs9PnTpVv1yj0Rhtt23bNgBAREQEh0MQERG1cLpmkOZONGy/fhkA0NnNx6z7JSIisgSrTDQAwCuvvAIAmD9/Pk6cOKF/PiMjQ98X4amnnoKrq6t+2dq1a9GxY0eMHj3aaH8ffvgh0tLSjJ7PysrCnDlzcOzYMbi6umLatGkGy++77z6EhYXh7NmzeOONNwySDTt37sSCBQsAAE8//XQDXi0RERE1B/qKBpgv0fD52d24kpcJAJi9bxWWXTpstn0TERFZgiSEmVPyZvTMM89g0aJFUCqVGD16NBwdHbF9+3ZkZ2dj8ODB2LZtG+zt7fXrr1ixAg888ACCg4MRFxdnsC9JkiCXyxEZGYmwsDAoFAokJSUhKioKKpUKrq6uWL16NW666SajOE6fPo2RI0ciMzMTYWFh6NGjB5KSknDkyBFoNBrMnDkTy5cvhyRJRttWJzc3F66ursjJyWElBBERtXqNdV40934TVdkIXPUelDI5SmZ+ZJb9Ba16H6JC4kIuSYib+qpZe0AQEREBTXcfqmi0PZvBwoULMXjwYHz11Vc4cOAASktLERYWhnnz5mHu3LmwsbGp874WL16MvXv3IioqCtu3b0d+fj6cnZ0RGRmJsWPH4vHHH4ePT9Xlit26dcPZs2fx/vvv4++//8aGDRvg4OCA4cOH45FHHsHdd99trpdMREREVszcPRou56YbJBkAoEwIROemM9FARETNllVXNLRkrGggIiK6oblUNKQU5sH3t7chQYLmgU8avD9tRcN7BqkGVjQQEVFjaar7UKvt0UBERERkbXQVDQIC5visJsDRzaABpFySsGTQFCYZiIioWbPqoRNERERE1kQ36wSgTTZIqHt/puqohbbR9Gd9b8e00O5MMhARUbPHigYiIiKiOpJVaPxsjj4NGqFBXH4WAGBScFcmGYiIqEVgooGIiIiojsydaEguzENxmRpySYZAJhmIiKiFYKKBiIiIqI7MnWiIy9NWMwQ4ukIhkzd4f0RERNaAiQYiIiKiOqrYo0GDhicaYvMzAQChTh4N3hcREZG1YKKBiIiIqI5k0o1LJ3NUNMTmlScanJloICKiloOJBiIiIqI6MvvQifKKhhAn9wbvi4iIyFow0UBERERUR+ZONNwYOuHZ4H0RERFZCyYaiIiIiOrI7D0aOHSCiIhaICYaiIiIiOrInBUNak0ZElTZADh0goiIWhYmGoiIiIjqSDJINGgatK+kghyohQY2Mjn8HFwaGhoREZHVYKKBiIiIyAS6qoaGVjTohk0EO7kbzGZBRETU3PGsRkRERGQCXZ+GhvZoiNXPOMH+DERE1LIw0UBERERkAnNVNMTlZwEAQploICKiFoaJBiIiIiITmHvoBGecICKiloaJBiIiIiIT6BINZQ1sBnlj6ARnnCAiopZFYeoGR48exfXr19GpUye0b9++xnUvXbqECxcuwM/PD3369Kl3kERERETWQlb+OU2Dh06wooGIiFookxIN6enpGD16NJydnXHy5Mla13d3d8cTTzyBgoICXLlyBW5ubvUMk4iIiMg6mGPoRHGZGkkFuQDYo4GIiFoek4ZO/Pzzz8jPz8fbb78Nb2/vWtf39vbGO++8g+zsbPz888/1DpKIiIjIWugTDQ2YdSI+PwsCAg4KJbztnMwVGhERkVUwKdGwadMmODo6YubMmXXeZsaMGXBycsLGjRtNDo6IiIjI2pijokE340SIkwek8v0RERG1FCYlGs6ePYv+/ftDqVTWeRulUol+/frhzJkzJgdHREREZG1kaHiiQdcIksMmiIioJTIp0ZCZmQlfX1+TD+Lj44OMjAyTtyMiIiKyNuaoaODUlkRE1JKZlGiwtbWFSqUy+SAFBQWwtbU1eTsiIiIia2OOHg1xnNqSiIhaMJMSDb6+vjh9+rTJBzl9+nS9KiGIiIiIrI1ZKhr0Qyc8zRITERGRNTEp0TBo0CDExcXhwIEDdd5m//79iI2NxaBBg0wOjoiIiMjamKVHA4dOEBFRC2ZSouHee++FEAKPPPIIcnJyal0/OzsbjzzyCCRJwj333FPvIImIiIisxY2KBk29ti9QlyC1KB8Ah04QEVHLZFKiYcyYMRg9ejT+++8/9O7dG+vXr4eoIpsvhMBff/2FPn364MKFCxgxYgRuvvlmswVNRGRuidmF2BmdjsTsQkuHQkRWrqE9GnT9GVxt7OBu62C2uIiIiKyFwtQNfvvtNwwePBiXLl3CpEmT4Obmhl69eqFNmzYAgNTUVJw4cQLZ2dkQQiA8PBy///672QMnIjKX/x26isfWnIZGADIJWDSxK54YHMK57YmoSjJJ+zlNfYdO6IdNcGpLIiJqoUxONHh6euLIkSN46qmn8OuvvyIrKwvbt2/XX5DrKhxkMhmmT5+OL7/8Em5ubmYNmoioIYQQ+C8lHzuj07HpfCo2X0jVL9MI4Km1Z/H0urNwt1fCrfyrusdu9ooql9kp5RZ8hdRSJGYX4nK6Cu29HBHgZm/pcKhcQ3s0xOVnAQBCmGggIqIWyuREAwC4uLjgxx9/xNtvv42NGzfi2LFjSEtLAwB4e3ujd+/euO2229CuXTuzBktEVB9CCFxIzcfO6AzsiknHrpgMpOWX1LiNRgAZBaXIKCit1zFtFbIqEhI2cLNX1Ji4cLdXwtVOAYXcpJFt1MIIITB/RzRe3XQBAtpKm6VTu2N2/yBLh0Zo+KwTbARJREQtXb0SDTqhoaGYM2eOuWIhIjILIQQupamwKya9PLmQgZS8YoN17BQyDA71QC9/F3y2+wo0Fe4X5BJw6JmhsFfKkV1Yqv/KqvKxGlmFJcguVGu/LyqFEECxWoPkvGIkVzpuXTnZyuFmp4S7Q3kSoqrHduUJjArPu9kr4WyrgEzGYR/NhRACsZkFOJ6Yg2MJ2TiemIOjCdnILVLr19EI4NHVpzE2wpuVDVagoT0abkxtyUQDERG1TA1KNBARWQMhBKLTVdgVk6GvWriea3iDb6uQYVCIO0aGe2FEmCf6BbnBVqEd3hDRxhmPrj6NMiEglyQsmdoNfQLd6hWLRiOQV6w2SkwYJCSKSpFVUKpPTFR8nF9cBgDILy5DfnEZEnOKTI5BJgGu5UkHg4RENRUVht8rYK+UN6g/Bcv9qyeEQHxWIY4lZuNYQg6Ol/+bVVh75UxZ+fucP1PLa2hFg64ZJGecICKilsqkREN8fHyt60iSBCcnJ7i78+RJRI1DCIErGQXliQXtUIikSjfkNnIZBoa4Y2SYJ0aEe6J/kHu1fRNm9w/C2AhvRKerEN7Am2OZTIKrvRKu9koE12P70jINcgpLkV2kNkhAZBeWJySMHquRVVCiX7+kTAONALLKEx2xmabHYCOX1TrEo7rExdoz1/H4H2f0jTVbc7m/EAKJ2UU4lpitr1Y4lpBd5XAcpVxCdz8X9A5wQ58AVwS62WP8d4cNKm0AIIhJBqvQ0B4NHDpBREQtnUmJhtDQ0Dqv6+TkhJEjR+LFF1/EoEGDTA6MiKii2IyCCkMh0pGQbZhYUMolDAh2x8gwL4wI98SAYHfYm9CQMcDN3io+KVbKZfBysoWXk229ti8sLbsxvKNCYqLyEI+qh4GUQiOAkjINUvNLkFpLH4vaaATw8KpTOJ6Qg65tnRHoZocgd3sEutnD3V7Zomb1EELgWm6RQZXC8cTsKn+GCpmEyLbO6BPohj4Bbugd4IqubZ31FTY6S6d211fa6Kw6dQ3zRrdv9NdDNWtIRUNOSSGySrTT6LIZJBERtVQmJRqECSfUvLw8rF+/Hhs3bsTChQvx5JNPmhwcEbVeVzMNKxauZhUaLFfKJfQLdNMPhRgY4g4HG44Gs1fKYa+Uo62LncnbCnFj2IdR74k69KjIK1Yb7xPANwfjjJ53tJEj0M0eQW7axIM2AWGn/b48GWFKoqipXc8tMqhSOJ6YU2U/DrlMQldfZ/QJcEOfQFf0DnBDZFvnOs1KUrHS5tS1HDz71394Z9sl3NPTH8EeDo3xsqiOGtKjQTfjhJetI5yU9UsoEhERWTuTrspjY2NrXUcIgfz8fFy+fBmrV6/G77//jrlz52Lo0KHo1q1bvQMlopYtIavQoHljbGaBwXKFTELfQDeMDPfEiDAvDApxh6MtEwvmJEkSXOyUcLFTIqgeo9/iMlQI+3CHQbm/JAGz+gQiq7AU8dmFiM8qRLqqBKqSMlxIzceF1Pxq9+flaKNNQLjaIcjdoTwpcaMqoq2LHeRN0PQyJa/YoErhWEIOruUa986QSUCX8qRC7wBX9Al0Qzc/lwYlTHSVNsPDPPHnmWTsuZKJZ/86h7UP9G3IS6IGulHRoDF5Ww6bICKi1sCkq/Tg4LqPOO7atSsmTZqEkSNH4tFHH8VXX32FJUuWmBwgEbVMSTmF2FWeVNgZnY6YDMPEglwmoU+Aq75iYXCoB5yYWLBqIZ6OBuX+usaalXs0FJaWISG7EAlZhYjPLkRCdhHiswqRkK37vhCqkjKkq0qQrirBicScKo8nl0nwd7W7URVRKRER5F7zEI2qmlam5RcbzP5wLCG7yoacMgno5OOMPgHaKoU+ga7o7ufSaFU1kiThq8mR6LlgD9adTcbf/6Xg1s4+jXIsql1Dhk5wxgkiImoNGv2q/eGHH8ZHH32E3bt3N/ahiMiKXc8tKk8saKsWLqerDJbLJKB3gK5iwRNDQj3hbMfEQnNTl8aa9ko5Ong7oYO3U5X7EEIgq7BUm3jIqjoRkZhThDKNdgaH+ErDaipysJHfSEC4OegTEWeT8/DFHu20phKAnv4uSC8orXJfkgR0bOOkrVIIcEOfQDf08HNp8oqarm1d8Oywdvh0VwzmrD2LUe29rHp4SUumLtPODpNaWH1FTnV0FQ2ccYKIiFqyJrlK6tWrFzZv3twUhyIiK5GSV4xd5f0Vdkan42KacWKhp/+NioWh7TzgYqe0ULRkTg1trClJEjwcbODhYIPufq5VrlOmEUjOKzJMRORokw66ZERafgkK6jBEQwA4kZSr/z7C21FfpaBNKrhaTdLrzZs74NeoJMRmFmD+9mi8fUuEpUNqdZZdOoyTWdcBALP3r4YGArM79K/z9rqpLTl0goiIWrImuXJSKBRQq42bhBFRy5GWX4xdMRnYFZ2BnTHpOJ9ieGMnSUAPP5cKiQVPuNkzsUD1ox02YQ9/V3sMrGadwtIyJFasiihPQJxMysGxKoZjfH5HZzzQLwiuVvy+dLJV4IsJXTD1x+OYvyMa9/X2R/tqKkPI/BJV2XjkwBr99wICjx5Yg7H+EQhwdKvTPm70aPBsjBCJiIisQpMkGqKjo+Ht7d0UhyKiJpKeX4zdV3SJhQycS84zWqe7n4u+eeOwdh5wd7CxQKTUWtkr5Wjv7WR0I56YXYjg9/41aFoplyRM6e5n1UkGnTu7tcXYCG9svZiGOWvPYvPD/VvUVKHW7HJuulFfhjIhEJ2bXqdEgxCCPRqIiKhVaPREw5EjR3D8+HFMnDixsQ/V6lXV2IzIXDILSrA75kbzxjPXjRMLkW2d9RULw9p5wtORiQWyPgFu9lU2rWwu/29KkoQvJ3VF1092Y+vFNPxx+jqmdPezdFitQnsXL8gkySDZIJckhLt41Wn7jOICqNQlAICgOlZAEBERNUeNlmjIycnBn3/+iZdffhkAMGvWrMY6FAFYdjgej6w+BY3Qjn1fOrW7Uad3IlNkFZRgz5VMffPG09dzUbnBehdfZ4wM88SIcE8Mb+cJLyfOCU/NQ12aVlqz9t5OmDcqHO9su4Rn/zqHsRFtrKaPREsW4OiGpYOm4OH9qyEASJCwZNAUE4ZNZAAA/BxcYKew/uoZIiKi+jLpqqRdu3a1riOEgEqlQkZGhv77u+++G7fffnv9IqRaJWYX6pMMAKARwKOrT2NshHezu3gmy8kpLMWeKzcqFk5eM04sdPJxwsgwL31ioY0zEwvUfDW0aaWlzRsdjp+OJyI2swDvbLuET27vbOmQWoXZHfrjn6RLWBV3Ci9GDjepESSHTRARUWthUqIhLi7OpJ27u7vjueee01c1UOO4nK4yGGsMlI8ZTVc164toaly5RaXYeyVTn1iISsoxeh9FeDvqh0KMCPeCDxMLRFbDXinH4sldcet3R/DFniuY2ScAXdu6WDqsVqGNvbbvh0IybXrRuPwsAEAIEw1ERNTCmZRoWL58ea3rSJIER0dHBAUFoWfPnlAoWMrZ2Np7OUImwaixWbiXo+WCIquTV6TG/rhM7IzWDoU4nphtlFho7+Wob944ItwTbV3sLBMsEdXJ+E4+mNjVF+vOJuPJP89g1xOD2BiyCTgqtP1ndP0W6urGjBNMNBARUctmUhZg5syZjRVHtVavXo2vvvoKp06dQklJCcLDw3Hvvfdi7ty5UCrrPr7xwIED+PnnnxEVFYX4+HhkZGRALpcjODgYo0ePxvPPP4+QkBCj7Xbt2oWRI0fWuO9vvvkGjz32mKkvzWwC3Ozx7LB2WLD7iv65b6dEspqhlVMV6xIL2uEQRxOyUVYpsxDm6VChYsET/q58zxA1N19M6IJ/LqVhz5VM/Hw8ETP6BFo6pBav3omGJh46wSbRRERkKU1SbqBWq7Fu3TpMmTLFpO2effZZLFy4EAqFAqNGjYKTkxN27NiBl156CRs2bMA///wDe/u6nTg3bdqEb775BkFBQYiIiICPjw9ycnJw4sQJLF68GMuXL8fGjRsxYsSIKrf38fHBLbfcUuWyiIgIk15XY+gf5G7wfXc/VwtFQk2l8gVkQYkaB+KysDM6HbtiMnAkPhvqSomFUA+H8ooFbdVCoDsvPImau2APB7w+pj1e3nQBL2z4D7d38YVbM5imsznTJRryTUw0xJUnGkKc3GtZs+HYJJqIiCypURMNZ8+exffff4+ff/4ZmZmZUKvVdd523bp1WLhwIZycnLB792706tULAJCeno5Ro0Zh3759eP311/Hpp5/WaX/33nsvHnroIaOqhZKSErz44otYuHAhZsyYgbi4OMjlxmMuO3bsiBUrVtQ5/qa28b9kg+9f3XQe/zw20ELRUGP7an8snl57FhoBSADCvBxxNasApWWGiYUgd3uMDPPUVy0EezhYJmAialTPDQ/DD8cScSE1H69tvoDFkyMtHVKLVlKqHZ6SXlBY5200QqPv0dDYQyeuZhbg4VWnoDsjaJtEn2KTaCIiajJmTzTk5eXh119/xbJly3Ds2DEA2pknvL29TdrPBx98AACYN2+ePskAAF5eXvj6668xdOhQLF68GK+//jpcXWv/9L5Tp05VPm9jY4NPPvkES5YsQWJiIv777z9ERjavC7TE7EL8fDzJ4Lltl9NxNbOAN5bNSH6xGil5xUjJK0ZyXjFS8is8rvD89bwiFJVq9NsJANHpKgBAgKsdRoZ76fsshHry90/UGtgoZPj6zkiM+uYgvj4Qhwf6BqJ3oJulw2qRlh2Oxyv/XAICgG3RyVh2OL5OlQLJhXkoLlNDLskQWMfpMOtj28U0PLbmNCq14EGZACavOIrpvQJwS4Q3Ito4sZ8HERE1GrMlGvbs2YNly5bhjz/+QGFhIYQQsLW1xe233477778f48aNq/O+kpKScPToUQDA9OnTjZYPGTIEgYGBSEhIwKZNm3DPPfc0KHZJkiCTyQAAtrbNr6v+5XSV0QUFAKw9ex3PDgtr8njoBlWx+kaiIN84aVDx+YKSsgYd6+fpPTG9lz8vHIlaqZHhXpje0x8ro5Lw+B9ncPDpIZDL+P+BOemmkxZO5ZWPsrI6Tycdl6etZghwdIVCZtpsFXVx9nou/m/jf9hyIa3adY4m5OBoQg7mAgh2t8ctHdvglghvjGrvBRc7DrchIiLzaVCi4fr16/jhhx/w/fffIyYmBkJob3d1N+0pKSlwcTF9qq2oqCgAgIeHB0JDQ6tcp0+fPkhISEBUVFSDEg1lZWV4++23UVBQgM6dOyM8PLzK9VJSUvDOO+8gKSkJdnZ26NixI2699VYEBVl+vGNVs04AwLGEHMsE1MKpitXa5EBuFVUHlb5XmZg8cLCRw9fZFj5OtvBxttU+1n05ab/XCIHhXx8wmmVkeJgnkwxErdynd3TGxvMpOJqQje8OX8WjA0MsHVKLop9OWmivcyDT1Hk66cZqBJmcW4Q3tl7EssPx0AhAKZfw5OAQhHo44Lm//kOZEJBLEt68uQMcbeXYciEVu2MycTWrEEsOXsWSg1ehkEkYHOqBWyK8cUvHNuju58LzCRERNYjJiYaysjJs2LABy5Ytw9atW1FWVgYhBDw8PDB9+nQ8+OCDmDNnDg4cOFCvJAMAxMbGAkCNN/GBgYEG69ZVfHw83njjDQBAZmYmoqKikJiYiPDwcKxatUqfJKnswoULePPNNw2eUygUmDNnDj7++ONap/EsLi5GcXGx/vvc3FyT4q5JgJs9lk7tjkdXn0aZEPqkw9//pWDbxVR08nHmmMxa6JIH1VUbVHy+PskDo8SBky18XYwTCk62dfuTrPj7lksSlkztxt8xEaGtix3evSUCz6w7h5f/voDJkW3h7dT8KvUaojHPt/rEvuZGRUNdp5M299SWqmI1Fuy5go92ROvPS3d2a4v5t3bSxzM5si2i01UIrzDrxHPDw6AqVmP3lQxsuZCGLRdScTldhd0xGdgdk4GXN12Ar7OtvtphTAdveDramCVmIiJqPUxKNPzf//0ffv75Z6SmpkIIAZlMhjFjxuDBBx/EpEmTYGNjnhNRXl4eAMDRsfoTt5OTEwDTLyAyMzPxww8/GDzXq1cvfP/99+jSpYvR+q6urnj22WcxadIkdOjQAS4uLoiJicHy5cuxePFifP7558jPz8fSpUtrPO6HH36It99+26RYTTG7fxDGRngjOl2FUA8HdPt0N7KL1Lh56eFW2226oESNlLwSJOcV3UgaVFOFkF9sWvLAXimDr7NdedLABr4udtVWITjZys3+yVDF33c4py0jogqeGBSC5UcScPJaLub9fR7L7uph6ZCaVGOebwPc7PHV5Eg8/vdB7RNSWZ0TveaacaJMI/DjsQS8tvkiruUWAQD6B7nhszu6YHCoYRIjwM2+ytgcbRUY38kH4zv5AABi0lXYelGbdNgenY7kvGKsOJqAFUcTIJOAfkHu+mqHPoFuHJJDRES1koRuvEMdyGQySJIEX19fPPbYY5g1a5a+sqCioUOH4sCBAygrq9+Y8w8++ACvvvoqBg8ejH379lW5zquvvooPPvgAN998M7Zu3WryMYQQuHbtGg4cOIA33ngDly9fxoIFC/D000/XeR9//vkn7rzzTgDa4R49evSodt2qPmEJDAxETk5OvSs/qpOYXYigd/816NsgAfh8Qmfc1KENIto4WewioaFzeuuSB9qkQVF50qCkPHGgSyhokwv1SR5oEwV28HGy0ScLtAkFm/IqBLtGSx4QEZnLwbhMDPpyPwBg31ODjW5ArVFubi5cXV0bfF5sivOtzWurUBp2BHYyJQpnflinbUZv+RY7rkfjx6H3YEZ473od999LaXhhw384dU37IUuIhz3mj++EaT38zHZOKlaXYX9sFrZcSMWWi6k4cz3PYLmHgxI3d9AmHcZGeMPXxc4sxyUioqZhrvNtberVoyEtLQ2HDx9GZGQk/Pz8qpwOsiGcnZ0BACqVqtp18vPzAaDePxxJkuDv74+pU6fipptuQpcuXTB37lwMHz4c3bt3r9M+Jk+ejB49euDkyZPYsGFDjYkGW1vbJms0WVVzSAHg2b/+A/AfHG3k6OHngl4Bbugd4IreAa7o2MYJCrnhsJGGJgUqq25O78LSsipnV7hRhVCElHxtMiGvuO5TpAKAnUJW5RAF46ELdkweEFGLMTDEA7P7BWHZkXg88ccZHJ871Oj/+Jaqsc+3yw7Ho7RU+7MsKivFd4eu4qEBwbVu15ChE+eS8/B/G/7D5gupAABXOwVev6kDnhoSAluFea/BbBVyjGrvhVHtvfDx7Z2RmF2or3bYdikNmQWl+O3kNfx28hoAoIefi3aYRUdvDArxgLKVvM+IiKhmJiUa1qxZg++//x5bt27Fpk2bsHnzZnh5eWHGjBl48MEH0blzZ7MEFRISAgBISEiodh3dMt26DeHm5oZJkybhq6++wvr16+ucaAC002aePHkSiYmJDY7DXKpqDikB6BPohv9S8qAqKcP+uCzsj8vSL7dXytDdz1WfeLiaVYB3t102SgoAgEYjUFhaBlVJGQp0/5aUoaBUXeFxmcHj6znF+PZgnMGc3g+tOoVn1p01ueeBnUJmNEShugaKzrYKJg+IqFWaf2tHrD17Haev52Lx/jg8O6ydpUNq9nSzTkAqv7mXgEf/OIlbOrapMSGv1pQhQZUNwLShE8m5RXhz60V8V97oUSHTNnp8/aYOTdY3IcDNHrP7B2F2/yCoyzQ4HJ+tr3Y4lpCDk9dycfJaLubviIazrQKj23thXHm1A6fYJiJqvUxKNEyePBmTJ0/GtWvXsHz5cixfvhxXrlzB559/js8//xz9+vXDAw88gIKCggYF1bNnTwBARkYGYmNjq5x54tixYwC0/RXMQdcPIjU11aTtMjIyANyowrAGlZtD6poFzu4fhDKNwKW0fBxPzMGJxBwcT8xGVFIu8orVOHQ1C4euZhntT5cUeGH9ORSpNShSa8wWqy7JoEseGFUdVOx3UL6MyQMiotp5Odli/q2d8Mjq03hjy0VM6+4HP1eWuTfEjVknblQRaKSyWmedSCrIgVpoYCOTw8+h9krMghI1Pttt2OhxcqQv5t/aCe29nRr8OupLIZdhcKgHBod64N1xHZGaV4xtl9Kw5WIqtl5MQ1p+CdadTca6s8kAgE4+TrglQlvtMKydJ+yU5p/Wk4iIrFO9hk74+fnh1VdfxauvvoqdO3fiu+++w9q1a3H48GEcOXJEv962bdswZswYk28KAwIC0LdvXxw9ehQrV67Eq6++arB83759SEhIgK2tLcaPH1+fl2Bkx44dAIAOHTrUeZukpCTs3bsXANCvXz+zxGEu1TULlMskdPJxRicfZ9zXOwCAtkLhcrpKn3j493K6fvxnRdlFxsMW7JUyONoo4GAjh4NSDkcbeZWPNUJbblpxSIdMAnY+PhDd/VzhYsfkARGRuc3uF4Rlh+NxOD4bz68/h19n1K83AGndqBiUAI0EyARkck2ts07ohk0EO7lDJlU9tCAxuxAXUvNx+louPtt9Rd/osV+QGz67vTOGtPM074sxgzbOtri3dwDu7R0AjUYgKikHWy6mYsuFNBy8moXzKfk4n5KPz/dcgb1ShhFhXrilo7a/Q3svR573iYhaMJOaQdYkJycHP/30E5YtW4ZTp05pdy5JaNu2LWbMmIH7778fnTp1qvP+1q1bh0mTJsHJyQm7d+/WVy5kZGRg5MiROHPmDJ5//nl8+umn+m3Wrl2Ll19+Gf7+/ti+fbvB/j788EM89NBD8Pb2Nng+KysLb7zxBhYvXgxXV1dcvHgRPj4++uULFy7EvffeCy8vL4PtTp8+jVmzZiEqKgphYWE4d+6cSWNCm6oJR30kZhci+L1/DYZeyCRg26MDEObpCAcbbRLBTiGHzISmkssOx1dZZUFERI0nKjEHfb7YA40A/n10AEZ38K59IwtorPOiufe77HA8Hl51CqLjfkChxjth9+L1YT1r3GbF5aN4YN/vuMmvA/4Z+0iV+9T1MNIJdrfH/Fs7YVp3P5POtdYiu7AU2y+nYfN5bcVDUk6RwfJQDwdt0iGiDUaGe8HZrl6ffRERkYma6j7UbImGiqKiovDdd9/h119/RXZ2NgDtjBVqtWmN/J555hksWrQISqUSo0ePhqOjI7Zv347s7GwMHjwY27Ztg739jVLFFStW4IEHHkBwcDDi4uIM9iVJEuRyOSIjIxEWFgaFQoGkpCRERUVBpVLB1dUVq1evxk033WSwnZubG/Lz89GjRw+EhoZCJpMhJiYGUVFR0Gg0CAoKwpYtW0xKogDWnWgAGi8pkJhdyCkZiYia2NNrz+LLfbGI8HbEqReGm72BoDk0l0QDADz311l8nvIroCzGkdueRl/vms+Pb0ZtxTsnt+GRDgOwZPAUg2VVJfclCbg0byTCvSw3TMKchBA4l5yHLRe0SYe9VzJRUnZjGKZSLmFIqEf5MIs2iGzrzGoHIqJGYtWzTtSmZ8+e+Oqrr/DZZ5/hjz/+wHfffYc9e/aYvJ+FCxdi8ODB+Oqrr3DgwAGUlpYiLCwM8+bNw9y5c2FjU/dGSIsXL8bevXsRFRWF7du3Iz8/H87OzoiMjMTYsWPx+OOPG1Qy6Lz66qvYv38/zp07h23btkGlUsHFxQWDBg3ChAkT8Oijj1pVfwZzqW7oRUNVN6c3ERE1nndvicCqU9dwMU2FBbuv4OXR7S0dUrPWztMRuK4dAqFSl9S6fk0zTuj7PlQgBJCYXdRiEg2SJKFrWxd0beuCF0aGIb9YjV0xGdhyIRWbL6TiSkYBdkZnYGd0Bl76+zz8XOz0QyzGtPeCu0PTNL4kIiLzaZSKhqpcuXIF7dqx47WOtVc0EBFRy/LL8UTctzIKdgoJ39/VA0PbeVpV4rc5VTT8HpWEuw/+D7DPx8YxD+LWwJpn3Rq26SvsTYnFb8Pvw13tehgsO5WUgx4LDD+MkUsS4l4bbVW/n8YUna7SzmRxIRU7otNRWHqj2kEmAQOC3bVTaEa0Qe8A12Y5lISIyFo024qG//u//8Off/6JmJgYg+eZZCAiIrKc6b388e62S7iYpsL0X6KMpi6muvNytAE02uEnplQ0hDgbT225MybD4HvdcMXWkmQAgHAvRzw1JBRPDQlFUWkZ9sVmlk+hmYZzyXk4EJeFA3FZeGPLRXg52uDmDt64paM3bo5oAx/nuvfHIiJqbhKzC3E5XYX2zXDYudkTDenp6Ub9EYiIiMiyknKKcDldpf9eI4BHV5/G2AjvZnfxYmleTjaApm5DJ4rL1Egq0M7kFOpkOHRCCIHlRxIAaIe3DAn1aPU9jOyUcozp4I0xHbzxKYCErMLymSxS8e/ldKSrSrAyKgkro5IAAL0CXHFLhHaYxYBgdyjlVc/qQUTU3FRsFNwcPxxgi18iIqJWoKpeAGVCIDpd1apvbOujYkVDfmnNiYb4/CwICDgolPC2M+y5EJWUg9PXc2Ejl+GJwSHwYC8CI4Hu9nh4QDAeHhCM0jINDl3N0lc7nEjM0X99sD0aLnYKjGnvpR9mEejO9zURNU+J2YUGsxE1xw8HmGggIiJqBdp7OUImwSDZIJckhHs5Wi6oZsrT4UaiIaOoqMZ14/KzAAAhTh5GMynoqhkmdvVlkqEOlHIZhrbzxNB2nnh/fCek5BXjn4up2HIhDVsvpiKjoBR/nknGn2eSAQCdfZxwS8c2GNexDYaEesBOaX2zrRARVaUlfDjARAMREVErEOBmj6VTu+Ph1acgBCABra4XgLnYKeVQSAqoAaQXFNa4bmx++YwTlYZNFKvL9OX/D/QLbJQ4WzofZ1vM6BOIGX0CUaYROJGYox9mcehqFv5Lycd/KflYsPsKHGzkGBnmqa126NiGCTYismrtvRwhAaiYa2huHw6YPdEghEATTWRBREREJpjdPwgSgNmrTqGrr3OzGutpbRwUSuQCyKyloqG6qS3Xn0tBZkEp/F3tcFMH78YKs9WQyyT0DXJD3yA3vH5TB2QVlODfy+nls1mk4VpuEf4+n4q/z6cCAMI8HfRJh5FhnnC05WdvRGQ9lHIZFHIJpWXa++rm2CjY7P+rrlixAitWrDB6Xq1WQ6Hgf+JERESW1DfIDQCQlFvzDTLVzFFhg1wAWcW1DZ0on3HCyXDGCd2wifv7BEDO6RrNzt3BBlO7+2Fqdz8IIXA2OQ+bz6diy8VU7IvNRExGAb7aH4ev9sfBRi7D0HYeuCWiDW7p6I0uvs5Gw1yIiJrSp7tiUFom0MPPBQvu6Iz23k7NKskANMHQCY1Gg++//x4ffPABrly50tiHIyIiohqEejgAADILSpFbVAoXO6WFI2qeXJS2uA4gp5ZEw42hE576567lFGHrRe0n67P6cthEY5MkCZFtXRDZ1gUvjgpHXpEaO6PTseViKjZfSEVcZiG2X07H9svp+L+NgL+rHW6JaINxnbwxur033OyVzXqKOSJqXlLzivH1gTgAwPvjO2Jk++ZZ9dZoiQaNRoOffvoJ7733HhMMREREVsLJVgEvRxukq0oQm1mA7n6ulg6pWXK1tQPUQF4ts05UNXTip+OJ0AhgUIg7Ong7VbcpNRJnOwXu6OqLO7r6QgiBy+kq/RCLndHpSMopwrIj8Vh2JB5ymYRQDwfEpKsgoJ1i7tsp3fDwgGBLvwwiJsBaqM92x6CgpAx9Al0xrmMbS4dTbyYnGjIyMvDWW29hw4YNSElJgY+PDyZMmIC33noL7u7assAtW7bgueeew8WLFyGEgJubG1566SWzB09ERESmC/Vw0CYaMphoqC83W1tAXfP0lgXqEqQW5QO4MXRCCIHlR+IBAA+wmsHiJElCB28ndPB2wtND26GwtAx7r2Rgy4U0bLmYivMp+YhOV+nX1wjgkdWn8eH2y+jYxhnhXo4I93JAmKcjwr0cEeJhD1sFZ7egxrfscLx++kOZBCyd2p19d1qA9PxifLU/DgDw5s0RzXoYl0mJBpVKhSFDhuDSpUv6ho/x8fFYvHgx9u/fj0OHDuG1117DJ598AiEE7O3tMWfOHMybNw9ubm6NET8RERGZKNTDAUcTshGbWWDpUJotTzt7QAUUlFWfaND1Z3C1sYO7rXbIyqGrWbiYpoK9UoZpPfyaJFaqO3ulHDdHtMHNEW2wAF3w64lETP8lymi92MxCxGYazzgiSUCQm3154uFGAiKs/LETm05SPaiK1YjPLkRcZgGuZhXizPU8fWk9oE2APbr6NMZGeLOyoZn7bPcVqErK0DvAFbd2ar7VDICJiYZFixbh4sWLsLGxwcyZM9GtWzfk5uZi48aNOHjwIKZMmYL169cDAO677z589NFHaNu2baMETkRERPWj69MQl1Xz1IxUPS97OwBAUVlptevoh01UmNpy+VFtE8gp3fzYH6MZGNrOEzIJBvPZyyXgl3t7IadIjZgMFaLTVYjJKEB0ugqqkjJczSrE1axC7Ig23p+vsy3CPB3Kkw+OCPfUJiHCvRzh4WDTdC+MrIYQAlmFpbiaWYirWQX690/Fx+mqmodoAUCZEDiXnMdEQzOWoSrB4v2xAIA3burQrKsZABMTDevXr4dMJsO///6LIUOG6J9/+eWX8cADD+CHH36AJElYtGgRnnzySbMHS0RERA0X4qG9EI3NYEVDfXk7aJM1JaL6RENcfhYAIKQ80VBQosZvUdcAAA/047CJ5iDAzR5Lp3bHo6tPo0wI/RRzd/X0N1pXCIHU/JLyxIM2ARGdXqB/nFlQiuS8YiTnFWN/XJbR9m72SoR7OZQnHwyTEL7Ots3+pqO10mgEkvOKq00iXM0qQH5xWa37cbVTINjdAcHu9vB0UOKHY4kQldZ59q+zWOfRDxFt2PulOVqwOwb5xWXo6e+C27v4WDqcBjMp0XDx4kX079/fIMmgM2/ePPzwww/o0KEDkwxERERWTFfRwKET9efrqE3WlAp1tetUbgT555lk5BWrEeJhj+HtPKvdjqzL7P5BGBvhjeh0FcJraLonSRJ8nG3h42yLwaEeRsuzCkoQk1GAmHQVojNUiEkv0P97LbcI2YWlOJaQg2MJOUbbOtjIb1RCVBqWEehmzylSLai0TIOE7PKkQRVVCQnZRSgp09S6nzZONvpEgvbLASEe9vrnXO0NK6CGtPPUJ8BkEuBko8CFVBV6f74HS6d2w/ReAY31kqkRZBaU4Mt9cQBaRjUDYGKiITc3F+3atatyWVhYGACge/fuDY+KiIiIGk2o541EgxCiRVzQNLW2To4AAA2q/yTyxtSW2pvO5Ue0wyZm9QmEjDeGzUqAm32DS9LdHWzQx8EGfQLdjJapitW4klmehKhQBRGTUYCrWQUoKCnDmet5OHM9z2hbpVw7M4Yu8cDmlOZVUKI2rkQoTyjEZRXiWm4RROXSgkpkkvY9VDGJUPFxkLs97JWm/Z4qJ8AUMgn3/hKFHdHpuPeXKOyKycDCiV1N3i9Zxue7ryCvWI3ufi6Y0NXX0uGYhUmJBo1GA6Wy6vGECoV2Vw7lpYRERERknYLd7SFJgKqkDOmqEng72Vo6pGbH31l7vSNkZVCXaaCQy4zW0TWDDHFyR1xmAXZEpwMAZnK2CarE0VaByLYuiGzrYrSsRK1BXFbVSYgrGQUoKdPgUpoKl9JURtuyOWXNzNUfwVYhQ5CbvUEFQsWEgr+rXZX/RzRU5QTYP48OwLvbLuGdbZfwv0PxOBKfjVX39+Y0ulYuq6AEi/a1nN4MOq37fxciIqJWyFYhh5+LHZJyihCbWcBEQz34O2srGiAJpKgK4e/iaLROxaETP5RXM4wK90KIBz+UobqzUcj0U3BWVqYRSMopNEhAsDnlDebqj+Bip6iyEkH3uI2TrVVUKcllEt4aG4EhoR6495cTOHUtF70/34Pvpnavsq8IWYcv9sQit0iNyLbOmNhCqhmAeiQatmzZglGjRpm8XJIkbN++3dTDERERUSMI9XDQJhoyCtAvyN3S4TQ7rrZ2+scJuSqjRENOSSGySrSzegQ5uGPFsXMA2ASSzEsukxDk7oAgdweMau9lsKyq5pS6vhAtpTllaZkGidlFVSYS4jIb1h8h2N0eweUVCm72zWuGmDEdvHHy+eGY/vMJ7IrJwN3l/34+oQvsOJTCqmQXlmLh3isAtNUM1pCwMheTEw3JyclITk42ebk1/GdEREREWqEeDtgXm8mGkPVkI5MDQgIkgcRc45J13YwT3naOOJ6Qj7jMQrjYKTA5suV8WkXWzVqbUyZmF+Jyugrta2isqVNTf4Sr5f0RNHXoj+DvaldtNUKQuz0cbFpekXdbFztse3QA3v7nEt7ffhnfHryKw/FZWHV/H4R7GVdgkWUs3HMFOUVqdPV1xuTItpYOx6xM+qt68803GysOIiIiakKceaJhJEmCHHKUQY3r+cY/Q92wiRAnD30TyLt6+LXIGxpqnmpqTllQosaV8uEX5mxOuTM6HU/+eQYaoU0AfD6hC4aGepqlP4I+geBh2HTR39UOykboj9AcKOQyvDuuI4a288C9v0QhKikXvRbswbK7umNqdz9Lh9fqZReW4vM92mqG11tYNQPARAMREVGrFOKh/SQxLrPQwpE0XwooUAY1UlRVJBrKG0EGOLhhzf5rAIAH2ASSmgkHGwW6tnVB12qaU17NKjDoBVHX5pQVaQTwzLpztcbibKvtjxDiYd39EazZzRFtcPL5Ybjn5xPYeyUT0348jicHZ+CzOzpzVhILWrQ3FjlFanT2ccKUbi2rmgGoZzPI6Oho/Pnnn4iLi4OtrS169uyJqVOnwt6+YdP+EBERUdNgRUPD2cqUKBZFSC0wTtbElVc0FKqUKCzVIMLbEQOC2QuDmj8bhQztvZ3Qvo7NKXXJiItp+SgqNe6X4GqnQHtvx2qHNrjZKzkE2wz8Xe2x47GBeGPrRXy4PRpf7Y/DoatZWHV/b7Tz5FCKppbTwqsZgHokGj7//HO89NJLKCsz7ND62muvYdOmTejatavZgiMiIqLGoUs0XM0qhEYjWuRFTmOzkymRWwakV5Fo0FU0XErSXi890C+IN0vU4tXUnDIhqwAh72836Kkgl4Cz/zei1l4NZB4KuQwfjO+EoaEemLEyCscTc9BrwR58f1cPTG6Bn6hbsy/3xSK7sBQd2zi12GEsJg1Y2rdvH1544QWo1Wo4ODigZ8+eCAsLgyRJSExMxJ133gmNpvbOrkRERGRZAW72UMgklJRpcC23yNLhNEv2Cm0n+owi45+frkfDlWQNZBIwo3dAk8ZGZG0C3R2wdGp3yMsTbnJJwpKp3ZlksIBxnXxw8vnhGBzijpwiNe784RieWXcWJWrexzWF3KJSLNitq2ZoX2Pz1ObMpETD4sWLIYTAzJkzkZycjGPHjuHSpUs4ceIEwsLCEB0djS1btjRWrERERGQm2k8etRf4HD5RP04KGwBAdrFhokEIoZ91AiV2uKVjG/i52lXenKjVmd0/CHGvjcbOxwci7rXRmN0/yNIhtVoBbvbY+cQgvDgyDIC2X8CQxfsRx/NBo1u8Lw5ZhaWI8HbEXT38LR1OozEp0XDw4EEEBARgyZIlcHS8MZanW7duWLhwIYQQOHTokNmDJCIiIvNjn4aGcbaxBQDklBQbPJ9RXIB8dflzpXaYxSaQRHoBbvYYEe7FSgYroJTL8NFtnbFhdj94OChxNCEbPRfswV9nky0dWouVV6TGZ7tjAACv3dShxVYzACYmGlJSUtCnTx/Y2NgYLRsyZAgAIDU11TyRERERUaPSJxoymGioD5fyRENepURDbF6G9kGpDTzsbXFHF5+mDo2IqM5u6+yDqOeGYWCwO7ILSzFx+VE899c5DqVoBF/tj0VmQSnaezni7h4tszeDjkmJhpKSEri5uVW5zMXFRb8OERERWT9WNDSMu612OES+2vDap+Kwiek9/Tl9HBFZvSB3B+x+chCeH94OAPD5nisY9tV+XOX5wWzyi9X4dJeumqE9FHKTbsWbnZb96oiIiKhaTDQ0jKe9NtFQoilFsfrGbFznMtO0D0rt8EA/DpsgouZBKZfh0zu64K8H+sLNXonD8dqhFBvOcSiFOXy9Pw4ZBaUI93LE9J4ttzeDjsnTW0ZHR+PHH3+s1/L777/f1MMRERFRIwnx0I6Rjssynp6RaqeraIBMg3RVCfxdtT/P7VcTAABtbFzR09/VUuEREdXLHV19EfXcMNz103Ecic/GHd8fxQsjwvDB+I5QtvBP4RuLqliNT8t7M7w6uuVXMwD1SDTs378f+/fvr3KZJEnVLpckiYkGIiIiK6KraEjMLkRpmYYXkCZyUmp7NEBWZpBoOJ2eCiiBUcEBkKSW2+iLiFquEA8H7H1yMF76+z98sScWn+6Kwf7YTPw+ozcC3dnI01TfHLiKtPwStPN0wH29W341A2BioiEoKIgnTCIiohbCx9kW9koZCks1iM8qRJiXY+0bkZ5j+fSWkJUhPV/bp+HM9VzkavIBAFM7h1oqNCKiBrNRyPD5hK4Y1s4TD/x2EgevZqHHgt34aXpPjO/EJrd1VVCixie7ogG0nmoGwMREQ1xcXCOFQURERE1NkiSEeDjgfEo+YjMLmGgw0Y1Eg3boBAB8fzgeUBYBAHp680KciJq/SZFt0cPPFdN+OoZjCTm49bsjeGlkON4dF8FKuDr49uBVpOaXINTDATP6BFg6nCbDdwYREVErxoaQ9WdQ0aAqQWmZBj+ejgFkAjJICHR0s2h8RETmEurpgH1PDcbTQ7WVWh/tjMbIrw8gMZs9fmpSUKLGxzu1vRleGR3eqhIzreeVEhERkREmGurPUWmYaPj7vxRkluQBAAId3aCQcVpLImo5bBVyLJzYFWtm9oaLnQL747LQc8EebLmQaunQrNbHO2OQklcMf1c73N+ndc1CxEQDERFRK6ZPNGQw0WAqfUWDpB06sfxoAmCjHTYR6uxhwciIiBrPnd38cGLuMPQKcEW6qgTj/ncYr246D3WZxtKhWZVvDsTh7X8uAQCu5RThp+OJFo6oaTHRQERE1IqxoqH+Kg6dOJeSh7/PpzLRQEStQpiXIw7MGYwnB4cAAD7YHo3R3x7EtZwiywZmJRKzC/HkH2f03wsAj64+3aqGmjDRQERE1IqFeGinKYvLaj0XP+ZSMdGwMzoDZRoBb3ftJ3ohTu4WjIyIqPHZKuRYPDkSv8/oDWdbBfZcyUSPBbux7WKapUOzuCPx2RCVnisTAtHpKovEYwlMNBAREbViuoqGlLxiFJSoLRxN81Ix0aDj4aZNNIQ6eVoiJCKiJjethx+Ozx2KHn4uSMsvwdj/HcIbWy6gTFP5Vrv1WHP6mtFzcklCeCua3YmJBiIiolbM3cEGrnba2a7jMlnVYIobzSAFAAFbuYQiSftpFYdOEFFr0t7bCQefHoLHBgZDCODdbZcx5tuDuJ7b+oZSRCXm4LeT2kSDTNI+J5ckLJnaDQFu9haMrGkpLB0AERERVS1RlY3Luelo7+KFADNNlVjVPkM9HHDyWi5iMwvQ2dfZLMdpDfQVDQAgK0NxGZCoygEAhDox0UBErYudUo5vpnTDsHaeeGTNKeyKyUDPBXvwxR2d4eNih/Zeji3+RlsIgWf+OgshgLt7+OGT2zsjOl2F8Fbw2itjooGIiMgKLbt0GI8cWAONEJBJEpYOmoLZHfo3yj5DPW8kGqjuMvLU2g5fErTDJ+RqlEEDG5kcbR2YsCGi1umeXv7oFeCKaT8ex+nrubjnlygA2k/3l07tjtn9gywcYeNZfeo69l7JhL1Sho9u64QAN/tWl2DQYaKBiIjIypzJvIaH96+BKG8lpRECD+1fjcXn98NOroQEQJIk7f2tJEGCBEmC9l/9MsPnijVq7E6+oj+GRgg8emANxvpHcOaJeorOKAA0MkCuAWQaQFkMAGhj6wKZxNGpRNR6RbRxwppZvRHx4U59U0SNAB5dfQpjI7xb5M13YWkZ/m/jfwCAl0aGI8jdwcIRWRYTDURERBaWX1qMvSlXsP1aNHZcj0ZUZlKV653MNG4u1RBlQiA6N52Jhnpq7+UICDkAjbaiQakdi9zOmY0giYgSs4uqmHkBWHEkHq/dHGGRmBrTp7tiEJ9ViEC3/2/vvuObrPY/gH+epE3apm3a0r0XLdAWsEzZQ1FQ2csFKCp4Ba+oIOBF9KqAgIiI+zrwKj9kyBDworJrZZdVVgstHazOdM+c3x8lgdBBR0rS9vN+vfqifcZ5znNa8n3yzRlWmNk/yNTVMTkmGoiIiO6xorJSHEi7jF1X47HzajwOpSWhTGhrPEcGCV/3HINWShtUTD0ooBUCAgJCwPBfVIwTFYD+mIyifLx2+FeDhz65JCHY3hn5ThWfxCdkMNFQF94O1nC2tkZ6aSkglUNSVDxUt3FwNnXViIhMrrWzCjKpoifD7ebtuICruSVYOrQdrC3lpqmckSVnFWLhzjgAwJJH28FGwbfZbAEiIqJGVqYtx7GMVOy8GoddV+MRdT0BReWGS0kG2DphgEcwBnq2Rn/3IGxLOYsp0etRLkTFbNU9RuPZkK4Nqoe9wko/RwMAfNR1GLxVDshxygXAHg314a6yQXp2DpYMC8UBzTlsSL7MFSeIiFCRjP1qTAdMWXdSH8seDHHG/86n4bPoROy7lIE1T3dCWDOYhHj2trMoLNWiV4ATxnb0NHV1zAITDUREREamFVrEZl2/2WMhDnuvXUJOqeESX+7WdhjgEaz/Criju/3kkG54yCsU8TnpCDbSqhO6MntsW4nk/Gw4W1Ws5+3nWDFWVlNUhuzCUjhYWzb4Wi2FbonLYFdrbEnLAcAVJ4iIdCZ388VDoS4GKy/8fv4GJvzfcZy+lovOH+3D8uFheKG7HyRJMnV16+WvhEysjkmFJAHLh4U12fswNrNPNKxbtw6ffvopTpw4gZKSEgQHB+PJJ5/EjBkzYGlZ+weh6Oho/Pjjj4iJiUFSUhIyMjIgl8vh5+eHgQMH4rXXXoO/v3+tyjpx4gS6dOmC0tJSBAUFIT4+vp53R0REzYEQApdyM7DzasUcC7uuxiGtKN/gGAeFNfq5B2GgRzAGeAajrdrtrg8j3ioHoy1reXuZTwZGYtGpXdicFIvHA++DSmkBV1sFbuSVICGjAPd5q416zeZMt8RlflkJEnIzAQD+do6mrBIRkVm5c+WFQaGuOPlaX0xcE4P/nUvD1PWn8Pv5NHw9tgOcbBQ1lGR+tFqBf246DQB4tosvOvk4mLZCZsSsEw2vvPIKPv74Y1hYWGDAgAGwtbXFrl278MYbb+DXX3/F77//Dmvr2s1Yun37dnz++efw9fVFaGgo3NzcoNFocOzYMaxcuRLfffcdtm7din79+tVYTklJCSZMmICysrIajyMiouYtNV+D3dfisfNKxXCIpPxsg/02Fpbo7RZYMRzCozU6OnlCLjOPlQiG+YZh0ald2J5yFsXlZVDKLRDgZFORaMhkoqEudImGrOICpBawRwMRUW242imxbXI3LN9/CbO3ncUvp67hcHI2fnoyEr0Dm86EuquOJONoigZ2Sgu8P6SNqatjVsw20bBp0yZ8/PHHsLW1xd69exEZGQkASE9Px4ABAxAVFYV58+Zh6dKltSrvySefxHPPPVep10JJSQlmzZqFjz/+GE8//TQSExMhl1c/Kcm///1vnDx5EtOmTcPKlSvrfX9ERNS0ZBTlY8+1i/rhEOc1aQb7LWVydHfxxUCP1hjgEYxuLr5QyM0zzHZ18YG7tR2uFeZiz7WL+iUuDyZlc56GOtIlGs5qbkBAwMbCEi5WtiauFRGR+ZPJJLzaNwh9A1vh8R+PIS49H/0+i8b8QaF484HWkMvMewhCTlEp5mw/BwB468EQuNkpTVwj82IeH61UYcGCBQCA2bNn65MMAODs7IzPPvsMALBy5UpoNJpalde2bdsqh0YoFAosWbIEVlZWSElJwZkzZ6ot4/Dhw1i0aBHGjBmDUaNG1eFuiIioqckrLcZvKWfx+qFfEbn5I7j839sYvfsHfHYuGuc1aZAgobOzN2aF98OOQc8j64l/Y9+QlzD/vkHo7R5otkkGAJBJMgz1CQMAbE6q6PIZ0IpLXNaHLtEQm30NAOBv68TxuUREddDJxwHHXu2DiZ29oRXA/B3nMeDzaCRnFZq6ajVa8Gc8rucWo7WzCi/3DjB1dcyOWT4Fpaam4vDhwwCAJ554otL+Xr16wcfHB8nJydi+fTsef/zxBl1PkiTIbnZnVSqrzkQVFRVh4sSJcHR0xMqVK2tMSBARkflLyc9GXE46Wt+caLE2S062c3DT91jo6x4IR6WNiWrfcMN8w/DVhQPYnBSLld1HIMCJiYb60CUaTmdVJBo4bIKIqO5slRb4/vH78GCIC17ccAr7LmWiw4d78c24DhgR4WHq6lUSn56Pj/ZdAgAsGxYGhYXZfn5vMmaZaIiJiQEAODk5ISCg6uxQ586dkZycjJiYmAYlGsrLy/HOO++goKAA7dq1Q3BwcJXHzZs3D2fPnsXq1avh6urKRAMRURP2zYWD+mUeJQBt1K5IyMu865KT7jb2pqlwIxjgEQxbCyWuFOTgaHoKEw31pFt1IqO4ot24tCURUf092ckb3f0c8fiPx3A4ORsjvz+CF3v44cOhYbC2rH54+732+pZYlJRrMSjEBY+0dTV1dcySWSYaEhISAAC+vr7VHuPj42NwbG0lJSXhrbfeAgBkZmYiJiYGKSkpCA4Oxtq1a/U9G24XHR2NZcuWYdiwYQ3uPUFERKZRVFaKQ+lJ2Jp8FktO79FvF6gYXw8AbtZ2FatCVLPkZHNiZWGJh71DsT7xJDYnxeIZvz4AgMTMAggh2P2/lnQ9GnT8bbniBBFRQwQ5qxA1rSfe+t95fLA7Hp9HX8a+S5lY81Qkwj1Mn/D/43waNsdeh1wm4SMuZ1kts0w05ObmAgBUKlW1x9jaVky0lJOTU6eyMzMzsWrVKoNtkZGR+PbbbxEWFlbp+IKCAkyaNAlqtRqff/55na51u+LiYhQXF+t/rmu9iYiobjQlhYi+kYj91xOw/3oCDqUloURbXu3x3/Uai4nBXVrUA8Mw3zCsTzyJTUmn8VaHhyBJQGGpFtdzi+Fub2Xq6tXLvY63dyYaAmybb3KKiOheUVjIsOjRthjY2hlP/18MYq/losvy/fhoWBim3O9nslhdVq7FjC2xAICXevqjnbudSerRFLS4wSQdO3aEEAJarRYpKSlYu3YtCgoK0KlTJ6xYsaLS8bNnz0ZcXByWL18OD4/6jw9auHAh1Gq1/kvXI4OIiIzjWkEO1iWcwMsHNuG+zcvgtPotDPnjGyw8uQtR1xNQoi2Hm7Udhni3wZ2PJ3JJwgOeIS0qyQAAj3i3hVySITb7OpILMuGtrkguJJr5BFw1udfxtlKigUMniIiM5sFQF5x8rS8ebuOCojItXtxwCqNXHUFmQYlJ6vPl35cRey0XrWws8fagEJPUoakwyx4NdnYVmaH8/Pxqj8nLywMA2NvXr/uMJEnw8vLCmDFj8OCDDyIsLAwzZsxA37590aFDBwDAnj17sHLlSgwZMgQTJkyo13V05syZg1dffVX/c05ODpMNRET1JITAxdwM7L9+qaLHwrUExOemVzouyK4VersFoLdbIHq7BSDY3hmSJOGbCwcxJXo9yoWAXJLwZY/R8FY53PsbMTFHpQ36ugdi19V4bE6KRYCTDZKzi5CQUYDufk1zCMC9jrccOkFE1Lhc7ZTYNrkbPt5/CW9sO4tfTl3DoaRs/PRkJPoE3bteZBn5JZj3v/MAgHcHt4GjjeIuZ7RsZplo0C1DmZycXO0xun1VLVlZVw4ODhgxYgQ+/fRTbNmyRZ9o2LRpE4QQSEpKQr9+/QzOyc7OBlCxQoZu3/Lly9GxY8cqr6FUKqtd0YKIiGpWrtXidPY17Lt2CfuvX0LU9URcLTTsEi9BQnsnj5uJhQD0cguAp426yvImh3TDQ16hiM9JR/DNVSdaqmG+Ydh1NR6bkk4j0KkH9l3KbNITQt7reHt7okGtsGrSK5EQEZkrmUzCjL5B6BvUCuP/ewxx6fno/3k05j0Ygn890BoW8sbvqP/2jvPIKixFhIcdnu9W/VyCVMEsEw333XcfACAjIwMJCQlVrjxx5MgRABXzKxiDbj6IGzduVNp3+vTpas8rKirC3r17AdxKPhARUcMUl5fhSHryzfkVLuGvG4nQlBQZHGMpk6OLs48+sdDTNQAOSutaX8Nb5dCiEww6w3zD8M+Dm/HXjUR0d+0FgCtP1IVu1QmAS1sSETW2SG8HHHu1D6ZvPI3vDyfjnd8vYFd8On584j74OjZeovf01Rx8/vdlAMDyYeH3JLHR1JllosHb2xtdunTB4cOHsXr1arz55psG+6OiopCcnAylUokhQ4YY5Zq7du0CAISE3Bprs3z5cixfvrzK4/fs2YP+/fsjKCgI8fHxRqkDEVFLlVtahOgbl7H/WsVQiIPpSSi+Y6lJWwslerj6VQyDcA9AV2dfWFtYmqjGzYefrRM6OnnieOYVZFleA8BEQ13YWtzqPeHPRAMRUaOzVVrgu/Ed8WCIM6auP4X9lzLR8cN9+M/YDhjZvv5z6lVHCIFXNseiXCswIsIdA1o7G/0azZFZJhoAYO7cuRgxYgQWLVqEwYMH63suZGRk4B//+AcAYNq0aVCrb3WL3bhxI+bMmQMvLy/s3LnToLyFCxfiueeeg4uLi8H2rKwsvPXWWzhy5AjUajXGjh3byHdGREQ3CnP1q0Hsv34JxzOvQCuEwTEuVir93Aq93QLRwckDFjLzWUO7ORnmG4bjmVdwrjARgCcTDXVw+9AJTgRJRHTvPBHpje5+jnj8x2M4lJSNUauOYOr9flg2LAzWlsZ7XtgSex0749KhkMuw9LF2Riu3uTPbRMPw4cPx8ssvY8WKFejevTsGDhwIlUqFnTt3Ijs7Gz179sS7775rcI5Go8H58+dRVFRUqby5c+di3rx5iIiIQFBQECwsLJCamoqYmBjk5+dDrVZj3bp1cHNzu1e3SETULKXkZyMuJx2tb859IIRAQl6mvrfC/usJuJCTVum8AFuniqSCe0VyIcTepcWtAmEqw33D8c7xP3AkOxGQ3JCUVYhyrYBcxva/m9uHTjhY1n7oDhERNVxgKxWipvXEvN/O44Pd8fji78vYn5CJNU9FItyjfosG3K64rByv3VzO8rV+gQhspWpwmS2F2SYaAODjjz9Gz5498emnnyI6OhqlpaUICgrC7NmzMWPGDCgUtZ/pc+XKldi/fz9iYmKwc+dO5OXlwc7ODhEREXjooYfw4osvMslARNRA31w4iBei10MrBCQAnZ19kFqgwZWCyhM3hju66+dX6O0WCC9V1RM3UuPr4OQJP1tHXM7Lgtw+G2WaVkjJLoSfEyc2vJvNl2/N4/T28d/hpbLH5JBuJqwREVHLYimXYdGjbTGwtTMm/F8MYq/losvy/Vg2LAxT7/dr0IcWH+9LwMWMAnjYKzFnQGsj1rr5k4S4o68q3RM5OTlQq9XQaDT1XqKTiMicXMzJQOsNC1FVULGQZLcmbnQPRE9Xf87Ob2ZePrAJn5yNgl2BN3IvBWH3i/ejX/C9G4faWHGxMeNtSn42/Na+D+1tf/VySULimDc50SgRkQncyC3GMz8fx/azFRP8Dw93xzfjOsCpHktRXsspQutFu5BXXI5Vj3fEhM6Nt1TyvXSv3oeadY8GIiIyf3mlxfj6wkG8f+LPKpMMy7o8hilt7oeNBdebNmfDfcPwydkoFFmnAQhEYmahqatk9uJy0g2SDABQLgTic9KZaCAiMgFXOyW2Tu6Kj/cnYNbWM9h0+hqOJGfjpycj0SeoVZ3Kmrv9HPKKy9HV1wFPRXo3Uo2bLyYaiIioXjKLC7DybBQ+PhOFzOKqJw+USxLGBHRgkqEJ6O0eCAeFNbJLCgGbHE4IiYqZxktLS6HVaqvcH6C0R4CVvcFEpnJJgr/Svsr5oohqSy6Xw9KSq+oQ1YckSXilTyD6BDrh8R+P4UJaPvp/Ho1/PRCCeQ+2rtXSlIeTsvHd4WQAwMfDwyHjnEV1xkQDERHVydWCHHwUuw+fn/sbeWXFAIBgO2fMbt8fpVotph34BeVCQC5J+LLHaH6y20RYyuR4xLstfrp0DLBPb9GJhpKSEty4cQMFBQUoLy+v8difw4Yi47ZEWyulDQqvZyABGY1dTWrmlEolnJ2dOcSWqJ4ivR1wdEYfvLzxNL47nIx//3EBu+LT8dOT98HXsfrhmxXLWVbMv/NUJy9093O8V1VuVphoICKiWknIzcCS03vwbdxhFJeXAaiYRHBOxACM9m8PuaziE4JHfdoiPicdwTdXnaCmY7hfWEWiwS4DlzLzTV0dkygoKEBycjLkcjkcHR1hbW0NuVxe42RiJeVlKC0vh6VcDoWcj1bUMLqeNBqNBqmpqQDAZANRPdkqLfDt+I54MMQFU9afRFRCJjp8uA//Gdseo9p7VnnOmpgriE7Mgo1CjkWPtL3HNW4+GA2JiKhGsVnXsOjULvzfpeMoFxVdyHu4+uPN9gMx2LtNpTdg3ioHJhiaqIe8QmEpyVGqLERcZuUlSFuC9PR0WFpaws/PD3J57dZht2rkOlHLY21tDTs7O6SkpCA9PZ2JBqIGejzSC938HPD4j8dwKCkbo1cdxZT707FsaDvYKG69Jc4vLsOsrWcAAHMHBsNLzWWL6+vuA1SoyUjJz8buq/FIyc82dVWIqBk4lJaE4Tu/Q/impfjx4jGUCy0e8grF3sEvImrISxji07ZBS0aR+bGztEJftyAAwA1ZKorLah420NyUlZUhPz8fTk5OtU4yEDUWSZKgVqtRXFyM0tJSU1eHqMkLbKVC1LSemD0gGJIEfPn3ZXRZvh+nrt5agnvx7otI0RTB38kar/YNMmFtmz72aGgmbl+7XiZJ+KrHaK7jTUR1JoTA7qvxWHByF3ZejQMASJAwyj8CsyMGoJMzZ11u7kYHRODPaxcAuwxczipEiIutqat0z5SVVQwJUiqVJq4JUQXdhJDl5eWcHJLICCzlMix8pC0GtnbG06tjcOZ6Hros349lQ8Nwn5c9Fu6sePZZ+lg7WFsy4dwQTDQ0Ayn52Xj+r3X6Bba0QmBK9Ho85BXK7stEVCtaocWvSWew8NQuHExLAgBYSDI8FRSJNyIGoI2Dq4lrSPfKUN8wTP17A2CTiyNXr7eoRIMOe+qQueDfIlHjeCDEBSdf74tJa45j+9kbeOmXUwb7swrYi6ihOHSiGYjLSa+0dr1uHW8iopqUacvx08VjaL/pQwzf9T0OpiXBSm6B6W17IX70bHzXezyTDC2Mh409HETFWuNbU86auDZERESNw8VWia2Tu+LtQSGV9k1dfwop2YUmqFXzwR4NzUBre2dIgEGyQS5JUFkosPtqPFpz5nciukNRWSm+jz+Mxaf2ICEvEwBgb2mFl9r2wCvtesPV2s60FSSTamvth7+LMnAgM97UVSEiImo0kiShT1CrStvLhUB8ej68HTgZZH0x0dAMeKscMC6gI9YkHAdQkWR4MjAS3baugAAgg4QPOg/BjLC++uXniKhlyi0twpfnDuDD2L24VpgLAHCxUmFGWB/8o00PqBUMqAT0cQ7B3ynHcLn0CnJKimCv4LoKRETUPLV2VkEmAdrbPrWVSxKCnVWmq1QzwERDM/GAZ2usSTiO7s6+WNF9uD7JAABaCMw8sg2zj2yHj60DfFWO8LN1hK/KAb62t32vcoDK8tYEWCn52YjLSWePCKJmIKMoHyvORuGTM1HIKqnoCuijcsDM8H6YHNIVNhYKE9eQzEk3N0/gojW0ykL8L/UcxgZ0NHWViIiIGoW3gzW+GtMBU9adRLkQkEsSvhzTnr0ZGoiJhmZCdfNNgpWFJZLysyvN2QAA5RBIzMtCYl4WcL3qclopbeCrcoSAwInMK/oeESu6D8dLbXs2Wv2JqHGk5muwLHYvvjx/APllJQCAEHsXzG7fH08GRkIhZxigygKdVUBOK8AlBZuTYplooEr8/f1x+fJlJCQkwN/fv8XWgYiah8ndfPFQqAvi0/MR7KxiksEI+ITZTOgSDQVlJVifcLLSfrkk4e9HpqNUq8Xl/Cwk5WUhKT8bl2/7N6e0CBnFBcgoLjA4VwuBaQc24uMz+9HD1R9dnH3QxdkH7R09YGXBpZaIzFF8TjoWn9qNVfFHUKItBwDc5+SFuR0GYIRvBIdRUY0CnGyAXGfAJQXbks+iVFsOSxmX+SIioubL28GaCQYjYqKhmdAlGs5p0nA4PQUAIJMkaHXdf3qMRhcXXwBAD/hXWYampBCX87KwNfks3jz2W6X9cTnpiMtJx6r4IwAAS5kcEY7u+sRDF2cftHNwgwUfRolM5mTmFSw6tRs/JxyHVlT0berjFog57QfgIa9QLpVGtWJvZQlHtEJWmSU0KMLeaxfxgGflWbmJiIiIqsJEQzOhG1+dU1oEAHgyMBKLOg9BfE46gms5x4JaYY32TtZwUtpgXsz/9G9SgIoeEV/3GIPE/CwcTkvG4fRkpBfn41hGKo5lpOLL8wcAANZyS9zXykufeOjs7I3W9s6QSRWfnnLeB6LG8feNRCw4uRNbk28tRzjEuw3mtB+IXm4BJqwZNVWBTioczWkFOF3D5qRYJhoIAPD999/jmWee0f8cEGD4+rJ7927069cPAHDlyhUsXboUv/32Gy5fvgy5XI42bdpg4sSJmDp1KiwsDB9Di4uLsWLFCqxZswYXLlxAcXExnJyc4OPjg/79+2P27NlwcnKqUx2IiMg0mGhoJnaknjf4uaOTB7xVDvV6M++tcsBXPUZjSvT6WxOi9BiNZ0K66o8RQuByXhaOZCTrEw9HM1KRU1qE6BuJiL6RqD9WrbBCp1beUMjk2JF6Xj/vw/Juw/Bim/vZA4KonoQQ+PNKHBac3Ik91y4CACRIGOPfHnPaD0DHVl4mriE1Zf5ONjiacCvRsKLbcPaIaUQp2YWIS89HazMfGxwcHIyJEydi/fr1yM/Px6hRo2Bra6vf7+7uDgDYt28fhg8fjqysLPj7++PBBx9EcXExDh06hOnTp+PXX3/F1q1bYWlZMQRTq9XikUcewc6dO2Fvb4/evXvDwcEBaWlpiIuLw5IlS/DEE0/Aycmp1nUgIiLTkYQQVc0bSI0sJycHarUaGo0G9vb2DSorJT8bfmvfhxaGPRASx7zZoF4DKfnZdeoRoRVaxOWk43B6RfLhSEYKYjJSUVheWuN5SrkFbC0UUFkoYGupNPi+qm22FgqoLBWwtdDtv/n9zX9tLSuOZQKDmiut0GJzUiwWnNyJIzeHSlnK5JgQ1AmzIvojRO1i4hpSczDz1zNYuvcCLMIOoAxlOPrYK4h09m606xkzLta33KKiIiQkJCAgIABWVpWX9BRCoKCk3Gh101l1JBnTN56GVgAyCfhkRDgmdvYx+nVsFHKjJYtqmojx2rVrCA8PR2ZmJj799FNMmTIFspvzwmRkZGDs2LHYtWsX3nnnHbz11lsAKhITffv2xX333Ye9e/fCzs7OoMwjR44gICAArVrdWu++JUwGebe/SSKiumqseHsn9mhoBuJy0g2SDABQLgTic9IblGioa48ImSRDqNoVoWpXPBXUCQBQpi1HbPZ1/Df+KD6M3VvlecXlZSguL6s0CWVD3ZnAUFkoYFtDAsMgacEEBpmhUm05/u9SDBad3IWzmhsAKoYrvRDaHa+F9YWPrYNpK0jNSoCTDSDkcBFuuCqlYnNybKMmGpqCgpJy2M6tPIeRMWkF8NIvp/HSL6eNXnbegsFQKRv/0W/58uXIyMjAtGnT8OKLLxrsa9WqFX744QcEBARg5cqVmDdvHiRJwvXrFcth9e7du1KSAQA6d+7c6PUmIiLjYaKhGaiYA0GqNKdCsL2zCWtVwUImRwcnT7QKs8FHZ/ZVqmPMsFehtrRCflkJ8kqLK/69/fvSEuSV3fH9zX/zSktuHq/bVnFemdACaNwERqWkxR0JjMrbDBMYqpuJi9u31TWBwfkuWo7CslJ8G3cIS07vweW8LAAVQ5Kmt+2Fl9v1gouV7V1KIKq7AKeK7vvyfBfANhWbk2Lxzn0PmbhW1BRs27YNADBu3Lgq93t5eaF169Y4c+YM4uLiEBISgsjISMjlcnz77bcICQnByJEj4eHhcS+rTURERsREQzNQ3ZwK5vTms7o6Rjga/yGipLzsjmRFMfLKSgy/r2UCQ/99FQmMzEZMYFQ1TOT2BMZZzQ1svHxKP9/F5z1G4oXQ+41aHzK9nJIifH4uGsti9+FGUR4AwNXKFq+G9cGLbXrAXsFutNR4ApxsAADZ1+0hs5NwIvMKEnIzEGDX6i5nNl82CjnyFgw2apmpmkK0XbwH2ts6Jsol4MysfvBSG3euBhvFvemRd+nSJQAVvRPuJi0tDSEhIQgKCsJHH32EmTNnYtq0aZg2bRr8/Pxw//3349FHH8WYMWOgUCgau+pERGQkTDQ0E5NDuuEhr9A6zalwr92rOirkFnCSW8BJaWO0MoUQKNGWG/a8qCGBYZisuJXQuDOBkVtajPIGJjC0EJgSvQHfXjiMPu6B6Obii24uvmb5N0C1k1aUhxVnovDJ2ShoSipWkvGzdcSs8H54pnVXWFtYmriG1BL430w05BVK6OHsj+i0BGxJOoN/ht39zWNzJUmS0YcehLja4asxHTBl3clbifgx7RHiWnn4QFOh1VbEtdGjR0OlUtV47O1zLkyfPh1jx47Fli1bEBUVhaioKKxZswZr1qzB/PnzsX//fvZyICJqIphoaEbqu8rEvdQU6lgVSZKglFtA2cgJjNuHgVSXwDivuYGNSZXH7h5MT8LB9CT9z5429ujm7Ivurn7o5uKLTq28YWupNFrdyfiS87LxYexefHX+gH4S1TZqV8xpPwCPB94HS84PQveQlaUcHvZKXM0pRjeHYESnJWBz0ukWnWhoLJO7+eKhUBfEp+cj2MxXnagNHx8fxMXF4Y033qjz3Apubm54/vnn8fzzzwMAzp07h2effRZ///03Zs+ejVWrVjVGlYmIyMiYaCAyofokMFLys7E5OdZgvguZJGFx50cQl5OOg2lJOJl1FVcKcrAx6bQ+KSGTJIQ7uOt7PHRz8UVbtRvkN2cCJ9O5oEnDB6d2478Xj6JUWzGjfadW3nizw0AM8w2DTOLviEwjwMkGV3OKEaioWAFh3/UEZBYXGDXhShW8HaybVIJBN4yhrKys0r7BgwcjLi4Oa9eubfAkjm3atMEbb7yB4cOH4/jx47WuAxERmRYTDURNTHXzXUwO6aY/Jr+0GEczUnAwLUn/lVKgwcmsqziZdRVfXzgIALCzVKKLs09F4sG5IvngbtN4y9yQoeMZqVh4chfWJZ6EuLlyTD/3IMxtPxAPeLY22jJ0RPXl72iD6MQsFOQpEOHogVNZV7Et+SyeDu5k6qqRiXl7eyMuLg6xsbEIDg422Ddz5kz88MMPWLZsGdzc3DB9+vRK8yskJCTgr7/+wlNPPQUA2LVrF4qKivDggw/C0vLW8DAhBLZu3QoA8PPzq3UdiIjItJhoIGqC7jbfhcpSiT7uQejjHqTflpqvuZl0uIyD6Uk4kp6C3NJi7Loaj11X4/XH+aoc0M3FF91dKoZcRLby5pwARhZ1PQELT+7E9pRz+m2P+rTFnIiB6OHmb7qKEd0hoFVFz4WEzAIMCwzDqayr2Jx0mokGwqhRo7B792489dRTGDRoEBwdHQFUJBlCQ0OxefNmjBo1Cq+//joWL16M8PBweHh4QKPR4OzZs7h48SK6deumTzScPHkSM2bMgL29PSIjI+Hp6YnCwkIcO3YMly9fhlqtxr///e861YGIiEyHiQaiJqqu8114qdQYqYrASP8IAECZthxnsq/rezwcSLuMM9k3kJSfjaT8bKxLPAkAsJBkaO/koe/10N3V7+aSquzOXxdCCOxIPY8FJ3di//UEABXDWcYFdMTsiP5o7+Rp4hoSVaZbeSIhswDP9QvHeyf+xP9Sz6OorBRWTEC2aC+++CJyc3Px448/Yvv27Sgqqpi49qmnnkJoaCj69OmD2NhYrFy5Etu2bcPhw4dRXFwMV1dX+Pr64qmnnsKoUaP05T322GPQaDTYv38/4uLicODAAVhbW8PHxwezZ8/GSy+9BG9v7zrVgYiITEcS4raB3nTP5OTkQK1WQ6PRwN6eXdXJPOSUFOFIenLFxJI3ExDXCnMrHeegsEZX3ZALF190c/GDs1XNM4u3VOVaLX65fAoLT+5CTGYqAEAhk2NScBfMjOiHYHtnE9eQqHq74tIx8Iu/Eeqiwtk3+sN37XtIKdBg6wPP4hGfdka9VmPFxbqUW1RUhISEBAQEBMDKisvHkunxb5KIjO1evQ9ljwYi0rNXWGGAZ2sM8GwNoOJT+KT8LIO5Ho5mpCC7pBC/X7mA369c0J8bZNfKYKLJjk5eUMpb7ktMSXkZfrp0DItO7saFnDQAgI2FJaaG3o9Xw/rCS6U2cQ2J7k7XoyExqxBCAMN8w/DpuWhsToo1eqKBiIiImo+W+y6AiO5KkiT42TrBz9YJYwM6AgBKteU4lXkVB9IuVyQf0pNwXpOGi7kZuJibgdWXYgBUfGrf0ckT3W7O9dDdxReBdq2a/QSHBWUl+M+Fg1h6ei+S87MBAI4Ka0xv1wsvt+2FVuz5QU2Ij4MV5DIJxWVaXMstxjDfcHx6Lhpbks/gC6HlECoiIiKqEhMNRFQnljI5Ip29EensjX+07QkAyCouwOH0ZIP5HjKKC3AoPRmH0pPxydmKc52VKnR18dEPt+jq7APHZrJMXnZxIT47F43lZ/YhrSgfAOBubYfXwvpiSpvusLNkl1dqeizkMvg4WCExsxAJmQXo6xsIe0srXC/MxWdnozHcL7xOc8UQERFRy8BEAxE1mKPSBoO8QjHIq2LyLSEELuVm3BpykZ6EmIxUpBfnY3vKOYPVFkLVLvqlNbu5+KG9kwcsZXJT3Uqd3SjMxfIz+/Hp2WjklFZMRBZg64RZEf0wKbgLJ8yjJi/AyUafaOgZ4IQ2alccSk/C9IOb8M9Dm/HVHcvrEhERETHRQERGJ0kSguydEWTvjCeCIgEAxeVlOJ6ZajDfw8XcDJzXpOG8Jg0/XDwKALCSWyCylTe63zbfg6/K0eyGXFzOy8TS03vxnwsHUVReBgAIc3DDnPYDMS6gAyyaULKEqCb+jjYAMpCQWYCU/GwcTk/S79MKgSnR6/GQVyh7NhAREZEeEw1EdE8o5RY352vw029LL8rHoZtDLQ6mJeFQejKySwoRfSMR0TcS9ce5Wdvd1uvBF12cfWCvMM1QhHPZN7Do1C78dPEYyoQWANDV2Qdz2w/EY77tOGadmp2AVjeXuMwoQFxOOu5cqqpcCMTnpDPRQERERHpMNBCRyThbqTDEpy2G+LQFAGiFFnE56QZzPZzMvIrrhbnYkhyLLcmxAAAJEto5uOonmuzm4oswB7dG7UVwND0FC0/uxC+XT0PcfKs10KM15rYfgP4ewWbX44LIWHQrTyRkFqC1vT9kkgTtbStjyyWJy7QSERGRASYaiMhsyCQZQtWuCFW7YkJwZwBAYVkpjmWkGMz3cDkvC7HZ1xGbfR3fxh0CAKgsFOjs7H2z50NFAqKhS0gKIbD/+iUsOLkLO1LP67cP8w3DnPYDDHpnEDVXtycavFUO+KrHaEyJXo9yISCXJHzZYzR7MxAREZEBJhqIyKxZW1iip1sAeroF6LddK8jRJx0OpiXhcHoyckuLsffaJey9dkl/nJeNWr+0ZjcXX3Rq5Q2VpfKu1xRCYHvKWSw4uUs/hEMuyfB4YEe8EdEf4Y4eRr9PInOlSzSkaIpQVq7F5JBueMgrFPE56Qi2d2aSgYiIiCphooGImhx3G3sM8wvHML9wAEC5Votzmhv64RYH05JwOvsaUgs0+OXyKfxy+RSAimRBuKO7wXwPbR1cIZNkSMnPxrnsGzifk4avzh/AyayrACrmlngmuAtmRvRDoF0rk90zkam42ymhtJChuEyL5OwiBLSygbfKgQkGIiIiqhYTDUTU5MllMoQ5uiPM0R3PhnQFAOSVFuPobUMuDqRdxpWCHJzIvIITmVfw1YUDAAA7SyW8bdQ4p7lhMMmdrYUSL7a5HzPC+sDDxt4Ed0VkHmQyCf6O1jiflo+EzAL95JBERERE1WGigYiaJVtLJfq6B6Gve5B+W0p+tsHymkcyKoZcnNXcMDhXAhD96DREcIgEEYCKlSd0iQYiIiKiu2GigYhaDF1371H+7QEAZdpyfBd3GC9Erzc4TgDIKMo3QQ2JzNPtE0ISERER3Q0XfCeiFstCJsdg7zaQ3bE0JZfrIzLk73gz0ZDBRAMRERHdHRMNRNSi6Zbrk99MNnC5PqLKdPMysEcDAYAkSZDuSNA2lsTEREiSBH9//zqf6+/vD0mSkJiYaPR6ERFRzTh0gohaPC7XR1QzDp0gqj9dUkYIcZcjiYiaDyYaiIgALtdHVANdouFabjEKS8thbSk3cY2IiIjInHHoBBEREdXIycYSdsqKzyYus1cDERER3QUTDURERFQjSZI4fIKqtGHDBvTq1Qv29vZQqVTo2bMntm/fXu3xmZmZmDt3LsLCwmBjYwM7Ozt06tQJixcvRmFhYZ2vf+bMGYwZMwbOzs6wtrZGeHg4li5divLy8hrPq2s9vv/+e0iShEmTJlVZXlVzSbz99tsGc1no5rbQfXHuCCJqzsw+0bBu3Tr069cPjo6OUKlU6NChAxYvXozS0tI6lRMdHY1//OMfuP/+++Hl5QUrKyuoVCq0a9cO06dPr/bF/syZM3jllVfQp08f+Pr6wsbGBtbW1ggODsbkyZNx6tQpI9wlERGReQtwsgYAJGTW/c0g3V1KfjZ2X41HSn62qatSa/Pnz8eYMWMAAEOGDEHr1q0RHR2NRx99FBs3bqx0/KVLlxAZGYmFCxciLS0NQ4YMwYABAxAXF4c33ngDvXr1QlZWVq2vHxUVha5du2L9+vVQq9UYPnw4PDw8MHfuXIwbN67a84xdj+p07NgREydO1P88ceJEgy9bW9sGX4OIyGwJM/bPf/5TABAWFhZi0KBBYuTIkcLBwUEAEL169RIFBQW1LuvNN98UAISvr6/o37+/GD9+vBg8eLBwc3MTAIRKpRK7d++udN7XX38tAAg3NzfRp08fMXbsWPHoo48KPz8/fd1Wr15d53vTaDQCgNBoNHU+l4iI6F57ZdMpgVe3iNe3xDZK+Y0VF+tSbmFhoThz5owoLCyscr9WqxV5JUVG//r0TJSQffu6wLevCdm3r4tPz0Q1ynW0Wq1R2hSAACAcHBzEgQMHDPbNnz9fABAhISGVzuvWrZsAIIYOHSry8vL022/cuCEiIyMFAPHEE08YnJOQkCAACD8/P4PthYWFwsfHRwAQr7zyiigrK9PvO3HihHB2dtbXMyEhocH1+O677wQAMXHixCrbpLp6CnGrverjbn+TRER1da/eh0pCmOcUuJs2bcKIESNga2uLvXv3IjIyEgCQnp6OAQMG4NSpU3jttdewdOnSWpV39uxZWFtbV1oeqaSkBLNmzcLHH38Mb29vJCYmQi6/NclVYmIiioqK0KZNG4PztFotli1bhpkzZ8LGxgYpKSlwdHSs9f3l5ORArVZDo9HA3t6+1ucRERGZwsf7LuGVzbEY1d4D6yd2Nnr5jRUX61JuUVEREhISEBAQACsrq0r780uLYfvjm0ar272W99T7UFkqG1yObjjAihUrMH36dIN9xcXFcHNzg0ajQVJSEnx8fABU9D7o3bs3bGxscOnSJbi5uRmcd/ToUXTu3BkymQyXL1+Gt7c3gIrnsICAAPj5+Rn0Pv3pp5/w1FNPwcfHBxcvXoSlpaVBecuXL8eMGTMAAAkJCfrnv/rW4/vvv8czzzyDiRMn4vvvv6/UJtXV8/b2qs8j993+JomI6upevQ8126ETCxYsAADMnj1bn2QAAGdnZ3z22WcAgJUrV0Kj0dSqvLZt21a5BrNCocCSJUtgZWWFlJQUnDlzxmC/v79/pSQDAMhkMrz++usIDAxEQUEBoqKiantrRERETQ7naKA7PfbYY5W2KZVKBAYGAgBSU1P12/fs2QMAePjhhyu9uQeATp06oUOHDtBqtdi7d+9dr60rb+zYsZWSDAAMhixUdZ6x6kFERFUzy+UtU1NTcfjwYQDAE088UWl/r1694OPjg+TkZGzfvh2PP/54g64nSRJksoqci1JZt0y/hYVFvc4jIiJqSgJa3Uw0ZLTcRIONhQJ5T71v1DJT8zVou3EJtLj1abdcknBm+Ex4qdRGvZaNhcKo5fn6+la5XfcJWVFRkX6bLukQEBBQbXlBQUE4ceKEQYKiOikpKTWW5+joqP/E7nbGrgcREVXNLBMNMTExAAAnJ6dqA0Hnzp2RnJyMmJiYBiUaysvL8c4776CgoADt2rVDcHBwrc/96quvcOHCBbi6uqJ79+71rgMREZG50/VoyCoshaawFGrryp8iN3eSJBll6MHtQhxc8VXP0ZgSvR7lQkAuSfiyx2iEOLga9TqNQfchDVUMqSUiolvMMtGQkJAAoPpMOQD9mD/dsbWVlJSEt956C0DF0kYxMTFISUlBcHAw1q5dW23QLCgowD/+8Q8AgEajwenTpxEfHw83NzesW7eO8ywQEVGzZqu0gLNKgfT8EiRkFqCjl3E/bW/JJod0w0NeoYjPSUewvTO8VQ6mrpLReXl5AahY8aE6un26Y2tTXnWrhmVnZ1c5vLa+9VAoKnqD5ObmVnnO5cuX71pnIqKWxCwTDboXcZVKVe0xuiWBcnJy6lR2ZmYmVq1aZbAtMjIS3377LcLCwqo9r6SkpNJ5gYGB+Oabb9C7d++7Xre4uBjFxcX6n+tabyIiIlMLcLJBen4JEs040dBU4623yqFZJhh0+vXrBwD43//+h+vXr1eaHyEmJgbHjx+HTCZDnz597lpe37598c0332Dt2rVYsGBBpXkafvjhB6PWQ5d0OHfuXJXlbtu2rdq6WlpaorS0FGVlZfoht0REzV2L6/PWsWNHCCGg1WqRkpKCtWvXoqCgAJ06dcKKFSuqPc/BwQFCCAghcO3aNWzfvh0uLi7o378/Zs6cedfrLly4EGq1Wv+l65FBRETUVDSFCSEZb81Tr1690K1bNxQWFmLKlCkoKLj1N5Seno4pU6YAAMaPH1+r39no0aPh5eWFpKQkzJkzx2DowunTp/Hee+8ZtR5du3aFvb09zpw5g//+978GZa5bt67GZ0jdyhWxsbF3vS8ioubCLBMNdnZ2AID8/Pxqj8nLywOAeg9ZkCQJXl5eGDNmDP7++2+4ublhxowZOHHixF3PdXNzw+DBg7Fv3z507NgRS5cuxdatW2s8Z86cOdBoNPqv5OTketWbiIjIVG4lGgpNXJPqMd6ar9WrV8PPzw+bN29GQEAAxowZg+HDhyMoKAiHDx9GZGQkVq5cWauyrK2t8dNPP8HGxgYffvghQkJC8Pjjj2PQoEGIjIxE79694efnZ7R6WFtb45133gEATJgwAT169MCYMWMQHh6OcePGYfbs2dXWddSoUQCABx54AOPGjcNzzz2H5557DhkZGbW6VyKipsgsEw26ZShrejjQ7atqycq6cnBwwIgRI6DVarFly5Zan6dQKPDkk08CADZu3FjjsUqlEvb29gZfRERETUlAK2sA5t2jgfHWfAUGBuLYsWOYM2cOWrVqha1bt+KPP/5AUFAQFi1ahKioKDg6Ota6vL59++LgwYMYOXIksrKysHHjRqSkpODf//43fv75Z6PX45VXXsGqVasQGRmJmJgY/P7773Bzc8Pvv/+OZ599ttrrvfvuu5g1axYcHBywadMmfPPNN/jmm2+qne+BiKg5MMuBYvfddx8AICMjAwkJCVWuPHHkyBEAFfMrGINuPogbN27ck/OIiIiaGn9H8x86QY1PCFHj/j179lS7z8nJCQsWLMCCBQtqdS1/f/8arxceHo4NGzZUua+6iSLrUw+dCRMmYMKECVXuq66eVlZW+OCDD/DBBx/U6VpERE2ZWfZo8Pb2RpcuXQBUdG+7U1RUFJKTk6FUKjFkyBCjXHPXrl0AgJCQkDqdt3PnznqdR0RE1NQEtLqVaLjbm00iIiJqucwy0QAAc+fOBQAsWrQIx44d02/PyMjQLzM5bdo0qNW3Zr3euHEj2rRpg4EDB1Yqb+HChUhLS6u0PSsrC9OnT8eRI0egVqsxduxYg/3Lly+vcghHQUEB3nvvPWzYsAEWFhZ45pln6nejRERETYSfozUkCSgoKUdaXompq0NERERmyiyHTgDA8OHD8fLLL2PFihXo3r07Bg4cCJVKhZ07dyI7Oxs9e/bEu+++a3CORqPB+fPnUVRUVKm8uXPnYt68eYiIiEBQUBAsLCyQmpqKmJgY5OfnQ61WY926dZWWOVq+fDleffVVtG3bFqGhobCyssLVq1dx4sQJZGVlQalU4uuvv0Z4eHijtgcREZGpKS3k8LS3QqqmCAmZBXC1U5q6SkRERGSGzDbRAAAff/wxevbsiU8//RTR0dEoLS1FUFAQZs+ejRkzZkChUNS6rJUrV2L//v2IiYnBzp07kZeXBzs7O0REROChhx7Ciy++WCnJAAALFizA77//jiNHjmDfvn3Izs6GSqVCUFAQJk+ejBdffBGBgYHGvG0iIiKzFeBko080dPOr/cR9RERE1HJIgoMsTSInJwdqtRoajYYzYhMRUZMxYXUM/ns0BQuHtMHsga2NVm5jxcW6lFtUVKSfhNrKyspodSCqL/5NEpGx3av3oWY7RwMRERGZnwAnrjxBRERENWOigYiIiGpNl2g4mpKNlOxCE9eGiIiIzBETDURERFRrsddzAABHU3Lg996f+OZgkolrZHwcVUrmgn+LRNRUMdFAREREtZKSXYhley/pf9YKYMq6k82mZ4NMVvFYVF5ebuKaEFXQ/S3q/jaJiJoKvmoRERFRrcSl50N7xwes5UIgPj3fNBUyMktLS1haWiIvL8/UVSECABQWFkIul8PS0tLUVSEiqhMmGoiIiKhWWjurIJMMt8klCcHOKtNUyMgkSYKdnR00Gg0KC5tHLw1qusrLy6HRaGBjYwNJku5+AhGRGbEwdQWIiIioafB2sMZXYzpgyrqTKBcCcknCl2Paw9vB2tRVMxpnZ2cUFhYiKSkJ9vb2sLOzg1wu5xs9umeEECguLkZmZia0Wi1cXV1NXSUiojpjooGIiIhqbXI3XzwU6oL49HwEO6uaVZIBAORyOXx8fJCeno7c3FxkZ2ebukrUQqlUKri7u0OhUJi6KkREdcZEAxEREdWJt4N1s0sw3E4ul8PNzQ2urq4oLS2FVqs1dZWohbGwsICFBR/Tiajp4isYERERURUkSeKnyURERPXAySCJiIiIiIiIyGiYaCAiIiIiIiIio2GigYiIiIiIiIiMhokGIiIiIiIiIjIaJhqIiIiIiIiIyGiYaCAiIiIiIiIio2GigYiIiIiIiIiMxsLUFWiphBAAgJycHBPXhIiIyPR08VAXH42F8ZaIiOiWxoq3d2KiwURyc3MBAD4+PiauCRERkfnIzc2FWq02ankA4y0REdHtjB1v7ySJxk5lUJW0Wi2uXLkCOzs7SJJklDJzcnLg4+OD5ORk2NvbG6XMloptaTxsS+NhWxoX29N4jNGWQgjk5ubC09MTMpnxRnYy3po3tqXxsC2Ni+1pPGxL4zHneHsn9mgwEZlMBm9v70Yp297env+JjYRtaTxsS+NhWxoX29N4GtqWjfHJCuNt08C2NB62pXGxPY2HbWk85hhv78TJIImIiIiIiIjIaJhoICIiIiIiIiKjYaKhGVEqlZg/fz6USqWpq9LksS2Nh21pPGxL42J7Gk9La8uWdr+NiW1pPGxL42J7Gg/b0niaUltyMkgiIiIiIiIiMhr2aCAiIiIiIiIio2GigYiIiIiIiIiMhokGIiIiIiIiIjIaJhrM1Lp169CvXz84OjpCpVKhQ4cOWLx4MUpLSysdm5+fj5deeglubm6wsrJCt27dsGvXrmrL3rp1KyRJwoIFCxrzFu6J0tJS7Ny5EzNnzkSXLl3g4OAAS0tLuLu7Y+jQodi2bVuN5//5558YMmQInJ2dYW1tjTZt2uDNN99EXl5elceXlZVh3rx58PHxgVKpREREBNatW1dt+cePH4elpSVeeOGFBt2nqcyaNQuSJEGSJLz33nvVHsd2rF5JSQlWrFiBXr16wcnJCVZWVvD29sbgwYPx888/V3kO27OypKQkTJs2DaGhobC2toaVlRUCAgIwceJEnDhxotrzWmpbnj9/Hp988gkmTZqEiIgIWFhY3PX/sU5d20wnPj4ekyZNgre3N5RKJby9vTFp0iRcunSp2nNWr16NsLAwKJVK+Pr64u2330Z5eXmVx+bn58Pf3x/h4eEoKSm5633UBWPu3THeNj7G3IZjzDUOxtzaY7ytgSCz889//lMAEBYWFmLQoEFi5MiRwsHBQQAQvXr1EgUFBQbHjxo1SgAQ4eHhYujQocLa2lpYWFiIv//+u1LZubm5wsfHR0RERIiSkpJ7dUuN5o8//hAABADh7u4uHnnkETF27FgRHh6u3/7CCy8IrVZb6dxly5YJAEKSJNGnTx8xZswY4e7uLgCI0NBQkZaWVumc1157TQAQgYGBYsSIEfrfy7p16yodW1ZWJjp37izc3d1FVlZWY9x+o/rrr7+ETCYTkiQJAOLdd9+t8ji2Y/WSk5NFu3btBADh7OwsHn30UTFu3DjRo0cPYWNjI0aNGlXpHLZnZQcOHBB2dnYCgPDy8hJDhw4VI0aMEAEBAfrXyrVr11Y6ryW3pS6O3PlV3f9jnfq0mRBCREVFCRsbGwFAhIWFiXHjxomwsDABQKhUqirj0a+//ioACEdHR4Pf57Rp06q8xiuvvCIkSRJ//fVX3RukBoy5tcN427gYcxuOMdc4GHPrhvG2ekw0mJmNGzcKAMLW1lYcPXpUvz0tLU1EREQIAOK1117Tb4+JiREAxMCBA0VZWZkQQog///xTABCPPPJIpfJffvllIZPJqvwjbIp27twpRo0aJfbt21dp35o1a4RcLhcAxKpVqwz2HTt2TEiSJORyudi+fbt+e35+vhg4cKAAUCkgXb9+XSgUCtGuXTuRn58vhBDi7NmzwtLSUoSFhVW6vu4FpKoXY3OXn58vWrduLby8vMTw4cOrfcFkO1avoKBAtGnTRgAQb7/9dqU3Gfn5+SImJsZgG9uzau3bt9e/ibm9HcvLy8W//vUvAUA4ODiIwsJC/b6W3pZff/21eP3118VPP/0kzp49K55++um7PvjUp810+z09PQUAMWfOHIN9c+bMEQCEj49PpTfsHTt2FAqFQpw/f14IIUReXp5o06aNkMvl4urVqwbHHj58WMjlcvGPf/yjzm1RE8bc2mO8bTyMuQ3HmGs8jLl1w3hbPSYazEyXLl0EAPHee+9V2rd//34BQCiVSpGdnS2EEOI///mPACB++ukng2ODg4OFo6OjwbZDhw4JmUxWbfaqOZo8ebL+ofB2Y8aMEQDEc889V+mcxMREIZPJBABx9uxZ/Xbdw+T7779vcPwDDzwgAAiNRqPfdvnyZaFSqcSjjz5q5Du6N15++WUBQGzbtk1MnDix2hdMtmP15s2bpw/UtcX2rCw9PV3/6cCNGzcq7S8rKxPW1tYCgDh27Jh+O9vSUE3/j3Xq02ZCCPHpp58KACIkJESUl5cb7CsvLxchISECgPjiiy/024uLi4VcLhcPPvigwfHvvfeeACA2b96s31ZWVibuu+8+4eXlZfA7MAbGXONhvK0/xtyGY8w1DsbchmO8vYVzNJiR1NRUHD58GADwxBNPVNrfq1cv+Pj4oLi4GNu3bwcAZGRkAACcnJwMjm3VqpXB+J6ysjK88MIL8PT0bPLjROvivvvuAwAkJyfrt5WUlOjHklbVzn5+fujZsycAYOPGjfrtNbU1AIP2fumllyBJEj777DNj3MY9tWfPHnzyySeYMGEChgwZUu1xbMfqlZaW4vPPPwcAzJw5s1bnsD2rplQqa32ss7MzALZlfdS3zW7/efz48ZDJDB8rZDIZxo0bBwD45Zdf9Nuzs7NRXl5eq7b+6KOPEBMTg5UrV8Le3r5e91cVxlzjYrytH8bchmPMNR7G3MbXkuItEw1mJCYmBkDFf7iAgIAqj+ncubPBsf7+/gCAs2fP6o8pLS1FfHw8vLy89NuWLVuG48eP49NPP4WdnV1jVN8sxcXFAQA8PDz02y5cuICCggIAt9rzTne2M1B1W+t+VigU+hfctWvXYuvWrXj//ffh4+NjnBu5R/Ly8vDss8/Czc0Ny5cvr/FYtmP1jh07hvT0dHh6eiI4OBinTp3CO++8gylTpmD27NnYtm0btFqtwTlsz6rZ2tqid+/eAIB//etfBpPzabVavP322ygsLMTgwYP198W2rLv6ttntP9flPFdXV9jY2FTZ1gD08SsxMRHz58/HiBEjMHz48Lrc0l0x5hoX423dMeYaB2Ou8TDmNr4WFW8b3CeCjGbFihUCgOjYsWO1x+i6140ePVoIUTGO1M7OTnh5eYlDhw6JrKws8eqrrwoA4uWXXxZCCHHp0iVhY2OjP6eluHr1qlCr1QKAWLFihX77li1b9OPLqqMbB9a5c2f9tuLiYuHn5yfs7OzEjh07hEajER9++KEAIIYOHSqEECIrK0u4u7uLrl27VurS1BRMmTJFABAbN27Ub6uuCxjbsXpfffWVACC6du0q3njjDf3kXrd/3XfffeLy5cv6c9ie1Tt37pwIDAwUQMXEVMOGDRMjR44UAQEBQqFQiKefftqgix/bsrK7deWsb5vl5OTo/6aPHz9e5XnHjh3TH5OXl6ffPn78eAFAfPjhh0Kj0YgdO3YIOzs74erqKoqKioQQQjz88MPC3t5epKam1ue2a8SYazyMt/XDmGscjLnGxZjbMIy3tzDRYEbef/99AUD07Nmz2mPmzp0rAIhBgwbpt3322WeVXlDbtm0rMjMzhRBCPPTQQ0KtVosrV64YlKWbeKU5Ki0t1U+mEhERIYqLi/X7fvrpJ/2LZ3V0QSskJMRg+5YtW4SFhYVBW3t4eIiLFy8KIYR44YUXhIWFhTh58qTBeU2hrXfs2CEAiPHjxxtsr+4Fk+1YvYULFwoAwtLSUgAQL730kjh//rzQaDTijz/+0I+hCw8P10+0xPas2fXr18WgQYMqvda1a9dOfPPNNwbHsi0ru9uDT33bLDU1Vd9OcXFxVZ534cIF/TG3x6GEhATh5uZm0NaWlpb68aKrV68WAMRnn31mUF5hYaF+IsaGYMw1Dsbb+mHMNR7GXONjzK0/xttbOHSiGXjxxRfx999/4/XXX8cLL7yATz/9FEeOHIGjoyN++ukn7NixA4sXL4aHhwcKCgrw0ksvQa1WQ6VSwcXFBW+99Va166g2VVOnTsXOnTvRqlUrrF+/HgqFwijlPvbYYzh+/Djmzp2L559/HkuWLMHp06cRGBiIqKgofP3115g5cyYiIiJQXl6Ot956Cy4uLlCpVFCr1XjppZf03aXMiUajweTJk+Hi4oJPPvmk0a/XXNtRRwgBoKJL9eOPP46VK1ciJCQE9vb2eOCBB/DHH3/AysoKp0+fxpo1axp8vebenn/99RciIiJw+vRprF69GteuXUNmZiZ+/fVXlJaWYvLkyZg8ebJRrtXc29Kc+Pv74/Tp0/jggw/w/PPP41//+hdOnjyJoUOHIisrC6+88gp69uyJqVOnAgB+/vln/Zru1tbWeOSRR5CQkHDP682Ya4jxtu4Yc42LMde4GHObH5PF23qnKMjo6tONsyYZGRnCxcVF9O7dW7+u9YgRI4RMJhOzZs0SW7ZsEc8++6wAIF599VWj3Yep6drI0dHRYEZcnfp2WapJcXGxaNu2rQgODtYv9zNjxgwBQEyePFls2bJFzJw5U0iSJEaOHFm/G2tEkyZNEgDEzz//XGmfsbtx1qSpt6POJ598os8Y79mzp8pjRo0aJQCICRMmCCHYntXJysoSLi4uQpIkceDAgUr7L168qF9PeteuXUIItmVVzLErZ02effZZoVAoRGxsrBBCiE2bNgkAolevXmLDhg1i5cqVQq1WCz8/P5Gbm1urMu/EmNtwjLf1w5hrXIy5xsOY23CMt7cw0WBGdH94rVq1qvaYESNGCADi9ddfv2t5kyZNEkqlUr80yrlz5wQA8cQTTxgc161bN6FQKOr9sGZOdGNlHRwcxOHDh6s85uTJk/r/hDk5OVUeo3sxq+0Y27ffflsAEDt37hRCVLwYKBQK0aNHD4Pjxo0bJwDo17E1F2q1WlhYWIi+fftW+tJ1tfL39xd9+/YV48aNE0KwHWuybds2fdskJSVVeczMmTMFAP1yQ2zPqv34448CgAgKCqr2mP79+wsAYu7cuUIItmVV7vbg05A2c3JyEgDEli1bqjxv8+bNAoBwdnauVV337NkjJEkSb731ln5b9+7dhUql0g9PEEKIzz//XAAQX375Za3KvRNjbsMw3tYfY65xMeYaD2NuwzHe3sKhE2ZEtzRURkZGtd1Tjhw5AgCIjIyssazdu3fj+++/x5w5c9CmTRsAwPHjxwEAPXr0MDi2V69eKCkpwZkzZxpSfZObNWsWli1bBrVajd9//73aGVlDQ0NhY2MD4FZ73qm27QwA58+fx8KFCzFp0iQMGDAAAHDmzBmUlJRU2dbArd+FOSkrK8PevXsrfV2/fh1AxWy0e/fuxYEDBwCwHWsSGRkJSZIAAOnp6VUeo9tua2sLgO1ZnaSkJACocZkltVoNAMjMzATAtqyPhrSZ7mdjtHVxcTGmTJmC0NBQzJ07V7/9+PHjaNu2LRwdHfXbGtrWjLn1x3jbcIy5xsOYazyMuY2vJcVbJhrMiLe3N7p06QIAWL16daX9UVFRSE5OhlKprHGt5eLiYkydOhVt27bFnDlz9Nt1L8L5+fkGx+t+1u1vimbPno0lS5ZArVbjjz/+0LdjVRQKBR555BEAVbfz5cuXER0dDQAYMWJEjdcVQmDKlClQq9X48MMP9dubWltnZ2dDVPRwqvQ1ceJEAMC7774LIQQSExMBsB1r4u7urn9R/vPPPyvtLy0txd69ewEAXbt2BcD2rI5u2aVz585Bo9FU2l9aWopjx44BgH6JQrZl3TWkzXQ/r1mzptISclqtFj///DMAYOTIkXetx/vvv48LFy7gyy+/NFjPXZIko7c1Y279MN42HGOucTHmGg9jbuNrUfG2Tv0fqNFt3LhRABC2trbi6NGj+u3p6ekiIiJCABCvvfZajWW8+eabQpIkERUVZbA9Li5OABDt27cXBQUFQoiKpbpcXV2FlZVVrcfymJs333xT333z0KFDtTrn6NGjQpIkIZfLxW+//abfnp+fr589e9SoUXct5+uvvxYAxE8//WSwPTc3VyiVSuHu7i4yMjL0ZYeHhwug+tlizVFNXcDYjtX7888/9WOX//77b/320tJSMX36dAFA2NnZiWvXrun3sT0ru3HjhlCpVAKAGDNmjEF38+LiYvHSSy8JoGL2ZN0s1UKwLe90t66cQtS/zfLz84Wnp6dBV1od3aoN3t7e+rhTndjYWKFQKMTzzz9faV/Pnj0FAIO4ppvv4D//+U+N5daEMbduGG8bH2Nu/TDmGgdjbsMx3t7CRIMZ0k2uZGlpKR5++GExatQo4eDgIICKZbhq+uM5ffq0sLS0FFOnTq1y/+OPPy4AiICAADFq1Cjh7u4uAIg5c+Y01u00Kt1YJKBiwpSJEydW+VXVg6JuohVJkkS/fv3E2LFjhYeHhwAgQkNDRVpaWo3XvnbtmnB0dBQPP/xwlfvnzJkjAAh3d3cxatQoERAQIACIxx9/3Cj3fq/c7QWT7Vi9d999VwAQFhYWokePHmLkyJHC399fABDW1tZi69atlc5he1b23//+V7/8lYuLixgyZIgYNmyY8PLyEgCETCYTn3/+eaXzWnJbHj16VHTr1k3/5ezsrH8AuX37nUsw1rfNoqKi9BOEhYeHi/Hjx+sfAFUqlcGDf1W0Wq3o2bOncHd3F1lZWZX2//bbb0KSJGFtbS2GDh0qunbtKgCI4ODgBi9zxphbO4y39wZjbv0x5hoHY27dMN5Wj4kGM/Xzzz+LPn36CHt7e2FtbS3Cw8PFokWLDNanvpNWqxU9evQQnp6eIjs7u8pjiouLxRtvvCG8vb2FpaWlCAwMFIsXL9bPkN3UfPfdd/oHn5q+/Pz8qjz/jz/+EA8//LBwcnISSqVStG7dWsyZM6fayVluN378eKFSqURCQkKV+7VarVi8eLEICgoSFhYWwsvLS7zxxhs1/g7NUW0ys2zH6u3YsUMMHjxYODk5CUtLS+Hj4yMmTZqknzCuKmzPyo4fPy4mTZokAgMDhVKpFAqFQvj5+Yknn3xSHDx4sNrzWmpb7t69u1avjVXdW33bLC4uTkyYMEF4enoKS0tL4enpKSZMmCDi4+PvWt8vvvhCABBr166t9phff/1VdOnSRSgUCuHg4CDGjx8vUlNT71p2bTDm3h3j7b3BmNswjLnGwZhbe4y31ZOEuLn4LBERERERERFRA3EySCIiIiIiIiIyGiYaiIiIiIiIiMhomGggIiIiIiIiIqNhooGIiIiIiIiIjIaJBiIiIiIiIiIyGiYaiIiIiIiIiMhomGggIiIiIiIiIqNhooGIiIiIiIiIjIaJBqIWwN/fH5IkITEx0Sjlvf3225AkCW+//bZRymsMkyZNgiRJ+P77701dFSIiaiEYb4mIKjDRQERNzvfffw9JkjBp0iRTV8Vs8cGPiIgaivH27hhviapmYeoKEFHj27lzJ0pLS+Hl5WWU8qZNm4bx48fD2dnZKOU1hoULF2L27Nnw8PAwdVWIiKiFYLwlIqrARANRCxAUFGTU8pydnc36oQcAPDw8+NBDRET3FOMtEVEFDp0gagGqGzPar18/SJKEPXv24Pjx4xg5ciScnZ2hVCrRrl07fPjhhxBCVCrvbmNGL1y4gClTpiAoKAhWVlZQq9Xo06cPfvzxx2rrKITAL7/8gkcffRTu7u5QKBRwd3dHr1698MEHH6CwsFB/L8888wwAYNWqVZAkSf/Vr18/fXl368q4Zs0aDBw4EE5OTlAqlfDz88Ozzz6LCxcu3LUNd+/ejUGDBsHR0RHW1taIjIzEDz/8UO29Vef2Op4+fRrjxo2Dh4cH5HK5vm1LS0vx448/4sknn0SbNm1gb28Pa2trhIaG4uWXX8aVK1cMykxMTIQkSVi1ahUA4JlnnjFoozt/Z4WFhfjwww/RvXt3ODg4wMrKCqGhoZg1axYyMjLqfE9ERC0Z421ljLcVGG+ppWGPBiLCjh07sGzZMgQFBeHBBx/E1atXERUVhddffx3JyclYvnx5rctat24dJkyYgKKiIrRp0wZDhgyBRqPBwYMH8fTTT2PXrl349ttvDc4pLS3F+PHj8csvv0Amk6Fr164YMGAA0tPTcebMGcyePRvjxo2Dv78/Ro8ejQMHDuCvv/5CUFAQevXqpS+nTZs2d62fEAKTJk3CDz/8AAsLC/Tp0weurq44duwYvvvuO/z888/YsGEDHn744SrP//bbb/Hee+8hMjISDz/8MBITE3HgwAFMnDgRmZmZeOWVV2rdVjrR0dGYOnUqPDw80KdPHxQWFsLOzg4AcP36dTz99NNQq9Vo27Yt2rdvj/z8fBw/fhyffPIJ1qxZg+joaAQHBwMAbG1tMXHiRERFReHixYvo2bOnfh8AdOzYUf/9lStX8PDDD+PUqVNwcnJCly5dYGdnh2PHjmHJkiVYt24d9uzZAz8/vzrfExERVcZ4y3jLeEsthiCiZs/Pz08AEAkJCQbb+/btKwAIAOKLL74w2Ldz504hSZKQy+UiOTnZYN/8+fMFADF//nyD7SdPnhRKpVJYWVmJDRs2GOxLTEwUERERAoBYtWqVwb5XX31VABD+/v7i+PHjBvu0Wq34888/RXZ2tn7bd999JwCIiRMnVnvPEydOFADEd999Z7D9888/FwCEs7OziImJMbiO7r4cHBzEjRs3DM7TtaGlpaX49ddfDfbp6qNWq0VBQUG1daqujgDE7NmzRXl5eaVjcnJyxObNm0VxcbHB9pKSEjFnzhwBQAwZMqTW93/7/fbs2VMAEJMnTxY5OTn6faWlpeK1114TAET//v1rfT9ERC0d4+0tjLe37pfxlloiJhqIWoC7PfiMHDmyyvMefvhhAUD88MMPBture/AZN26cACCWLl1aZXmHDh0SAESnTp30265fvy4UCoUAII4cOVKr+2nIg09QUJAAIFasWFHpHK1WK9q3by8AiPfff99gn64NX3311Sqv16ZNGwFA7Nu3r1b3cHsdQ0JCRFlZWa3Pu52np6eQyWQGDy63l13dg89vv/0mAIiOHTuK0tLSSvvLy8tFeHi4ACBOnTpVr7oREbU0jLe3MN5WYLyllopzNBARHnvssSq3t23bFgCQmpp61zK0Wi1+++03AMC4ceOqPKZz586wtbVFTEwMioqKAAC7d+9GSUkJOnXqhE6dOtWn+rWWkpKCixcvAgAmTpxYab8kSfrxqLt3766yDGO01Z2GDx8OuVxe4zEnTpzAsmXLMH36dDz77LOYNGkSJk2ahLKyMmi1WsTHx9fpmtu2bQMAjBo1ChYWlUfRyWQy9OnTB0BFV1MiImo4xtsKjLe3MN5Sc8U5GogIvr6+VW63t7cHAP1DSk0yMjKQk5MDAPDx8anV8V5eXrh8+TKA2o33bCjdQ0mrVq3093Yn3Yzh1T3AGKOt7uTv71/tvvz8fDz99NPYuHFjjWXo2r62Ll26BACYN28e5s2bV+OxaWlpdSqbiIiqxnh7C+NtZYy31Jww0UBEkMka3rlJq9Xqv6/q04s7KZXKBl/TFIzRVneytraudt+cOXOwceNGtGnTBosWLUKXLl3g7OwMhUIBAOjRowf+/vvvKmcrr4nu99WrV6+7LscWFhZWp7KJiKhqjLe1x3hL1LQx0UBERuHs7Axra2sUFhZi6dKltV73W/eJxblz5xqzegAALy8vALc+DarqUxbdJw+6Y01t7dq1AICff/4Z7du3r7Q/Li6uXuXqPgUbNmwYXn/99fpXkIiI7inG28bBeEtkXJyjgYiMQi6X48EHHwRwK1jXxoABA6BQKHD06FEcO3asVufoPl0oKyurUx29vb31nyZUtd63EEK/vX///nUqu7FkZmYCQJVLXu3YsQPp6elVnne3Nho8eDCAiuXR6vrpDBERmQ7jbeNgvCUyLiYaiMho5s+fD4VCgZkzZ2LVqlUG3Tt1Tp8+jV9++UX/s6urK1588UUAwJgxY3D69GmD44UQ2LVrFzQajX6bt7c3AODMmTN1rqPu04R3330XJ06cMLjOe++9h+PHj8PBwQHPP/98nctuDLpJrz755BOD7efPn8fUqVOrPU/XRrGxsVXuHzZsGLp06YJDhw7hmWeeqXJcaFZWFr744os6P2ASEVHjYrw1PsZbIuPi0AkiMprIyEj8+OOP+hma//Wvf6Fdu3ZwcXFBZmYmTp06hZSUFIwbNw4jR47Un7d48WIkJCRgy5Yt6NChA7p164aAgACkp6cjNjYWqampSEhIgFqtBgB0794dnp6eiImJQWRkJCIiImBpaYnQ0FDMnDmzxjpOmTIF0dHR+O9//4vOnTujb9++cHV1xbFjx3D+/HlYW1tj9erVcHFxadS2qq358+dj9OjRmDdvHtauXYuwsDDcuHED+/fvR+/eveHp6VnlLNXDhw/HO++8gxUrVuD06dPw8fGBTCbD0KFDMXToUMhkMmzatAmPPPIIVq1ahfXr16NDhw7w9fVFSUkJLl26hFOnTqG8vByTJk2qcqZsIiIyDcZb42O8JTIu9mggonqTJKnStjFjxiA2NhYzZsyAg4MD/vrrL2zYsAFnzpxBcHAwFi1ahPfff9/gHIVCgU2bNmH16tV44IEHcOHCBaxbtw4nT55EYGAglixZAnd3d4Pjd+zYgaFDhyIlJQU//vgjvvnmG/0SUner8w8//IDVq1ejV69eOHr0KNavX4+CggJMmjQJMTEx+m6O5mDkyJHYu3cvBg4ciKtXr2LLli24ceMG3n77bfz222+wtLSs8rz27dtjw4YNuP/++3Hw4EF8//33+Oabbwy6y3p6euLAgQP44osv0LVrV5w/fx7r169HVFQUAGDq1KnYsWMHrKys7sm9EhFR1RhvGx/jLZFxSYKDhYiojmbNmoUlS5Zg8eLFd/1Eg4iIiOqH8ZaImir2aCCiOhFC4NChQwCA0NBQE9eGiIioeWK8JaKmjD0aiKjWnn76aZw6dQonTpyAl5cX4uLialyTmoiIiOqO8ZaImjomGoio1iRJgp2dHXr16oWlS5eiXbt2pq4SERFRs8N4S0RNHRMNRERERERERGQ0nKOBiIiIiIiIiIyGiQYiIiIiIiIiMhomGoiIiIiIiIjIaJhoICIiIiIiIiKjYaKBiIiIiIiIiIyGiQYiIiIiIiIiMhomGoiIiIiIiIjIaJhoICIiIiIiIiKjYaKBiIiIiIiIiIzm/wFhBeQ7ZnfcqQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2, figsize=(12,4), sharey=True, sharex=True)\n",
    "sbnn_plot.plot_metric_dependence(inj_metrics_ds[inj_metrics_ds['model'] == 'ds irloop naics data gen'],\n",
    "                                 xformatter = ticker.PercentFormatter(1),\n",
    "                                 ylabel='PR-AUC',\n",
    "                                 title='NAICS only',\n",
    "                                 xlabel='injection rate', ax=ax[0],\n",
    "                                marker='.')\n",
    "sbnn_plot.plot_metric_dependence(inj_metrics_ds[inj_metrics_ds['model'] == 'ds irloop naics+hier data gen'],\n",
    "                                 xformatter = ticker.PercentFormatter(1),\n",
    "                                 ylabel='PR-AUC',\n",
    "                                 title='NAICS+Hierarchy',\n",
    "                                 xlabel='injection rate', ax=ax[1],\n",
    "                                marker='.')\n",
    "ax[0].legend().set_visible(False)\n",
    "plt.subplots_adjust(wspace=0.05, bottom=0.1)\n",
    "fig.suptitle('Downsampled Data')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('40_PLOT_injection_rate_comparison_ds.png'),\n",
    "        bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a2a595fd-87ae-4fed-a892-8d13835b3043",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8e629f33-0ba8-41e6-9567-64a4fface6eb",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tf_p39",
   "language": "python",
   "name": "tf_p39"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}