02b_Eval_Hodgkin.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prospective Evaluation of Targeted Literature Search: Hodgkin Lymphoma"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from dotenv import load_dotenv\n",
    "load_dotenv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from api.app import app\n",
    "from fastapi.testclient import TestClient\n",
    "import os\n",
    "\n",
    "with TestClient(app) as client:\n",
    "    # Initialize state\n",
    "    pass    \n",
    "\n",
    "session = app.state.session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "OVERWRITE_LATEX_OUTPUT = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "GUIDELINE_ID = \"hodgkin-lymphom\"\n",
    "SCREENING_TF_MIN = \"01/01/2016\"\n",
    "SCREENING_TF_MAX = \"06/01/2023\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "PATH_PREFIX = \"hodgkin_lymphoma\"\n",
    "PATH_GGPONC_LITERATURE = Path(\"data/screening/ggponc_v2.1_2023_03_30_literature.csv\")\n",
    "PATH_RIS_CONTENT = Path(f\"data/screening/{PATH_PREFIX}/ris_with_matched_ids.csv\")\n",
    "PATH_MANUAL_REVIEW = Path(f\"data/screening/{PATH_PREFIX}/manual_review.csv\")\n",
    "PATH_QUERY_RESULT = Path(f\"data/screening/{PATH_PREFIX}/retrieved_evidence.parquet\")\n",
    "PATH_UNSCREENED_HQ = Path(f\"data/screening/{PATH_PREFIX}/unscreened_hq.csv\")\n",
    "PATH_LATEX_TABLE_OUTPUT = Path(\n",
    "    f\"thesis/from_python/tables/evaluation/{PATH_PREFIX}/\"\n",
    ")\n",
    "PATH_LATEX_PLOT_OUTPUT = Path(\n",
    "    f\"thesis/from_python/figures/evaluation/{PATH_PREFIX}/\"\n",
    ")\n",
    "PATH_LATEX_TABLE_OUTPUT.mkdir(exist_ok=True, parents=True)\n",
    "PATH_LATEX_PLOT_OUTPUT.mkdir(exist_ok=True, parents=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Screened evidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_s = pd.read_csv(\n",
    "    PATH_RIS_CONTENT,\n",
    "    parse_dates=[\"publication_date\", \"article_date_api\"],\n",
    ").astype({\"pm_id\": \"Int64\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import ast\n",
    "\n",
    "df_s[\"publication_types\"] = df_s[\"publication_types\"].apply(ast.literal_eval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "tags": []
   },
   "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>is_rct_api</th>\n",
       "      <th>False</th>\n",
       "      <th>True</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>is_relevant</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>False</th>\n",
       "      <td>79</td>\n",
       "      <td>35</td>\n",
       "      <td>114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>40</td>\n",
       "      <td>31</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>119</td>\n",
       "      <td>66</td>\n",
       "      <td>185</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "is_rct_api   False  True  All\n",
       "is_relevant                  \n",
       "False           79    35  114\n",
       "True            40    31   71\n",
       "All            119    66  185"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(df_s[\"is_relevant\"], df_s[\"is_rct_api\"], margins=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "tags": []
   },
   "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>screening_origin</th>\n",
       "      <th>manual</th>\n",
       "      <th>original</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>is_relevant</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>False</th>\n",
       "      <td>11</td>\n",
       "      <td>103</td>\n",
       "      <td>114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>6</td>\n",
       "      <td>65</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>17</td>\n",
       "      <td>168</td>\n",
       "      <td>185</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "screening_origin  manual  original  All\n",
       "is_relevant                            \n",
       "False                 11       103  114\n",
       "True                   6        65   71\n",
       "All                   17       168  185"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(df_s[\"is_relevant\"], df_s[\"screening_origin\"], margins=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "screening_origin  is_relevant  is_rct_api\n",
       "manual            False        True          11\n",
       "                  True         True           6\n",
       "original          False        False         79\n",
       "                               True          24\n",
       "                  True         False         40\n",
       "                               True          25\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_s.groupby(['screening_origin', 'is_relevant', 'is_rct_api']).size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Retrieved Evidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from api.models import EvidenceQuery\n",
    "from api.utils import query_api_for_evidence, evidence_to_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from integration.citation_utils import pm_id_to_publication_date\n",
    "from evaluation.matching import is_in_ggponc\n",
    "from tqdm.auto import tqdm\n",
    "import os\n",
    "from requests.exceptions import ConnectionError\n",
    "\n",
    "if PATH_QUERY_RESULT.exists():\n",
    "    df_query = pd.read_parquet(PATH_QUERY_RESULT).astype({\"pm_id\": \"Int64\"})\n",
    "else:\n",
    "    query = EvidenceQuery(guideline_id=GUIDELINE_ID)\n",
    "    try:\n",
    "        # fetch evidence from API\n",
    "        print(\"Fetching evidence from API\")\n",
    "        evidence = query_api_for_evidence(query=query)\n",
    "        df_query = evidence_to_df(evidence)\n",
    "        df_query[\"guideline_id\"] = GUIDELINE_ID\n",
    "        print(\"Done.\")\n",
    "\n",
    "        # retrieve exact article dates from Pubmed API\n",
    "        tqdm.pandas(desc=\"Fetching article dates\")\n",
    "        df_query[\"article_date_api\"] = pd.to_datetime(\n",
    "            df_query[\"pm_id\"].progress_apply(\n",
    "                lambda pm_id: pm_id_to_publication_date(\n",
    "                    pm_id=pm_id,\n",
    "                    email=os.environ.get(\"PUBMED_USER\"),\n",
    "                    api_key=os.environ.get(\"PUBMED_API_KEY\"),\n",
    "                    cache_path=\"data/literature/pm_id_to_entrez_xml.json\",\n",
    "                )\n",
    "            )\n",
    "        )\n",
    "\n",
    "        # exclude articles outside of screening timeframe\n",
    "        df_query[\"publication_date_combined\"] = df_query[\"article_date_api\"].fillna(\n",
    "            df_query[\"publication_date\"]\n",
    "        )\n",
    "        df_query = df_query.query(\n",
    "            \"publication_date_combined >= @SCREENING_TF_MIN & publication_date_combined <= @SCREENING_TF_MAX\"\n",
    "        )\n",
    "\n",
    "        # check presence in GGPONC\n",
    "        df_ggponc = (\n",
    "            pd.read_csv(PATH_GGPONC_LITERATURE)\n",
    "            .astype({\"pm_id\": \"Int64\"})\n",
    "            .query(\"guideline_id == @GUIDELINE_ID\")\n",
    "        )\n",
    "        ggponc_pm_ids = set(df_ggponc[\"pm_id\"].dropna())\n",
    "        ggponc_titles = set(df_ggponc[\"title\"])\n",
    "\n",
    "        tqdm.pandas(desc=\"Checking for presence in GGPONC\")\n",
    "        df_query = df_query.progress_apply(  # type: ignore\n",
    "            lambda row: is_in_ggponc(\n",
    "                row=row, ggponc_pm_ids=ggponc_pm_ids, ggponc_titles=ggponc_titles\n",
    "            ),\n",
    "            axis=1,\n",
    "        )\n",
    "\n",
    "        df_query.to_parquet(PATH_QUERY_RESULT)\n",
    "    except ConnectionError:\n",
    "        print(\"Can't connect to the API, is it running?\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some of the retrieved results are already in GGPONC:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "tags": []
   },
   "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>is_in_ggponc</th>\n",
       "      <th>False</th>\n",
       "      <th>True</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>source</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Civic</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ClinicalTrials</th>\n",
       "      <td>138</td>\n",
       "      <td>0</td>\n",
       "      <td>138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pubmed</th>\n",
       "      <td>225</td>\n",
       "      <td>26</td>\n",
       "      <td>251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>364</td>\n",
       "      <td>26</td>\n",
       "      <td>390</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "is_in_ggponc    False  True  All\n",
       "source                          \n",
       "Civic               1     0    1\n",
       "ClinicalTrials    138     0  138\n",
       "Pubmed            225    26  251\n",
       "All               364    26  390"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(df_query[\"source\"], df_query[\"is_in_ggponc\"], margins=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Retrieval Evaluation using the Pubmed data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the evaluation, we only look at results retrieved from Pubmed:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_r = df_query.query(\"source == 'Pubmed'\").copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "Check if CIViC results are already in PubMed result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "29351986 True\n"
     ]
    }
   ],
   "source": [
    "df_civic = df_query.query(\"source == 'Civic'\")\n",
    "for pm_id in df_civic.pm_id:\n",
    "    print(pm_id, pm_id in df_r.pm_id.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Furthermore, we consider all results that we can retrieve from GGPONC as having been screened and included, thus making them relevant. We therefor add these records to the dataframe containing the screened records:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from evaluation.matching import add_unscreened_ggponc_citations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_s_with_ggponc = add_unscreened_ggponc_citations(df_s, df_r)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from evaluation.metrics import FILTERS_SCREENED, FILTERS_RETRIEVED"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Original Screening"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'All'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f5093540f82e45bca925b66679627d5c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/14 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'Only RCTs'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6362538c9904bd294e50a82ecbc655d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/14 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from evaluation.metrics import run_evaluation\n",
    "\n",
    "metrics = run_evaluation(\n",
    "    df_s_with_ggponc.query(\"screening_origin == 'original'\"),\n",
    "    df_r,\n",
    "    filters_screened=FILTERS_SCREENED,\n",
    "    filters_retrieved=FILTERS_RETRIEVED,\n",
    "    col_id=\"pm_id\",\n",
    "    col_relevant=\"is_relevant\",\n",
    "    compute_combinations=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from evaluation.plotting import draw_venn_rsi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### All evidence with a Pubmed ID"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/Florian.Borchert/miniconda3/envs/nge/lib/python3.10/site-packages/matplotlib_venn/_venn3.py:117: UserWarning: Bad circle positioning\n",
      "  warnings.warn(\"Bad circle positioning\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draw_venn_rsi(\n",
    "    title=\"All\",\n",
    "    ids_retrieved=metrics[\"All\"][\"All\"][\"ids_retrieved\"],\n",
    "    ids_screened=metrics[\"All\"][\"All\"][\"ids_screened\"],\n",
    "    ids_screened_relevant=metrics[\"All\"][\"All\"][\"ids_screened_relevant\"],\n",
    "    add_pn_labels=True,\n",
    ");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from evaluation.metrics import metrics_to_df, highlight_metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_e7f9c_row0_col1, #T_e7f9c_row0_col2, #T_e7f9c_row0_col3, #T_e7f9c_row0_col5, #T_e7f9c_row2_col1, #T_e7f9c_row2_col2, #T_e7f9c_row2_col3, #T_e7f9c_row2_col5, #T_e7f9c_row5_col0, #T_e7f9c_row5_col4, #T_e7f9c_row6_col1, #T_e7f9c_row6_col2, #T_e7f9c_row6_col3, #T_e7f9c_row6_col5 {\n",
       "  font-weight: bold;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_e7f9c\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_e7f9c_level0_col0\" class=\"col_heading level0 col0\" >Precision</th>\n",
       "      <th id=\"T_e7f9c_level0_col1\" class=\"col_heading level0 col1\" >Recall</th>\n",
       "      <th id=\"T_e7f9c_level0_col2\" class=\"col_heading level0 col2\" >$F_1$</th>\n",
       "      <th id=\"T_e7f9c_level0_col3\" class=\"col_heading level0 col3\" >TP</th>\n",
       "      <th id=\"T_e7f9c_level0_col4\" class=\"col_heading level0 col4\" >FP</th>\n",
       "      <th id=\"T_e7f9c_level0_col5\" class=\"col_heading level0 col5\" >FN</th>\n",
       "      <th id=\"T_e7f9c_level0_col6\" class=\"col_heading level0 col6\" >TN</th>\n",
       "      <th id=\"T_e7f9c_level0_col7\" class=\"col_heading level0 col7\" >?</th>\n",
       "      <th id=\"T_e7f9c_level0_col8\" class=\"col_heading level0 col8\" >Retrieved</th>\n",
       "      <th id=\"T_e7f9c_level0_col9\" class=\"col_heading level0 col9\" >Screened</th>\n",
       "      <th id=\"T_e7f9c_level0_col10\" class=\"col_heading level0 col10\" >Relevant</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row0\" class=\"row_heading level0 row0\" >All</th>\n",
       "      <td id=\"T_e7f9c_row0_col0\" class=\"data row0 col0\" >0.9000</td>\n",
       "      <td id=\"T_e7f9c_row0_col1\" class=\"data row0 col1\" >0.5714</td>\n",
       "      <td id=\"T_e7f9c_row0_col2\" class=\"data row0 col2\" >0.699000</td>\n",
       "      <td id=\"T_e7f9c_row0_col3\" class=\"data row0 col3\" >36</td>\n",
       "      <td id=\"T_e7f9c_row0_col4\" class=\"data row0 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row0_col5\" class=\"data row0 col5\" >27</td>\n",
       "      <td id=\"T_e7f9c_row0_col6\" class=\"data row0 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row0_col7\" class=\"data row0 col7\" >211</td>\n",
       "      <td id=\"T_e7f9c_row0_col8\" class=\"data row0 col8\" >251</td>\n",
       "      <td id=\"T_e7f9c_row0_col9\" class=\"data row0 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row0_col10\" class=\"data row0 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row1\" class=\"row_heading level0 row1\" >Is RCT</th>\n",
       "      <td id=\"T_e7f9c_row1_col0\" class=\"data row1 col0\" >0.8621</td>\n",
       "      <td id=\"T_e7f9c_row1_col1\" class=\"data row1 col1\" >0.3968</td>\n",
       "      <td id=\"T_e7f9c_row1_col2\" class=\"data row1 col2\" >0.543500</td>\n",
       "      <td id=\"T_e7f9c_row1_col3\" class=\"data row1 col3\" >25</td>\n",
       "      <td id=\"T_e7f9c_row1_col4\" class=\"data row1 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row1_col5\" class=\"data row1 col5\" >38</td>\n",
       "      <td id=\"T_e7f9c_row1_col6\" class=\"data row1 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row1_col7\" class=\"data row1 col7\" >58</td>\n",
       "      <td id=\"T_e7f9c_row1_col8\" class=\"data row1 col8\" >87</td>\n",
       "      <td id=\"T_e7f9c_row1_col9\" class=\"data row1 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row1_col10\" class=\"data row1 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row2\" class=\"row_heading level0 row2\" >Non-empty Abstract</th>\n",
       "      <td id=\"T_e7f9c_row2_col0\" class=\"data row2 col0\" >0.9000</td>\n",
       "      <td id=\"T_e7f9c_row2_col1\" class=\"data row2 col1\" >0.5714</td>\n",
       "      <td id=\"T_e7f9c_row2_col2\" class=\"data row2 col2\" >0.699000</td>\n",
       "      <td id=\"T_e7f9c_row2_col3\" class=\"data row2 col3\" >36</td>\n",
       "      <td id=\"T_e7f9c_row2_col4\" class=\"data row2 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row2_col5\" class=\"data row2 col5\" >27</td>\n",
       "      <td id=\"T_e7f9c_row2_col6\" class=\"data row2 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row2_col7\" class=\"data row2 col7\" >189</td>\n",
       "      <td id=\"T_e7f9c_row2_col8\" class=\"data row2 col8\" >229</td>\n",
       "      <td id=\"T_e7f9c_row2_col9\" class=\"data row2 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row2_col10\" class=\"data row2 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row3\" class=\"row_heading level0 row3\" >Non-pediatric population</th>\n",
       "      <td id=\"T_e7f9c_row3_col0\" class=\"data row3 col0\" >0.8947</td>\n",
       "      <td id=\"T_e7f9c_row3_col1\" class=\"data row3 col1\" >0.5397</td>\n",
       "      <td id=\"T_e7f9c_row3_col2\" class=\"data row3 col2\" >0.673300</td>\n",
       "      <td id=\"T_e7f9c_row3_col3\" class=\"data row3 col3\" >34</td>\n",
       "      <td id=\"T_e7f9c_row3_col4\" class=\"data row3 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row3_col5\" class=\"data row3 col5\" >29</td>\n",
       "      <td id=\"T_e7f9c_row3_col6\" class=\"data row3 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row3_col7\" class=\"data row3 col7\" >185</td>\n",
       "      <td id=\"T_e7f9c_row3_col8\" class=\"data row3 col8\" >223</td>\n",
       "      <td id=\"T_e7f9c_row3_col9\" class=\"data row3 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row3_col10\" class=\"data row3 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row4\" class=\"row_heading level0 row4\" >Phase >= 2</th>\n",
       "      <td id=\"T_e7f9c_row4_col0\" class=\"data row4 col0\" >0.9259</td>\n",
       "      <td id=\"T_e7f9c_row4_col1\" class=\"data row4 col1\" >0.3968</td>\n",
       "      <td id=\"T_e7f9c_row4_col2\" class=\"data row4 col2\" >0.555600</td>\n",
       "      <td id=\"T_e7f9c_row4_col3\" class=\"data row4 col3\" >25</td>\n",
       "      <td id=\"T_e7f9c_row4_col4\" class=\"data row4 col4\" >2</td>\n",
       "      <td id=\"T_e7f9c_row4_col5\" class=\"data row4 col5\" >38</td>\n",
       "      <td id=\"T_e7f9c_row4_col6\" class=\"data row4 col6\" >93</td>\n",
       "      <td id=\"T_e7f9c_row4_col7\" class=\"data row4 col7\" >139</td>\n",
       "      <td id=\"T_e7f9c_row4_col8\" class=\"data row4 col8\" >166</td>\n",
       "      <td id=\"T_e7f9c_row4_col9\" class=\"data row4 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row4_col10\" class=\"data row4 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row5\" class=\"row_heading level0 row5\" >Phase >= 3</th>\n",
       "      <td id=\"T_e7f9c_row5_col0\" class=\"data row5 col0\" >1.0000</td>\n",
       "      <td id=\"T_e7f9c_row5_col1\" class=\"data row5 col1\" >0.2698</td>\n",
       "      <td id=\"T_e7f9c_row5_col2\" class=\"data row5 col2\" >0.425000</td>\n",
       "      <td id=\"T_e7f9c_row5_col3\" class=\"data row5 col3\" >17</td>\n",
       "      <td id=\"T_e7f9c_row5_col4\" class=\"data row5 col4\" >0</td>\n",
       "      <td id=\"T_e7f9c_row5_col5\" class=\"data row5 col5\" >46</td>\n",
       "      <td id=\"T_e7f9c_row5_col6\" class=\"data row5 col6\" >95</td>\n",
       "      <td id=\"T_e7f9c_row5_col7\" class=\"data row5 col7\" >29</td>\n",
       "      <td id=\"T_e7f9c_row5_col8\" class=\"data row5 col8\" >46</td>\n",
       "      <td id=\"T_e7f9c_row5_col9\" class=\"data row5 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row5_col10\" class=\"data row5 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row6\" class=\"row_heading level0 row6\" >Sample Size >= 5</th>\n",
       "      <td id=\"T_e7f9c_row6_col0\" class=\"data row6 col0\" >0.9000</td>\n",
       "      <td id=\"T_e7f9c_row6_col1\" class=\"data row6 col1\" >0.5714</td>\n",
       "      <td id=\"T_e7f9c_row6_col2\" class=\"data row6 col2\" >0.699000</td>\n",
       "      <td id=\"T_e7f9c_row6_col3\" class=\"data row6 col3\" >36</td>\n",
       "      <td id=\"T_e7f9c_row6_col4\" class=\"data row6 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row6_col5\" class=\"data row6 col5\" >27</td>\n",
       "      <td id=\"T_e7f9c_row6_col6\" class=\"data row6 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row6_col7\" class=\"data row6 col7\" >200</td>\n",
       "      <td id=\"T_e7f9c_row6_col8\" class=\"data row6 col8\" >240</td>\n",
       "      <td id=\"T_e7f9c_row6_col9\" class=\"data row6 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row6_col10\" class=\"data row6 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row7\" class=\"row_heading level0 row7\" >Sample Size >= 10</th>\n",
       "      <td id=\"T_e7f9c_row7_col0\" class=\"data row7 col0\" >0.8947</td>\n",
       "      <td id=\"T_e7f9c_row7_col1\" class=\"data row7 col1\" >0.5397</td>\n",
       "      <td id=\"T_e7f9c_row7_col2\" class=\"data row7 col2\" >0.673300</td>\n",
       "      <td id=\"T_e7f9c_row7_col3\" class=\"data row7 col3\" >34</td>\n",
       "      <td id=\"T_e7f9c_row7_col4\" class=\"data row7 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row7_col5\" class=\"data row7 col5\" >29</td>\n",
       "      <td id=\"T_e7f9c_row7_col6\" class=\"data row7 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row7_col7\" class=\"data row7 col7\" >194</td>\n",
       "      <td id=\"T_e7f9c_row7_col8\" class=\"data row7 col8\" >232</td>\n",
       "      <td id=\"T_e7f9c_row7_col9\" class=\"data row7 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row7_col10\" class=\"data row7 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row8\" class=\"row_heading level0 row8\" >Sample Size >= 15</th>\n",
       "      <td id=\"T_e7f9c_row8_col0\" class=\"data row8 col0\" >0.8919</td>\n",
       "      <td id=\"T_e7f9c_row8_col1\" class=\"data row8 col1\" >0.5238</td>\n",
       "      <td id=\"T_e7f9c_row8_col2\" class=\"data row8 col2\" >0.660000</td>\n",
       "      <td id=\"T_e7f9c_row8_col3\" class=\"data row8 col3\" >33</td>\n",
       "      <td id=\"T_e7f9c_row8_col4\" class=\"data row8 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row8_col5\" class=\"data row8 col5\" >30</td>\n",
       "      <td id=\"T_e7f9c_row8_col6\" class=\"data row8 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row8_col7\" class=\"data row8 col7\" >185</td>\n",
       "      <td id=\"T_e7f9c_row8_col8\" class=\"data row8 col8\" >222</td>\n",
       "      <td id=\"T_e7f9c_row8_col9\" class=\"data row8 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row8_col10\" class=\"data row8 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row9\" class=\"row_heading level0 row9\" >Known Intervention</th>\n",
       "      <td id=\"T_e7f9c_row9_col0\" class=\"data row9 col0\" >0.8974</td>\n",
       "      <td id=\"T_e7f9c_row9_col1\" class=\"data row9 col1\" >0.5556</td>\n",
       "      <td id=\"T_e7f9c_row9_col2\" class=\"data row9 col2\" >0.686300</td>\n",
       "      <td id=\"T_e7f9c_row9_col3\" class=\"data row9 col3\" >35</td>\n",
       "      <td id=\"T_e7f9c_row9_col4\" class=\"data row9 col4\" >4</td>\n",
       "      <td id=\"T_e7f9c_row9_col5\" class=\"data row9 col5\" >28</td>\n",
       "      <td id=\"T_e7f9c_row9_col6\" class=\"data row9 col6\" >91</td>\n",
       "      <td id=\"T_e7f9c_row9_col7\" class=\"data row9 col7\" >169</td>\n",
       "      <td id=\"T_e7f9c_row9_col8\" class=\"data row9 col8\" >208</td>\n",
       "      <td id=\"T_e7f9c_row9_col9\" class=\"data row9 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row9_col10\" class=\"data row9 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row10\" class=\"row_heading level0 row10\" >Unknown Intervention</th>\n",
       "      <td id=\"T_e7f9c_row10_col0\" class=\"data row10 col0\" >0.8889</td>\n",
       "      <td id=\"T_e7f9c_row10_col1\" class=\"data row10 col1\" >0.2540</td>\n",
       "      <td id=\"T_e7f9c_row10_col2\" class=\"data row10 col2\" >0.395100</td>\n",
       "      <td id=\"T_e7f9c_row10_col3\" class=\"data row10 col3\" >16</td>\n",
       "      <td id=\"T_e7f9c_row10_col4\" class=\"data row10 col4\" >2</td>\n",
       "      <td id=\"T_e7f9c_row10_col5\" class=\"data row10 col5\" >47</td>\n",
       "      <td id=\"T_e7f9c_row10_col6\" class=\"data row10 col6\" >93</td>\n",
       "      <td id=\"T_e7f9c_row10_col7\" class=\"data row10 col7\" >110</td>\n",
       "      <td id=\"T_e7f9c_row10_col8\" class=\"data row10 col8\" >128</td>\n",
       "      <td id=\"T_e7f9c_row10_col9\" class=\"data row10 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row10_col10\" class=\"data row10 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row11\" class=\"row_heading level0 row11\" >K+U Intervention</th>\n",
       "      <td id=\"T_e7f9c_row11_col0\" class=\"data row11 col0\" >0.8824</td>\n",
       "      <td id=\"T_e7f9c_row11_col1\" class=\"data row11 col1\" >0.2381</td>\n",
       "      <td id=\"T_e7f9c_row11_col2\" class=\"data row11 col2\" >0.375000</td>\n",
       "      <td id=\"T_e7f9c_row11_col3\" class=\"data row11 col3\" >15</td>\n",
       "      <td id=\"T_e7f9c_row11_col4\" class=\"data row11 col4\" >2</td>\n",
       "      <td id=\"T_e7f9c_row11_col5\" class=\"data row11 col5\" >48</td>\n",
       "      <td id=\"T_e7f9c_row11_col6\" class=\"data row11 col6\" >93</td>\n",
       "      <td id=\"T_e7f9c_row11_col7\" class=\"data row11 col7\" >88</td>\n",
       "      <td id=\"T_e7f9c_row11_col8\" class=\"data row11 col8\" >105</td>\n",
       "      <td id=\"T_e7f9c_row11_col9\" class=\"data row11 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row11_col10\" class=\"data row11 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row12\" class=\"row_heading level0 row12\" >Significant Finding</th>\n",
       "      <td id=\"T_e7f9c_row12_col0\" class=\"data row12 col0\" >0.8696</td>\n",
       "      <td id=\"T_e7f9c_row12_col1\" class=\"data row12 col1\" >0.3175</td>\n",
       "      <td id=\"T_e7f9c_row12_col2\" class=\"data row12 col2\" >0.465100</td>\n",
       "      <td id=\"T_e7f9c_row12_col3\" class=\"data row12 col3\" >20</td>\n",
       "      <td id=\"T_e7f9c_row12_col4\" class=\"data row12 col4\" >3</td>\n",
       "      <td id=\"T_e7f9c_row12_col5\" class=\"data row12 col5\" >43</td>\n",
       "      <td id=\"T_e7f9c_row12_col6\" class=\"data row12 col6\" >92</td>\n",
       "      <td id=\"T_e7f9c_row12_col7\" class=\"data row12 col7\" >141</td>\n",
       "      <td id=\"T_e7f9c_row12_col8\" class=\"data row12 col8\" >164</td>\n",
       "      <td id=\"T_e7f9c_row12_col9\" class=\"data row12 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row12_col10\" class=\"data row12 col10\" >63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e7f9c_level0_row13\" class=\"row_heading level0 row13\" >No Significant Finding</th>\n",
       "      <td id=\"T_e7f9c_row13_col0\" class=\"data row13 col0\" >0.9412</td>\n",
       "      <td id=\"T_e7f9c_row13_col1\" class=\"data row13 col1\" >0.2540</td>\n",
       "      <td id=\"T_e7f9c_row13_col2\" class=\"data row13 col2\" >0.400000</td>\n",
       "      <td id=\"T_e7f9c_row13_col3\" class=\"data row13 col3\" >16</td>\n",
       "      <td id=\"T_e7f9c_row13_col4\" class=\"data row13 col4\" >1</td>\n",
       "      <td id=\"T_e7f9c_row13_col5\" class=\"data row13 col5\" >47</td>\n",
       "      <td id=\"T_e7f9c_row13_col6\" class=\"data row13 col6\" >94</td>\n",
       "      <td id=\"T_e7f9c_row13_col7\" class=\"data row13 col7\" >48</td>\n",
       "      <td id=\"T_e7f9c_row13_col8\" class=\"data row13 col8\" >65</td>\n",
       "      <td id=\"T_e7f9c_row13_col9\" class=\"data row13 col9\" >158</td>\n",
       "      <td id=\"T_e7f9c_row13_col10\" class=\"data row13 col10\" >63</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f12b92b2ef0>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "highlight_metrics(\n",
    "    metrics_to_df(metrics[\"All\"], drop_zero_tp=False).head(len(FILTERS_RETRIEVED))\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### RCTs only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/Florian.Borchert/miniconda3/envs/nge/lib/python3.10/site-packages/matplotlib_venn/_venn3.py:117: UserWarning: Bad circle positioning\n",
      "  warnings.warn(\"Bad circle positioning\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draw_venn_rsi(\n",
    "    title=\"RCTs only\",\n",
    "    ids_retrieved=metrics[\"Only RCTs\"][\"Is RCT\"][\"ids_retrieved\"],\n",
    "    ids_screened=metrics[\"Only RCTs\"][\"Is RCT\"][\"ids_screened\"],\n",
    "    ids_screened_relevant=metrics[\"Only RCTs\"][\"Is RCT\"][\"ids_screened_relevant\"],\n",
    "    add_pn_labels=True,\n",
    ");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_49f56_row0_col1, #T_49f56_row0_col2, #T_49f56_row0_col3, #T_49f56_row0_col5, #T_49f56_row1_col1, #T_49f56_row1_col2, #T_49f56_row1_col3, #T_49f56_row1_col5, #T_49f56_row2_col1, #T_49f56_row2_col2, #T_49f56_row2_col3, #T_49f56_row2_col5, #T_49f56_row5_col0, #T_49f56_row5_col4, #T_49f56_row6_col1, #T_49f56_row6_col2, #T_49f56_row6_col3, #T_49f56_row6_col5, #T_49f56_row7_col1, #T_49f56_row7_col2, #T_49f56_row7_col3, #T_49f56_row7_col5, #T_49f56_row8_col1, #T_49f56_row8_col2, #T_49f56_row8_col3, #T_49f56_row8_col5 {\n",
       "  font-weight: bold;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_49f56\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_49f56_level0_col0\" class=\"col_heading level0 col0\" >Precision</th>\n",
       "      <th id=\"T_49f56_level0_col1\" class=\"col_heading level0 col1\" >Recall</th>\n",
       "      <th id=\"T_49f56_level0_col2\" class=\"col_heading level0 col2\" >$F_1$</th>\n",
       "      <th id=\"T_49f56_level0_col3\" class=\"col_heading level0 col3\" >TP</th>\n",
       "      <th id=\"T_49f56_level0_col4\" class=\"col_heading level0 col4\" >FP</th>\n",
       "      <th id=\"T_49f56_level0_col5\" class=\"col_heading level0 col5\" >FN</th>\n",
       "      <th id=\"T_49f56_level0_col6\" class=\"col_heading level0 col6\" >TN</th>\n",
       "      <th id=\"T_49f56_level0_col7\" class=\"col_heading level0 col7\" >?</th>\n",
       "      <th id=\"T_49f56_level0_col8\" class=\"col_heading level0 col8\" >Retrieved</th>\n",
       "      <th id=\"T_49f56_level0_col9\" class=\"col_heading level0 col9\" >Screened</th>\n",
       "      <th id=\"T_49f56_level0_col10\" class=\"col_heading level0 col10\" >Relevant</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row0\" class=\"row_heading level0 row0\" >All</th>\n",
       "      <td id=\"T_49f56_row0_col0\" class=\"data row0 col0\" >0.8621</td>\n",
       "      <td id=\"T_49f56_row0_col1\" class=\"data row0 col1\" >1.0000</td>\n",
       "      <td id=\"T_49f56_row0_col2\" class=\"data row0 col2\" >0.925900</td>\n",
       "      <td id=\"T_49f56_row0_col3\" class=\"data row0 col3\" >25</td>\n",
       "      <td id=\"T_49f56_row0_col4\" class=\"data row0 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row0_col5\" class=\"data row0 col5\" >0</td>\n",
       "      <td id=\"T_49f56_row0_col6\" class=\"data row0 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row0_col7\" class=\"data row0 col7\" >222</td>\n",
       "      <td id=\"T_49f56_row0_col8\" class=\"data row0 col8\" >251</td>\n",
       "      <td id=\"T_49f56_row0_col9\" class=\"data row0 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row0_col10\" class=\"data row0 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row1\" class=\"row_heading level0 row1\" >Is RCT</th>\n",
       "      <td id=\"T_49f56_row1_col0\" class=\"data row1 col0\" >0.8621</td>\n",
       "      <td id=\"T_49f56_row1_col1\" class=\"data row1 col1\" >1.0000</td>\n",
       "      <td id=\"T_49f56_row1_col2\" class=\"data row1 col2\" >0.925900</td>\n",
       "      <td id=\"T_49f56_row1_col3\" class=\"data row1 col3\" >25</td>\n",
       "      <td id=\"T_49f56_row1_col4\" class=\"data row1 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row1_col5\" class=\"data row1 col5\" >0</td>\n",
       "      <td id=\"T_49f56_row1_col6\" class=\"data row1 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row1_col7\" class=\"data row1 col7\" >58</td>\n",
       "      <td id=\"T_49f56_row1_col8\" class=\"data row1 col8\" >87</td>\n",
       "      <td id=\"T_49f56_row1_col9\" class=\"data row1 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row1_col10\" class=\"data row1 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row2\" class=\"row_heading level0 row2\" >Non-empty Abstract</th>\n",
       "      <td id=\"T_49f56_row2_col0\" class=\"data row2 col0\" >0.8621</td>\n",
       "      <td id=\"T_49f56_row2_col1\" class=\"data row2 col1\" >1.0000</td>\n",
       "      <td id=\"T_49f56_row2_col2\" class=\"data row2 col2\" >0.925900</td>\n",
       "      <td id=\"T_49f56_row2_col3\" class=\"data row2 col3\" >25</td>\n",
       "      <td id=\"T_49f56_row2_col4\" class=\"data row2 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row2_col5\" class=\"data row2 col5\" >0</td>\n",
       "      <td id=\"T_49f56_row2_col6\" class=\"data row2 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row2_col7\" class=\"data row2 col7\" >200</td>\n",
       "      <td id=\"T_49f56_row2_col8\" class=\"data row2 col8\" >229</td>\n",
       "      <td id=\"T_49f56_row2_col9\" class=\"data row2 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row2_col10\" class=\"data row2 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row3\" class=\"row_heading level0 row3\" >Non-pediatric population</th>\n",
       "      <td id=\"T_49f56_row3_col0\" class=\"data row3 col0\" >0.8571</td>\n",
       "      <td id=\"T_49f56_row3_col1\" class=\"data row3 col1\" >0.9600</td>\n",
       "      <td id=\"T_49f56_row3_col2\" class=\"data row3 col2\" >0.905700</td>\n",
       "      <td id=\"T_49f56_row3_col3\" class=\"data row3 col3\" >24</td>\n",
       "      <td id=\"T_49f56_row3_col4\" class=\"data row3 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row3_col5\" class=\"data row3 col5\" >1</td>\n",
       "      <td id=\"T_49f56_row3_col6\" class=\"data row3 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row3_col7\" class=\"data row3 col7\" >195</td>\n",
       "      <td id=\"T_49f56_row3_col8\" class=\"data row3 col8\" >223</td>\n",
       "      <td id=\"T_49f56_row3_col9\" class=\"data row3 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row3_col10\" class=\"data row3 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row4\" class=\"row_heading level0 row4\" >Phase >= 2</th>\n",
       "      <td id=\"T_49f56_row4_col0\" class=\"data row4 col0\" >0.9048</td>\n",
       "      <td id=\"T_49f56_row4_col1\" class=\"data row4 col1\" >0.7600</td>\n",
       "      <td id=\"T_49f56_row4_col2\" class=\"data row4 col2\" >0.826100</td>\n",
       "      <td id=\"T_49f56_row4_col3\" class=\"data row4 col3\" >19</td>\n",
       "      <td id=\"T_49f56_row4_col4\" class=\"data row4 col4\" >2</td>\n",
       "      <td id=\"T_49f56_row4_col5\" class=\"data row4 col5\" >6</td>\n",
       "      <td id=\"T_49f56_row4_col6\" class=\"data row4 col6\" >22</td>\n",
       "      <td id=\"T_49f56_row4_col7\" class=\"data row4 col7\" >145</td>\n",
       "      <td id=\"T_49f56_row4_col8\" class=\"data row4 col8\" >166</td>\n",
       "      <td id=\"T_49f56_row4_col9\" class=\"data row4 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row4_col10\" class=\"data row4 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row5\" class=\"row_heading level0 row5\" >Phase >= 3</th>\n",
       "      <td id=\"T_49f56_row5_col0\" class=\"data row5 col0\" >1.0000</td>\n",
       "      <td id=\"T_49f56_row5_col1\" class=\"data row5 col1\" >0.5200</td>\n",
       "      <td id=\"T_49f56_row5_col2\" class=\"data row5 col2\" >0.684200</td>\n",
       "      <td id=\"T_49f56_row5_col3\" class=\"data row5 col3\" >13</td>\n",
       "      <td id=\"T_49f56_row5_col4\" class=\"data row5 col4\" >0</td>\n",
       "      <td id=\"T_49f56_row5_col5\" class=\"data row5 col5\" >12</td>\n",
       "      <td id=\"T_49f56_row5_col6\" class=\"data row5 col6\" >24</td>\n",
       "      <td id=\"T_49f56_row5_col7\" class=\"data row5 col7\" >33</td>\n",
       "      <td id=\"T_49f56_row5_col8\" class=\"data row5 col8\" >46</td>\n",
       "      <td id=\"T_49f56_row5_col9\" class=\"data row5 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row5_col10\" class=\"data row5 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row6\" class=\"row_heading level0 row6\" >Sample Size >= 5</th>\n",
       "      <td id=\"T_49f56_row6_col0\" class=\"data row6 col0\" >0.8621</td>\n",
       "      <td id=\"T_49f56_row6_col1\" class=\"data row6 col1\" >1.0000</td>\n",
       "      <td id=\"T_49f56_row6_col2\" class=\"data row6 col2\" >0.925900</td>\n",
       "      <td id=\"T_49f56_row6_col3\" class=\"data row6 col3\" >25</td>\n",
       "      <td id=\"T_49f56_row6_col4\" class=\"data row6 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row6_col5\" class=\"data row6 col5\" >0</td>\n",
       "      <td id=\"T_49f56_row6_col6\" class=\"data row6 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row6_col7\" class=\"data row6 col7\" >211</td>\n",
       "      <td id=\"T_49f56_row6_col8\" class=\"data row6 col8\" >240</td>\n",
       "      <td id=\"T_49f56_row6_col9\" class=\"data row6 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row6_col10\" class=\"data row6 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row7\" class=\"row_heading level0 row7\" >Sample Size >= 10</th>\n",
       "      <td id=\"T_49f56_row7_col0\" class=\"data row7 col0\" >0.8621</td>\n",
       "      <td id=\"T_49f56_row7_col1\" class=\"data row7 col1\" >1.0000</td>\n",
       "      <td id=\"T_49f56_row7_col2\" class=\"data row7 col2\" >0.925900</td>\n",
       "      <td id=\"T_49f56_row7_col3\" class=\"data row7 col3\" >25</td>\n",
       "      <td id=\"T_49f56_row7_col4\" class=\"data row7 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row7_col5\" class=\"data row7 col5\" >0</td>\n",
       "      <td id=\"T_49f56_row7_col6\" class=\"data row7 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row7_col7\" class=\"data row7 col7\" >203</td>\n",
       "      <td id=\"T_49f56_row7_col8\" class=\"data row7 col8\" >232</td>\n",
       "      <td id=\"T_49f56_row7_col9\" class=\"data row7 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row7_col10\" class=\"data row7 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row8\" class=\"row_heading level0 row8\" >Sample Size >= 15</th>\n",
       "      <td id=\"T_49f56_row8_col0\" class=\"data row8 col0\" >0.8621</td>\n",
       "      <td id=\"T_49f56_row8_col1\" class=\"data row8 col1\" >1.0000</td>\n",
       "      <td id=\"T_49f56_row8_col2\" class=\"data row8 col2\" >0.925900</td>\n",
       "      <td id=\"T_49f56_row8_col3\" class=\"data row8 col3\" >25</td>\n",
       "      <td id=\"T_49f56_row8_col4\" class=\"data row8 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row8_col5\" class=\"data row8 col5\" >0</td>\n",
       "      <td id=\"T_49f56_row8_col6\" class=\"data row8 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row8_col7\" class=\"data row8 col7\" >193</td>\n",
       "      <td id=\"T_49f56_row8_col8\" class=\"data row8 col8\" >222</td>\n",
       "      <td id=\"T_49f56_row8_col9\" class=\"data row8 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row8_col10\" class=\"data row8 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row9\" class=\"row_heading level0 row9\" >Known Intervention</th>\n",
       "      <td id=\"T_49f56_row9_col0\" class=\"data row9 col0\" >0.8571</td>\n",
       "      <td id=\"T_49f56_row9_col1\" class=\"data row9 col1\" >0.9600</td>\n",
       "      <td id=\"T_49f56_row9_col2\" class=\"data row9 col2\" >0.905700</td>\n",
       "      <td id=\"T_49f56_row9_col3\" class=\"data row9 col3\" >24</td>\n",
       "      <td id=\"T_49f56_row9_col4\" class=\"data row9 col4\" >4</td>\n",
       "      <td id=\"T_49f56_row9_col5\" class=\"data row9 col5\" >1</td>\n",
       "      <td id=\"T_49f56_row9_col6\" class=\"data row9 col6\" >20</td>\n",
       "      <td id=\"T_49f56_row9_col7\" class=\"data row9 col7\" >180</td>\n",
       "      <td id=\"T_49f56_row9_col8\" class=\"data row9 col8\" >208</td>\n",
       "      <td id=\"T_49f56_row9_col9\" class=\"data row9 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row9_col10\" class=\"data row9 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row10\" class=\"row_heading level0 row10\" >Unknown Intervention</th>\n",
       "      <td id=\"T_49f56_row10_col0\" class=\"data row10 col0\" >0.8462</td>\n",
       "      <td id=\"T_49f56_row10_col1\" class=\"data row10 col1\" >0.4400</td>\n",
       "      <td id=\"T_49f56_row10_col2\" class=\"data row10 col2\" >0.578900</td>\n",
       "      <td id=\"T_49f56_row10_col3\" class=\"data row10 col3\" >11</td>\n",
       "      <td id=\"T_49f56_row10_col4\" class=\"data row10 col4\" >2</td>\n",
       "      <td id=\"T_49f56_row10_col5\" class=\"data row10 col5\" >14</td>\n",
       "      <td id=\"T_49f56_row10_col6\" class=\"data row10 col6\" >22</td>\n",
       "      <td id=\"T_49f56_row10_col7\" class=\"data row10 col7\" >115</td>\n",
       "      <td id=\"T_49f56_row10_col8\" class=\"data row10 col8\" >128</td>\n",
       "      <td id=\"T_49f56_row10_col9\" class=\"data row10 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row10_col10\" class=\"data row10 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row11\" class=\"row_heading level0 row11\" >K+U Intervention</th>\n",
       "      <td id=\"T_49f56_row11_col0\" class=\"data row11 col0\" >0.8333</td>\n",
       "      <td id=\"T_49f56_row11_col1\" class=\"data row11 col1\" >0.4000</td>\n",
       "      <td id=\"T_49f56_row11_col2\" class=\"data row11 col2\" >0.540500</td>\n",
       "      <td id=\"T_49f56_row11_col3\" class=\"data row11 col3\" >10</td>\n",
       "      <td id=\"T_49f56_row11_col4\" class=\"data row11 col4\" >2</td>\n",
       "      <td id=\"T_49f56_row11_col5\" class=\"data row11 col5\" >15</td>\n",
       "      <td id=\"T_49f56_row11_col6\" class=\"data row11 col6\" >22</td>\n",
       "      <td id=\"T_49f56_row11_col7\" class=\"data row11 col7\" >93</td>\n",
       "      <td id=\"T_49f56_row11_col8\" class=\"data row11 col8\" >105</td>\n",
       "      <td id=\"T_49f56_row11_col9\" class=\"data row11 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row11_col10\" class=\"data row11 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row12\" class=\"row_heading level0 row12\" >Significant Finding</th>\n",
       "      <td id=\"T_49f56_row12_col0\" class=\"data row12 col0\" >0.8421</td>\n",
       "      <td id=\"T_49f56_row12_col1\" class=\"data row12 col1\" >0.6400</td>\n",
       "      <td id=\"T_49f56_row12_col2\" class=\"data row12 col2\" >0.727300</td>\n",
       "      <td id=\"T_49f56_row12_col3\" class=\"data row12 col3\" >16</td>\n",
       "      <td id=\"T_49f56_row12_col4\" class=\"data row12 col4\" >3</td>\n",
       "      <td id=\"T_49f56_row12_col5\" class=\"data row12 col5\" >9</td>\n",
       "      <td id=\"T_49f56_row12_col6\" class=\"data row12 col6\" >21</td>\n",
       "      <td id=\"T_49f56_row12_col7\" class=\"data row12 col7\" >145</td>\n",
       "      <td id=\"T_49f56_row12_col8\" class=\"data row12 col8\" >164</td>\n",
       "      <td id=\"T_49f56_row12_col9\" class=\"data row12 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row12_col10\" class=\"data row12 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_49f56_level0_row13\" class=\"row_heading level0 row13\" >No Significant Finding</th>\n",
       "      <td id=\"T_49f56_row13_col0\" class=\"data row13 col0\" >0.9000</td>\n",
       "      <td id=\"T_49f56_row13_col1\" class=\"data row13 col1\" >0.3600</td>\n",
       "      <td id=\"T_49f56_row13_col2\" class=\"data row13 col2\" >0.514300</td>\n",
       "      <td id=\"T_49f56_row13_col3\" class=\"data row13 col3\" >9</td>\n",
       "      <td id=\"T_49f56_row13_col4\" class=\"data row13 col4\" >1</td>\n",
       "      <td id=\"T_49f56_row13_col5\" class=\"data row13 col5\" >16</td>\n",
       "      <td id=\"T_49f56_row13_col6\" class=\"data row13 col6\" >23</td>\n",
       "      <td id=\"T_49f56_row13_col7\" class=\"data row13 col7\" >55</td>\n",
       "      <td id=\"T_49f56_row13_col8\" class=\"data row13 col8\" >65</td>\n",
       "      <td id=\"T_49f56_row13_col9\" class=\"data row13 col9\" >49</td>\n",
       "      <td id=\"T_49f56_row13_col10\" class=\"data row13 col10\" >25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f12b92b3c70>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "highlight_metrics(\n",
    "    metrics_to_df(metrics[\"Only RCTs\"], drop_zero_tp=False).head(len(FILTERS_RETRIEVED))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'All'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c6c866ef91f34e0bb40188d3b244dee3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/12 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'Only RCTs'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b7a3a27b7bef4207afc363fb5763e386",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/12 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_4e149_row0_col1, #T_4e149_row0_col2, #T_4e149_row0_col3, #T_4e149_row0_col5, #T_4e149_row2_col0, #T_4e149_row2_col4 {\n",
       "  font-weight: bold;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_4e149\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_4e149_level0_col0\" class=\"col_heading level0 col0\" >Prec.</th>\n",
       "      <th id=\"T_4e149_level0_col1\" class=\"col_heading level0 col1\" >Rec.</th>\n",
       "      <th id=\"T_4e149_level0_col2\" class=\"col_heading level0 col2\" >$F_1$</th>\n",
       "      <th id=\"T_4e149_level0_col3\" class=\"col_heading level0 col3\" >TP</th>\n",
       "      <th id=\"T_4e149_level0_col4\" class=\"col_heading level0 col4\" >FP</th>\n",
       "      <th id=\"T_4e149_level0_col5\" class=\"col_heading level0 col5\" >FN</th>\n",
       "      <th id=\"T_4e149_level0_col6\" class=\"col_heading level0 col6\" >TN</th>\n",
       "      <th id=\"T_4e149_level0_col7\" class=\"col_heading level0 col7\" >?</th>\n",
       "      <th id=\"T_4e149_level0_col8\" class=\"col_heading level0 col8\" >Retr.</th>\n",
       "      <th id=\"T_4e149_level0_col9\" class=\"col_heading level0 col9\" >Scr.</th>\n",
       "      <th id=\"T_4e149_level0_col10\" class=\"col_heading level0 col10\" >Rel.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_4e149_level0_row0\" class=\"row_heading level0 row0\" rowspan=\"3\">RCTs</th>\n",
       "      <th id=\"T_4e149_level1_row0\" class=\"row_heading level1 row0\" ></th>\n",
       "      <th id=\"T_4e149_level2_row0\" class=\"row_heading level2 row0\" ></th>\n",
       "      <th id=\"T_4e149_level3_row0\" class=\"row_heading level3 row0\" ></th>\n",
       "      <th id=\"T_4e149_level4_row0\" class=\"row_heading level4 row0\" ></th>\n",
       "      <th id=\"T_4e149_level5_row0\" class=\"row_heading level5 row0\" ></th>\n",
       "      <td id=\"T_4e149_row0_col0\" class=\"data row0 col0\" >0.862100</td>\n",
       "      <td id=\"T_4e149_row0_col1\" class=\"data row0 col1\" >1.000000</td>\n",
       "      <td id=\"T_4e149_row0_col2\" class=\"data row0 col2\" >0.925900</td>\n",
       "      <td id=\"T_4e149_row0_col3\" class=\"data row0 col3\" >25</td>\n",
       "      <td id=\"T_4e149_row0_col4\" class=\"data row0 col4\" >4</td>\n",
       "      <td id=\"T_4e149_row0_col5\" class=\"data row0 col5\" >0</td>\n",
       "      <td id=\"T_4e149_row0_col6\" class=\"data row0 col6\" >20</td>\n",
       "      <td id=\"T_4e149_row0_col7\" class=\"data row0 col7\" >58</td>\n",
       "      <td id=\"T_4e149_row0_col8\" class=\"data row0 col8\" >87</td>\n",
       "      <td id=\"T_4e149_row0_col9\" class=\"data row0 col9\" >49</td>\n",
       "      <td id=\"T_4e149_row0_col10\" class=\"data row0 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e149_level1_row1\" class=\"row_heading level1 row1\" rowspan=\"2\">Abstr., NC</th>\n",
       "      <th id=\"T_4e149_level2_row1\" class=\"row_heading level2 row1\" ></th>\n",
       "      <th id=\"T_4e149_level3_row1\" class=\"row_heading level3 row1\" ></th>\n",
       "      <th id=\"T_4e149_level4_row1\" class=\"row_heading level4 row1\" ></th>\n",
       "      <th id=\"T_4e149_level5_row1\" class=\"row_heading level5 row1\" ></th>\n",
       "      <td id=\"T_4e149_row1_col0\" class=\"data row1 col0\" >0.857100</td>\n",
       "      <td id=\"T_4e149_row1_col1\" class=\"data row1 col1\" >0.960000</td>\n",
       "      <td id=\"T_4e149_row1_col2\" class=\"data row1 col2\" >0.905700</td>\n",
       "      <td id=\"T_4e149_row1_col3\" class=\"data row1 col3\" >24</td>\n",
       "      <td id=\"T_4e149_row1_col4\" class=\"data row1 col4\" >4</td>\n",
       "      <td id=\"T_4e149_row1_col5\" class=\"data row1 col5\" >1</td>\n",
       "      <td id=\"T_4e149_row1_col6\" class=\"data row1 col6\" >20</td>\n",
       "      <td id=\"T_4e149_row1_col7\" class=\"data row1 col7\" >41</td>\n",
       "      <td id=\"T_4e149_row1_col8\" class=\"data row1 col8\" >69</td>\n",
       "      <td id=\"T_4e149_row1_col9\" class=\"data row1 col9\" >49</td>\n",
       "      <td id=\"T_4e149_row1_col10\" class=\"data row1 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e149_level2_row2\" class=\"row_heading level2 row2\" >Phase >= 2</th>\n",
       "      <th id=\"T_4e149_level3_row2\" class=\"row_heading level3 row2\" ></th>\n",
       "      <th id=\"T_4e149_level4_row2\" class=\"row_heading level4 row2\" ></th>\n",
       "      <th id=\"T_4e149_level5_row2\" class=\"row_heading level5 row2\" ></th>\n",
       "      <td id=\"T_4e149_row2_col0\" class=\"data row2 col0\" >0.900000</td>\n",
       "      <td id=\"T_4e149_row2_col1\" class=\"data row2 col1\" >0.720000</td>\n",
       "      <td id=\"T_4e149_row2_col2\" class=\"data row2 col2\" >0.800000</td>\n",
       "      <td id=\"T_4e149_row2_col3\" class=\"data row2 col3\" >18</td>\n",
       "      <td id=\"T_4e149_row2_col4\" class=\"data row2 col4\" >2</td>\n",
       "      <td id=\"T_4e149_row2_col5\" class=\"data row2 col5\" >7</td>\n",
       "      <td id=\"T_4e149_row2_col6\" class=\"data row2 col6\" >22</td>\n",
       "      <td id=\"T_4e149_row2_col7\" class=\"data row2 col7\" >20</td>\n",
       "      <td id=\"T_4e149_row2_col8\" class=\"data row2 col8\" >40</td>\n",
       "      <td id=\"T_4e149_row2_col9\" class=\"data row2 col9\" >49</td>\n",
       "      <td id=\"T_4e149_row2_col10\" class=\"data row2 col10\" >25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f12b70d00d0>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from evaluation.metrics import build_filter_combinations_table\n",
    "\n",
    "combinations_table, metrics_combinations = build_filter_combinations_table(\n",
    "    df_retrieved=df_r,\n",
    "    df_screened=df_s_with_ggponc.query(\"screening_origin == 'original'\"),\n",
    "    latex_output_path=PATH_LATEX_TABLE_OUTPUT / \"final_metrics_original_only.tex\",\n",
    "    stop_at_line=3,\n",
    "    overwrite_latex_output=OVERWRITE_LATEX_OUTPUT,\n",
    "    use_abbreviations=True,\n",
    "    column_format=\"llllllS[table-format=1.2]S[table-format=1.2]S[table-format=1.2]S[table-format=2]S[table-format=3]S[table-format=1]S[table-format=3]S[table-format=2]S[table-format=3]\",\n",
    ")\n",
    "\n",
    "highlight_metrics(combinations_table)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Including Results of Manual Review"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'All'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0240fd4055fe4cb590a69d90b38fcd72",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/14 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'Only RCTs'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3d0b71745ffb421686fc4022841cde7d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/14 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "metrics_w_manual = run_evaluation(\n",
    "    df_s_with_ggponc,\n",
    "    df_r,\n",
    "    filters_screened=FILTERS_SCREENED,\n",
    "    filters_retrieved=FILTERS_RETRIEVED,\n",
    "    col_id=\"pm_id\",\n",
    "    col_relevant=\"is_relevant\",\n",
    "    compute_combinations=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### All"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb0AAAFgCAYAAAAvjqe1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABXpElEQVR4nO3dd3xV9f3H8dfdI3tDQiAk7I1MWYKCewAOql1atdplbWurdqiora1atbVa9WctirTuQS0OFCoyVJCprLAhCdk7d5/z++NKNCSBJNzknHPv5/l45AHJPfecTyC57/v9nu8wqaqqIoQQQsQAs9YFCCGEED1FQk8IIUTMkNATQggRMyT0hBBCxAwJPSGEEDFDQk8IIUTMkNATQggRMyT0hBBCxAwJPSGEEDFDQk8IjZlMJu66667mzxctWoTJZOLAgQOa1SREtJLQE6KbPf7445hMJiZNmqR1KULEPAk9IbrZkiVLyMvL49NPP2XPnj1alyNETJPQE6Ib7d+/n7Vr1/LQQw+RkZHBkiVLtC5JiJgmoSdEN1qyZAkpKSlccMEFXHbZZRJ6QmhMQk+IbrRkyRLmz5+P3W7nyiuvpLCwkPXr12tdlhAxS0JPiG7y2WefsXPnTr7xjW8AMG3aNPr06SOtPSE0JKEnRDdZsmQJWVlZzJo1CwhPTViwYAEvvPACoVBI4+qEiE0SekJ0g1AoxAsvvMCsWbPYv38/e/bsYc+ePUyaNInS0lI++OADrUsUIiZZtS5AiGi0YsUKSkpKeOGFF3jhhRdaPb5kyRLOPvtsDSoTIrZJ6AnRDZYsWUJmZiaPPfZYq8dee+01Xn/9dZ544glcLpcG1QkRuyT0hIgwj8fDa6+9xuWXX85ll13W6vHs7Gz+/e9/s3TpUhYsWKBBhULELrmnJ0SELV26lPr6ei6++OI2H588ebJMVBdCIxJ6QkTYkiVLcDqdzJkzp83HzWYzF1xwAe+88w6VlZU9XJ0Qsc2kqqqqdRFCCCFET5CWnhBCiJghoSeEECJmSOgJIYSIGRJ6QgghYoaEnhBCiJghoSeEECJmSOgJIYSIGRJ6QgghYoaEnhBCiJghoSeEECJmSOgJIYSIGRJ6QgghYoaEnhBCiJghoSeEECJmSOgJIYSIGRJ6QgghYoaEnhBCiJghoSeEECJmSOgJIYSIGRJ6QgghYoaEnhBCiJgRE6GXl5fH1VdfrXUZJ2Qymbjrrru0LkMIIaJat4beokWLMJlMzR9Wq5WcnByuvvpqioqKOn2+7du3c9ddd3HgwIHIFyuEECLqWXviInfffTf9+/fH6/Xy8ccfs2jRIlavXs3nn3+O0+ns8Hm2b9/OwoULmTlzJnl5eR1+3q5duzCbY6JRK4QQ4gR6JPTOO+88xo8fD8B1111Heno6f/rTn1i6dClXXHFFt1xTVVW8Xi8ulwuHw9Et1xBCCGEsmjR/pk+fDsDevXubv7Zz504uu+wyUlNTcTqdjB8/nqVLlzY/vmjRIi6//HIAZs2a1dxl+r///Q8I37e78MILeffddxk/fjwul4snn3yy+bHj7+nV1NRw8803k5ubi8PhYMCAAfzpT39CURQAAoEAqampXHPNNa3qr6urw+l0cssttzR/zefzceeddzJgwAAcDge5ubn86le/wufztXiuz+fjZz/7GRkZGSQkJHDxxRdz5MiRLv5LCiGE6Iweaekd79g9uZSUFAC++OILpk6dSk5ODrfddhtxcXG89NJLzJ07l1dffZV58+YxY8YMbrrpJv7617/y61//mqFDhwI0/wnhbswrr7ySG264geuvv57Bgwe3ef2mpibOOOMMioqKuOGGG+jbty9r167l9ttvp6SkhEceeQSbzca8efN47bXXePLJJ7Hb7c3Pf+ONN/D5fHzjG98AQFEULr74YlavXs33v/99hg4dyrZt23j44YfZvXs3b7zxRvNzr7vuOp5//nmuuuoqpkyZwooVK7jgggsi+c8rhBCiPWo3+uc//6kC6vvvv6+Wl5erhw8fVl955RU1IyNDdTgc6uHDh1VVVdWzzjpLHTlypOr1epufqyiKOmXKFHXgwIHNX3v55ZdVQF25cmWra/Xr108F1HfeeafNx7773e82f37PPfeocXFx6u7du1scd9ttt6kWi0U9dOiQqqqq+u6776qA+p///KfFceeff76an5/f/PnixYtVs9msfvTRRy2Oe+KJJ1RAXbNmjaqqqrp582YVUH/4wx+2OO6qq65SAfXOO+9sVbsQQojI6ZHuzdmzZ5ORkUFubi6XXXYZcXFxLF26lD59+lBVVcWKFSu44oorqK+vp6KigoqKCiorKznnnHMoLCzs8EjP/v37c84555z0uJdffpnp06eTkpLSfL2Kigpmz55NKBRi1apVAJx55pmkp6fz4osvNj+3urqa5cuXs2DBghbnGzp0KEOGDGlxvjPPPBOAlStXArBs2TIAbrrpphb13HzzzR36/oQQQpyaHunefOyxxxg0aBC1tbU888wzrFq1qnlwyZ49e1BVld/97nf87ne/a/P5ZWVl5OTknPQ6/fv371A9hYWFbN26lYyMjHavB2C1Wrn00kv517/+hc/nw+Fw8NprrxEIBFqEXmFhITt27Djp+Q4ePIjZbKagoKDF4+11wwohhIisHgm9iRMnNo/enDt3LtOmTeOqq65i165dzQNHbrnllnZbaQMGDOjQdVwuV4eOUxSFOXPm8Ktf/arNxwcNGtT892984xs8+eSTvP3228ydO5eXXnqJIUOGMHr06BbnGzlyJA899FCb58vNze1QXUIIIbpXjw9ksVgs3HfffcyaNYu//e1vfO973wPAZrMxe/bsEz7XZDJFpIaCggIaGhpOej2AGTNm0Lt3b1588UWmTZvGihUr+M1vftPqfFu2bOGss846YY39+vVDURT27t3bonW3a9eurn8zQgghOkyTKQszZ85k4sSJPPLIIyQmJjJz5kyefPJJSkpKWh1bXl7e/Pe4uDggPN3gVFxxxRWsW7eOd999t9VjNTU1BIPB5s/NZjOXXXYZ//nPf1i8eDHBYLBF1+ax8xUVFfF///d/rc7n8XhobGwEwvMVAf7617+2OOaRRx45pe9HCCFEx2gyZQHgl7/8JZdffjmLFi3iscceY9q0aYwcOZLrr7+e/Px8SktLWbduHUeOHGHLli0AjBkzBovFwp/+9Cdqa2txOByceeaZZGZmdvraS5cu5cILL+Tqq69m3LhxNDY2sm3bNl555RUOHDhAenp68/ELFizg0Ucf5c4772TkyJEtpkkAfPvb3+all17ixhtvZOXKlUydOpVQKMTOnTt56aWXmucOjhkzhiuvvJLHH3+c2tpapkyZwgcffMCePXtO/R9UCCHESWkWevPnz6egoIAHH3yQ66+/ng0bNrBw4UIWLVpEZWUlmZmZjB07ljvuuKP5Ob169eKJJ57gvvvu49prryUUCrFy5cpOh57b7ebDDz/kD3/4Ay+//DLPPfcciYmJDBo0iIULF5KUlNTi+ClTppCbm8vhw4dbtfIg3Bp84403ePjhh3nuued4/fXXcbvd5Ofn89Of/rTFPcJnnnmGjIwMlixZwhtvvMGZZ57Jf//736i+76eqKt6gl6ZAE56gB0/AQ1AJoqgKITWEoirhvytf/V1RFaxmKzaLDZvZhs1iC3/+5d9tZht2ix23zY3NYtP6WxRCGIRJVVVV6yKEsYWUEDXeGqq91dR4a2jwN+AJeFqEnEr3/ZjZzDbi7HHE2eKIt8eT4Egg0ZFIgj2BJGcSTmvH13cVQkQ3CT3RYSElRLW3mmpPdYs/63313Rpqp8ptc5PuTifNlRb+051GoiNR67KEEBqQ0BPt8of8HG04Skl9CUcbjlLeVI6iKlqXFRF2i500Vxpp7jQy4zLJTsjGbXNrXZYQ4jgHDhygf//+/POf/4zIvqia3dMT+tPobwyHXEM45Ko8VVqX1G38IT8lDSWUNHw1YjjFmUJOYg45CTlkJ2TLvULRY7Zt28bChQtZv349paWlpKWlMWzYMC6++GJ+8pOfaF1eVJHQi3EVTRUcqDnAgZoDUR1yHVHtDXfXfl72OWaTmQx3BjmJOfRJ7ENmXCZmk+zJKCJv7dq1zJo1i759+3L99dfTq1cvDh8+zMcff8xf/vIXCb0Ik9CLMaqqUtJQ0hx0Df4GrUvSJUVVKG0spbSxlI0lG3FanfRP7k9BagG943tHbKEEIX7/+9+TlJTE+vXrSU5ObvHYsSUMu0r92r6iIkzeusYAVVU5VHuI/x34H4u3Luat3W/xednnEnid4A162VGxg7d2v8WSbUtYe3gtpQ2lWpclosDevXsZPnx4q8ADWk3Hev7555k4cSJut5uUlBRmzJjBe++91/z4ifYVPdkeoscoisIjjzzC8OHDcTqdZGVlccMNN1BdXd3iuGPXWr16NRMnTsTpdJKfn89zzz3X6vvo6LVramq4+uqrSUpKIjk5me9+97unvBjJ8WQgSxSr89Wxq2IXuyp30RRo0rqcqBRvjyc/JZ+BqQNJc6dpXY4woHPOOYd169axdu1aRowY0e5xCxcu5K677mLKlCnMmzcPu93OJ598Qm5uLn/84x+BcBDZbDYqKyu54YYbyMvLY/DgwUycOJHTTz+91R6iixcv5qabbmqxKtT111/PokWLuOaaaxg3bhz79+/nb3/7G8OGDWPNmjXYbLbmazmdTmpqarj22mvJzs7mmWeeYdOmTWzbto3hw4cD4f1LO3JtVVWZOXMmq1ev5sYbb2To0KG8/vrrVFRUsHXr1ogNZJHQizIhJcSBmgPsrNhJUX3HtmQSkZEZl8mIzBHkp+TL/T/RYcuXL29eonDixIlMnz6ds846i1mzZjUHzJ49exg8eDCXXHIJr7zyCmbzVz9fqqo2d7fn5eVx8OBB3nnnnRYL+N9777388Y9/ZNOmTQwcOLD567fffjsPPPAA+/fvJzc3l9WrVzN9+nSWLFnCVVdd1Xzcu+++y7nnntvi68eutWrVKqZPnw6El43Mzc3lxz/+MQ8++GCnrv3mm28yd+5c7r//fn75y18CEAqFmDVrFh999FHEQk9+M6NEtaeadYfXsWTbEj7Y/4EEngbKGstYsX8FS7YuYUPxBhr9jVqXJAxgzpw5rFu3josvvpgtW7Zw//33c84555CTk8PSpUsBeOONN1AUhTvuuKNF4EHrhfjb2le0o3uIvvzyyyQlJTFnzpwWx40bN474+PjmvUGPGTZsWHPgAWRkZDB48GD27dvX6WsvW7YMq9XKD37wg+bnWiyWiA/kkYEsBldUV8SW0i0cqTuidSniS56gh40lG9l8dDN5yXkMzxhO74TeWpcldGzChAm89tpr+P1+tmzZwuuvv87DDz/MZZddxubNm9m7dy9ms5lhw4ad9Fxt7Sva0T1ECwsLqa2tbXdpx+MH1vTt27fVMSkpKS3u/3X02gcPHqR3797Ex8e3eDzS+41K6BmQqqrsr9nP5qObqWiq0Loc0Q5FVdhXvY991ftIc6UxtvdY8lPytS5L6JjdbmfChAlMmDCBQYMGcc011/Dyyy936hxtjdTs6B6iiqKQmZnJkiVL2jzu+OCyWCxtHvf1u2ad2b+0J0joGYiqqhRWFbL56GZqvDValyM6odJTyfv73ifNlca47HHkJedpXZLQuWMbb5eUlDBgwAAURWH79u2MGTOm0+fq6B6iBQUFvP/++0ydOjVi0xw6eu1+/frxwQcf0NDQ0KK1F+n9RuWengEoqsKuil28+MWL/O/A/yTwDKzSU8l7e9/jtR2vcbDmoNblCB1YuXIlbY0nXLZsGRDu3ps7dy5ms5m777671TD/joxF7OgeoldccQWhUIh77rmn1XHBYLBL0wc6eu3zzz+fYDDI3//+9+bHQ6EQjz76aKeveSLS0tO5w7WHWXdknQRdlKloquDdve+S4c5gXPY4+ia1vjciYsNPfvITmpqamDdvHkOGDMHv97N27VpefPFF8vLyuOaaa0hOTuY3v/kN99xzD9OnT2f+/Pk4HA7Wr19PdnY299133wmv0dE9RM844wxuuOEG7rvvPjZv3szZZ5+NzWajsLCQl19+mb/85S9cdtllnfr+Onrtiy66iKlTp3Lbbbdx4MABhg0bxmuvvUZtbe2p/PO2IqGnU9WeatYdWScDVKJceVM57+x5h6y4LKbkTiEjru2b/VFFVcHrBY8HmprCfwaDEAq1/WEygdkc/rBawx82W/jD4QC3G+LiwGnMLaQefPBBXn75ZZYtW8ZTTz2F3++nb9++/PCHP+S3v/1t86T1u+++m/79+/Poo4/ym9/8BrfbzahRo/j2t7990mt0Zg/RJ554gnHjxvHkk0/y61//GqvVSl5eHt/61reYOnVqp7+/jl7bbDazdOlSbr75Zp5//nlMJhMXX3wxf/7znxk7dmynr9semaenM96glw3FG9hRvkPX2/WIyDNhYkj6ECbkTDD+HoBeL9TUtPxobAwHnMcTDr5IM5vDAXjsIyEBUlK++rDbI39NYTgSejqhqAqfl33OxpKN+EN+rcsRGnJanUzInsCQ9CHGWOOzpgbKysIfVVXhz71eratqze3+KgDT0yErC77WwhGxQUJPB4rri1l1cBV1vjqtSxE6kuHOYFrfafrq8gwGw+FWWhr+KCvTZ8B1lMsVDr9evcIf6enhFqOIWhJ6GgqEAnxS9Anby7drXYrQKRMmBqcPZmLORO26PGtr4dCh8EdJCSjRsZFwmywW6N0b+vaFfv3CXaQiqkjoaaSorogPD34oOx2IDnFZXZyRd0bPjPJUFCgu/iro6mK4ByIlJRx+ffuGW4RG6G4WJySh18MCoQAfH/mYHRU7tC5FGNCwjGFM7jMZq7kbBl4fPQqFhbB3L/jlvnIrTifk58OgQdDOMl1C/yT0etCRuiOsOrhKWnfilCQ5kpjVfxaZcRF44a2tDQddYSHU15/6+WJFcjIMHBj+OG6tSKFvEno9QFEV1h1exxflX2hdiogSZpOZsb3Gclrv0zo/wjMUgj17YMeO8EAU0XUmE2Rnw+DB4VagDILRPQm9btbgb+D9fe9T1igvLiLyMuMyObP/mSQ6Ek9+sMcDX3wRDjuPp/uLizVuNwwfDsOGhSfNC12S0OtGR+qOsGL/CrxBAw/pFrpnM9s4I++M9ndwqKiAbdvC9+qieeSlXlit4ft+I0fKPEAdktDrBqqqsrFkIxtLNsqqKqLHjM4azcSciV91dxYXw8aN4T+FNvr2hbFjwyM/hS5I6EWYN+hlxf4Vsmam0EROQg5z4kZh37g5PKdO6EPfvjBxIqSmal1JzJPQi6DyxnKW71suozOFJtJUO0P9ZpI8QTKOmrA3BrQuSRxvwAAYPx4SO3APVnQLCb0IOVR7iPf3vU9QCWpdiogxiVgZ6reR/LXBKWbMpFfZcVXJ/WTdMZvD9/zGjQvvDiF6lIReBOwo38HqQ6vl/p3oUVZMjAi6yGpsaveYlEY3iSXtPy40ZLWG7/eNHi1THXqQhN4p2lC8gY0lG7UuQ8SYHNXJ4KYgtuDJexYSAi5SDnkwyW+6PiUlwZQpkJurdSUxQUKvixRV4aODH7GrcpfWpYgY4sTMaL+jRVdmR7hDTtIPeCX49Cw/Pxx+brfWlUQ1Cb0uCCpBlu9dzuG6w1qXImJIgeKif6MfixLq0vNdIScZEnz6ZreHR3kOG6Z1JVFLQq+TvEEvbxe+TXlTudaliBiRoFoZ7bcSF4F96yT4DCI3F844Q1p93UBCrxN8QR9v7X6LSk+l1qWIGDFUcZPb4MEUwV9TV8hJxkEfJkV+9XXN6YQZMyAvT+tKooqEXgf5Q37e2v0WFU0VWpciYoAdM+N9dhK6aVdyl+Ik44AEnyEMHhy+12ezaV1JVJDQ6wB/yM+ywmWyaLToEWmqndFNSodGZp4KCT4DSUyEM8+UffwiQELvJAKhAMsKl1HaWKp1KSIGDFCc5Nf7MPXQnE+n4iRTgs8YzGaYOhWGDtW6EkOT0DuBoBLk7cK3KWmQNQxF97KoJk4LOklt6vktf5xftvjMEnzGMGQITJsmE9q7SEKvHUElyDt73qG4XlaoF90rESunNZlxBPya1SDBZzBZWTBnjozu7AIJvTaoqsryfcs5UHNA61JElMtVnQxp8GPWwT53rpCTjP1eOrkPu9CK2x0OPtm2qFMk9Nrw8ZGP2Vq6VesyRJQbqLjIr9fXDuYJPheph/VVkzgBsxlmzgzv3iA6RDqFj7OzYqcEnuh2wxW37gIPoN7hob6XS+syREcpCqxYAZ9/rnUlhiGh9zVFdUWsPrRa6zJElBsTdNOnXr87H1TFe/CkOrUuQ3TG2rWwfr3WVRiChN6Xarw1LN+3HEXV/t6KiE4mFSYETrwVkF5UpPrxx8lkaEPZtAlWrQK5Y3VCEnqE19N8Z887+EPajZ4T0c2impgc0GZKQlcoKJT3VgnZ5CXCUHbuhOXLIdS1RcljQcz/RCuqwnt736POV6d1KSJK2VQTp/scJHqMtYt5kCAVuXbZGtloDhyA994L3+8TrcR86H185GOONhzVugwRpeyYmeK1EeczVuAd4zV7qcmVuWCGc/gwvP++BF8bYjr0DtYc5PMyGfUkuodFNTHJa8PpN3a3eZ2jiaZ0GdhiOAcOwMqVco/vODEbeo3+Rv534H9alyGilEmFiX4Hbp9P61IiojLZT0AGthjP3r3w4YdaV6ErMRl6qqrywf4P8IWi4wVJ6M+4kIvEbtoWSAsKCuW9QJXlWoxn92746COtq9CNmAy9z0o+k/t4otuMDLlJazTGKM3OCJgC1ObI/T1D2rFD5vF9KeZCr7i+mE0lm7QuQ0SpQYqL7Ab9z8PrqjpnE/4Eu9ZliK7YtCnc6otxMRV63qCXFftXoMogbNEN+qlO+utwabFIUoGKTOnmNKxVq6AktrdKi6nQ++jgRzQFovdduNBOLxwMrouNe8QBk1+6OY1KUcKT1+tid15yzITevup97K/Zr3UZIgrFqRZG1Ad7bLdzPZBuTgPzeuHttyFKRhZ3VkyEni/oY82hNVqXIaKQSYVxPisWJbaWfZJuToOrrQ23+GJwDl9MhN66I+vwBKP7XovQxmjFjStG3zEHTH5qs2UbIsMqLoYNG7SuosdFfegV1xezu1JGLInIy1WdZEXxSM2OqHN58MdLN6dhbd4MRUVaV9Gjojr0FFWR/fFEt4jHyuCGgNZlaE4FKrOkm9OwVDW8CW1T7Lx5i+rQ23J0CzXeGq3LEFHGDJzmscTcfbz2+GU0p7F5POHgi5H7e1EbevW+ejYdlUnoIvJGBV24/LF5H689dU4PQadV6zJEVxUXhyevx4CoDb1Piz4lqAS1LkNEmVzVSVYULjF2qlRUanrJgtSG9tlnUFqqdRXdLipDr6Kpgr3Ve7UuQ0QZO2a5j3cCjVaPzN0zMlUN78gQ5buuR2XofVr0qdYliCg0KuiI6vt4KmZU7IRwESSBIAmEcKJ24mWiOiMqX1JiR01N1HdzRl0nfFFdEUfqjmhdhogymdiNu3OCaiJAHEHiCShOAiEbIcWEophQVRMhBZQQJ1xPxmwGqwUsZgWLRcVsVrCYQpjNIax4sZkasdCE1+zFk+rEVRU92yrFnM2boX9/SEvTupJuEXWhJ608EWlmYJhBRnSr2AkQT1B14ws58Put+PynvkCaooBfga86hyzAsXt4cUAaFjM4HApNCUH6ehSc/jIsIYP8w4mvKEq4m3PePDBF31yUqAq9fdX7KG8q17oMEWWGKG4cel2oXDXhN6XgCSXR6HUQ0PCWY0iBJo+ZJuyYHVk4PYOxWoM4bXW41FJcviJMavR2D0eVigrYuhVGj9a6koiLmtBTVIX1RbJJooiseKz0adBXt6aqWvCZ0mgKJNLksRJStK6otRp7DVlkEQxaaQim0kAqJtMQXK463JTh8h3BrPi1LlOcyIYNkJ8PCQlaVxJRURN6uyp2Ueur1boMEWVG+ayYVB1MfVFN+EilMZhMo8eGosOg+zqf4sOb7MVV89XanKpqosmTRBNJmBiIw9lIvKkIt+cgJnT+DcWiUAg++QRmz9a6koiKitBTVZUtpVu0LkNEmX6qkwSvtgMyVNVKo5pNbaOboMF6Bqvt1bhMrjZHyKiA1xuHl0FYLAXEO8pI8O2Re4B6s29feNPZ3r21riRiomJ88YGaA9T5YndTRBF5FtXEgEbtWniKaqcu1I+i+gFU1hkv8AACSgBv0snfNIRCFmqbelMUmk6FcxI+R1YPVCc6bO1arSuIqKho6W0t3ap1CSLKDFRdWDVodYRw0hDqTV2DAyUKlkKss9fhxNmhY1Wg0ZtMI2Nw2L0kmwpx+oq7t0BxcpWVUFgIAwdqXUlEGL6lV9ZYRmlj9C+dI3qORYU+TT27tmYINzXBfIpq8qipj47AA/AoHgLuzg8p9fmdlPpGUuacit8enfPFDGX9+qhZqcXwoSetPBFpg1Q3lh76BVdUO9XBAo7U9KW2wX7K8+n0qCG+ocvP9XjjKfGPp8I5kaA1ukYRGkpDA3zxhdZVRIShQ6/eV8/+6v1alyGiiBUTOT3UyvOoWRQ35FPXEN0LNdcr9Si2Uxud2ehNoTg4hWr3WBRTVNyVMZ6tW6OitWfo0NtWtg01Kt8bC60MUlzd3soL4aTSP5Cy2pRoeA05KRWVxsTGCJwH6poyKbHMwOvodeqFic5paoKdO7Wu4pQZNvT8IT+7KnZpXYaIIlZMZDd2vJVXXdvAuEtv4fDRig4/p0npTUldHg1NljYf/8FtP+DDdR92+HxGUWepgwitaBUM2ij1jabKfZq0+nrali3ofpLoSRj2J2ZP1R4CimzzIk5s98bdvLf4PQ7tOERtRS0/ePAHjJk5psUxi+5axLq31rX42owJw3jujz894bn/tmQZc6aMJrdXOgCHj1Yw/Zu/aXXc3LMm8udf/4TPD1m55JqLm7+eGJ/IoIJB3PCtGxiUPwiAb1/2bf72z78xfdJ0zGbDvidtJagE8SR6cNW6Tn5wB9U3ZeCxziDNsgOnryRi5xUn0NAQHsk5eLDWlXSZYUNvd+VurUsQBuD3+OkzsA9TL57KE798ot3jRkwZztO/+B7mL7cOcthO/Kvh8fp56Z01PNtGMC554GYG5mU3f67Y+lFc2wuPN/zC/NCdD5GXm0d5ZTl/feav3Pr7W3nur8+REJfApLGTeODvD/DJpk84fdzpXfmWdavOUYeLyIUefNnqC44i0Z1FctMWTn1pbXFSmzfDoEGGXYzakG8la721lDWWaV2GMIARU0cw94dzGTtr7AmPS7Y66JUcT2ZqEpmpSSQlxJ3w+JWfbMNus3HasPzW50r88jwpydgST8MX6tViCkJiQiJpKWkMGTCEH37nh1TVVLFj9w4ALBYLk06bxAerP+j8N6tzXsVL0NU9E/7rmrIoc04lZHZ0y/nF19TWwn7jDiA0ZOgVVhVqXYKIMls27mDcpbdw5nfv4DePLKG69sTD7D/dtocRA/u2+7iCjXLfwJOOzLTbwzuNB4JfddUPHTCUbTu2daJ642iK674J/15vHEfNU/HbUrrtGuJLW407VcyQ3ZuFlRJ6InKmnT6G700aS26vdA4Wl/PAP97g6tsf5bVHb8Viaft9YVFpJVlpyW0+dulN92MymTk2cuOv9/y1+Z7d19U31vPcK8/hcroYOnBo89fTU9MpqyxDUZSouq8H0GhuJJHEbjt/MGij1DyBNOcu3N6D3XadmFdWBlVVkJqqdSWdZrjQK6kvod5fr3UZIop8e9ZkEj3hNSKH5OcwND+HGd/+LR9v2cXU04a2+RyvP4DD3vavz12/uIvc7H7Nn2emZ7Z4/Ee//hFmsxmP10N2VjZ3/fwuUpO/evGw2+0oikIgEMDhiK7uOr/iJ+AOYGvqvrmJimKi3DuEJHcSyU3GbZHo3vbtMG2a1lV0muFCTwawiEiKx9oceMf0zc4gNSmeA0Xl7YZealI8tQ0tu+p8hEdxpqdm0qd3n3aveefP7yQvN4/EhEQS4lqvMlLfUI/L6Yq6wDvGE+fp1tA7prapN6rbSkrTxm6/VkzaswcmTwarsWLEUH0nQSXI/hrj3kAV+tM/ZG/1tZLyaqrrGslMS2r3ecMH5FJ48Kth8g1KLhV17R//dZnpmeT0ymkz8AD2H9rPgP4DOnQuI6o39VxPTV1TBlWu8T12vZji94eDz2AMFXqHaw/jD8luy6LjvE1eDu86zOFdhwGoKKrg8K7DVB2tAiCuso4/PPkKG7fv4/DRCtZs3MH1v3ucvOwMZowf1u55Z4wfTuGBYmrrG6kN5VFZd+LRnp2xdcdWJoyeELHz6U1QCeKP77nf43pPGpWuCaiRmh0vvrJ9u9YVdJqh2qUHa+XGtOicg9sP8tCNDzV//vLDLwNw+oWnc/ud38eBnx37inj1vY+pa2giMy2ZGeOH8vOrL8Fhb78Lbkh+DsMH9uWVlfuYPWNcxOotryzn812f85uftp7kHk08Lg/2htat7O7S4ElFdU0gzbNe5vJFUkVF+CM9XetKOsykqqphfgKe3/o8TQHZWVlExsSAi5QmT5efv2xdMX96YhGLHl4UsVGWTyx+gvqGen75g19G5Hx6ZTVbyS7LxtTDra84Vw3pnk969JpRb/RomDRJ6yo6zDDdmxVNFRJ4ImKsmEj2nHxX7/Z41F4MH3omF825iIqqjq+9eTIpSSlce+W1ETufXgWVIMH4nt+ZvtGTTI17dI9fN6rt26d1BZ1imJbexpKNbCjeoHUZIkr0U50Mqeta6PlJ5WhNpnSSnaJUUkmo0GaPvDTXXuI9xhuEoVvz5xumi9MwLb1DtYe0LkFEkcwuNjJCxFFWJ4EXCR5z17uWT1WVpwCvo7dm1486BmrtGSL0vEEv5Y3lWpchokiSr/M7dKjYKW3MJWTsnVV0w6t4Uc3avH1QgYrgcALW7lsdJqZI6EXWkbojslmsiJh07J3fKFY1Ue7NIyC7WUWMikogXrt/0FDIQrlpvCxSHQl1deFRnAZgiNCTrk0RSb1DnZ+p06Dm4vEa4tfFULz2rg8mioRAwEaV/TRNa4gaBw5oXUGHGOK3uKReNogUkZPq79wNvSAJVNW5u6ma2Kblfb1jmryJ1LuMuymqbhQVaV1Bh+g+9JoCTTQGGrUuQ0QJl2rB6e/EaiCqiYqmbOlc7yZexYti1f4mabW3n2xJdKrKyzFC/7/uQ082ixWRlEPnVgGpU3Lx+WX5qu4UiNP+hVJVTVSaRqPq/yVRvxQFSvTfK6f7/2EJPRFJ6Z14fQ2QRHW9dGt2N63v6x3j9zuocw/XugxjM0AXp4SeiCkJPl+HjlNVC5WNvbq5GgHgNekj9ABqm7Lx24y3MapuSOidGlVVZX6eiJhkbJiVjt0/qlNy8QWkW7Mn+FW/bqYkqUC1WVp7XVZVBR7tByediK5Dr9pbTUDRvr9fRIdUtWNTFfwkU1Pv7OZqxDGKqqA4tR/McozX56bJ2U/rMozr6FGtKzghXYeedG2KSEoKdazlVuPN7OZKxPECDn29ua0OFsiglq7S+SR1Xf+vStemiKT44MlXYfGRJpPQNRCw6iv0gkEb9e4hWpdhTBJ6XVfjrdG6BBFFnB2YQ1TrTeuBSsTxAhZ9hR5ArTeHkFm6uTtNQq/ran21WpcgokQi1pMOYpFWnnb8dGLBgB6iKGZqHSO0LsN4PB5o0u/ep7r9DQ8qQdk0VkRMmmo76TG1HmnlacWn+FBN+hjB+XUN3lRCFpfWZRiPjlt7ug29Wq+08kTknGwQi480PD7d/jrEhKCr53dSPxlVNVHvGKh1GcYjodd5db46rUsQUSQ+eOKuTWnlaS/o0F/oATT4MlFMnd+ZI6ZVVWldQbt0G3pyP09E0okGsUgrTx+CZn2GXihkodFZoHUZxlJfr3UF7dLtb7q09ESkmAGL0v50BRmxqQ+KWT8T1I9XH8xBRVbo6TAJvc6Te3oiUty03zUVwiUjNnUiaNJnSw/Cm816nH21LsM4vF7dbjOk29/2er9+3ykIY4lT2/8x96jpPViJOJEQJ188QEsN9NG6BGPRaWtPt6HnCeh70VJhHC4s7T7W4JGtg/QiiH5begBeb7xMVu+MOn3eotJl6AWVICFV3+/6hHG41LbvxQRN8bJBrI4EFX2Hngo02XO1LsM4pKXXcb5gx/Y8E6Ij2lvA3xOSASx6oqKiWPQ7mAWgSZU9FjussVHrCtqkz9ALSeiJyHEoba/0Ue+RlTb0RrHrO/S8Pres0NJRHdywuafpMvS8Qf3spCyMz97GmpsBEvU6uCymKVZ9hx5Ak1322usQCb2Ok+5NEUm2NrYU8oRSNahEnEzIov97+Y2K7LfYIX79LSIOeg29GO7e3PHpDu687E6U0Mnf8RbvK+bW82/F54ndf6+OMKut/y3rPTIKT49Us/4WnT6e3+eSZck6QqctPV3+z2nZ0tu9cTfvLX6PQzsOUVtRyw8e/AFjZo5pccwN429o87nzb5rPOd85h10bdvHQjQ+1ecztz95O3vC8dq//2l9f4/xrz8dsCb8f2fy/zby3+D2K9hRhd9o5/cLTmffjeZhMJrLzs8kfkc/7S97ngusu6NL3GwtMassX0hBxBPU9UDB26T/zUAGfPROXr1jrUvRNQq/jtGzp+T1++gzsw9SLp/LEL59o85j737m/xeefr/2cxfcs5rQzTwOgYHRBq2OWPrGUnet30m9Y+/cD9mzeQ/mR8ubzAOz4ZAdnXHYG/Uf0p2hPEf/47T/o3b83p194OgBTLp7C4nsXc+7V52Kxtj8fLZYdH3p+EjWqRJyMHrcXaovPnIYLCb0T0mn3pi5DT2mjO6qnjJg6ghFTT7xxZFJ6UovPt3y4hUHjB5HRJwMAq83a4phQMMSWD7cwa8EsTKb254Wtf3c9QycNxeb4au+3K2+9svnvmbmZZOZmUnX0qxXMh04aSmNdI7s37mboxKEd+yZjnC8oo+90yyDTJn1KstYl6J9OQ0+X9/RU1Rjv9gDqKuvYtnob0y6Z1u4xWz7cQkNtA1MumnLCc+3ZvId+Q9tvCa5ZuoaK4grGzxnf/DWrzUruoFz2bNrT+eJjQFsNB4/v5BvKCm2oRujfBHx+F6o+Xz71RYev5br8XzPKDz7AurfW4YxzMnbW2HaPWfPmGoZPHk5KVsoJz1VZUklyRnK713npzy/x44d/TFa/rBaPJWUkUXm0stO1xwLzcU0HVbXiDxikORGLDPJfo6om/I4MrcvQPwm9jjFSS2/N0jVMPHdiiy7Jr6sureaLj79g6iVTT3qugC/Q5nmUkMILD7zApTddyqBxg1o9bnfY8Xv12ZWgNdNxb6ACpgSNKhEdYaQ3vD6LrOhzUm3MkdWaLkPPKAo3FVJ6sJRpc9vv2lz7n7XEJ8Uz+ozRJz1ffHI8jXWtl+7xeXx4G72tWnjHNNY1kpAsL+ZtMZta/ogHVGMtMF1bX8sl11xCSVlJxM75yaZPuPYX16Lo8AXJSKEXUOO0LgF/MMiA3/6WtXv3Rvzc24uL6XPrrTTqdBRmV+lyIMuJBnvoyZo319B3aF9yB7W9CK2qqqz9z1omXzC5QyMrcwfnUrK/9Yubw+Xg9udup1e/ttf9K95b3GLEp/jK8e/qAiFHxK/xzxf/yaKXFrX4Wt/svix+dHHz50VHi3j82cfZtnMbgUCAiWMm8tPrfkpq8oknyS9+ZTFTJ06ld2bvNq9zvA9f/ZD7Hr2Pd/73DgBWq5Ws9CzOPuNsvnXpt7BarEwaO4l//PsfLF+1nHNmntOl77nbGONXH4Cg0rG5nvVeL797801e37yZsvp6xubm8pcFC5iQl9d8zGsbN/LEqlV8dugQVY2NbPrtbxmTe/LFrZ9YtYr+6elMKShg0dq1XPPssyc8fv/vf8+idetY+NZb3DBjBk9885vNj20+fJix997L/t//nrz0dIZlZzM5P5+H3n+f313QxSlRZv21q/QZehr+5HubvJQfLm/+vKKogsO7DhOXFEdqr69eoDwNHj57/zMuu/myds+1c/1OKooqTtgS/Lphk4ex7r/rWn29trKWZ373DNcsvIb+I/q3eKyiuIKashqGTpKRm205/p6eL9A9g1j65/bnz3f+uflzi+WrNzker4db7r6FgrwCHr7rYQCe+fcz3H7f7fz9vr9jbueFwevzsmzFMh743QMALLh4AReffXHz4zfcegMXzbmIC2df2Oq5E8dO5LYf3UYgEODjjR/zyNOPYLVa+db8bwFw3qzzeHXZq7oLPUO19EL2Dh133XPP8XlxMYuvuYbs5GSe/+QTZj/8MNvvuouclPB9/ka/n2kDBnDF+PFcv3jxSc4Ypqoqf1u5krsvDv9MLBg/nnOHD29+fP4TTzAiO7v5cYCMhHCPkNNm4x+rV/OL2bMZmNV2DxLANVOmcP3ixdx+7rlYLV2YEqXDBoz+YhhtW3oHtx/k3m/ey73fvBeAlx9+mXu/eS9Ln1ja4rj1761HVVUmnjux3XOteXMNBaMK6JXXsZXZJ503iZJ9JRw9cLTF10PBEKUHS9u8b7f+3fUMnTyUtN5yf6EtynEvov5u2krIYrGQlpLW/JGcmNz82Oc7P+do+VFu//HtFPQroKBfAbf/5HZ27d3Fxm0b2z3nxxs/xma1MXxQ+IXM7XK3uIbFbGn1tWPsVjtpKWn0yuzF3HPnMm7UONasX9P8+JTxU9i1dxdFR4si/49xCswn2PBXb0JB60lXZvH4/by6aRP3X3opMwYNYkBmJndddBEDMjP5+4cfNh/37cmTuePCC5k9ZEiHr//ZwYPsLS/ngpEjAXDZ7fRKSmr+sFutuI/7muXLN1iDs7KYNXgwv3nzzRNeY87QoVQ1NvLh7t0drqsFHYaeLlt6x9+H6UmDxw/myQ1PnvS4GfNnMGP+jBMec93vr+vUteOS4ph5xUzeX/I+3/rNt5q/np6d3mZNwUCQVa+u4tp7r+3UdWKJ72tzPlXVQjsbLpyyIyVHmH/dfOw2O8MHD+f73/w+WRnhd9D+gB8TJmy2r1qZdrsds8nMtp3bGD96fJvn3Lp9K4MKWg9c6gqH3UFd/VebemZlZJGanMrW7VvJ6ZUTkWtEgknR34vkiQStidgDVe0/riiEFAWnteVLrctmY/Up3of7aM8eBmVlkeDs2pJ6f5w/nwl/+AMbDhxg/Ne6Wr/ObrUyJjeXj/bs4ayhnexNsnesJdzTdPm2ym7R5z9WTzj/e+eT2ju1Q4MMqo5Wcd415zFgzIAeqMyYVBMoX767VUzd07U5dOBQbvvxbTzw2wf4+fd/TklZCT/57U9o8jQBMHzQcJxOJ08ufhKvz4vH6+HxZx8npISorG5/qklpeSnpKemnVJuqqmzYsoH1m9dz2oiW933TUtIorSg9pfNHmpFaegBBy4lX90lwOjk9P597li2juKaGkKLw/Mcfs27fPkpqa0/p2gcrK8lOSjr5ge04rW9frhg/nltfe+2Ex2UnJXGwsgtTonQaerps6TmtsbsYsDvBzfnfO79Dxx5boUWcWMhsxqwoqHRP6E0+bXLz3wvyChg6aCgLblzAyjUruWD2BSQnJbPwFwt56KmHeHXZq5hNZs6cdiaD8gedsCvf5/dh7+ILx7rP1nHuN88lGAyiqAqzp8/m6gVXtzjGYXfg9elrGy+zYrDQM598dZ/F3/se33v2WXJuvRWL2cxpffty5YQJfHbo0Cld2xMI4LSd2s/0vZdcwtA77+S97dvJTGh7BLjLbqepK6urOCI/aCwSdBl6Dos+/7GEMQXNZmyA2kM/7glxCfTp3afF/bIJYybw78f/TU1dDRaLhYS4BOZdO4/srOx2z5OUmER9Q32XahgzYgw///7PsVltpKWmYbW0/t7rGupa3HvUA6N1byodeCNVkJHBh7fcQqPPR53XS++kJBY89RT56afWik+Pj2db0andky3IyOD6adO47bXX+Md3vtPmMVWNjRRkdGEivk5berp8W+WwSuiJyAkd697sodBr8jRRXFpMakrr6QjJickkxCWwcdtGqmurmTqh/UULBvYfyMEjB7tUg8vhok/vPmRlZLUZeD6/j+LSYgb2H9il83cXo7X0VDo+ojHO4aB3UhLVjY28u307l4w++dzdExmbm8vOo0dPeTGPOy68kN1lZbywfn2bj39eXMzYDkyfaEVaeh0Xy92bIvKCX3YhdlfoPf7s40wZP4WsjCwqqyp55sVnMJvNzJ42u/mYZSuW0a9PP5ITk/li1xc8+syjXH7h5fTN6dvueSeMmcBTS56ivqGehPjILj6wfff28MjQwcNPfnAPMoeMFXqKevLQe/eLL1BVlcG9erGnrIxfvvoqQ3r14pqpX73hqWps5FBVFcU1NQDsOhoewd0rMZFe7dy3mzV4MA0+H18UFzMip+uDkbISE/n57Nk88N57rR47UFFBUU0Nszs7iAV029LTZehJ96aIpID5y9BTumfrpfLKcu5++G7q6sPdhSOHjuTv9/2d5KTk5mMOFx3m/5b8H3UNdfTK6MW3Lv0WV1x0xQnPW9CvgEH5g1i5dmWL+XmR8MHqD5gzYw5Oh77eYJpCxure7EhLr9bj4fbXX+dITQ2pbjeXnnYav587F9vX5r0t3bKlxcTybzz9NAB3Xnghd110UZvnTYuPZ96YMSz59FPumzfvlL6PW+bM4e8ffog3EGjx9X+vX8/ZQ4fSL60LU6J02tIzqTpc6DKoBHlm0zNalyGixKiQm94NTdQr/aiqM9a2Qus+W8ffn/s7ix5e1O4k9s6qqavh2z/5Nk/d/xS9s3pH5JyR0reqr6Hu67mc9WR612p2/a1HjjDnkUfYe++9xHdx6kJ7/MEgA3/3O/517bVMHdCFEeKnnw5fziHUE132JVjNViwm2RBVRIb/y/2FFIPdLwI4fdzpXDTnIiqqKiJ2zqNlR/nZ93+mu8AzYTJU4EF47qeWRvXpw5/mz2d/V6YUnMShqip+fd55XQs8gHZGg2pNly09gH9t+xcN/gatyxBRoJ/qZEidl+pgAXUNspeeXtnMNrLL2h/NqkcOh4devlVal6FPl14KXekW7Wa6feubYNfnuwRhPDWEADCb9LergPiKrRsWD9h7dC/jfzmeBm9k30D7g36m/noqWw5sjeh5o4pOW3q6HMgCkORMoqQhctupiNhVR/jmvNkU0rgS0ZbnX3ueVR+v4nDRYVw2F6fln8Zt82+joFcBAIcrDjP9N9PbfO5j33+MC8a1vwPA/a/fz9WzribeGQ/Aul3ruPKhK1sd9+Pzfswtc29pfnxg74G8c8c7WMxfdV+OvHkkd1xxB5dPuRy71c7353yf37/0Z875qSwD2IrDIaM3OyvRceLlfYToKNUEPpsdi19aenq05YstzDt3HpMHTMZeaeeBNx7gO3/5DsvvWo7b4SY7NZtP7/+0xXP+/dG/eeq9p5g5fGa75y2qKmLFthUs/MbCVo+tuHtFcxACxDla7o13qOIQr657lSumtj/Cdu7Eufz+lXv5oriY4dnG6pbtdjpt5YGOuzcl9EQkeS1mTAS1LkO04YHfPcB5Z57HkOwhDMsdxoNXP0hRVRHbDm4DwGK2kJmU2eLj3c3vcsH4C4hztr+R6383/JehfYbSK6X1LidpCWktznf8ea6edTWP/OcRfIH2N1BNikti4sDR7U7qjmmJ+n391m3oJTm6vpCqEMdrtJqke1PnLP5wV2K9J7z0WnJccpvHbTu4je2Ht7Ng6oITnu/TPZ8ysl/Xhsx/76zvEVSCPLvyxJuyjisYxkd79nTpGlHty30C9Ui/oeeU0BORU29GWno6Z/VZURSFu1+6m/EF4xmcM7jN415c8yIDeg9gXMG4E56vqLKIrOS2N0g9/bbTGXbTsOaP6obqFo877U5+euFPefydx6nz1LV5DoDeyWld24Eg2p3iuqLdSbf39KxmK26bm6ZAk9aliChQSxCzhJ5uWc1WTCETv/33b9lVvItXfvlKm8d5/V7e/PRNbrrgppOe0xfwtbuO70u3vNTinl6Su/Wb7AVTF/D08qd54p0n+NW8X7V5Hpfd2rUdCKKdjkNPty09kPt6InJqTQEgcNLjhDbsJjt3/PsOVmxbwQs/f4HeKW1PnF+2cRlev5f5k+ef9Jwp8SnUNrW9Z11uei55mXnNH22tdmO1WLll7i38c8U/Ka1pe9/B2oZKMnQ8aEMTLhfEtX+vVWu6Dr1kZ7LWJYgooQAhq7FW+4gVqqry4JMP8u7md/nXz/5Fbnr7K/q/uOZFZo+eTVrCySc9D88dTmFJ4SnVdsG4CxiUPYhH3nqkzcd3HN7btR0IopkOJ6R/na5DLzNONkgVkdNkB4uuf+Jj08P/9zDLPlzGX679C3HOOMpqyyirLcPrb7nB7YGyA3xa+OlJB7AcM2PYDDbt20RIObUBTLfOu5WX17yMx+dp9dja3V9w9rBhp3T+qKPjrk3Q8T09kNATkVVlhUSHQpNHkk9P3nz3TQC+8edvtPj6A999gMunXN78+UtrXqJ3cm9mDJvRofPOHDETi8XC6h2rOWP4GV2ub8qQKZw+5HQ+2v5Ri69v2r+BWk8Tl512WpfPHZV0Hnq6XXsTwt0e/9z8T4KKDEAQpy4eKyOr+lBdr6/tdGKdxWShT3mfbjn3cyufY/nW5Sz+6eKIn/ump3/ApGw7vz7//Iif29C+/e3wfT2d0nVLz2QykeHOkOXIREQ0EES1+QEJPT1xmrrv/+OqGVdR56mjwdvQYrTmqfIH/Qzr25+fzezC5qrRLDVV14EHOm/pAXxy5BO2lG7RugwRJUb7U/GVSLe5nqSSSkKF8UZAprgPk9i0Xesy9GXECJgyResqTkj3Nzfkvp6IpFprExHai1VEiN2nz4WJT8YRjNweh1EjJ0frCk5K97/+EnoikkpNARx2WXhaT+xNxgs9k0nF7pfQa8Fkgt762pi4LboPvTh7HG6bW+syRJTwmEIoDpmkrhcOswNTyHjzJx12DybkzVMLGRm63U7o63QfegDZCbJth4gcr7P9lfNFz3KobS8TpncOS9srvcS0Pt0zAjfSDBF6uYmy4oGInFpro9YliC85g8YcSetQZJHpVvLytK6gQ4wRekm5mDBeF4jQpwprPRZdT9aJHY4Gg7b0fGVal6AviYm6n5R+jCFCz2l1khGXoXUZIkoEUfAnyMr4WnOZXZgDhngJasHh8GBW5b5wC/37a11BhxnmJ65vUl+tSxBRpMYl92S05g4Zc4BanFlaea3k52tdQYcZJvTkvp6IpAprJVi0riK2OZuMeT/P7T+odQn6kpAQHrlpEIYJvYy4DFxWfS9vI4xDNUEgUbqotGIz27B6jHdj1elowhJqvdtCTDNQ1yYYKPQgPKBFiEipcddoXULMilP1u8noibjNbW8mG9MGDNC6gk4xVOjJfT0RSZWWSkx2GRWsBZfHmL02bt8hrUvQl/R0w4zaPMZwoWcz27QuQ0QLE3iSpKuqp1lMFuyN+l+543hOZyMWxXvyA2PJUOPtMmGo0LOarfRPMVb/sdC3MlcZJpO09npSnCkOdL23S9sSkFZeCzab4bo2wWChBzAobZDWJYgo0mRqIpQQ0rqMmOL2GG+qgs0WwO2V0Gth4MBw8BmM4UIvOyGbeHvkNoMUoiq+SusSYobNbMNeb7yuzQRrkdYl6I8BuzbBgKEHMDB1oNYliChSYatAdRqwv82AEpQEwy0paDYrxHn3al2GvmRmQlqa1lV0iTFDL01CT0RWfZKs0NIT4uqMN1Uh3lGGWQ1qXYa+jBihdQVdZsjQS3Ymy+ayIqJKHaXIwODuFWeOw+w31kuOCUj0F2pdhr4kJkJBgdZVdJmxfgK/Rro4RSSFTCqelHqty4hq8T7j3YuPd1VgCTVpXYa+jBkT3iXdoAwbegNSB2A1G28ZI6FfJc5S2XKom1hMFpx1xlpr02xWSPJ9oXUZ+hIfD4OMPYLesKHnsDoYnDZY6zJEFPGagvjT67QuIyolkmi4uXlJziKZjH68UaPAbNjYAAwcegAjs0YabiSY0LeDjqO44hSty4g6cfXGGsBitQZIaNqpdRn64nLBkCFaV3HKDB16iY5E8pLztC5DRBE/Co3JZZjlvVTExJnjsHiNtY9TsnUfJuTNTwujRoHV+P3/hg49gFFZo7QuQUSZvZZa4pNk8EKkJDYlal1CpzgcHuK8B7QuQ1/i4w09TeHrDB96WfFZMn1BRFTIpFIVX4HDbrCbUDrktrixNxhnBRYTkKp8rnUZ+jN+PFiM1Vpvj+FDD6S1JyJvr7mJ+MQKrcswPKO18pLdR7AHZFm6FlJTw+tsRomoCL3+yf1JsCdoXYaIIqoJjrgbSYr3a12KYTnNThz1Dq3L6DCHw0NC03aty9CfKVMMPS/veFEReiaTidG9RmtdhogyB01e7O5SrNHRq9PjkrxJWpfQYWazSnpoEyajzavobnl5kJ2tdRURFRWhBzAkfQiJDmN1pQj92+4MkCbdnJ3mMDsMNRk9xbEfa1BW5GnBYoHJk7WuIuKiJvTMJjMTsidoXYaIMpUmP9VxTaQlNmhdiqEk+5O1LqHDXM564j2yvmYr48aF19mMMlETegAFqQWku9O1LkNEmc8tTTgcZSTGB7QuxRDsZjuOGmPcy7PZ/KT7N2hdhv6kp4fn5UWhqAo9gIk5E7UuQUQZBfjcDcnW/bicMmH5ZFJ9qYZYKcliCZGhfoZZkcFKLZjNMGOG4Zcba0/UfVd9EvuQk5CjdRkiypThpyLOSbrjADbjL0rRbdwWN45a/bfyTEC69QtsQVlrtZVRo8ItvSgVdaEH0toT3WOr1UPIppAZXyTLlLXBhImUuhSty+iQVNc+nL4SrcvQn+Tk8L28KBaVoZcRl0F+Sr7WZYgoE0RlZ5wFK/VkJMkE5uMlmZKwevTfDE50l8nAlbaYTOFuzShZeaU9URl6EG7tWUzR/Z8nel4xPqrcLpyUkZYo63MeYzFZSKzS/0i/OFcNyU2btS5Dn8aOhV69tK6i20Vt6CU6Ehnbe6zWZYgotMnmxW+zEW8+REqCBB9AaigVU0jffb5xrmrSPJ/KBPS2ZGdHfbfmMVEbegBjeo0hyWGcVSGEMQRR2egCxWwm0XKIjKTYnsPnNDtxVbm0LuOE4l1VpEvgtc3phDPPjKqlxk4kqkPPbDIzre80rcsQUajWFGB3fHj3ALfpCJlJNQYYpN89UppSdD1FId5VSZpnvdZl6NeZZ4LbrXUVPSaqQw8gJzGHganRs0K40I+DJi+l8eEXC5fpKFnJFTE3qjPRnKjrrYMSXJWkeWTyebvGjIE+fbSuokdFfegBnJ57Oi6rvrtfhDFtMTfR6AivMemggqyko1hi4rcKrGYrSVX6vX2Q5C4hVQKvfb17h/fJizEx8evptDqZkjtF6zJEFFJNsMHpJ2gJD9W3U0OvxCKs+h+5f8rSPemYg/p7CTGZVNJdhSQ3bdW6FP1KTIQ5c6J21ZUTiZnvuCC1gH5J/bQuQ0QhLwpb48yoX97XslJPr/iD2GwaF9aNkk3Jutwrz2IJkWXbTJxnn9al6JfDAeedFx7AEoNiJvQAZvSbgdsWOzdsRc8pN/nZn/DVi4gFD73de0iIC2pYVfdwmB0kVupvTp7D4aE3a3H4y7QuRb/M5nALL0m/3dLdLaZCz2VzMStvlq5HmgnjKjR7KIn/6k2VyRQk1baHjKSaqBngYsJEWkMaJkVf31Ciu5ws32osIZk3eULTp0fdprCdFVOhB+HRnLLLuuguWy1NVMa1HDTlNh0lO+kwTrvx54ilkIKtST/9tlZrgCzHVlKaNmJCdsA4odGjYfBgravQXMyFHsD47PFkxWVpXYaIUp9ZPNQdd7/EQiNZrt2kJnoM28/gNDtJqEzQuoxmCa4KeodWycLRHTFkCEyapHUVuhCToWc2mTkr/ywcFv3diBfGp5rgU7uPBsdxP18mlQTzQbKTiwzX6jObzKTXpaOHBU2s1iBZjm2kej7DrEbfPdOIGzAg3K0pgBgNPYB4ezwz+s3QugwRpUImlU+cfjzHBx/h0Z1Zrt2kJTYa5l5fhj8Di1fbBdxNhO/dhVt3xZrWYhj9+8OsWTGzxFhHmFRV1cF7N+2sPrSa7eXbtS5DRCmbamKK14bT3/bu3Cp26kPZ1DY4UXT6m5hKKgkV2nZrxrlqSA5sxxqs17QOQ+nbF84+Oybn4p1IzIdeSAnx1u63KG0s1boUEaUcqpnTvVYc7QQfgIKD+mBv6hr1FX7x5njSytI0u77T2UiKshO7v0KzGgypTx8455yo3xuvK2I+9AA8AQ9v7HyDer+8ixTdw6aamOR3EOf1nvA4BUe45Vfv0Pz2mcPsIKsiS5PpCXa7l2RTIS7pxuy8fv3grLOIiWWBukBC70vVnmre3PUm/lD778aFOBVmYILfRbLHc9JjFVzUBXpR16hN+FlMFnrX9cbi67mWgglwu2pJUPbj8EnPS5cMHhze/Vzu4bVLQu9rjtQd4Z0976CoMt9HdJ/Tgm4yGjs2iVrBRaOSQaPHhS/Qcy9kvX29sdf3zO4JFkuIeEcZCb49Mrn8VIwZAxMnal2F7knoHWd7+XZWH1qtdRkiyg0PuenT0LkX+IApEU8wlXqPk2A3jtTPUDNwV3bvcn0mwOlswG06Spxnv0wsP1Wnnw4jR2pdhSFI6LVh7eG1fF72udZliCg3QHVRUHfyrs62+EnGE0yh3uMgFIpcTUnmJJLLkiN3wq8xmVRcjnrcplJcviOYFbmVcMrMZjjjDBgoe4Z2lIReG1RV5d2973Ko9pDWpYgo10d1Mqzeh6mrv4aqCb8plaZgEj6/DZ/f1OV7gAnmBFLKIrsLus0WwG5twK2W4vQVyWTySHK5wotH9+qldSWGIqHXjqAS5O3CtylpkCWORPdKxcaYRrAFA6d8LhUzQRIIqPH4Qk58fis+/8lDLM4cR3r5qa24Yjar2O1NOEx1ONRq7P5yLMqJR6uKLsrMDAdeXJzWlRiOhN4JBEIB3t7zNkcbjmpdiohyNtXEuICTpA6M7OwsVbUQIIEAcQRCDkKKhVDIRDBkIhQy4TC5yKjI6NDUBIslhNXix2r2YTH7sKpNWNVGrKEGrIE6ut7OFB02eDBMmyZz8LpIQu8kAqEA/y38L2WNskeX6H6DFRf96r09Fh521U7mYSBoRjXbUEzhHRRMahCTGmr5IYNNtGU2w5QpMGyY1pUYmoReB/hDft7a/RYVTbIqhOh+6dgZ3aBgDXXv/S+baifrUBBLQMJM9xISwmtoyv27Uyah10G+oI+3dr9FpadS61JEDLBjZpzfTqKne+6J2VQbWYcVLP4IDv0U3WPQoHALz94z8yajnYReJ3iDXt7a/RZVniqtSxExYpjiJrc+shO2w12aIQk8vXM4wlsC5edrXUlUkdDrJE/Aw9t73pauTtFjkrEx0mvG7fOd8rkcqp3Mg0HMQenS1LU+fWDmTHB37yIBsUhCrwsCoQDL9y3nSN0RrUsRMWSg4iKv0YdZ6VpgORUnGQd9mEPyK69bNlt4KbHhw7WuJGpJ6HWRoiqsOriK3ZW7tS5FxBCXamF0wN7pqQ3ukIv0Ax5M8tuuX3l5MHWqzL3rZhJ6p+jTok/ZfHSz1mWIGNNHdTK4MdihEZ4Jfjcph5oiuM6KiKjExPDamf36aV1JTJAtdU/RxJyJTOs7LaJLNwlxMkdMXj6MD1EW1/49n3Wbd5F31g1sXb5Jfjr1yGqF8ePh8ssl8HqQ7DIYAcMyhuG2uVmxfwVBRdYWFF2z9j9reXbhs82fmy1mElMTGTppKJf88BJSMlNaHB9EZZO1iaREG0MD1hZdniZMJDcYc4j7H5YtY1h2NnPHjNG6lO5hMoWnIYwbB/HxWlcTcyT0IiQvOY8LB13I8r3LaQw0al2OMLCLb7yYtOw0gv4g+7btY91b69izeQ93vngnNoet1fG1pgAf2wNk2OwM9plJ8gXJOGpmX60xdzH4wzvvcNlpp0Vn6OXnh1t3yclaVxKzJPQiKDMuk0uHXcoH+z6gqL5I63KEQQ2fMpy8YXkATJs7jfjkeN599l22rNrC+Dnj231eucmPLSObOa4R2Bu29lC1okNyc2HCBEhP17qSmCehF2FOq5PzB57PhuINbDq6SetyRBQYOHYg7z77LuVHypu/dvTAUd58/E12btiJ3+snuyCbH/3iR1z3veswm8yQmwftzOv7ZP9+7ly6lHX79hEIhZiQl8cf5s5l6oABALzy2Wdc/tRT/O8Xv+CMQYNaPPfJVau4cckStt1xByNycth65AgPvf8+qwoLKa6pIdnt5vwRI3jg0ktJ+1rX3V3/+Q8L33qLwnvu4d5ly3hj82ZUVWX+2LE8dtVVuL9cbcR0ww0APLtuHc+uWwfAd08/nUVXXx2pf86elZsLY8fK8mE6IqHXDUwmExNyJpAVn8XK/SvxhU59UrGIXRXF4YUQ3AnhQSvFe4u5/9r7Sc5M5tzvnovL7WL3R7u57frbGJg6kHnz5oWfmJER/nP69PAIwX37WLF9O+c9+ijj+vblzgsvxGwy8c916zjz4Yf56JZbmNi/PxeMHEm8w8FLGza0Cr0XN2xgeHY2I3JyAFi+Ywf7ysu5ZsoUeiUm8kVxMU999BFfFBfz8W23YTK1HEJzxVNP0T89nfvmzWPjoUM8vXo1mQkJ/OnSSwFYfM01XLd4MRPz8vj+9OkAFBz7PozCYglv6jpqlHRj6pCEXjfqm9SX+UPns3zfclnBRXSYp8FDQ00DAV+A/Z/v57//91+sdiujpo8C4MU/v0hqr1Ruf+52spOzmdV/FkmOJKZNm8att976Vegdk5wMM2eiTprEjUOHMmvoUN7+0Y+aA+mGGTMYvnAhv33zTd67+WZcdjsXjRrFKxs38tdvfAOLOTzI+2htLR/u3s1dF13UfOofnnEGv5gzp8XlJufnc+XTT7N6zx6mH7ej99i+ffnHd77T/HllQwP/WLOmOfS+NXkyN/7rX+RnZPCtyZMj8u/ZY1yu8A4Iw4eD06l1NaIdMmWhmyU4Erhk8CUMy5DtQETHPPLDR/jF7F9w2wW38eStT2J32fnRQz8iJSuFxtpGdq3fxfjZ4xmcMJip6VMJNgSprKzknHPOobCwkKKitu8nb969m8LDh7nqZz+jctQoKux2KhoaaPT7OWvIEFYVFqJ8udrLgvHjKauv53+7djU//5WNG1FUlQXjv7qv6PraIsjeQICKhgYm9+8PwMZDh1rVcOOMGS0+nz5wIJWNjdR1wz6CPaZ37/CSYVddFR6RKYGna9LS6wEWs4VpfafRL6kfqw6uktGd4oSuvPVKsvpm4WnwsGbpGgo3FWK1hX9Vyw6Xoaoqbz7xJm8+8Wabzy8rKyPny+7HryssLATgu9/7XrvXrvV4SImL49zhw0lyuXhxwwbOGjoUCHdtjsnNZVBWVvPxVY2NLHzrLV5Yv56y+vpW5zpe39TUFp+nfLm2ZHVTE4kuV7t16U5iYnjawcCB4W1/hGFI6PWg3KRcLh9+OesOr2NX5a6TP0HEpLzhec2jN8fMHMP9193PP377Dxa+upC+iX0BuOWWWzjnnHPafP6ALwekHO9YK+6BBx5gTFvTASoqiM/IgMOHcQBzx4zh9c2befyqqyitq2PN3r38Ye7cFk+54qmnWLt3L788+2zG5OYS73CgqCrn/vWvKG0s9nSsq/R4hlgYyuGA/v3DYScDUwxLQq+H2S12zsg7g4LUAlYdXEWDv0HrkoSOmS1m5v1oHg/d+BBF7xfxix/9gh/yQ2w2G7Nnz+7UuQoKCgBITEw88XNDISguZkFdHc+uW8cHO3eyo6QE9biuzerGRj7YuZOFF13EHRde2Pz1wtLSzn2Tx9HV6jHJydC3b3jFlKys8O7lwtDkf1AjfRL7cNmwyxiSPkTrUoTOXXzOxYyfMJ5nn3yWxMREZs6cyZNPPklJSUmrY8vLy9s4Q9i4ceMoKCjgwQcfpKGh9Zut5udaLJCby+yf/pTU1FRerKzkxV27mDhoEP0zM5uPP9ZqO76V9sgHH3Tl22wW53BQ0xTZPQQ7zGyG7GyYPBkWLIArrgj/vXdvCbwoIS09Ddktdmb0m0F+Sj4fHfyIen/9yZ8kYka6O52puVPJis/i1l/dyuWXX86iRYt47LHHmDZtGiNHjuT6668nPz+f0tJS1q1bx5EjR9iyZUub5zObzTz99NOcd955DB8+nGuuuYacnByKiopYuXIliYmJ/Oc//2k+3mazMX/+fF544QUaGxt58MEH4bvfhbIyKC0lsbSUGYMHc/977xEIhchJSeG97dvZX3FqI5XH9e3L+zt38tDy5WQnJ9M/PZ1JXw6OiTiHI9yC69Ur/JGREQ59EbUk9HSgT2Ifrhh+BdvKtrGpZBMBJaB1SUJDdrOdaX2nMTR9aPO0gvnz5ze30q6//no2bNjAwoULWbRoEZWVlWRmZjJ27FjuuOOOE5575syZrFu3jnvuuYe//e1vNDQ00KtXLyZNmsQNX04M/7oFCxbw9NNPYzKZuOKKK8L7veXkhD+Af40axU9uvJHHVq1CVRTOHjWKt2+5heybb+7y9//Q5Zfz/eef57dvvoknEOC7p58emdBzuSAlJdxlmZ4eDjmZRxdzZGshnfEEPKwvXs+uil2oyH9NrBmSPoSJORNxWg0+7N3jgZoaqK4O/1lbC42N0NQEXm/3XNNkCgeb2x3eky4hIRxqKSnhD5lKIJDQ060qTxXrDq+TNTxjRGZcJlNyp5AZl3nyg41OUcLB19QU/vB4IBgMD6A5/uPYLvEWS/iemsUSbm1areE/nc5wyLnd4cAz6WoYjNAhCT2dO1hzkI+PfEytr1brUkQ3SHenM673OPoly35qQvQECT0DUFSFXRW72Hx0swx2iRJprjTGZY8jLzlP61KEiCkSegaiqAqFlYVsOrqJOl+d1uWILkhxpjAuexz5KflalyJETJLQMyBVVdlbvZctR7dQ6anUuhzRASnOFE7rfRr5Kfmtdh4QQvQcCT2DO1J3hM1HN1NcX6x1KeI4Jkz0S+7HiMwRZCdka12OEAIJvahR5aliR/kO9lTtkf37NOa0OhmSPoRhGcOIt8ef/AlCiB4joRdlQkqIfdX72Fmxk5KG1stUie6T4c5geOZwClIKsJhlVQ8h9EhCL4rVemvZWbGT3ZW78QQNvF+ZjrltbgpSChiQOoCMOIPt8C1EDJLQiwGKqnCo9hD7qvdxqPYQ/pBf65IMzWV10T+lPwUpBfSK7yUDU4QwEAm9GKOoCiX1Jeyv2c/BmoOyoW0HOSyO5qDLTsiWoBPCoCT0Ylx5YzkHag6wv2Y/Nd4arcvRDRMmMuIyyEnIIScxh17xvTCbZGsZIYxOQk80a/A3cLThKCX1JRxtOEq1t1rrknpUsjO5OeSyE7KxW+xalySEiDAJPdEub9DL0YajzR8VTRUoqqJ1WRFhNVtJc6WR7k5vbtHF2eO0LksI0c0k9ESHBZUgVZ4qarw1VHuqqfZWU+2p1v16oE6rk3R3OunudNJcaaS500hyJMl9OSFikISeOGVBJUiNt6Y5EBv9jXiCHpoCTTQFmvAGu2n/tC9ZzVbibHHE2eOIt8eTYE8gwZFAoiORREcibpu7W68vhDAOCT3R7RRVwRv0tgjBoBJEURVCSij8pxr+89jXVFSsZitWsxWb2YbNYmv+89jX7BY7bpsbh9Wh9bcohDAICT0hhBAxQ8ZgCyGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZkjoCSGEiBkSekIIIWKGhJ4QQoiYIaEnhBAiZvw/Ml/BduuAz0IAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draw_venn_rsi(\n",
    "    title=\"All\",\n",
    "    ids_retrieved=metrics_w_manual[\"All\"][\"All\"][\"ids_retrieved\"],\n",
    "    ids_screened=metrics_w_manual[\"All\"][\"All\"][\"ids_screened\"],\n",
    "    ids_screened_relevant=metrics_w_manual[\"All\"][\"All\"][\"ids_screened_relevant\"],\n",
    "    add_pn_labels=True,\n",
    ");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_2c3ff_row0_col1, #T_2c3ff_row0_col2, #T_2c3ff_row0_col3, #T_2c3ff_row0_col5, #T_2c3ff_row2_col1, #T_2c3ff_row2_col2, #T_2c3ff_row2_col3, #T_2c3ff_row2_col5, #T_2c3ff_row5_col0, #T_2c3ff_row5_col4 {\n",
       "  font-weight: bold;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_2c3ff\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_2c3ff_level0_col0\" class=\"col_heading level0 col0\" >Precision</th>\n",
       "      <th id=\"T_2c3ff_level0_col1\" class=\"col_heading level0 col1\" >Recall</th>\n",
       "      <th id=\"T_2c3ff_level0_col2\" class=\"col_heading level0 col2\" >$F_1$</th>\n",
       "      <th id=\"T_2c3ff_level0_col3\" class=\"col_heading level0 col3\" >TP</th>\n",
       "      <th id=\"T_2c3ff_level0_col4\" class=\"col_heading level0 col4\" >FP</th>\n",
       "      <th id=\"T_2c3ff_level0_col5\" class=\"col_heading level0 col5\" >FN</th>\n",
       "      <th id=\"T_2c3ff_level0_col6\" class=\"col_heading level0 col6\" >TN</th>\n",
       "      <th id=\"T_2c3ff_level0_col7\" class=\"col_heading level0 col7\" >?</th>\n",
       "      <th id=\"T_2c3ff_level0_col8\" class=\"col_heading level0 col8\" >Retrieved</th>\n",
       "      <th id=\"T_2c3ff_level0_col9\" class=\"col_heading level0 col9\" >Screened</th>\n",
       "      <th id=\"T_2c3ff_level0_col10\" class=\"col_heading level0 col10\" >Relevant</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row0\" class=\"row_heading level0 row0\" >All</th>\n",
       "      <td id=\"T_2c3ff_row0_col0\" class=\"data row0 col0\" >0.7973</td>\n",
       "      <td id=\"T_2c3ff_row0_col1\" class=\"data row0 col1\" >0.6860</td>\n",
       "      <td id=\"T_2c3ff_row0_col2\" class=\"data row0 col2\" >0.737500</td>\n",
       "      <td id=\"T_2c3ff_row0_col3\" class=\"data row0 col3\" >59</td>\n",
       "      <td id=\"T_2c3ff_row0_col4\" class=\"data row0 col4\" >15</td>\n",
       "      <td id=\"T_2c3ff_row0_col5\" class=\"data row0 col5\" >27</td>\n",
       "      <td id=\"T_2c3ff_row0_col6\" class=\"data row0 col6\" >91</td>\n",
       "      <td id=\"T_2c3ff_row0_col7\" class=\"data row0 col7\" >177</td>\n",
       "      <td id=\"T_2c3ff_row0_col8\" class=\"data row0 col8\" >251</td>\n",
       "      <td id=\"T_2c3ff_row0_col9\" class=\"data row0 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row0_col10\" class=\"data row0 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row1\" class=\"row_heading level0 row1\" >Is RCT</th>\n",
       "      <td id=\"T_2c3ff_row1_col0\" class=\"data row1 col0\" >0.7273</td>\n",
       "      <td id=\"T_2c3ff_row1_col1\" class=\"data row1 col1\" >0.4651</td>\n",
       "      <td id=\"T_2c3ff_row1_col2\" class=\"data row1 col2\" >0.567400</td>\n",
       "      <td id=\"T_2c3ff_row1_col3\" class=\"data row1 col3\" >40</td>\n",
       "      <td id=\"T_2c3ff_row1_col4\" class=\"data row1 col4\" >15</td>\n",
       "      <td id=\"T_2c3ff_row1_col5\" class=\"data row1 col5\" >46</td>\n",
       "      <td id=\"T_2c3ff_row1_col6\" class=\"data row1 col6\" >91</td>\n",
       "      <td id=\"T_2c3ff_row1_col7\" class=\"data row1 col7\" >32</td>\n",
       "      <td id=\"T_2c3ff_row1_col8\" class=\"data row1 col8\" >87</td>\n",
       "      <td id=\"T_2c3ff_row1_col9\" class=\"data row1 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row1_col10\" class=\"data row1 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row2\" class=\"row_heading level0 row2\" >Non-empty Abstract</th>\n",
       "      <td id=\"T_2c3ff_row2_col0\" class=\"data row2 col0\" >0.7973</td>\n",
       "      <td id=\"T_2c3ff_row2_col1\" class=\"data row2 col1\" >0.6860</td>\n",
       "      <td id=\"T_2c3ff_row2_col2\" class=\"data row2 col2\" >0.737500</td>\n",
       "      <td id=\"T_2c3ff_row2_col3\" class=\"data row2 col3\" >59</td>\n",
       "      <td id=\"T_2c3ff_row2_col4\" class=\"data row2 col4\" >15</td>\n",
       "      <td id=\"T_2c3ff_row2_col5\" class=\"data row2 col5\" >27</td>\n",
       "      <td id=\"T_2c3ff_row2_col6\" class=\"data row2 col6\" >91</td>\n",
       "      <td id=\"T_2c3ff_row2_col7\" class=\"data row2 col7\" >155</td>\n",
       "      <td id=\"T_2c3ff_row2_col8\" class=\"data row2 col8\" >229</td>\n",
       "      <td id=\"T_2c3ff_row2_col9\" class=\"data row2 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row2_col10\" class=\"data row2 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row3\" class=\"row_heading level0 row3\" >Non-pediatric population</th>\n",
       "      <td id=\"T_2c3ff_row3_col0\" class=\"data row3 col0\" >0.8209</td>\n",
       "      <td id=\"T_2c3ff_row3_col1\" class=\"data row3 col1\" >0.6395</td>\n",
       "      <td id=\"T_2c3ff_row3_col2\" class=\"data row3 col2\" >0.719000</td>\n",
       "      <td id=\"T_2c3ff_row3_col3\" class=\"data row3 col3\" >55</td>\n",
       "      <td id=\"T_2c3ff_row3_col4\" class=\"data row3 col4\" >12</td>\n",
       "      <td id=\"T_2c3ff_row3_col5\" class=\"data row3 col5\" >31</td>\n",
       "      <td id=\"T_2c3ff_row3_col6\" class=\"data row3 col6\" >94</td>\n",
       "      <td id=\"T_2c3ff_row3_col7\" class=\"data row3 col7\" >156</td>\n",
       "      <td id=\"T_2c3ff_row3_col8\" class=\"data row3 col8\" >223</td>\n",
       "      <td id=\"T_2c3ff_row3_col9\" class=\"data row3 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row3_col10\" class=\"data row3 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row4\" class=\"row_heading level0 row4\" >Phase >= 2</th>\n",
       "      <td id=\"T_2c3ff_row4_col0\" class=\"data row4 col0\" >0.7636</td>\n",
       "      <td id=\"T_2c3ff_row4_col1\" class=\"data row4 col1\" >0.4884</td>\n",
       "      <td id=\"T_2c3ff_row4_col2\" class=\"data row4 col2\" >0.595700</td>\n",
       "      <td id=\"T_2c3ff_row4_col3\" class=\"data row4 col3\" >42</td>\n",
       "      <td id=\"T_2c3ff_row4_col4\" class=\"data row4 col4\" >13</td>\n",
       "      <td id=\"T_2c3ff_row4_col5\" class=\"data row4 col5\" >44</td>\n",
       "      <td id=\"T_2c3ff_row4_col6\" class=\"data row4 col6\" >93</td>\n",
       "      <td id=\"T_2c3ff_row4_col7\" class=\"data row4 col7\" >111</td>\n",
       "      <td id=\"T_2c3ff_row4_col8\" class=\"data row4 col8\" >166</td>\n",
       "      <td id=\"T_2c3ff_row4_col9\" class=\"data row4 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row4_col10\" class=\"data row4 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row5\" class=\"row_heading level0 row5\" >Phase >= 3</th>\n",
       "      <td id=\"T_2c3ff_row5_col0\" class=\"data row5 col0\" >0.8788</td>\n",
       "      <td id=\"T_2c3ff_row5_col1\" class=\"data row5 col1\" >0.3372</td>\n",
       "      <td id=\"T_2c3ff_row5_col2\" class=\"data row5 col2\" >0.487400</td>\n",
       "      <td id=\"T_2c3ff_row5_col3\" class=\"data row5 col3\" >29</td>\n",
       "      <td id=\"T_2c3ff_row5_col4\" class=\"data row5 col4\" >4</td>\n",
       "      <td id=\"T_2c3ff_row5_col5\" class=\"data row5 col5\" >57</td>\n",
       "      <td id=\"T_2c3ff_row5_col6\" class=\"data row5 col6\" >102</td>\n",
       "      <td id=\"T_2c3ff_row5_col7\" class=\"data row5 col7\" >13</td>\n",
       "      <td id=\"T_2c3ff_row5_col8\" class=\"data row5 col8\" >46</td>\n",
       "      <td id=\"T_2c3ff_row5_col9\" class=\"data row5 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row5_col10\" class=\"data row5 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row6\" class=\"row_heading level0 row6\" >Sample Size >= 5</th>\n",
       "      <td id=\"T_2c3ff_row6_col0\" class=\"data row6 col0\" >0.7945</td>\n",
       "      <td id=\"T_2c3ff_row6_col1\" class=\"data row6 col1\" >0.6744</td>\n",
       "      <td id=\"T_2c3ff_row6_col2\" class=\"data row6 col2\" >0.729600</td>\n",
       "      <td id=\"T_2c3ff_row6_col3\" class=\"data row6 col3\" >58</td>\n",
       "      <td id=\"T_2c3ff_row6_col4\" class=\"data row6 col4\" >15</td>\n",
       "      <td id=\"T_2c3ff_row6_col5\" class=\"data row6 col5\" >28</td>\n",
       "      <td id=\"T_2c3ff_row6_col6\" class=\"data row6 col6\" >91</td>\n",
       "      <td id=\"T_2c3ff_row6_col7\" class=\"data row6 col7\" >167</td>\n",
       "      <td id=\"T_2c3ff_row6_col8\" class=\"data row6 col8\" >240</td>\n",
       "      <td id=\"T_2c3ff_row6_col9\" class=\"data row6 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row6_col10\" class=\"data row6 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row7\" class=\"row_heading level0 row7\" >Sample Size >= 10</th>\n",
       "      <td id=\"T_2c3ff_row7_col0\" class=\"data row7 col0\" >0.7887</td>\n",
       "      <td id=\"T_2c3ff_row7_col1\" class=\"data row7 col1\" >0.6512</td>\n",
       "      <td id=\"T_2c3ff_row7_col2\" class=\"data row7 col2\" >0.713400</td>\n",
       "      <td id=\"T_2c3ff_row7_col3\" class=\"data row7 col3\" >56</td>\n",
       "      <td id=\"T_2c3ff_row7_col4\" class=\"data row7 col4\" >15</td>\n",
       "      <td id=\"T_2c3ff_row7_col5\" class=\"data row7 col5\" >30</td>\n",
       "      <td id=\"T_2c3ff_row7_col6\" class=\"data row7 col6\" >91</td>\n",
       "      <td id=\"T_2c3ff_row7_col7\" class=\"data row7 col7\" >161</td>\n",
       "      <td id=\"T_2c3ff_row7_col8\" class=\"data row7 col8\" >232</td>\n",
       "      <td id=\"T_2c3ff_row7_col9\" class=\"data row7 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row7_col10\" class=\"data row7 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row8\" class=\"row_heading level0 row8\" >Sample Size >= 15</th>\n",
       "      <td id=\"T_2c3ff_row8_col0\" class=\"data row8 col0\" >0.7857</td>\n",
       "      <td id=\"T_2c3ff_row8_col1\" class=\"data row8 col1\" >0.6395</td>\n",
       "      <td id=\"T_2c3ff_row8_col2\" class=\"data row8 col2\" >0.705100</td>\n",
       "      <td id=\"T_2c3ff_row8_col3\" class=\"data row8 col3\" >55</td>\n",
       "      <td id=\"T_2c3ff_row8_col4\" class=\"data row8 col4\" >15</td>\n",
       "      <td id=\"T_2c3ff_row8_col5\" class=\"data row8 col5\" >31</td>\n",
       "      <td id=\"T_2c3ff_row8_col6\" class=\"data row8 col6\" >91</td>\n",
       "      <td id=\"T_2c3ff_row8_col7\" class=\"data row8 col7\" >152</td>\n",
       "      <td id=\"T_2c3ff_row8_col8\" class=\"data row8 col8\" >222</td>\n",
       "      <td id=\"T_2c3ff_row8_col9\" class=\"data row8 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row8_col10\" class=\"data row8 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row9\" class=\"row_heading level0 row9\" >Known Intervention</th>\n",
       "      <td id=\"T_2c3ff_row9_col0\" class=\"data row9 col0\" >0.7917</td>\n",
       "      <td id=\"T_2c3ff_row9_col1\" class=\"data row9 col1\" >0.6628</td>\n",
       "      <td id=\"T_2c3ff_row9_col2\" class=\"data row9 col2\" >0.721500</td>\n",
       "      <td id=\"T_2c3ff_row9_col3\" class=\"data row9 col3\" >57</td>\n",
       "      <td id=\"T_2c3ff_row9_col4\" class=\"data row9 col4\" >15</td>\n",
       "      <td id=\"T_2c3ff_row9_col5\" class=\"data row9 col5\" >29</td>\n",
       "      <td id=\"T_2c3ff_row9_col6\" class=\"data row9 col6\" >91</td>\n",
       "      <td id=\"T_2c3ff_row9_col7\" class=\"data row9 col7\" >136</td>\n",
       "      <td id=\"T_2c3ff_row9_col8\" class=\"data row9 col8\" >208</td>\n",
       "      <td id=\"T_2c3ff_row9_col9\" class=\"data row9 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row9_col10\" class=\"data row9 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row10\" class=\"row_heading level0 row10\" >Unknown Intervention</th>\n",
       "      <td id=\"T_2c3ff_row10_col0\" class=\"data row10 col0\" >0.8333</td>\n",
       "      <td id=\"T_2c3ff_row10_col1\" class=\"data row10 col1\" >0.2907</td>\n",
       "      <td id=\"T_2c3ff_row10_col2\" class=\"data row10 col2\" >0.431000</td>\n",
       "      <td id=\"T_2c3ff_row10_col3\" class=\"data row10 col3\" >25</td>\n",
       "      <td id=\"T_2c3ff_row10_col4\" class=\"data row10 col4\" >5</td>\n",
       "      <td id=\"T_2c3ff_row10_col5\" class=\"data row10 col5\" >61</td>\n",
       "      <td id=\"T_2c3ff_row10_col6\" class=\"data row10 col6\" >101</td>\n",
       "      <td id=\"T_2c3ff_row10_col7\" class=\"data row10 col7\" >98</td>\n",
       "      <td id=\"T_2c3ff_row10_col8\" class=\"data row10 col8\" >128</td>\n",
       "      <td id=\"T_2c3ff_row10_col9\" class=\"data row10 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row10_col10\" class=\"data row10 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row11\" class=\"row_heading level0 row11\" >K+U Intervention</th>\n",
       "      <td id=\"T_2c3ff_row11_col0\" class=\"data row11 col0\" >0.8276</td>\n",
       "      <td id=\"T_2c3ff_row11_col1\" class=\"data row11 col1\" >0.2791</td>\n",
       "      <td id=\"T_2c3ff_row11_col2\" class=\"data row11 col2\" >0.417400</td>\n",
       "      <td id=\"T_2c3ff_row11_col3\" class=\"data row11 col3\" >24</td>\n",
       "      <td id=\"T_2c3ff_row11_col4\" class=\"data row11 col4\" >5</td>\n",
       "      <td id=\"T_2c3ff_row11_col5\" class=\"data row11 col5\" >62</td>\n",
       "      <td id=\"T_2c3ff_row11_col6\" class=\"data row11 col6\" >101</td>\n",
       "      <td id=\"T_2c3ff_row11_col7\" class=\"data row11 col7\" >76</td>\n",
       "      <td id=\"T_2c3ff_row11_col8\" class=\"data row11 col8\" >105</td>\n",
       "      <td id=\"T_2c3ff_row11_col9\" class=\"data row11 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row11_col10\" class=\"data row11 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row12\" class=\"row_heading level0 row12\" >Significant Finding</th>\n",
       "      <td id=\"T_2c3ff_row12_col0\" class=\"data row12 col0\" >0.7561</td>\n",
       "      <td id=\"T_2c3ff_row12_col1\" class=\"data row12 col1\" >0.3605</td>\n",
       "      <td id=\"T_2c3ff_row12_col2\" class=\"data row12 col2\" >0.488200</td>\n",
       "      <td id=\"T_2c3ff_row12_col3\" class=\"data row12 col3\" >31</td>\n",
       "      <td id=\"T_2c3ff_row12_col4\" class=\"data row12 col4\" >10</td>\n",
       "      <td id=\"T_2c3ff_row12_col5\" class=\"data row12 col5\" >55</td>\n",
       "      <td id=\"T_2c3ff_row12_col6\" class=\"data row12 col6\" >96</td>\n",
       "      <td id=\"T_2c3ff_row12_col7\" class=\"data row12 col7\" >123</td>\n",
       "      <td id=\"T_2c3ff_row12_col8\" class=\"data row12 col8\" >164</td>\n",
       "      <td id=\"T_2c3ff_row12_col9\" class=\"data row12 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row12_col10\" class=\"data row12 col10\" >86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2c3ff_level0_row13\" class=\"row_heading level0 row13\" >No Significant Finding</th>\n",
       "      <td id=\"T_2c3ff_row13_col0\" class=\"data row13 col0\" >0.8485</td>\n",
       "      <td id=\"T_2c3ff_row13_col1\" class=\"data row13 col1\" >0.3256</td>\n",
       "      <td id=\"T_2c3ff_row13_col2\" class=\"data row13 col2\" >0.470600</td>\n",
       "      <td id=\"T_2c3ff_row13_col3\" class=\"data row13 col3\" >28</td>\n",
       "      <td id=\"T_2c3ff_row13_col4\" class=\"data row13 col4\" >5</td>\n",
       "      <td id=\"T_2c3ff_row13_col5\" class=\"data row13 col5\" >58</td>\n",
       "      <td id=\"T_2c3ff_row13_col6\" class=\"data row13 col6\" >101</td>\n",
       "      <td id=\"T_2c3ff_row13_col7\" class=\"data row13 col7\" >32</td>\n",
       "      <td id=\"T_2c3ff_row13_col8\" class=\"data row13 col8\" >65</td>\n",
       "      <td id=\"T_2c3ff_row13_col9\" class=\"data row13 col9\" >192</td>\n",
       "      <td id=\"T_2c3ff_row13_col10\" class=\"data row13 col10\" >86</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f12b7099db0>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "highlight_metrics(\n",
    "    metrics_to_df(metrics_w_manual[\"All\"], drop_zero_tp=False).head(\n",
    "        len(FILTERS_RETRIEVED)\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RCTs only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draw_venn_rsi(\n",
    "    title=\"RCTs only\",\n",
    "    ids_retrieved=metrics_w_manual[\"Only RCTs\"][\"Is RCT\"][\"ids_retrieved\"],\n",
    "    ids_screened=metrics_w_manual[\"Only RCTs\"][\"Is RCT\"][\"ids_screened\"],\n",
    "    ids_screened_relevant=metrics_w_manual[\"Only RCTs\"][\"Is RCT\"][\n",
    "        \"ids_screened_relevant\"\n",
    "    ],\n",
    "    add_pn_labels=True,\n",
    ");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_84041_row0_col1, #T_84041_row0_col3, #T_84041_row0_col5, #T_84041_row1_col1, #T_84041_row1_col3, #T_84041_row1_col5, #T_84041_row2_col1, #T_84041_row2_col3, #T_84041_row2_col5, #T_84041_row3_col2, #T_84041_row5_col0, #T_84041_row5_col4, #T_84041_row6_col1, #T_84041_row6_col3, #T_84041_row6_col5, #T_84041_row7_col1, #T_84041_row7_col3, #T_84041_row7_col5, #T_84041_row8_col1, #T_84041_row8_col3, #T_84041_row8_col5 {\n",
       "  font-weight: bold;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_84041\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_84041_level0_col0\" class=\"col_heading level0 col0\" >Precision</th>\n",
       "      <th id=\"T_84041_level0_col1\" class=\"col_heading level0 col1\" >Recall</th>\n",
       "      <th id=\"T_84041_level0_col2\" class=\"col_heading level0 col2\" >$F_1$</th>\n",
       "      <th id=\"T_84041_level0_col3\" class=\"col_heading level0 col3\" >TP</th>\n",
       "      <th id=\"T_84041_level0_col4\" class=\"col_heading level0 col4\" >FP</th>\n",
       "      <th id=\"T_84041_level0_col5\" class=\"col_heading level0 col5\" >FN</th>\n",
       "      <th id=\"T_84041_level0_col6\" class=\"col_heading level0 col6\" >TN</th>\n",
       "      <th id=\"T_84041_level0_col7\" class=\"col_heading level0 col7\" >?</th>\n",
       "      <th id=\"T_84041_level0_col8\" class=\"col_heading level0 col8\" >Retrieved</th>\n",
       "      <th id=\"T_84041_level0_col9\" class=\"col_heading level0 col9\" >Screened</th>\n",
       "      <th id=\"T_84041_level0_col10\" class=\"col_heading level0 col10\" >Relevant</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row0\" class=\"row_heading level0 row0\" >All</th>\n",
       "      <td id=\"T_84041_row0_col0\" class=\"data row0 col0\" >0.7273</td>\n",
       "      <td id=\"T_84041_row0_col1\" class=\"data row0 col1\" >1.0000</td>\n",
       "      <td id=\"T_84041_row0_col2\" class=\"data row0 col2\" >0.842100</td>\n",
       "      <td id=\"T_84041_row0_col3\" class=\"data row0 col3\" >40</td>\n",
       "      <td id=\"T_84041_row0_col4\" class=\"data row0 col4\" >15</td>\n",
       "      <td id=\"T_84041_row0_col5\" class=\"data row0 col5\" >0</td>\n",
       "      <td id=\"T_84041_row0_col6\" class=\"data row0 col6\" >20</td>\n",
       "      <td id=\"T_84041_row0_col7\" class=\"data row0 col7\" >196</td>\n",
       "      <td id=\"T_84041_row0_col8\" class=\"data row0 col8\" >251</td>\n",
       "      <td id=\"T_84041_row0_col9\" class=\"data row0 col9\" >75</td>\n",
       "      <td id=\"T_84041_row0_col10\" class=\"data row0 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row1\" class=\"row_heading level0 row1\" >Is RCT</th>\n",
       "      <td id=\"T_84041_row1_col0\" class=\"data row1 col0\" >0.7273</td>\n",
       "      <td id=\"T_84041_row1_col1\" class=\"data row1 col1\" >1.0000</td>\n",
       "      <td id=\"T_84041_row1_col2\" class=\"data row1 col2\" >0.842100</td>\n",
       "      <td id=\"T_84041_row1_col3\" class=\"data row1 col3\" >40</td>\n",
       "      <td id=\"T_84041_row1_col4\" class=\"data row1 col4\" >15</td>\n",
       "      <td id=\"T_84041_row1_col5\" class=\"data row1 col5\" >0</td>\n",
       "      <td id=\"T_84041_row1_col6\" class=\"data row1 col6\" >20</td>\n",
       "      <td id=\"T_84041_row1_col7\" class=\"data row1 col7\" >32</td>\n",
       "      <td id=\"T_84041_row1_col8\" class=\"data row1 col8\" >87</td>\n",
       "      <td id=\"T_84041_row1_col9\" class=\"data row1 col9\" >75</td>\n",
       "      <td id=\"T_84041_row1_col10\" class=\"data row1 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row2\" class=\"row_heading level0 row2\" >Non-empty Abstract</th>\n",
       "      <td id=\"T_84041_row2_col0\" class=\"data row2 col0\" >0.7273</td>\n",
       "      <td id=\"T_84041_row2_col1\" class=\"data row2 col1\" >1.0000</td>\n",
       "      <td id=\"T_84041_row2_col2\" class=\"data row2 col2\" >0.842100</td>\n",
       "      <td id=\"T_84041_row2_col3\" class=\"data row2 col3\" >40</td>\n",
       "      <td id=\"T_84041_row2_col4\" class=\"data row2 col4\" >15</td>\n",
       "      <td id=\"T_84041_row2_col5\" class=\"data row2 col5\" >0</td>\n",
       "      <td id=\"T_84041_row2_col6\" class=\"data row2 col6\" >20</td>\n",
       "      <td id=\"T_84041_row2_col7\" class=\"data row2 col7\" >174</td>\n",
       "      <td id=\"T_84041_row2_col8\" class=\"data row2 col8\" >229</td>\n",
       "      <td id=\"T_84041_row2_col9\" class=\"data row2 col9\" >75</td>\n",
       "      <td id=\"T_84041_row2_col10\" class=\"data row2 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row3\" class=\"row_heading level0 row3\" >Non-pediatric population</th>\n",
       "      <td id=\"T_84041_row3_col0\" class=\"data row3 col0\" >0.7600</td>\n",
       "      <td id=\"T_84041_row3_col1\" class=\"data row3 col1\" >0.9500</td>\n",
       "      <td id=\"T_84041_row3_col2\" class=\"data row3 col2\" >0.844400</td>\n",
       "      <td id=\"T_84041_row3_col3\" class=\"data row3 col3\" >38</td>\n",
       "      <td id=\"T_84041_row3_col4\" class=\"data row3 col4\" >12</td>\n",
       "      <td id=\"T_84041_row3_col5\" class=\"data row3 col5\" >2</td>\n",
       "      <td id=\"T_84041_row3_col6\" class=\"data row3 col6\" >23</td>\n",
       "      <td id=\"T_84041_row3_col7\" class=\"data row3 col7\" >173</td>\n",
       "      <td id=\"T_84041_row3_col8\" class=\"data row3 col8\" >223</td>\n",
       "      <td id=\"T_84041_row3_col9\" class=\"data row3 col9\" >75</td>\n",
       "      <td id=\"T_84041_row3_col10\" class=\"data row3 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row4\" class=\"row_heading level0 row4\" >Phase >= 2</th>\n",
       "      <td id=\"T_84041_row4_col0\" class=\"data row4 col0\" >0.7111</td>\n",
       "      <td id=\"T_84041_row4_col1\" class=\"data row4 col1\" >0.8000</td>\n",
       "      <td id=\"T_84041_row4_col2\" class=\"data row4 col2\" >0.752900</td>\n",
       "      <td id=\"T_84041_row4_col3\" class=\"data row4 col3\" >32</td>\n",
       "      <td id=\"T_84041_row4_col4\" class=\"data row4 col4\" >13</td>\n",
       "      <td id=\"T_84041_row4_col5\" class=\"data row4 col5\" >8</td>\n",
       "      <td id=\"T_84041_row4_col6\" class=\"data row4 col6\" >22</td>\n",
       "      <td id=\"T_84041_row4_col7\" class=\"data row4 col7\" >121</td>\n",
       "      <td id=\"T_84041_row4_col8\" class=\"data row4 col8\" >166</td>\n",
       "      <td id=\"T_84041_row4_col9\" class=\"data row4 col9\" >75</td>\n",
       "      <td id=\"T_84041_row4_col10\" class=\"data row4 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row5\" class=\"row_heading level0 row5\" >Phase >= 3</th>\n",
       "      <td id=\"T_84041_row5_col0\" class=\"data row5 col0\" >0.8571</td>\n",
       "      <td id=\"T_84041_row5_col1\" class=\"data row5 col1\" >0.6000</td>\n",
       "      <td id=\"T_84041_row5_col2\" class=\"data row5 col2\" >0.705900</td>\n",
       "      <td id=\"T_84041_row5_col3\" class=\"data row5 col3\" >24</td>\n",
       "      <td id=\"T_84041_row5_col4\" class=\"data row5 col4\" >4</td>\n",
       "      <td id=\"T_84041_row5_col5\" class=\"data row5 col5\" >16</td>\n",
       "      <td id=\"T_84041_row5_col6\" class=\"data row5 col6\" >31</td>\n",
       "      <td id=\"T_84041_row5_col7\" class=\"data row5 col7\" >18</td>\n",
       "      <td id=\"T_84041_row5_col8\" class=\"data row5 col8\" >46</td>\n",
       "      <td id=\"T_84041_row5_col9\" class=\"data row5 col9\" >75</td>\n",
       "      <td id=\"T_84041_row5_col10\" class=\"data row5 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row6\" class=\"row_heading level0 row6\" >Sample Size >= 5</th>\n",
       "      <td id=\"T_84041_row6_col0\" class=\"data row6 col0\" >0.7273</td>\n",
       "      <td id=\"T_84041_row6_col1\" class=\"data row6 col1\" >1.0000</td>\n",
       "      <td id=\"T_84041_row6_col2\" class=\"data row6 col2\" >0.842100</td>\n",
       "      <td id=\"T_84041_row6_col3\" class=\"data row6 col3\" >40</td>\n",
       "      <td id=\"T_84041_row6_col4\" class=\"data row6 col4\" >15</td>\n",
       "      <td id=\"T_84041_row6_col5\" class=\"data row6 col5\" >0</td>\n",
       "      <td id=\"T_84041_row6_col6\" class=\"data row6 col6\" >20</td>\n",
       "      <td id=\"T_84041_row6_col7\" class=\"data row6 col7\" >185</td>\n",
       "      <td id=\"T_84041_row6_col8\" class=\"data row6 col8\" >240</td>\n",
       "      <td id=\"T_84041_row6_col9\" class=\"data row6 col9\" >75</td>\n",
       "      <td id=\"T_84041_row6_col10\" class=\"data row6 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row7\" class=\"row_heading level0 row7\" >Sample Size >= 10</th>\n",
       "      <td id=\"T_84041_row7_col0\" class=\"data row7 col0\" >0.7273</td>\n",
       "      <td id=\"T_84041_row7_col1\" class=\"data row7 col1\" >1.0000</td>\n",
       "      <td id=\"T_84041_row7_col2\" class=\"data row7 col2\" >0.842100</td>\n",
       "      <td id=\"T_84041_row7_col3\" class=\"data row7 col3\" >40</td>\n",
       "      <td id=\"T_84041_row7_col4\" class=\"data row7 col4\" >15</td>\n",
       "      <td id=\"T_84041_row7_col5\" class=\"data row7 col5\" >0</td>\n",
       "      <td id=\"T_84041_row7_col6\" class=\"data row7 col6\" >20</td>\n",
       "      <td id=\"T_84041_row7_col7\" class=\"data row7 col7\" >177</td>\n",
       "      <td id=\"T_84041_row7_col8\" class=\"data row7 col8\" >232</td>\n",
       "      <td id=\"T_84041_row7_col9\" class=\"data row7 col9\" >75</td>\n",
       "      <td id=\"T_84041_row7_col10\" class=\"data row7 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row8\" class=\"row_heading level0 row8\" >Sample Size >= 15</th>\n",
       "      <td id=\"T_84041_row8_col0\" class=\"data row8 col0\" >0.7273</td>\n",
       "      <td id=\"T_84041_row8_col1\" class=\"data row8 col1\" >1.0000</td>\n",
       "      <td id=\"T_84041_row8_col2\" class=\"data row8 col2\" >0.842100</td>\n",
       "      <td id=\"T_84041_row8_col3\" class=\"data row8 col3\" >40</td>\n",
       "      <td id=\"T_84041_row8_col4\" class=\"data row8 col4\" >15</td>\n",
       "      <td id=\"T_84041_row8_col5\" class=\"data row8 col5\" >0</td>\n",
       "      <td id=\"T_84041_row8_col6\" class=\"data row8 col6\" >20</td>\n",
       "      <td id=\"T_84041_row8_col7\" class=\"data row8 col7\" >167</td>\n",
       "      <td id=\"T_84041_row8_col8\" class=\"data row8 col8\" >222</td>\n",
       "      <td id=\"T_84041_row8_col9\" class=\"data row8 col9\" >75</td>\n",
       "      <td id=\"T_84041_row8_col10\" class=\"data row8 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row9\" class=\"row_heading level0 row9\" >Known Intervention</th>\n",
       "      <td id=\"T_84041_row9_col0\" class=\"data row9 col0\" >0.7222</td>\n",
       "      <td id=\"T_84041_row9_col1\" class=\"data row9 col1\" >0.9750</td>\n",
       "      <td id=\"T_84041_row9_col2\" class=\"data row9 col2\" >0.829800</td>\n",
       "      <td id=\"T_84041_row9_col3\" class=\"data row9 col3\" >39</td>\n",
       "      <td id=\"T_84041_row9_col4\" class=\"data row9 col4\" >15</td>\n",
       "      <td id=\"T_84041_row9_col5\" class=\"data row9 col5\" >1</td>\n",
       "      <td id=\"T_84041_row9_col6\" class=\"data row9 col6\" >20</td>\n",
       "      <td id=\"T_84041_row9_col7\" class=\"data row9 col7\" >154</td>\n",
       "      <td id=\"T_84041_row9_col8\" class=\"data row9 col8\" >208</td>\n",
       "      <td id=\"T_84041_row9_col9\" class=\"data row9 col9\" >75</td>\n",
       "      <td id=\"T_84041_row9_col10\" class=\"data row9 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row10\" class=\"row_heading level0 row10\" >Unknown Intervention</th>\n",
       "      <td id=\"T_84041_row10_col0\" class=\"data row10 col0\" >0.7619</td>\n",
       "      <td id=\"T_84041_row10_col1\" class=\"data row10 col1\" >0.4000</td>\n",
       "      <td id=\"T_84041_row10_col2\" class=\"data row10 col2\" >0.524600</td>\n",
       "      <td id=\"T_84041_row10_col3\" class=\"data row10 col3\" >16</td>\n",
       "      <td id=\"T_84041_row10_col4\" class=\"data row10 col4\" >5</td>\n",
       "      <td id=\"T_84041_row10_col5\" class=\"data row10 col5\" >24</td>\n",
       "      <td id=\"T_84041_row10_col6\" class=\"data row10 col6\" >30</td>\n",
       "      <td id=\"T_84041_row10_col7\" class=\"data row10 col7\" >107</td>\n",
       "      <td id=\"T_84041_row10_col8\" class=\"data row10 col8\" >128</td>\n",
       "      <td id=\"T_84041_row10_col9\" class=\"data row10 col9\" >75</td>\n",
       "      <td id=\"T_84041_row10_col10\" class=\"data row10 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row11\" class=\"row_heading level0 row11\" >K+U Intervention</th>\n",
       "      <td id=\"T_84041_row11_col0\" class=\"data row11 col0\" >0.7500</td>\n",
       "      <td id=\"T_84041_row11_col1\" class=\"data row11 col1\" >0.3750</td>\n",
       "      <td id=\"T_84041_row11_col2\" class=\"data row11 col2\" >0.500000</td>\n",
       "      <td id=\"T_84041_row11_col3\" class=\"data row11 col3\" >15</td>\n",
       "      <td id=\"T_84041_row11_col4\" class=\"data row11 col4\" >5</td>\n",
       "      <td id=\"T_84041_row11_col5\" class=\"data row11 col5\" >25</td>\n",
       "      <td id=\"T_84041_row11_col6\" class=\"data row11 col6\" >30</td>\n",
       "      <td id=\"T_84041_row11_col7\" class=\"data row11 col7\" >85</td>\n",
       "      <td id=\"T_84041_row11_col8\" class=\"data row11 col8\" >105</td>\n",
       "      <td id=\"T_84041_row11_col9\" class=\"data row11 col9\" >75</td>\n",
       "      <td id=\"T_84041_row11_col10\" class=\"data row11 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row12\" class=\"row_heading level0 row12\" >Significant Finding</th>\n",
       "      <td id=\"T_84041_row12_col0\" class=\"data row12 col0\" >0.6875</td>\n",
       "      <td id=\"T_84041_row12_col1\" class=\"data row12 col1\" >0.5500</td>\n",
       "      <td id=\"T_84041_row12_col2\" class=\"data row12 col2\" >0.611100</td>\n",
       "      <td id=\"T_84041_row12_col3\" class=\"data row12 col3\" >22</td>\n",
       "      <td id=\"T_84041_row12_col4\" class=\"data row12 col4\" >10</td>\n",
       "      <td id=\"T_84041_row12_col5\" class=\"data row12 col5\" >18</td>\n",
       "      <td id=\"T_84041_row12_col6\" class=\"data row12 col6\" >25</td>\n",
       "      <td id=\"T_84041_row12_col7\" class=\"data row12 col7\" >132</td>\n",
       "      <td id=\"T_84041_row12_col8\" class=\"data row12 col8\" >164</td>\n",
       "      <td id=\"T_84041_row12_col9\" class=\"data row12 col9\" >75</td>\n",
       "      <td id=\"T_84041_row12_col10\" class=\"data row12 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_84041_level0_row13\" class=\"row_heading level0 row13\" >No Significant Finding</th>\n",
       "      <td id=\"T_84041_row13_col0\" class=\"data row13 col0\" >0.7826</td>\n",
       "      <td id=\"T_84041_row13_col1\" class=\"data row13 col1\" >0.4500</td>\n",
       "      <td id=\"T_84041_row13_col2\" class=\"data row13 col2\" >0.571400</td>\n",
       "      <td id=\"T_84041_row13_col3\" class=\"data row13 col3\" >18</td>\n",
       "      <td id=\"T_84041_row13_col4\" class=\"data row13 col4\" >5</td>\n",
       "      <td id=\"T_84041_row13_col5\" class=\"data row13 col5\" >22</td>\n",
       "      <td id=\"T_84041_row13_col6\" class=\"data row13 col6\" >30</td>\n",
       "      <td id=\"T_84041_row13_col7\" class=\"data row13 col7\" >42</td>\n",
       "      <td id=\"T_84041_row13_col8\" class=\"data row13 col8\" >65</td>\n",
       "      <td id=\"T_84041_row13_col9\" class=\"data row13 col9\" >75</td>\n",
       "      <td id=\"T_84041_row13_col10\" class=\"data row13 col10\" >40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f12b70a3d90>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "highlight_metrics(\n",
    "    metrics_to_df(metrics_w_manual[\"Only RCTs\"], drop_zero_tp=False).head(\n",
    "        len(FILTERS_RETRIEVED)\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Final Results (for RCTs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from evaluation.metrics import build_filter_combinations_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Queries for combinations table\n",
    "Q_RCTS = \"is_rct_pt == True & abstract.str.len() > 1\"\n",
    "Q_RCTS_P2 = Q_RCTS + \" & phase_int >= 2\"\n",
    "Q_RCTS_P3 = Q_RCTS + \" & phase_int >= 3\"\n",
    "Q_RCTS_P3_NC = Q_RCTS_P3 + \" & has_pediatric_population == False\"\n",
    "Q_SIG = \"(has_significant_finding == True)\"\n",
    "Q_INSIQ = \"(has_significant_finding == False)\"\n",
    "Q_KNOWN = \"(has_known_intervention == True)\"\n",
    "Q_UNKNOWN = \"(has_unknown_intervention == True)\"\n",
    "Q_REC = \"(has_recommended_intervention == True)\"\n",
    "Q_UNREC = \"(has_not_recommended_intervention == True)\"\n",
    "\n",
    "# Filters for combinations table\n",
    "FILTERS_COMBINATIONS = {\n",
    "    \"RCTs\": Q_RCTS,\n",
    "    \"Phase >= 2\": Q_RCTS_P2,\n",
    "    \"Phase >= 3\": Q_RCTS_P3,\n",
    "    \"Exclude Children\": Q_RCTS_P3_NC,\n",
    "    \n",
    "    \"Significant\" : Q_RCTS_P3_NC + \"&\" + Q_SIG,\n",
    "    \n",
    "    \"R\" :  Q_RCTS_P3_NC + \"&\" + Q_REC + \"&\" + Q_SIG,\n",
    "    \"NR\" :  Q_RCTS_P3_NC + \"&\" + Q_UNREC + \"&\" + Q_SIG,\n",
    "}\n",
    "\n",
    "# Hierarchy for combinations table\n",
    "FILTERS_HIERARCHY = {\n",
    "    \"RCTs\": {\n",
    "        \"Phase >= 2\": {\n",
    "            \"Phase >= 3\": {\n",
    "                \"Exclude Children\": \n",
    "                    {\n",
    "                        \"Significant\" : [\"R\", \"NR\" ],\n",
    "                    }\n",
    "                }\n",
    "            }\n",
    "        }\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Without manual review"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'All'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3cb9d2c39f5c4e04b206340794054fa2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'Only RCTs'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2e87676425e940cabe7f5cc4d5bf939a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_aee67_row0_col1, #T_aee67_row0_col2, #T_aee67_row0_col3, #T_aee67_row0_col5, #T_aee67_row2_col0, #T_aee67_row2_col4, #T_aee67_row3_col0, #T_aee67_row3_col4, #T_aee67_row4_col0, #T_aee67_row4_col4, #T_aee67_row5_col0, #T_aee67_row5_col4, #T_aee67_row6_col0, #T_aee67_row6_col4 {\n",
       "  font-weight: bold;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_aee67\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_aee67_level0_col0\" class=\"col_heading level0 col0\" >Prec.</th>\n",
       "      <th id=\"T_aee67_level0_col1\" class=\"col_heading level0 col1\" >Rec.</th>\n",
       "      <th id=\"T_aee67_level0_col2\" class=\"col_heading level0 col2\" >$F_1$</th>\n",
       "      <th id=\"T_aee67_level0_col3\" class=\"col_heading level0 col3\" >TP</th>\n",
       "      <th id=\"T_aee67_level0_col4\" class=\"col_heading level0 col4\" >FP</th>\n",
       "      <th id=\"T_aee67_level0_col5\" class=\"col_heading level0 col5\" >FN</th>\n",
       "      <th id=\"T_aee67_level0_col6\" class=\"col_heading level0 col6\" >TN</th>\n",
       "      <th id=\"T_aee67_level0_col7\" class=\"col_heading level0 col7\" >?</th>\n",
       "      <th id=\"T_aee67_level0_col8\" class=\"col_heading level0 col8\" >Retr.</th>\n",
       "      <th id=\"T_aee67_level0_col9\" class=\"col_heading level0 col9\" >Scr.</th>\n",
       "      <th id=\"T_aee67_level0_col10\" class=\"col_heading level0 col10\" >Rel.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_aee67_level0_row0\" class=\"row_heading level0 row0\" rowspan=\"7\">RCTs</th>\n",
       "      <th id=\"T_aee67_level1_row0\" class=\"row_heading level1 row0\" ></th>\n",
       "      <th id=\"T_aee67_level2_row0\" class=\"row_heading level2 row0\" ></th>\n",
       "      <th id=\"T_aee67_level3_row0\" class=\"row_heading level3 row0\" ></th>\n",
       "      <th id=\"T_aee67_level4_row0\" class=\"row_heading level4 row0\" ></th>\n",
       "      <th id=\"T_aee67_level5_row0\" class=\"row_heading level5 row0\" ></th>\n",
       "      <td id=\"T_aee67_row0_col0\" class=\"data row0 col0\" >0.862100</td>\n",
       "      <td id=\"T_aee67_row0_col1\" class=\"data row0 col1\" >1.000000</td>\n",
       "      <td id=\"T_aee67_row0_col2\" class=\"data row0 col2\" >0.925900</td>\n",
       "      <td id=\"T_aee67_row0_col3\" class=\"data row0 col3\" >25</td>\n",
       "      <td id=\"T_aee67_row0_col4\" class=\"data row0 col4\" >4</td>\n",
       "      <td id=\"T_aee67_row0_col5\" class=\"data row0 col5\" >0</td>\n",
       "      <td id=\"T_aee67_row0_col6\" class=\"data row0 col6\" >20</td>\n",
       "      <td id=\"T_aee67_row0_col7\" class=\"data row0 col7\" >51</td>\n",
       "      <td id=\"T_aee67_row0_col8\" class=\"data row0 col8\" >80</td>\n",
       "      <td id=\"T_aee67_row0_col9\" class=\"data row0 col9\" >49</td>\n",
       "      <td id=\"T_aee67_row0_col10\" class=\"data row0 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_aee67_level1_row1\" class=\"row_heading level1 row1\" rowspan=\"6\">Phase >= 2</th>\n",
       "      <th id=\"T_aee67_level2_row1\" class=\"row_heading level2 row1\" ></th>\n",
       "      <th id=\"T_aee67_level3_row1\" class=\"row_heading level3 row1\" ></th>\n",
       "      <th id=\"T_aee67_level4_row1\" class=\"row_heading level4 row1\" ></th>\n",
       "      <th id=\"T_aee67_level5_row1\" class=\"row_heading level5 row1\" ></th>\n",
       "      <td id=\"T_aee67_row1_col0\" class=\"data row1 col0\" >0.904800</td>\n",
       "      <td id=\"T_aee67_row1_col1\" class=\"data row1 col1\" >0.760000</td>\n",
       "      <td id=\"T_aee67_row1_col2\" class=\"data row1 col2\" >0.826100</td>\n",
       "      <td id=\"T_aee67_row1_col3\" class=\"data row1 col3\" >19</td>\n",
       "      <td id=\"T_aee67_row1_col4\" class=\"data row1 col4\" >2</td>\n",
       "      <td id=\"T_aee67_row1_col5\" class=\"data row1 col5\" >6</td>\n",
       "      <td id=\"T_aee67_row1_col6\" class=\"data row1 col6\" >22</td>\n",
       "      <td id=\"T_aee67_row1_col7\" class=\"data row1 col7\" >24</td>\n",
       "      <td id=\"T_aee67_row1_col8\" class=\"data row1 col8\" >45</td>\n",
       "      <td id=\"T_aee67_row1_col9\" class=\"data row1 col9\" >49</td>\n",
       "      <td id=\"T_aee67_row1_col10\" class=\"data row1 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_aee67_level2_row2\" class=\"row_heading level2 row2\" rowspan=\"5\">Phase >= 3</th>\n",
       "      <th id=\"T_aee67_level3_row2\" class=\"row_heading level3 row2\" ></th>\n",
       "      <th id=\"T_aee67_level4_row2\" class=\"row_heading level4 row2\" ></th>\n",
       "      <th id=\"T_aee67_level5_row2\" class=\"row_heading level5 row2\" ></th>\n",
       "      <td id=\"T_aee67_row2_col0\" class=\"data row2 col0\" >1.000000</td>\n",
       "      <td id=\"T_aee67_row2_col1\" class=\"data row2 col1\" >0.520000</td>\n",
       "      <td id=\"T_aee67_row2_col2\" class=\"data row2 col2\" >0.684200</td>\n",
       "      <td id=\"T_aee67_row2_col3\" class=\"data row2 col3\" >13</td>\n",
       "      <td id=\"T_aee67_row2_col4\" class=\"data row2 col4\" >0</td>\n",
       "      <td id=\"T_aee67_row2_col5\" class=\"data row2 col5\" >12</td>\n",
       "      <td id=\"T_aee67_row2_col6\" class=\"data row2 col6\" >24</td>\n",
       "      <td id=\"T_aee67_row2_col7\" class=\"data row2 col7\" >15</td>\n",
       "      <td id=\"T_aee67_row2_col8\" class=\"data row2 col8\" >28</td>\n",
       "      <td id=\"T_aee67_row2_col9\" class=\"data row2 col9\" >49</td>\n",
       "      <td id=\"T_aee67_row2_col10\" class=\"data row2 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_aee67_level3_row3\" class=\"row_heading level3 row3\" rowspan=\"4\">Exclude Children</th>\n",
       "      <th id=\"T_aee67_level4_row3\" class=\"row_heading level4 row3\" ></th>\n",
       "      <th id=\"T_aee67_level5_row3\" class=\"row_heading level5 row3\" ></th>\n",
       "      <td id=\"T_aee67_row3_col0\" class=\"data row3 col0\" >1.000000</td>\n",
       "      <td id=\"T_aee67_row3_col1\" class=\"data row3 col1\" >0.520000</td>\n",
       "      <td id=\"T_aee67_row3_col2\" class=\"data row3 col2\" >0.684200</td>\n",
       "      <td id=\"T_aee67_row3_col3\" class=\"data row3 col3\" >13</td>\n",
       "      <td id=\"T_aee67_row3_col4\" class=\"data row3 col4\" >0</td>\n",
       "      <td id=\"T_aee67_row3_col5\" class=\"data row3 col5\" >12</td>\n",
       "      <td id=\"T_aee67_row3_col6\" class=\"data row3 col6\" >24</td>\n",
       "      <td id=\"T_aee67_row3_col7\" class=\"data row3 col7\" >12</td>\n",
       "      <td id=\"T_aee67_row3_col8\" class=\"data row3 col8\" >25</td>\n",
       "      <td id=\"T_aee67_row3_col9\" class=\"data row3 col9\" >49</td>\n",
       "      <td id=\"T_aee67_row3_col10\" class=\"data row3 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_aee67_level4_row4\" class=\"row_heading level4 row4\" rowspan=\"3\">Significant</th>\n",
       "      <th id=\"T_aee67_level5_row4\" class=\"row_heading level5 row4\" ></th>\n",
       "      <td id=\"T_aee67_row4_col0\" class=\"data row4 col0\" >1.000000</td>\n",
       "      <td id=\"T_aee67_row4_col1\" class=\"data row4 col1\" >0.280000</td>\n",
       "      <td id=\"T_aee67_row4_col2\" class=\"data row4 col2\" >0.437500</td>\n",
       "      <td id=\"T_aee67_row4_col3\" class=\"data row4 col3\" >7</td>\n",
       "      <td id=\"T_aee67_row4_col4\" class=\"data row4 col4\" >0</td>\n",
       "      <td id=\"T_aee67_row4_col5\" class=\"data row4 col5\" >18</td>\n",
       "      <td id=\"T_aee67_row4_col6\" class=\"data row4 col6\" >24</td>\n",
       "      <td id=\"T_aee67_row4_col7\" class=\"data row4 col7\" >3</td>\n",
       "      <td id=\"T_aee67_row4_col8\" class=\"data row4 col8\" >10</td>\n",
       "      <td id=\"T_aee67_row4_col9\" class=\"data row4 col9\" >49</td>\n",
       "      <td id=\"T_aee67_row4_col10\" class=\"data row4 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_aee67_level5_row5\" class=\"row_heading level5 row5\" >R</th>\n",
       "      <td id=\"T_aee67_row5_col0\" class=\"data row5 col0\" >1.000000</td>\n",
       "      <td id=\"T_aee67_row5_col1\" class=\"data row5 col1\" >0.280000</td>\n",
       "      <td id=\"T_aee67_row5_col2\" class=\"data row5 col2\" >0.437500</td>\n",
       "      <td id=\"T_aee67_row5_col3\" class=\"data row5 col3\" >7</td>\n",
       "      <td id=\"T_aee67_row5_col4\" class=\"data row5 col4\" >0</td>\n",
       "      <td id=\"T_aee67_row5_col5\" class=\"data row5 col5\" >18</td>\n",
       "      <td id=\"T_aee67_row5_col6\" class=\"data row5 col6\" >24</td>\n",
       "      <td id=\"T_aee67_row5_col7\" class=\"data row5 col7\" >3</td>\n",
       "      <td id=\"T_aee67_row5_col8\" class=\"data row5 col8\" >10</td>\n",
       "      <td id=\"T_aee67_row5_col9\" class=\"data row5 col9\" >49</td>\n",
       "      <td id=\"T_aee67_row5_col10\" class=\"data row5 col10\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_aee67_level5_row6\" class=\"row_heading level5 row6\" >NR</th>\n",
       "      <td id=\"T_aee67_row6_col0\" class=\"data row6 col0\" >1.000000</td>\n",
       "      <td id=\"T_aee67_row6_col1\" class=\"data row6 col1\" >0.280000</td>\n",
       "      <td id=\"T_aee67_row6_col2\" class=\"data row6 col2\" >0.437500</td>\n",
       "      <td id=\"T_aee67_row6_col3\" class=\"data row6 col3\" >7</td>\n",
       "      <td id=\"T_aee67_row6_col4\" class=\"data row6 col4\" >0</td>\n",
       "      <td id=\"T_aee67_row6_col5\" class=\"data row6 col5\" >18</td>\n",
       "      <td id=\"T_aee67_row6_col6\" class=\"data row6 col6\" >24</td>\n",
       "      <td id=\"T_aee67_row6_col7\" class=\"data row6 col7\" >3</td>\n",
       "      <td id=\"T_aee67_row6_col8\" class=\"data row6 col8\" >10</td>\n",
       "      <td id=\"T_aee67_row6_col9\" class=\"data row6 col9\" >49</td>\n",
       "      <td id=\"T_aee67_row6_col10\" class=\"data row6 col10\" >25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f12b71f3970>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combinations_table, metrics_combinations_final = build_filter_combinations_table(\n",
    "    df_retrieved=df_r,\n",
    "    df_screened=df_s_with_ggponc.query(\"screening_origin != 'manual' and screening_origin != 'ggponc'\"),\n",
    "    latex_output_path=PATH_LATEX_TABLE_OUTPUT / \"final_metrics.tex\",\n",
    "    overwrite_latex_output=False,\n",
    "    use_abbreviations=True,\n",
    "    column_format=\"llllllS[table-format=1.2]S[table-format=1.2]S[table-format=1.2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]\",\n",
    "    filters_retrieved = FILTERS_COMBINATIONS,\n",
    "    filters_retrieved_hierarchy = FILTERS_HIERARCHY,\n",
    ")\n",
    "highlight_metrics(combinations_table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrrrrrr}\n",
      "\\toprule\n",
      "{} &  Retr. &  TP &   ? &  FP &  TN &  FN &  Rec. \\\\\n",
      "\\midrule\n",
      "0 &     80 &  25 &  51 &   4 &  20 &   0 &  1.00 \\\\\n",
      "1 &     45 &  19 &  24 &   2 &  22 &   6 &  0.76 \\\\\n",
      "2 &     28 &  13 &  15 &   0 &  24 &  12 &  0.52 \\\\\n",
      "3 &     25 &  13 &  12 &   0 &  24 &  12 &  0.52 \\\\\n",
      "4 &     10 &   7 &   3 &   0 &  24 &  18 &  0.28 \\\\\n",
      "5 &     10 &   7 &   3 &   0 &  24 &  18 &  0.28 \\\\\n",
      "6 &     10 &   7 &   3 &   0 &  24 &  18 &  0.28 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_3057478/569929526.py:1: FutureWarning: In future versions `DataFrame.to_latex` is expected to utilise the base implementation of `Styler.to_latex` for formatting and rendering. The arguments signature may therefore change. It is recommended instead to use `DataFrame.style.to_latex` which also contains additional functionality.\n",
      "  print(combinations_table.reset_index()[['Retr.', 'TP', '?', 'FP', 'TN', 'FN', 'Rec.']].round(2).to_latex())\n"
     ]
    }
   ],
   "source": [
    "print(combinations_table.reset_index()[['Retr.', 'TP', '?', 'FP', 'TN', 'FN', 'Rec.']].round(2).to_latex())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## After manual review"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_manual = pd.read_csv(PATH_MANUAL_REVIEW)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False non-Hodgkin Lymphome\n",
      "False Children\n",
      "False für Leitlinie nicht relevant, da cost-effectiveness Analyse\n",
      "False Children\n",
      "False irrelevant in RCT Suche: sekundäre Analyse eines RCTs, die nicht randomisiert war \n",
      "False irrelevant in RCT Suche: sekundäre Analyse eines RCTs, die nicht randomisiert war \n",
      "False non-Hodgkin Lymphome \n",
      "False Studienprotokoll, nicht relevant \n",
      "False irrelevant in RCT Suche: \"healthy controls\", deshalb Fall-Kontroll-Studie\n",
      "False irrelevant in RCT Suche: sekundäre Analyse eines RCTs (NIVAHL), die nicht randomisiert war\n",
      "False Children\n",
      "True nan\n",
      "True grundsätzlich relevant, fälschlicherweise als \"Kohortenstudie\" verstanden\n",
      "True Grundsätzlich relevant, im letzten Update zur Verfügung gestellt, von Autor*innen scheinbar nicht genutzt\n",
      "True Grundsätzlich relevant, im letzten Update zur Verfügung gestellt, von Autor*innen scheinbar nicht genutzt\n",
      "True Grundsätzlich relevant, im letzten Update zur Verfügung gestellt, von Autor*innen scheinbar nicht genutzt\n",
      "True grundsätzlich relevant, zum Zeitpunkt des Screenings kein Zugriff auf den Volltext, deshalb zunächst als Abstract ausgeschlossen\n"
     ]
    }
   ],
   "source": [
    "for _, p in df_manual[df_manual.pm_id.isin(df_r.pm_id)].sort_values('is_included').iterrows():\n",
    "    print(p.is_included, p['Ein- / Ausschlussgrund'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'All'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0b7085913d31495fbc353c9f4249ce86",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating screened document subset 'Only RCTs'\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "45b13a43dcdc44f2a55a7590c77760e1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating filter combinations:   0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_644c4_row0_col1, #T_644c4_row0_col2, #T_644c4_row0_col3, #T_644c4_row0_col5, #T_644c4_row4_col0, #T_644c4_row4_col4, #T_644c4_row5_col0, #T_644c4_row5_col4, #T_644c4_row6_col0, #T_644c4_row6_col4 {\n",
       "  font-weight: bold;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_644c4\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank\" >&nbsp;</th>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_644c4_level0_col0\" class=\"col_heading level0 col0\" >Prec.</th>\n",
       "      <th id=\"T_644c4_level0_col1\" class=\"col_heading level0 col1\" >Rec.</th>\n",
       "      <th id=\"T_644c4_level0_col2\" class=\"col_heading level0 col2\" >$F_1$</th>\n",
       "      <th id=\"T_644c4_level0_col3\" class=\"col_heading level0 col3\" >TP</th>\n",
       "      <th id=\"T_644c4_level0_col4\" class=\"col_heading level0 col4\" >FP</th>\n",
       "      <th id=\"T_644c4_level0_col5\" class=\"col_heading level0 col5\" >FN</th>\n",
       "      <th id=\"T_644c4_level0_col6\" class=\"col_heading level0 col6\" >TN</th>\n",
       "      <th id=\"T_644c4_level0_col7\" class=\"col_heading level0 col7\" >?</th>\n",
       "      <th id=\"T_644c4_level0_col8\" class=\"col_heading level0 col8\" >Retr.</th>\n",
       "      <th id=\"T_644c4_level0_col9\" class=\"col_heading level0 col9\" >Scr.</th>\n",
       "      <th id=\"T_644c4_level0_col10\" class=\"col_heading level0 col10\" >Rel.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_644c4_level0_row0\" class=\"row_heading level0 row0\" rowspan=\"7\">RCTs</th>\n",
       "      <th id=\"T_644c4_level1_row0\" class=\"row_heading level1 row0\" ></th>\n",
       "      <th id=\"T_644c4_level2_row0\" class=\"row_heading level2 row0\" ></th>\n",
       "      <th id=\"T_644c4_level3_row0\" class=\"row_heading level3 row0\" ></th>\n",
       "      <th id=\"T_644c4_level4_row0\" class=\"row_heading level4 row0\" ></th>\n",
       "      <th id=\"T_644c4_level5_row0\" class=\"row_heading level5 row0\" ></th>\n",
       "      <td id=\"T_644c4_row0_col0\" class=\"data row0 col0\" >0.727300</td>\n",
       "      <td id=\"T_644c4_row0_col1\" class=\"data row0 col1\" >1.000000</td>\n",
       "      <td id=\"T_644c4_row0_col2\" class=\"data row0 col2\" >0.842100</td>\n",
       "      <td id=\"T_644c4_row0_col3\" class=\"data row0 col3\" >40</td>\n",
       "      <td id=\"T_644c4_row0_col4\" class=\"data row0 col4\" >15</td>\n",
       "      <td id=\"T_644c4_row0_col5\" class=\"data row0 col5\" >0</td>\n",
       "      <td id=\"T_644c4_row0_col6\" class=\"data row0 col6\" >20</td>\n",
       "      <td id=\"T_644c4_row0_col7\" class=\"data row0 col7\" >25</td>\n",
       "      <td id=\"T_644c4_row0_col8\" class=\"data row0 col8\" >80</td>\n",
       "      <td id=\"T_644c4_row0_col9\" class=\"data row0 col9\" >75</td>\n",
       "      <td id=\"T_644c4_row0_col10\" class=\"data row0 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_644c4_level1_row1\" class=\"row_heading level1 row1\" rowspan=\"6\">Phase >= 2</th>\n",
       "      <th id=\"T_644c4_level2_row1\" class=\"row_heading level2 row1\" ></th>\n",
       "      <th id=\"T_644c4_level3_row1\" class=\"row_heading level3 row1\" ></th>\n",
       "      <th id=\"T_644c4_level4_row1\" class=\"row_heading level4 row1\" ></th>\n",
       "      <th id=\"T_644c4_level5_row1\" class=\"row_heading level5 row1\" ></th>\n",
       "      <td id=\"T_644c4_row1_col0\" class=\"data row1 col0\" >0.711100</td>\n",
       "      <td id=\"T_644c4_row1_col1\" class=\"data row1 col1\" >0.800000</td>\n",
       "      <td id=\"T_644c4_row1_col2\" class=\"data row1 col2\" >0.752900</td>\n",
       "      <td id=\"T_644c4_row1_col3\" class=\"data row1 col3\" >32</td>\n",
       "      <td id=\"T_644c4_row1_col4\" class=\"data row1 col4\" >13</td>\n",
       "      <td id=\"T_644c4_row1_col5\" class=\"data row1 col5\" >8</td>\n",
       "      <td id=\"T_644c4_row1_col6\" class=\"data row1 col6\" >22</td>\n",
       "      <td id=\"T_644c4_row1_col7\" class=\"data row1 col7\" >0</td>\n",
       "      <td id=\"T_644c4_row1_col8\" class=\"data row1 col8\" >45</td>\n",
       "      <td id=\"T_644c4_row1_col9\" class=\"data row1 col9\" >75</td>\n",
       "      <td id=\"T_644c4_row1_col10\" class=\"data row1 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_644c4_level2_row2\" class=\"row_heading level2 row2\" rowspan=\"5\">Phase >= 3</th>\n",
       "      <th id=\"T_644c4_level3_row2\" class=\"row_heading level3 row2\" ></th>\n",
       "      <th id=\"T_644c4_level4_row2\" class=\"row_heading level4 row2\" ></th>\n",
       "      <th id=\"T_644c4_level5_row2\" class=\"row_heading level5 row2\" ></th>\n",
       "      <td id=\"T_644c4_row2_col0\" class=\"data row2 col0\" >0.857100</td>\n",
       "      <td id=\"T_644c4_row2_col1\" class=\"data row2 col1\" >0.600000</td>\n",
       "      <td id=\"T_644c4_row2_col2\" class=\"data row2 col2\" >0.705900</td>\n",
       "      <td id=\"T_644c4_row2_col3\" class=\"data row2 col3\" >24</td>\n",
       "      <td id=\"T_644c4_row2_col4\" class=\"data row2 col4\" >4</td>\n",
       "      <td id=\"T_644c4_row2_col5\" class=\"data row2 col5\" >16</td>\n",
       "      <td id=\"T_644c4_row2_col6\" class=\"data row2 col6\" >31</td>\n",
       "      <td id=\"T_644c4_row2_col7\" class=\"data row2 col7\" >0</td>\n",
       "      <td id=\"T_644c4_row2_col8\" class=\"data row2 col8\" >28</td>\n",
       "      <td id=\"T_644c4_row2_col9\" class=\"data row2 col9\" >75</td>\n",
       "      <td id=\"T_644c4_row2_col10\" class=\"data row2 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_644c4_level3_row3\" class=\"row_heading level3 row3\" rowspan=\"4\">Exclude Children</th>\n",
       "      <th id=\"T_644c4_level4_row3\" class=\"row_heading level4 row3\" ></th>\n",
       "      <th id=\"T_644c4_level5_row3\" class=\"row_heading level5 row3\" ></th>\n",
       "      <td id=\"T_644c4_row3_col0\" class=\"data row3 col0\" >0.920000</td>\n",
       "      <td id=\"T_644c4_row3_col1\" class=\"data row3 col1\" >0.575000</td>\n",
       "      <td id=\"T_644c4_row3_col2\" class=\"data row3 col2\" >0.707700</td>\n",
       "      <td id=\"T_644c4_row3_col3\" class=\"data row3 col3\" >23</td>\n",
       "      <td id=\"T_644c4_row3_col4\" class=\"data row3 col4\" >2</td>\n",
       "      <td id=\"T_644c4_row3_col5\" class=\"data row3 col5\" >17</td>\n",
       "      <td id=\"T_644c4_row3_col6\" class=\"data row3 col6\" >33</td>\n",
       "      <td id=\"T_644c4_row3_col7\" class=\"data row3 col7\" >0</td>\n",
       "      <td id=\"T_644c4_row3_col8\" class=\"data row3 col8\" >25</td>\n",
       "      <td id=\"T_644c4_row3_col9\" class=\"data row3 col9\" >75</td>\n",
       "      <td id=\"T_644c4_row3_col10\" class=\"data row3 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_644c4_level4_row4\" class=\"row_heading level4 row4\" rowspan=\"3\">Significant</th>\n",
       "      <th id=\"T_644c4_level5_row4\" class=\"row_heading level5 row4\" ></th>\n",
       "      <td id=\"T_644c4_row4_col0\" class=\"data row4 col0\" >1.000000</td>\n",
       "      <td id=\"T_644c4_row4_col1\" class=\"data row4 col1\" >0.250000</td>\n",
       "      <td id=\"T_644c4_row4_col2\" class=\"data row4 col2\" >0.400000</td>\n",
       "      <td id=\"T_644c4_row4_col3\" class=\"data row4 col3\" >10</td>\n",
       "      <td id=\"T_644c4_row4_col4\" class=\"data row4 col4\" >0</td>\n",
       "      <td id=\"T_644c4_row4_col5\" class=\"data row4 col5\" >30</td>\n",
       "      <td id=\"T_644c4_row4_col6\" class=\"data row4 col6\" >35</td>\n",
       "      <td id=\"T_644c4_row4_col7\" class=\"data row4 col7\" >0</td>\n",
       "      <td id=\"T_644c4_row4_col8\" class=\"data row4 col8\" >10</td>\n",
       "      <td id=\"T_644c4_row4_col9\" class=\"data row4 col9\" >75</td>\n",
       "      <td id=\"T_644c4_row4_col10\" class=\"data row4 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_644c4_level5_row5\" class=\"row_heading level5 row5\" >R</th>\n",
       "      <td id=\"T_644c4_row5_col0\" class=\"data row5 col0\" >1.000000</td>\n",
       "      <td id=\"T_644c4_row5_col1\" class=\"data row5 col1\" >0.250000</td>\n",
       "      <td id=\"T_644c4_row5_col2\" class=\"data row5 col2\" >0.400000</td>\n",
       "      <td id=\"T_644c4_row5_col3\" class=\"data row5 col3\" >10</td>\n",
       "      <td id=\"T_644c4_row5_col4\" class=\"data row5 col4\" >0</td>\n",
       "      <td id=\"T_644c4_row5_col5\" class=\"data row5 col5\" >30</td>\n",
       "      <td id=\"T_644c4_row5_col6\" class=\"data row5 col6\" >35</td>\n",
       "      <td id=\"T_644c4_row5_col7\" class=\"data row5 col7\" >0</td>\n",
       "      <td id=\"T_644c4_row5_col8\" class=\"data row5 col8\" >10</td>\n",
       "      <td id=\"T_644c4_row5_col9\" class=\"data row5 col9\" >75</td>\n",
       "      <td id=\"T_644c4_row5_col10\" class=\"data row5 col10\" >40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_644c4_level5_row6\" class=\"row_heading level5 row6\" >NR</th>\n",
       "      <td id=\"T_644c4_row6_col0\" class=\"data row6 col0\" >1.000000</td>\n",
       "      <td id=\"T_644c4_row6_col1\" class=\"data row6 col1\" >0.250000</td>\n",
       "      <td id=\"T_644c4_row6_col2\" class=\"data row6 col2\" >0.400000</td>\n",
       "      <td id=\"T_644c4_row6_col3\" class=\"data row6 col3\" >10</td>\n",
       "      <td id=\"T_644c4_row6_col4\" class=\"data row6 col4\" >0</td>\n",
       "      <td id=\"T_644c4_row6_col5\" class=\"data row6 col5\" >30</td>\n",
       "      <td id=\"T_644c4_row6_col6\" class=\"data row6 col6\" >35</td>\n",
       "      <td id=\"T_644c4_row6_col7\" class=\"data row6 col7\" >0</td>\n",
       "      <td id=\"T_644c4_row6_col8\" class=\"data row6 col8\" >10</td>\n",
       "      <td id=\"T_644c4_row6_col9\" class=\"data row6 col9\" >75</td>\n",
       "      <td id=\"T_644c4_row6_col10\" class=\"data row6 col10\" >40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f12b722e290>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combinations_table, metrics_combinations_final = build_filter_combinations_table(\n",
    "    df_retrieved=df_r,\n",
    "    df_screened=df_s_with_ggponc,\n",
    "    latex_output_path=PATH_LATEX_TABLE_OUTPUT / \"final_metrics.tex\",\n",
    "    overwrite_latex_output=False,\n",
    "    use_abbreviations=True,\n",
    "    column_format=\"llllllS[table-format=1.2]S[table-format=1.2]S[table-format=1.2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]\",\n",
    "    filters_retrieved = FILTERS_COMBINATIONS,\n",
    "    filters_retrieved_hierarchy = FILTERS_HIERARCHY,\n",
    ")\n",
    "\n",
    "highlight_metrics(combinations_table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrrrrrrrr}\n",
      "\\toprule\n",
      "{} &  Retr. &  TP &   ? &  FP &  TN &  FN &  Prec. &  Rec. &  \\$F\\_1\\$ \\\\\n",
      "\\midrule\n",
      "0 &     80 &  40 &  25 &  15 &  20 &   0 &   0.73 &  1.00 &   0.84 \\\\\n",
      "1 &     45 &  32 &   0 &  13 &  22 &   8 &   0.71 &  0.80 &   0.75 \\\\\n",
      "2 &     28 &  24 &   0 &   4 &  31 &  16 &   0.86 &  0.60 &   0.71 \\\\\n",
      "3 &     25 &  23 &   0 &   2 &  33 &  17 &   0.92 &  0.57 &   0.71 \\\\\n",
      "4 &     10 &  10 &   0 &   0 &  35 &  30 &   1.00 &  0.25 &   0.40 \\\\\n",
      "5 &     10 &  10 &   0 &   0 &  35 &  30 &   1.00 &  0.25 &   0.40 \\\\\n",
      "6 &     10 &  10 &   0 &   0 &  35 &  30 &   1.00 &  0.25 &   0.40 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_3057478/1384292062.py:1: FutureWarning: In future versions `DataFrame.to_latex` is expected to utilise the base implementation of `Styler.to_latex` for formatting and rendering. The arguments signature may therefore change. It is recommended instead to use `DataFrame.style.to_latex` which also contains additional functionality.\n",
      "  print(combinations_table.reset_index()[['Retr.', 'TP', '?', 'FP', 'TN', 'FN', 'Prec.', 'Rec.', '$F_1$']].round(2).to_latex())\n"
     ]
    }
   ],
   "source": [
    "print(combinations_table.reset_index()[['Retr.', 'TP', '?', 'FP', 'TN', 'FN', 'Prec.', 'Rec.', '$F_1$']].round(2).to_latex())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Remaining False Positives"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "ids_false_positives = (\n",
    "    metrics_to_df(metrics_combinations_final[\"Only RCTs\"], drop_columns=[])\n",
    "    .sort_values([\"Recall\", \"Precision\", \"Retrieved\"], ascending=[False, False, True])\n",
    "    .iloc[0][\"Ids_fp\"]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "false_positives = df_s_with_ggponc[\n",
    "    df_s_with_ggponc[\"pm_id\"].apply(lambda pm_id: pm_id in ids_false_positives)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ex_<50%:proportion of HL not reported                                                         3\n",
       "Children                                                                                      3\n",
       "irrelevant in RCT Suche: sekundäre Analyse eines RCTs, die nicht randomisiert war             2\n",
       "ex_abstract                                                                                   1\n",
       "non-Hodgkin Lymphome                                                                          1\n",
       "irrelevant in RCT Suche: \"healthy controls\", deshalb Fall-Kontroll-Studie                     1\n",
       "Studienprotokoll, nicht relevant                                                              1\n",
       "irrelevant in RCT Suche: sekundäre Analyse eines RCTs (NIVAHL), die nicht randomisiert war    1\n",
       "non-Hodgkin Lymphome                                                                          1\n",
       "für Leitlinie nicht relevant, da cost-effectiveness Analyse                                   1\n",
       "Name: exclusion_reason, dtype: int64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import ast\n",
    "\n",
    "false_positives[\"exclusion_reason\"].apply(\n",
    "    lambda r: ast.literal_eval(r) if r.startswith(\"{\") else r\n",
    ").apply(lambda d: d[\"subgroup\"] if isinstance(d, dict) else d).value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Missing RCTs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We should be able to find all included RCTs. If not, we need to find out why we did not retrieve them:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from api.utils import query_api_for_concepts, concepts_to_df\n",
    "from api.models import ConceptsQuery\n",
    "\n",
    "missing_rcts_df = concepts_to_df(\n",
    "    query_api_for_concepts(\n",
    "        ConceptsQuery(\n",
    "            source=\"pubmed\",\n",
    "            ids=[\n",
    "                int(pm_id) for pm_id in metrics_w_manual[\"Only RCTs\"][\"All\"][\"ids_fn\"]\n",
    "            ],\n",
    "            kind=\"population\",\n",
    "        )\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like it is due to vague population concepts:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "tags": []
   },
   "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>UMLS Term</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  UMLS Term\n",
       "0      None"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from evaluation.metrics import prepare_missing_rcts_concepts_df\n",
    "\n",
    "missing_rcts_concepts_df = prepare_missing_rcts_concepts_df(\n",
    "    missing_rcts_df=missing_rcts_df\n",
    ")\n",
    "missing_rcts_concepts_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Unscreened High Quality Evidence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this section, we consider all sources. However, we exclude duplicates that are both in Pubmed and in Trialstreamer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_query.reset_index(drop=True, inplace=True)\n",
    "df_query = df_query[\n",
    "    (\n",
    "        ~df_query.sort_values([\"pm_id\", \"source\"]).duplicated(\n",
    "            subset=\"pm_id\", keep=\"first\"\n",
    "        )\n",
    "    )\n",
    "    | (df_query[\"pm_id\"].isnull())\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We then set boolean flags with which we can filter out the already screened evidence:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from evaluation.matching import check_for_matches_in_screened\n",
    "\n",
    "df_query = check_for_matches_in_screened(\n",
    "    df_query=df_query, df_screened=df_s_with_ggponc\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_query_unscreened = df_query.query(\"is_screened == False\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "tags": []
   },
   "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>has_title_match_in_screened</th>\n",
       "      <th>False</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>source</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Civic</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ClinicalTrials</th>\n",
       "      <td>136</td>\n",
       "      <td>136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pubmed</th>\n",
       "      <td>176</td>\n",
       "      <td>176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>313</td>\n",
       "      <td>313</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "has_title_match_in_screened  False  All\n",
       "source                                 \n",
       "Civic                            1    1\n",
       "ClinicalTrials                 136  136\n",
       "Pubmed                         176  176\n",
       "All                            313  313"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(\n",
    "    df_query_unscreened[\"source\"],\n",
    "    df_query_unscreened[\"has_title_match_in_screened\"],\n",
    "    margins=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "FILTER_HQ = (\n",
    "    \"is_in_ggponc == False\"\n",
    "    \"&\"\n",
    "    \"has_title_match_in_screened == False\"    \n",
    "    \"&\"\n",
    "    \"(\"\n",
    "    \"(source == 'ClinicalTrials' & phase_int >= 3 & results_available)\"\n",
    "    \"|\"\n",
    "    \"(source == 'Civic' & evidence_rating >= 4)\"\n",
    "    \"|\"\n",
    "    \"(source == 'Pubmed' & abstract.str.len() > 1 & is_rct_pt == True & phase_int >= 2)\"\n",
    "    \")\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "256 2023-01-31 00:00:00 <NA> Tac, Mini-MTX, MMF Versus Tac, MTX for GVHD Prevention\n"
     ]
    }
   ],
   "source": [
    "df_query_unscreened_hq = df_query_unscreened.query(FILTER_HQ)\n",
    "\n",
    "manual_review = df_query_unscreened_hq.sort_values(\n",
    "    [\"source\", \"publication_date\"], ascending=False\n",
    ")\n",
    "for i, r in manual_review.iterrows():\n",
    "    print(i, r.publication_date, r.pm_id, r.title)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_query_unscreened_hq.to_csv(PATH_UNSCREENED_HQ, index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Latex Output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from evaluation.metrics import remove_last_cline_from_tex_table"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Missing RCTs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    missing_rcts_concepts_df[[\"UMLS Term\"]].style.format(escape=\"latex\").to_latex(\n",
    "        PATH_LATEX_TABLE_OUTPUT / f\"missing_rcts.tex\",\n",
    "        hrules=True,\n",
    "        clines=\"skip-last;data\",\n",
    "    )\n",
    "    remove_last_cline_from_tex_table(PATH_LATEX_TABLE_OUTPUT / f\"missing_rcts.tex\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Retrieval Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    for subset_screened_desc in metrics_w_manual.keys():\n",
    "        if subset_screened_desc == \"All\":\n",
    "            column_format = \"lS[table-format=1.2]S[table-format=1.2]S[table-format=1.2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=3]S[table-format=2]S[table-format=3]\"\n",
    "        else:\n",
    "            column_format = \"lS[table-format=1.2]S[table-format=1.2]S[table-format=1.2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=2]S[table-format=3]\"\n",
    "        highlight_metrics(\n",
    "            metrics_to_df(\n",
    "                metrics_w_manual[subset_screened_desc],\n",
    "                drop_zero_tp=False,\n",
    "                use_abbreviations=True,\n",
    "            )\n",
    "            .head(len(FILTERS_RETRIEVED))\n",
    "            .drop(columns=[\"Scr.\", \"Rel.\"])\n",
    "        ).applymap_index(lambda v: \"font-weight: bold;\", axis=\"columns\").to_latex(\n",
    "            PATH_LATEX_TABLE_OUTPUT / f\"prec_rec_{subset_screened_desc}.tex\",\n",
    "            hrules=True,\n",
    "            convert_css=True,\n",
    "            siunitx=True,\n",
    "            clines=\"all;data\",\n",
    "            column_format=column_format,\n",
    "        )\n",
    "        remove_last_cline_from_tex_table(\n",
    "            PATH_LATEX_TABLE_OUTPUT / f\"prec_rec_{subset_screened_desc}.tex\"\n",
    "        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAFgCAYAAABNIYvfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABpgklEQVR4nO3dd5xU9bn48c+ZvrO9L7v0pS9SpVcFCwqiYrALahLTvMaoMYktyY3Ra34pRmO8N1GxkUSQGEABkSq9L3Xpnd1le50+5/fHwMqybbaeKc/79drXwpkz5zzb5plve76KqqoqQgghhAhqOq0DEEIIIUTrSUIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQoAkdCGEECIESEIXQgghQkCzE3r37t1RFKXWh9lspnPnzsycOZOlS5e2R5ytcjnmU6dOaR1Km1q7di2KojB58mStQxFCCKGxFrfQx40bx5w5c5gzZw633HILBoOBxYsXM2PGDH7yk5+0SXChmoiFEEKItmZo6RO//e1vM3fu3Jr/u91unnzySd58803++Mc/cu+99zJixIi2iLHVVq1ahcvlIiMjQ+tQhBBCiHbRZmPoBoOB3/3ud8TExACwZMmStrp0q2VmZtKvXz+MRqPWoQghhBDtok0nxVksFnr37g1Afn5+ncdXrVrFnXfeSadOnTCZTKSkpHDHHXewefPmWufNmzcPRVE4ffo0AD169Kg1Zr927Vqg9hhydXU1L774Iv3798dqtdK9e/ea6zXVdb9w4UJuvvlmkpOTMZlMZGRk8MADD3Dw4MFa561YsQJFUejfv3+D3wO3201aWhqKopCdnV3rMZvNxu9//3tGjx5NXFwcFouFvn378tOf/pSioqIGr/nBBx8wYsQIrFYrCQkJ3HzzzXz99dcNni+EECL8tLjLvSHl5eUApKam1jr+9NNP8/vf/x6dTse1117LhAkTOHPmDP/5z39YsmQJf/vb33j44YcB6NWrF3PmzGHhwoVUVVUxa9YsoqKiaq6VlpZW69p2u53Jkydz8OBBJk6cyODBgxtNkJe53W7uv/9+PvnkE8xmM8OHDycjI4MjR47w8ccfs2jRIhYtWsTNN98MwA033EDnzp3Jyclhy5YtjB49us41ly1bRn5+PsOGDWPw4ME1xy9cuMDNN9/Mvn37SEhIYMSIEURHR7Nr1y5+97vfsWDBAtauXUu3bt1qXe+JJ57gz3/+MzqdjvHjx5Oens7evXuZPHkyjz/+eJNfoxBCiDChNlO3bt1UQH3vvffqPHbw4EFVr9ergLp9+/aa4//3f/+nAmqvXr3U7OzsWs9Zt26dGh0drZpMJvXIkSP13uvkyZP1xrJmzRoVUAF10KBBam5ubqMxX32dX/ziFyqgjho1Sj1x4kStxxYsWKDq9Xo1Pj5eLSkpqTn+3HPPqYD62GOP1XuvO+64QwXUN954o+aY1+tVx40bpwLqo48+qpaXl9c85nK51KeeekoF1Ouuu67WtZYuXaoCamRkpLp+/fpaj/32t7+t+donTZpUbyxCCCHCR5sk9NLSUnXFihVqv379VEB9/vnnax7zeDxqenq6Cqg7duyo95qvvfaaCqhPPfVUvffyJ6FfnfCauk5RUZEaERGhWiwW9dy5c/U+7wc/+EGd5Hzs2DEVUGNjY1WbzVbr/IsXL6pGo1E1m81qUVFRzfFly5apgDpkyBDV5XLVuY/H41EHDhyoAuq+fftqjk+dOlUF1Geffbbe+IYMGSIJXQghhKqqqtriMfSHH364Zkw7Li6Om266iaNHj/LRRx/x3//93zXn7d69mwsXLpCZmcnw4cPrvdblddSbNm1qUSwpKSlMmDChWc9Zs2YNNpuNcePGNTj7vb64MjMzmThxImVlZfz73/+udf7HH3+My+Vi5syZJCQk1Bz//PPPAZg1axYGQ91RDp1Ox8SJE2vdy+12s2HDBgAeeOCBeuN76KGH/PlShRBChIEWj6GPGzeOXr16AVBQUMDXX39NRUUF3//+9+nduzcjR44E4MSJEwAcP34cRVEavWZBQUGLYrlyApy/Lse1atWqZsf1yCOPsH79et577z3uvffemuPvvfceQM1cgKvv9cILL/DCCy/4da+ioiLsdjvgmxRYn4aOCyGECD9ttg69rKyMO+64gzVr1jB79mwOHjyI1WrF6/UCvolsN910U6PXTEpKalEsERERzX7O5bh69erFuHHjGj23X79+tf7/rW99i8cff5xVq1Zx7tw5OnfuzK5du9i7dy8ZGRnceOON9d5r/PjxZGZmNnqvrKys5n4pQgghRNvNco+NjeVf//oX/fr14/Tp0/zhD3/g+eefp0uXLgAkJiYyb968trpdq12Oq2/fvs2Oy2q1Mnv2bN555x3ef/99nnvuuZprzJkzB52u9kjG5XvNnDmTp59+2q97JCYmYjabcTgcnDp1qt5ELxX0hBBCXNam69CTk5N5/vnnAfh//+//UVpayogRI0hKSuLgwYMcOHCgWdczmUyAbzy5rU2ZMgWTycTatWu5ePFis5//yCOPAPD+++/jcDiYP38+QK1ei8umTZsGwIIFC1BV1a/rGwyGmp6Djz/+uN5zPvzww+aGLYQQIkS1+W5rP/jBD+jatStlZWX8/ve/x2g08tJLL6GqKnfccUfNRK8reTweVq9ezZYtW2od79y5M0Cz3wj4IzU1lccff5yqqipmzJjBvn376pzjcDhYvHgxOTk5dR4bO3Ysffv25ejRozz77LMUFRUxfvz4msI6V5o5cyYjRoxg27ZtPPzww/XOFSgpKeHtt9+u9eblxz/+MQBvvPFGnQmDr732Grt27Wruly2EECJUNXdafGPr0C979913VUCNjo6uWb71zDPP1Cwxy8rKUmfOnKnec8896uTJk9W4uDgVUP/617/Wus6bb76pAmpUVJR65513qo8++qj66KOPqjk5OaqqfrNsrallWw0tf3O5XOp9992nAqpOp1OHDh2qzpo1S7377rvVcePGqZGRkSqgLlu2rN7rvvrqqzVfE6C+++67DcZw/vz5mmVmkZGR6tixY9V77rlHvfPOO9UhQ4bUrN+/eincD3/4w5r4Jk+erN57771qVlaWqtPp1CeeeEKWrQkhhFBVtY3WoV/N7XarAwYMUAH1Zz/7Wc3xjRs3qvfff7/arVs31Ww2q9HR0WqfPn3U22+/Xf373/+uFhcX17qOx+NRX3nlFTUrK0u1WCw1iXPNmjWqqrY+oV/2xRdfqHfeeaeakZGhGo1GNS4uTu3fv796zz33qPPnz1erqqrqfd6FCxdqEnFkZKRaUVHRaBx2u119++231euuu05NTExUDQaDmpKSog4ZMkT94Q9/qK5YsaLe57377rvq8OHDVYvFosbGxqpTp05V16xZ4/fXL4QQIvQpqurnoK4QGlFVFZvbhs1lo9pVTbWrGpv7m387PU48Xg9e1YtH9X32qt46x/SKHqPeiEFnwKgzYtQb63w26U1YjVYijZFEmiKJNEZi1MumPkKIwCcJXQQEVVUpd5RTYi+hxFZCib2EUnspVc4q7G47Ktr9mpr0JqJN0cSYY4g2+z7HWeJIjEjEbDBrFpcQQlxJErrocFXOKgqqCyi2FVNqL6XYVkyZvQyP6tE6tGaLMkWRZE0iyZpEYkQiSdYkIk2RWoclRIsdPXqU119/ndWrV3P69Gk8Hg9JSUl06tSJUaNGcd111zFr1iytwwxZkydPZt26daxZs6amWqm/JKGLdldqLyWvMo/cilzyKvOocFZoHVK7shgsJEYkkhyZTHp0OmlRaRh0bb6xoRBtbtGiRdx33304HA4SExMZNmwYycnJlJSUsGfPHnJzc0lMTKSwsFDrUENWaxK6vMqINqWqKoXVhb4EXulL4Ha3XeuwOpTdbed8xXnOV5xnT94e9Iqe1KhUMqIzyIjJINma3GS5YSE6Wn5+PnPmzMHhcPDUU0/xm9/8BovFUuucnTt3snDhQo0iFE2RhC5aze11c678HKdKT3G69DQOj0PrkAKKR/VwoeICFyousP3Cdkx6E+nR6TUJPs4Sp3WIQrB06VIqKytJT0/n//2//1fvOcOHD29wky2hvTYvLCPCg91t53DhYVYcW8EH2R/w5fEvOVJ0RJK5H5weJ6dKT7Hx7EY+OfAJCw4sYFfuLkrtpVqHJsJYfn4+4Kv42Vxut5t3332XqVOnkpSUhNlspnPnzkydOpU33nij1rlz585FURTmzZvH/v37ufvuu+nUqRN6vZ5f/vKXta7597//ncmTJ5OQkIDZbKZHjx58//vf5+zZsw3GcuHCBX7yk5/Qv39/rFYr0dHRjBgxgjfffLPeqqNXxnPy5EkefPBB0tLSMJvNZGZm8vzzz+NwNPy6tnPnTu6//366du2K2WwmISGBm266iS+++KLB55w9e5ZHHnmETp06YbFY6N27N8899xw2m62R73LTpIUu/FbprORkyUlOlZ4irzJP05nnoaTEXsKOCzvYcWEHiRGJZCZkkhmfSbQ5WuvQRBjp2rUrAPv372fVqlVMmTLFr+eVlZUxffp0NmzYgNFoZOzYsaSnp5OXl8fevXtZtWoVjz/+eJ3nbdq0ie9973t06tSJiRMnYrPZiI72/c5XVFRw2223sXbtWqKiohg+fDjJycns27ePt99+mwULFrBy5UqGDh1a65rr16/n9ttvp6SkhO7du3PDDTfgcDjYtm0bjz/+OEuWLGHp0qUYjXWXou7Zs4cnnniC+Ph4Jk2aRHFxMRs3buTll1/mwIEDdbbLBnj99df5yU9+gtfrZciQIYwaNYq8vDzWrl3Ll19+ya9+9StefPHFWs/Jyclh0qRJXLx4kU6dOnHbbbdRVVXFH//4R9asWePX97whMilONMqrejlVeoqcwhzOl5+XJN6BUiJT6Bnfk8z4TJk5L9pdZWUl/fr14/z58yiKwqRJk5gyZQrDhg1jxIgRDbbcZ82axaJFixg6dCiLFi2qtZ212+3m888/Z+bMmTXH5s6dy/vvvw/Az372M15++eU6G1rdf//9zJ8/n+nTp/POO++QkpJS89if/vQnnnzySXr37s2hQ4fQ6/UA5OXlMXDgQIqLi/nLX/7CY489VnPdoqIiZs+ezerVq+sk2Svjee655/jVr35Vc839+/czevRoqqqq2LRpE2PGjKl53ooVK5g2bRqJiYl8+umnTJw4seaxffv2ccstt3Du3DnWrl3LpEmTah4bOXIk27dvZ/bs2bz//vs18xTOnDnD9ddfz/HjxwFklrtoO6X2UnIKczhadBSbu3XdQKL1usR0ISsli66xXbUORYSww4cPM2fOHLZu3VrnsSFDhvDYY4/xne98pybhZWdnM2TIECwWC8eOHSMjI6PJe1xOoH369OHgwYM117rs0KFDZGVl0alTJ3Jycmpa7Ve69dZb+eKLL1iyZAnTp08HfG8O/ud//ocf/ehHdbr5Ac6fP0+PHj2Ii4sjPz+/ZmLq5XiGDx/O9u3b60xY/f73v8/bb7/Nr3/9a1544YWa46NHj2br1q0sXLiw3mV8CxYsYPbs2cyaNatmIuHGjRsZP348kZGRnD59msTExFrP+eyzz7jjjjuAliV0GUMXNdxeN0eLjrL48GI+OfAJe/P3SjIPEGfLz7L82HL+uf+f7M3fi9Pj1DokEYL69u3Lli1b2Lp1Ky+++CI33XRTTct8z549fP/73+fmm2/G6fT9/i1fvhzwJVh/kvmVbr/99jrJHOCLL75AVVWmTZtWbzIHahLdlZtWff755wDcfffd9T4nIyOD3r17U1BQwNGjR+s8Pn369HpXn/Tv3x/wvSG4rLCwkG3bthEREcGMGTP8jnHt2rUA3HzzzXWSOfg28oqNja33ev6QMXRBpbOSffn7OFx0WBJFgCt3lLPl3BZ2XNhBr4ReDEwZSEJEgtZhiRAzcuRIRo4cCfiWou7evZvf/e53/POf/+Srr77i9ddf55lnnuH06dMA9OvXr9n3uLJr/konTpwA4J133uGdd95p9BpX7lx5+XkTJkxo8t4FBQX06dOn1rHLcwiuFhMTA4Dd/s3y25MnT/pKUttsmM2NV4u8MsZz584B0KNHj3rPVRSF7t27k52d3eTXUB9J6GGs2FbMnrw9nCg5gVf1ah2OaAa3101OYQ45hTmkRaUxMGUgPeJ6yPp20eYURWHYsGH84x//oLq6msWLF/PZZ5/xzDPPtOq6ERER9R73en2vRUOGDGHw4MGNXmPUqFF1nnfXXXcRGdn4nJP6WsdXj+M35vK9oqKiAqpqniT0MJRXmcfu3N2cLW946YcIHnmVeeRV5hFviWd4+nB6xvfUOiQRom688UYWL15cUynucqs2Jyenze7RpUsXAMaNG8ebb77ZrOcdPXqUZ599lmuvvbbN4mnoXuB7s/Puu+/6/Wbg8rDEqVOnGjzncq9HS8gYehi5UHGBpUeWsvjwYknmIajEXsJXJ75i4cGFnCw5qXU4Isj4Mz/6zJkzAHTu3BnwjQWDb9z7woULbRLHtGnTAFi8eHGtbm5/n/fJJ5+0SRyNSU9PZ9CgQVRUVNTMI/DH5dnuy5cvp7i4uM7jixcvprS0tMVxSUIPA3mVeSw5vISlR5ZyoaJt/uhE4Cq2FbPyxEoWHVrE6dKWv9sX4eWtt95izpw5tSZxXaaqKosWLappMd9zzz2Ar1t85syZ2Gw2Zs6cWZPwL3O73SxevLhZcQwdOpRZs2Zx9uxZ7rzzznpbs1VVVXz88cc1xXAAnnnmGeLi4vjDH/7A73//+5qJe1c6efIkH330UbPiachvfvMbAB5++GGWLFlS53FVVdm6dStffvllzbEJEyYwbNgwKisr+eEPf1irYM3Zs2d5+umnWxWTdLmHsApHBVvPb+VEyQmtQxEaKKwuZMXxFSRbkxmePlyWvIlGuVwuPvjgAz744AOSk5MZOnQoSUlJlJaWcvDgwZrE+sADD/Doo4/WPO+9997jlltuYcuWLfTu3btWYZl9+/ZRUFDgV+v/Su+99x6lpaUsW7aMvn37MnjwYHr06IGqqpw6dYrs7GycTieHDh0iNTUV8PUa/Oc//2HWrFk8/fTTvPbaawwcOJBOnTpRVlbGoUOHOH78OKNGjeKBBx5o9fdrxowZvP766zz11FPcdttt9OrVi759+xIbG0tBQQHZ2dlcvHiRZ599lhtvvLHmeR9++CGTJ0/mn//8J+vXr2f8+PFUV1ezevVqBg0aRFJSEps3b25RTJLQQ5DT42RP3h725e8Lyi1JRdsqqC5g+bHlpEWlMb7r+PCbFW+3Q3U12Gy1P7tc4PGA1wtud+3PXi8oCuj1oNOBweD7MBp9n81msFp9H5GRvs8NTPIKFo8++ig9evRg1apVbN26lYMHD5Kfn4/BYCA9PZ17772Xhx56qKab/bL4+HjWrVvHu+++y/z589mzZw+bNm0iJSWFIUOGcPvttzc7lujoaL788kv+9a9/8dFHH7Fz50727NlDTEwMnTp14v777+e2224jMzOz1vMmTpzIgQMHePPNN/n888/Zvn07DoeDlJQUunbtygMPPNCmk9j+67/+i+uvv5433niDNWvWsGrVKnQ6HWlpaQwdOpRbb721zv0GDBjAjh07eOmll1i2bBmfffYZnTt35vHHH+fFF1/klltuaXE8UlgmhKiqSk5hDjsu7JD146JeCgpZKVlcm34tJr1J63DaTmUllJRAaek3H+XlvuTt7aAVHDqdL6lHRkJMDMTHf/MRE+N7gyBEO5KEHiLOl59n87nNFNvqTrQQ4moRhghGdR5Fn8Q+TZ8cSFQViorg4kXIz4fiYigr87WsA5leD7GxvuSemAipqZCS4jsuRBuRhB7kql3VbDizgVOlp7QORQShgO+Gdzh8ifvyR0GBr6s8FOh0kJQEaWm+j9TUoO+2F9qShB7EjhQdYfPZzbJlqWiVgOqGV1Vf0j5zxvdxab1z2IiNhS5doGtXSE/3JX0h/CQJPQhVOav4+szXnCk70/TJQvjJarQyuftkOsd07tgbO51w9qzv48wZ3yQ24ZuA17kzdOvmS/CXduUSoiGS0IPM4cLDbD63WWqui3aTlZzFqM6jMOjacRGM2w0nT8LRo3DhQsdNXAtWiuIbc+/ZE3r1kq55US9J6EGiylnF+tPrpcKb6BBxljiu73E9SdaktruoqsL5874kfvJk4E9kC1Q6na/l3qePr/UuE+vEJZLQg8CRoiNsOrtJWuWiQ+kUHcM7DWdI2pDWbfpSUgKHD8OxY77136LtmM2QmelL7ikpWkcjNCYJPYB5vB42nNnA4aLDWociwlhaVBrXdb+OaHP9e1M36MwZ2LfP1yoX7S8hAa65xtclL632sCQJPUCVO8pZeXwlRbYirUMRAqPOyPiu4+md2LvxE10uX2v8wAHf+nDR8SIiICsLBgyQiXRhRhJ6ADpdepo1p9ZIF7sIOFnJWYzpMgadctVyqspK2L8fcnJ8s9aF9gwGX2t90CCIi9M6GtEBJKEHEFVV2X5hO3vy9mgdihANSotKY2rPqViNVl8i37ULjhyRmeqBrHt3GDHCV6lOhCxJ6AHC5rKx6uQq2d5UBIVkYww3OzsTsSdHEnmwUBRfi334cF9teRFyJKEHgItVF/ny+JdUu2QGsAhsJnQM8FhIqbKhUyGx3ELkRdkIKKjodNCvHwwb5tslToQMSegaO1N2hq9OfIXbK2tyReDSqwr9vBGkV9vRXdUij7VbiTsnb0aDjsHgmzw3dCiYQmjnvTAmCV1DhwoOseHMBlTkRyACVxfVQp8qNwZPw286rR4Liacd6Lzyuxx0LBYYOdLXahdBTRK6RnZc2MGu3F1ahyFEgyJUPYNdJmJt/nWpm1QTyec8GByedo5MtIuUFBg/3rcDnAhKktA7mFf18vXpr6VYjAhovb0RdK9y1Oleb4oBA6nnwWCTIaSgpCi+bvgRI3ybw4igIgm9A7m9blYeXyn12EXAilWNDHLosDpaviWvAQMp58EoST14RUbCuHG+5W4iaEhC7yA2l43lx5ZTUF2gdShC1KEDBnispFfaUNpgTock9RDRp48vsUtrPShIQu8AdredJYeXUGIv0ToUIepIVk1k2cDsatsKb/pL3e+S1INcdDRcdx2kpWkdiWiCJPR25nA7WHJkCcW2Yq1DEaKOAZ4IulS23zpySeohQlFg8GC49lrfOnYRkCShtyOnx8nSI0sprC7UOhQhajGhY7jDRIzd3u730qMn9YIOY7Wr3e8l2llSElx/vdSGD1CS0NuJ0+Pk8yOfy5i5CDgJGBlSBUZ3xyVYPXpSc3UYqySpBz2DASZO9JWRFQFFEno7cHlcfHH0C/Kr8rUORYhaeqkR9Kywo2jwZy9JPcQMGgSjRvm640VAkITextxeN18c/YK8yjytQxGihl5VGOq2kFitbd11SeohJiMDpkyRfdcDhCT0NuTxelh2bJnsmCYCSrRqYLhdhzlA9imXpB5ioqPhxhshMVHrSMKeJPQ2tPrkao4VH9M6DCFqpGMmq8LV7Ipv7c2AgbQzKnqnlIkNCTKuHhBk/UEb2XFhhyRzEVC6qRYGljkDLpkDuHFT0EWPqpPx15DgdsPq1bBnj9aRhDVJ6G3gWPEx2WhFBJTe3gj6ldvbpOpbe3EoToq6ythrSNm2DTZv1jqKsCUJvZXyKvNYd2qd1mEIUWOA10rPCm0nv/mrymCjtLNV6zBEW9q3z9daD8CeoVAnCb0Vyh3lfHn8SzyqjAOKwDDEY6VLRbXWYTRLmaWaypQIrcMQbenYMVixwtcVLzqMJPQWcnqcLD+2HLu7/SttCdEURYURrghSK4MrmV9WHGPHEWPSOgzRls6ehaVLoRU794nmkYTeAl7Vy8rjKym1l2odihDogNFOCwkarzFvDRWVghQvHpNe61BEW7p4EZYskaTeQSSht8D289s5X3Fe6zCEwIDCWLulQ2qytzcPbgo7G1Bl4ntoKS6Gzz+HAKmDEMokoTfT2bKzZOdnax2GEOiAUQ4zkY7gT+aX2XUOSrrIJLmQU1gIy5aBS4oJtSdJ6M1Q7apmzak1WochBADXuiKICoGW+dUqTNVUySS50JOfD8uXy0S5diQJ3U+qqrLm5BqZBCcCwhCPlfggHjNvSlGMHWekUeswRFvLzYUvvwSPrAxqD5LQ/bQnb4+Mm4uA0M9rDdrZ7P5SUSlKC+SyOKLFzp2Dr74CqTre5iSh+yGvMo+duTu1DkMIeqgRdAuydeYt5VSclGfIeHpIOn1aKsq1A0noTXC4Haw+uRqvKlWPhLbSVTN9ykO3m70+ZRHV0vUeqvbvhwMHtI4ipEhCb8K60+uodFZqHYYIc0mYGFgRfjOEVZCu91C2aZOvAI1oE5LQG3Gs+BinSk9pHYYIc9GqgSEVHpQw7SWSrvcQpqq+8fTiYq0jCQmS0Btgd9vZdHaT1mGIMKcDhtl16L3hPSu4LMKGS7reQ5PL5VvOZguv4aT2IAm9ARvPbJQlakJzQ10RWKTClsx6D3WVlb7NXGSHtlaRhF6PM2VnOF5yXOswRJjrrkaQFMJrzZvLoTipSJeu95B18aJvP3XRYpLQr+L2utlwZoPWYYgwF4OB3pWyocXVSq3S9R7S9u6FM2e0jiJoSUK/ys4LO2VWu9CUXlUYZtOhk+7HOqTrPQysWePrghfNJgn9CsW2YvZd3Kd1GCLMDfVYMMu4eYOk6z3EORywerWMp7eAJPQrbDizQQrICE31UCNIrJJx86aUWm14zLJ3esjKy4MdO7SOIuhIQr/kePFx8irztA5DhLFY1UjvChk394eKSmmaSeswRHvaswfOy/4ZzSEJHfCqXrZf2K51GCKMKSoMsSthWzymJSqNNpxRMkEupK1fL9utNoMkdOBQwSHKHeVahyHCWF/VKuvNW6A0WbrdQ1pFhSxlawaD1gFoze11syt3l9ZhiDAWqerpWhliRYxUBbcShRcTXvR4VT2gx6vqaj5UVbn0bwVVBb1ORafzolc86HRu32fc6HCiw4VOdaIotVtrNr0de5wZS6kMVYSsAwcgMxNSU7WOJOCFfULfm78Xm1smIQntDHIaUdTgTegqJlxE4fJacXpMOFxGnM6WLi3TAw13oyuAyaRiNrox6R0YddUUJdrJKG3RzUQwUFVYtw5mzQK99Mg0JqwTut1tJzsvW+swRBjrolqIsQdPMldVAw4lAafHgstjxu7Ud+gQpwo4nAoOpxFf4o8CwJ6QQKzNjYUSzO4CjK7SjgtKtL/SUti1C0aM0DqSgBbWCX1X7i5c3vDbklIEBgMKfaoCf8KPByt2NZ5qpxWbTR+QRV0KDVXobalUEQ/0xGBwYTWVEOHJxezIR0rRhIDsbOjZExITtY4kYCmqqoblb3qFo4J/HfiXrDsXmhnitpJaVa11GPXyYMXmSaTSYcXhVLQOxy+p7lQspZY6x/V6DxGmUqxqLhZ7LgryNx+0UlNh5kytowhYYdtC35m7U5K50EwCxoBL5l7VhI1kquyR2BzBtwCmxFRCGmko1H4D4vHoqbQlUkkiOt0AIs0lRLmPY3KVaBSpaLH8fDh2DHr10jqSgBSWCb3KWcWx4mNahyHC2EBb4CRMD5FUuFMprzQFdce00+vEHm8noiSiwXO8Xh0VtkQqSMRiriZaOUOE/Yx0yQeTrVuhe3cwhGX6alRYfkcOFByQ1rnQTC81ggin9isr3ERT7kqloip0XgbKjGVE0HBCv5LdYcVOPwyGXkSZ8oiyH0fvDZ4JimGrqsq3K9uwYVpHEnBC5y/ZT26vm4MFB7UOQ4QpHdC1WtuJmC5iKXemUFkdekuAHF4Hzmgnpgr/y8K63QZK3Z0pUzKIthYQYz+I3ivr2gNadjb07w8R/r15CxeB0+/XQXIKc3B6pCKX0EamNwKjRqUsnSRQ4OjNhdJOIZnML6uIqGjR81RVobw6hQtMpMyahVcJu/ZO8HC5YOdOraMIOGGV0FVVZf/F/VqHIcKUVq1zJ3FctPchtzSFalvoJvLLKr2VeMyeFj/f69VRWt2ZC7rJVFh6oxIcs/zDTk4OlJVpHUVACauEfqr0lNRsF5rJ9EZg8HRc61xV9ZS6u5NbmobNHlZ/6lTGVLb6Gh6PnmJ7T3KNk7FZurRBVKJNeb2we7fWUQSUsPor35u/V+sQRJhqSev8Oy+8xZSHX8Ljaf4ETicJ5FX3pqyy7rrshvzm9d8w+7HZOJzBP35cQQWqrm1mrrtcJi7aB1AYMQqPztwm1xRt5Ngx3wYuAgijSXH5lfnkV+VrHYYIElu+2MKx3cc4nXOaC8cu4Ha5mfPSHMbOGFvv+Uv+dwlL/7a0wev947PfY4iO8v/+2UdYuSmbPz/3bfT6b953//H9Jbz+QcP3uWHsEH730u8oqzSxe/9ufvzSj2s9bjKaSIxPZNigYTw460E6pXSqeWzOt+Yw54k5LFy6kPvvvN/vWAORR/Vgi7NhLba22TWrbHHY9RNIsBzFaj/dZtcVreD1+ibIjR+vdSQBIWwS+oGCA1qHIILI4r8upii3iKi4KGKTYinKLfLreWOmjyGxU+3SlArQV+//rGuAP7z3HzJSE5k+eXi9j0+bMIw+PdJrHfNgJTm5H2WVte/VN7MvY4aPAaCyqpI9B/bw+Vefs37Let5+5W06p3cGoEt6F8aNGMf8f8/nzlvuJMIS3DOIywxlWGm7hA6+bvgCTz8iLWnEO7NlmVsgOHzYt4TN2rY/62AUFgnd6XFyqvSU1mGIIPLg8w+S0jWFxE6JLJ+3nH+/+W+/njdm+hj6Xtu31rG+3ggSKvxfd37k1AW27TvGj+6fhk5X/6jYtInDuO1630YVKiZKnF2oqK5/l7K+mX15+O6Ha/6vqiq/feO3fLnuSz789EN+/vjPax67YdINrN+6ntUbVnPr1Fv9jjkQOb1OnDFOTOXNezPljyp7HHb9eGmtBwKPx7cuffRorSPRXFiMoZ8oOYHbG/ibYIjA0X9U/zot7ZbQAZ2bOXa+YPkmAG6ZVH/r/Eou4jlf0bPBZF4fRVG44+Y7AMg5llPrsTHDxmAxW1i2dlkzIg5cFZb2G1/1ePQU2PtRYh0iM+G1dvAgBNGuhe0lLFroR4uOah2CCBNHdx/l5P6TKDqF1K6p3HLt0GbPbN+4OwerxUzf7hmNnmdT0ygoi2tV0VJFqZ2IjEYjfXr24cCRA9jstqDvdq/0VpJgSEBxt1/CLa9OxWUZS5JzOzqv1LjQhNsNBw7A8KbfBIeykE/oFY4KcitztQ5DhIkl/7uk1v8/jrLy0g9nM+vGMX49v8pmJ+fEOYYN6FlrMlyd+6w/xK6jdd8o3HfHfZhNjc/EVlWV/6z4DwD9ever83jfzL7sPbSXQ0cPMeya4C+vaYtp28lx9d7DHkWecTzJ+l2yF7tWDh2CoUOhgWGqcBDyCf1osbTORfvr3KczD734EH2H9yU2KZayojJOfH2QhW//m6dfe5+YKCs3jB3c5HXyC8vwelWS4mPqP+FSc3zl1xuBjXUevmv6XXUS+uHjh3nvX+8BUFVdxe4Duzl28hgxUTE8OOvBOtdIiEsAoKCooMl4g0GVsarNJ8fVx+UykqcbQaLlMFb7mXa/n7hKdTWcPg09emgdiWZCP6FLd7voAEOvG1rr/0npSdx8501cn5LMAz99nd+/+x+/EnpJua8gSkxk3QTkxUi1xzeu/+KTLzJl/BS/Yjt8/DCHjx8GwGgwkpSQxPSp03lw1oOkpaTVOT86KhqAsorQqMJV7anGa/Sic7V/y83r1VFg70+8NZqYallZ0+EOHZKEHqryK/Mpc4TGi5IILiZ0JNjsjBvWn27pyeScPE9FlY3oyMbHpC1m34xsh6v2RDoPVi5WdcHlXt/sWG678Taeeuwpv8+/XFjGYva/KE2gs0XbiCyO7LD7lVR3xmvVE1ctxaw61LlzUF4OMQ30cIW4kB5skO52oZXuXjOK6usfj4/xJRKbo+kJU4mxvuIzpeVVNcecxJFb0RWnq2NmUldU+maGx8XEdcj9OkKVoarpk9pYWXUnSq1N98qINpaT0/Q5ISpkE7qqqpwoOaF1GCJMdXL4Ngeptjk4ejoXq8VMQmzTleJSk+KIj4nkxDlfVUMnCeSXpeFp+V4jzXbmvG/8t2fXnh1303Zm89rwmppfQre1yqrTKLEObfpE0XYOH/ZVkAtDIZvQL1ZdxO6WdYmi/dmr7OSf/qascBxGLE4ndoeTn/3hQyqr7dw6eTgGfdM7nSmKwohrenM2t5C8UoX88hS8bVOS3G+Hjh4iMT6xpoJcqKiOrtbkvuXVKRRbg3+1QNCw2eDUKa2j0ETIjqGfKZNZpqLlNny2gWN7jgFw/tj5mmNHdh4BoNeQXoy/3Vc/urKskpfueoluA7rRqUcneicksiC/iI27csgtKKFfjwx+8d1Zft/7pvFD+HLjHr7YcIYp4/s2/YQ2dD7vPLkXc5l508wOvW9HqNJXEYX/9fTbUkV1MkRcS4Jthyb3DzvHjkHP0Olh8lfIJvSz5We1DkEEsWN7jrF56eZax45nH+d49vGa/19O6JExkUy6axKnDpxi/8b9bC+vxmI2ktk1jbl3XMec26+rmezmj2mTx/OrtxaxYu1Kpoy/oW2+ID+tXL8S8E2kCzV2rx2P2YPeoc2e8BW2RPQRWcTaZPZ7uzt3zldsxhCyKa5eiqqqHdyh1/6qXdV8tPcjrcMQYaiTamZQecu3H/ViJL+qJ3/94F0+/vfHzH9zfr1Ly9qD2+Pm/h/dT6eUTvzpV3/qkHt2tGRvcrsXmWlKUsRhIm2nNI0hLEyZApmZWkfRoUJyDP1smbTOhTZSPa1o/akKRfYeOF0K995+LzFRMXz46YdtF1wTVqxZQX5BPj+Y84MOu2dHsxn83ySnvRTZ++AwpWgdRug7EX6TokMyocv4udBKvLN5G7FcqcTTk2q770/SGmHluf96jrSUNDwdNcVdgWe+9wx9evbpmPtpwKZqn9BVVaHAMwi3QZvx/LBx9qyv2z2MhFyXu1f18kH2Bzg9skmC6FjRqoGx5S17Aan0dKGoouMKn4SzjKoMDDbtx1aNJgdpro3o1Ja/CRRNmDo1rCbHhVwLPa8yT5K50ESG6v8WpldykiDJvAM5rYHx+uBymik0j9A6jNAWZt3uIZfQz5Wf0zoEEaYSXc3v7FJVA4WVMp7akar12qxHr4/NHk25te6Od6KNnDsHodUJ3aiQS+j5lflNnyREGzOgEGlv/uz2Mk9XXOE1zKc5u2pHbdUu8m2r1NYVpylR6zBCk9MJBaGxa6A/Qiqhq6pKQXX4/PBE4OikmlCamSQcJFFW6f/6dNE2PKoHd2TgvItSVYVCdRCqos36+JB3/rzWEXSYkEroJfYS3N7A+UMV4SPV3bw/JS9GCiuT2ika0RSHpeW1AtqDy2WiNGKQ1mGEJknoweli1UWtQxBhKtbRvJnKpa5u4baiJqDY9YG3z0N5dQp2c8cUEQor+fl06O5GGpKELkQrxWHE4PE/O9tJoaJK+2VT4cyuBl5CByjyDJCu97bm8UBentZRdAhJ6EK0UqLX/+TsVU0UViS0YzTCHx7Vo8l2qk1xu42UR/TXOozQEybd7iGT0F0eFyW2Eq3DEGEophm9eSWuruHS+xfwXJbALOhSbk/Ho9e23nzIkRZ6cCmoLgiopSgifER7/GvpuYinslq62gOFyxiYCd3rVSg1ZWkdRmgpLAyL9eghk9Clu11oQVHB4mf99lJHcjtHI5rDpQ/MhA5QaUuQteltye2GsjKto2h3IZPQpbtdaCFWMaKoTbfQXcRTbQuZP7eQ4CCwlq5drUQZoHUIoaWwUOsI2l3IvMKUOUL/3ZcIPAmqf13o0joPPE6vExSto2iY3WGl2tJN6zBChyT04FHuKNc6BBGGYj1NZwRpnQcmFRW3JbCLAZR6wmensHYnCT04OD1O7O7AXFcqQlukH9VhpHUeuNzmwE7oLpcJm6Wz1mGEBknowaHMLt3tQhsRTeysIq3zwOYyBO7EuMvK6a51CKHB6YSKCq2jaFch8Uoj3e1CCzEY0HkbnxBXapd67YHMqQuMvdEbY7dHyoz3thLiM91DIqHLhDihhZgmJsQ5iafaLmU8A5mbwO5yv6xcn6l1CKFBWuiBT1roQgsRauMT4iqc0joPdB6Co2xftS1eqse1BUnogU/G0IUWGkvoXoxUVUvrPNAFy3bLKlBu7qN1GMGvPLQbfyGR0Cucof2uSwQms7fhUpJ2NUkKEQcBFRXVEBw/qSpHEmogL5wPBtJCD3yyZE1owdRIQq9yRndgJKI1PMbg6Hb3ePQ4zJ20DiO4SUIPbC6PC68fpTeFaGvGBjZl8WKWpWpBxGMIjoQOUKVL1zqE4Ga3gyvwlyq2VNC/6jg8gV2PWYQug6f+8VebKpPhgolXHzwNApszDjX4X7a1VVmpdQTtJuh/MxxuSeii4+lVBX0Da9Cr7JEdHI1oDY8+eFroHo8eu0Va6a3iCN2cEfwJXVroQgORSv0z2D1EYHME/Z9VWPHogiehA1QpktBbxRn4xYRaKuhfeWRCXP2+mv8VPxj9AwovNL9+scft4bmZz/F/P/u/dogsNESq9Sd06W4PPh4luBK6zREr3e6tEcItdP/2fgxgHdHlvuWLLRzbfYzTOae5cOwCbpebOS/NYeyMsfWe/9i1jzV5zVeWvkJCWgIA8345j81LNzd6/m3fu41bv32rX/FWlVfxxTtfMO62cSSlf5NgygrLWPPJGvZt2Ef+6XxMFhMZmRnc8u1b6D+yf815eoOeaQ9P48PffMiJfSfoeY3s+HQ1SwMvqFU2Kf4RbLwEzxg6gNerw2FOw+K4oHUowUkSeuDqiC73xX9dTFFuEVFxUcQmxVKUW9To+dO/M73e4xfPXWTbsm106tmpJpkDDJk8hMRO9ddqXvnRShw2BwNGD/A73lXzV1FVVsWND91Y6/j8V+ez9+u9DBg9gKwxWZQVlLHjqx28/sPX+d7vvseQyUNqzh0zfQyfvfUZi/+6mB+/9WO/7x0u6mufe4nA7pR1wsFGVYJjHfqVHPpELEhCb5EQ7nIP/oTeAS30B59/kJSuKSR2SmT5vOX8+81/N3r+jMdm1Hv8H6/9A4Bxt42rdXzI5CG1kullpw+dZunflpLRK4MeA3v4FavH7WHDZxvIHJxJcufa23YOHDeQ2U/NrvXmYcRNI3jjiTdY+fHKWjHoDXpG3DiCNf9aw8WzF0npkuLX/cOGWjcJOInRIBARjuzeWGK1DiJYhXALPegHYjqidGP/Uf0bbEH7y+VwsW35NgxGA6NvHe3Xczb+ZyMA42aOa+LMbxzYfICywjKGTx1e57EJd0yo83X0HtYbgMqSuks5hk8djqqqTQ4HhCNdPRW7HB7pbg9Gaj1vzgKdw2GVcfSWCuEWetD/RgRLUZnda3ZTXV7NoImDiI5vuoqY0+70vQEwGRh9i39vAABytuUA+D3ufflNQ31d+t36d0Nv0JOzPcfv+4eL+jrW7U5Th8chWk8NwiK9qqrgNMkEzBZxB0f9/pYI+oQeLH+MG/6zAYDxt4/36/xdq3Zhq7QxdPJQImP9X9d8PPs4ik6hc5/OTZ67eelmFvxxAemZ6dz2vdvqPG40G0nPTOf0wdO4nKFbXakllKtSuqrqsTtk/Fx0HIdeEnqLBGGPjL+CP6EHwQ+n8HwhR3YcISEtgf6j+jf9BK7obr/d/+52gJKLJVijrRhNxkbP27psK+//+n3SuqXx5FtPEhEVUe95MYkxeNweKopDuwZyc139h+NWgnv8/LlXn+PB/3oQj6djlnC53W7u+f49vPT/XuqQ+zUmWBoFV7N747QOoYaqqgx/+WVu/NOfNLn/4bw8DN//Pm+tXdv0yUGQM1oq6CfFBcMf48bFG1FVlbEzxqLTNf0e6uLZixzdfZSkjCT6jejXrHtVlVURlxLX6DnlxeV89PJHxCbF8uTbTxKT0HAyiozx9Q5UllbWmpkf7q5uizvVlo+f3/29u8kryKv3sSFZQ3j916/XOX7q3Ck+XPghu/btoqKygoS4BMaPHM/c2XOJiW7em4s9B/awYfsGXnzyRfR6PctWL+PVv7zq9/NvnnwzP3/857z3r/eY98m8Wo9ZzBbSU9OZOHoi98y8hwiL742jwWDgwVkP8tpfX+PAkQNk9clqVswCnO5vfue8Xi9vrVvHuxs3kpOXh0GvZ0jnzjx9443cNnhwnefuOXuWT3bsYOeZM+w6c4bCykom9enD2qeealEsH2zZwq4zZ9j87LMA/HLJEn61dKnfz39p+nR+OWMGc+fN4/3Nvjk7m376U8ZkZtY59+bXX2fFwYOcfPlluif5ein6pqVx74gR/GrpUh4cPZpoi6Xhm0lCFy3l9XrZvHQzik7xe3LblW8AFKV53bhGsxG3s/ExopP7T+K0O5k4a2KjyRzA6fBNIDFZZHz4SspVO625POZWXS/KGsVd0++qczwtOa3OsQNHDvDUr57C4XQwbsQ4MlIzOHrqKJ9+8Snb9mzjL7/9C7HR/s+Bfucf75CWnMZ1Y68DoFePXsydPbfWOXkX81i+djm9uvdi/Mjaw0a9uveq9f9JoyfRo6tvVUZRSREbt29k3ifz2LRjE2/99i2MRl/v0U3X3cTf5v+Nd/7xDn946Q9+x9vWgm0d+mUejx6vzoTicTD7b3/j0127yExO5tFx43C43fwnO5uZb73FG/fcw4+uu67Wcz/bs4dXli/HZDDQJyWFwlbUN/d6vfxyyRIm9OrF6J6+uTuT+/SB6bWX7+45e5b/ZGczqU8f3+NXuPr/AM8uWsT6Z57xO46f3nQTH23dyp9Xr+a5W25pwVcS/II+oV89lhloDmw6QEl+CQPGDPCrhev1eNmydAs6vY6xt9VfuKYx0fHRlFwsafQcp82XpC3WRt7FXlJdVl1zXfEN3VVvtByuxoc4mhIVGcXDdz/s17m/++vvsNlt/PZnv2XciG/eJP7js3/w9odv8/f5f+epx/xraZ08c5K9h/by4KwHa3qPevfoTe8evWudt3v/7pqE3lSck8ZMYsr4KTX//8GcH/DYs49x5MQRvvr6K6ZdPw0Ag97A9eOuZ9GyRZzLPUfnTk3P+xC1uQ0xLN6+gk937WJcZiYrf/xjIky+N9+/veMOrv3tb3l64UKmX3NNTWsW4FvDh3Pb4MFck5FBUWUlnX760xbHsOzAAU4VFdVKopP79mVy3761zpu3aRP/yc5mcp8+/HJG/Ut7L8tMTubrY8dYkp3NjHp6GOpzTUYGgzp35m8bNvDzm29uuDe0mY2kYBL0Y+jNbcF2tMtj4eNn+jcZbt/GfZQWlJI1Jov4lPhm3y+9Vzouh4vivOIGz+ncpzN3/OgOv8bz807nEZcS16yJeeHgylokqgrODioocz7vPCfPnKRfr361kjnA3bfdTWx0LF+u+xKb3ebX9ZatWQbA5LGT2zrUGtYIK9Ou8yXxnOO1V0xcN+46VFVlxdoV7Xb/pgTDsF1DXPpo/pOdDcAvpk2rSeYASVFRPDllCg63m/c2bar1vKz0dIZ17YpRX38J4+Z4b9MmFEVh1tChrb7WZS9Nn45Bp+MXn32Gt4FNkOoze/hwThcVsebw4YZPCvCc0RpBn9B1SuB+CRUlFez9ei/R8dEMnuTfu8yWrD2/Up9hvq6rk/tPNnhOXHIcQyYPIa173e7cKxXnFVNeVE7vob0bPS8cqVe8JngVK95W5gSny8my1cv48NMPWfTFIg4eOVjvecWlvjdqnVI61XlMp9ORkpSC3WFv8PlX27lvJxGWCHp08a9wUWtd3aPWp2cfDAYDO/ft7JD710cXxC+DbiWKvPJyAHok1Z31fvnY6sYSXCuoqsqaw4fpm5pKfGTbvenvnZLCdyZMYP+FCzVj6v4Yc6nLf1VOI0tt/ZjHFKyky90PGz7bwLE9xwA4f+x8zbEjO48A0GtIr3qXo235fAset4dRt4zCYGz6W11eVM6+DfuISYxh0IRBLYp18KTBLPjjAg5uPVhvcRnwrYl//1fvM/070xusagdwcIsvKdRXxS7cXdkg99L6+QXFpcV1JqL169WPF598kYy0jJpjl8fGcy/m1rmG1+vlYuFFAM5eOMvwQfX//C+rtlVz4vQJsvpkoW+Dllpj97ncAu/Xu/YkT7PJTI8uPTh8/DBOlxOTsePnagRzQnepESRFRQFwsrCQ/p1qv9E7WejbnOlIfn673P9Qbi7FVVVMy2r7SY0vTZ/Oh1u28OKSJdw7ciQWY9PDWtd26wbAxuPHGz7Jj+sEq6BP6CZ9+78AHNtzrE61tOPZxzme/c0vTX0Jvaa73c+155uXbsbr8TL61tHoDS17gU1KTyJrdBa7Vu3inmfuaXL5WmO2LttKdEI0Q69ru660UOG6opvW28o/o2nXT2NQ/0H06NqDCEsEZy+c5ZMln/Dlui/5yS9/wnt/fA9rhG9Gc5f0LqSnppNzLIfNOzczZviYmussWLqAsooyACqrm57kVFRShNfrJT62+UM7jVm3eR1nzp8BfG9UNu3YRGFxIX0z+zJl3JQ658fHxXP05FFKy0pJSer4EsOB3MvXFLfXwrSsLP65fTuvLl/O9f361SS+ospK/rRqFQClNv+GYJrrXGkpAKkxbb9sMzUmhp9MncqvP/+cP69ezU9vuqnJ58RERGAxGjlX0sg8InPrJrAGsqBP6GZD+/9w5v5yLnN/ObfZz/vlgl826/yb5tzETXOa/qVtytQHpvKnH/yJXat2MWraqDqPj50xtsGd4i7LP5PP0V1Hmf7d6S1+cxHKHFfMjPY2sJWqv66eUd67R2+e+6/nAPhy3ZcsWbmEu2+7G/DNGXnyO0/y81d/zi9e/QXjR4wnPS2d46eOsz17Oz279uTEmRN+JanLyT8qMqpV8V9t3ZZ1rNuyDvhm2dqMG2Zw921318xwv1JMlC8ZlJZrlNDVIE7obhP3jRzJvM2bWXP4MNf8+tfcPGAALq+Xz/bsqUm0V0/ibCtFl2bHx1nbp+zx0zfeyF/Xr+fV5cv5zvjxfnXrJ0RGNj5r3xS6K3aC9zf5Eouh6Zna4ab/yP5kjc3ii3e+aNaEkit9/rfPiU2K5YYHbmjj6EKDU9d2LfSG3Hajr3rf/pz9tY6PHDqSN/77DUYNHcWu/btY9MUiyivL+c1Pf8PgLN9cjbjYuCavbzb53gw7XW1b2/rFJ19k3afrWPfpOlbMX8F7f3yPubPn1qxBv5rD6dssw2LW5m9ZX+/eecHB69Fj0OtZ9vjj/HL6dHSKwv9t2MCi3buZOXgwC7/7XQBSottnlcrlSXh2V/tUkoy2WHjhllsoqa7mleXL/XqOzenE2ljSlhZ64DLrQ/eH0xp3P3U325Zvo/RiabMLwnjcHlK7pTJmxhjMEfL9rY+9DVvoDbk8Xm532Os8NqDPAF79Rd3iLws/XwhA38y+dR67WlxMHADlleWtiLL1Kip9VQgvx9PRgrmFrgKqosdshJdmzOClq5aDrb00Ge7y2HJbS740fl9cVdUu1wf43qRJvL56NW+sWcPjV62nv5rX66XMZiMrPb3hk6SFHrg6oss9GKV2S2XGYzNaVN1Nb9Bz67dvpf9I/8rUhiOH2v4J/eBR36TEtJTGVyNclncxj32H9tG9c3cyu9WtsHW1pIQkYqNjOXv+bKvibK0z58+QnJDc7Ap3bSWYEzqAV2l4nszH27YBcM+IEe1y76z0dHSKwuF2mnQHYNTr+c3MmdhdLl5cvLjRc49evIhXVbkmI6Phk0K4hR7cv8lIC11ow3lFC93jbfmf0elzp+ttgZ8+d5r//fB/AZg6fmqtx6pt1XX2MKisquTlP7+Mx+vhuw981697K4rCNf2vIfdiLqVlpS37AlopvyCf4tJiBg1o2aqOtqB4g3tdsqozUl7PpLeFO3fy7saNjOjenTvbcI34leKsVgZ17syO06dbPLznj7uvvZbhXbvywZYtHLl4scHztp70Lded1LuRpbYh3EIP/i53aaELDagKeHR69F4PaitaeKs3ruaTJZ8weMBgUpNTsZgtnLtwji27t+B2u7n/zvtrxsUv27BtA3+b/zeGDRxGYkIipWWlbNy+kdLyUh6999E6BWcaM2HUBDZs28D2vdu5YULHz5fYkb0DoE452Y4UCi30Ua++Spf4ePp36oTFYGDbqVOsPXKEnklJLPjud9FftfY6Jy+PVy+NSdsu7Q+ek5fH3Hnzas6ZN3euX/e/Y8gQXlqyhC0nTzK2ntrrbUFRFF69805u+NOfapbi1WfloUMYdDqmD2rkDWJjdd6DXNAndJkUJ7Ti1enQez14PC1v4Q0dOJTT505z9ORR9h7ai91hJzY6ltFDR3P7zbczYkjdrtKe3XqS2S2T7dnbKasoI9IayYDeA5g9YzbDrhnWrPtfN/Y6/vLeX1i5bqUmCf3L9V8SHxvPxFETO/zel+la0cMSCFSdkbuvvZZFu3ez5eRJXB4PPZKSeP6WW3jmxhuJiag7GTGvrKxOwZb88vJax/xN6N8eP57//vxzPtq6td0SOsDU/v25ccAAvjxYf9GkaqeTz/bsYfqgQaTHxdV/EUWBqLZd1RFIFDUY9h9twru738XtDd1N60VgmmQzYXE6ya3qi9MVvN227/zjHT7+98fMf3O+3+P1beHchXM88F8PMHf23DpL9zpSuj0dY2XbFxtRVZUZv51BfGQ8H/74wza//mUplv1E2M/XOubyeOj74otc260bn3zXvyGY1njw3Xf5fN8+Tr/ySuM7nbWjv2/YwHc+/JB1Tz3FxHo2ewF8yfy++zo2sA4U9C10AKvRSrlD25m6IvzY9XosgBrgGwQ15d7b72XJyiV8+OmHPPN9/3e3aq15C+aRGJ9Ys8a+LRUUFbB281q27NrCmfNnKC4tJiYqhoF9B3Lv7fcyoM+AmnP1Tt+kxu6PdW/yupte2UR6QiMzqK/w6ZZP2X9mP4ueXVTr+FPznuLTzZ82+LyHr3+Yl+727RO/YNMCnnnf9zN55vZn+OG0H9Y5//8tXcCLC97nvTlzmDvWV1/CqNfz3LRpfPvDD9ly4kTNLmjt5TczZ7Jw1y7eWL2aX2iw05nb4+G3y5Zx2+DBDSdzgHZavhcoQiKhx5hjJKGLDlelgzhAr1NxBXFSt0ZYee6/niPneA4ej6ddy8Be5na76ZrRlWnXTWtwfXprLPpiEfM/m09GWgYjBo8gLiaOc7nn2LBtAxu2b+CFH7/A9eOuR0FBuVTH94npT9R7rdMXT/PZts/o3am338nc6/XypyV/YmSvkQzrWf8wyN3j7iYtvm6PyNAe9U9ge3vF29w/8X7iIuNqP9BAJ+ucMWP4xWef8cLixaz88Y/9iruluiUm8v7cueSXa/M6fKa4mIdGj+bB0aMbP1ESeuCLMWuz3EWEtwq974VUp3ghiIuTAIwYMqLe8fr2YjAYeOiuh9rt+v179+f1X7/OkKwhtY5nH8zmJ7/6CX/4vz8wfuR4Ik2RNftBPDnjyXqv9dI/fK3lu8f535Ow9sBazhWd40e3/KjBc+4ef3eDyf5q3ZK7cbrgNG9+8SbPf+v5Wo/p8NT7HINezz0jRvDGmjUcu3iRXintW4Vv9rXXtuv1G9MzObnJLVkBaIcStYEkuGeDXBJrjtU6BBGGyhTfC6leF/TTUELOxNET6yRzgMEDBjM0aygVlRWcOH0Co67xsXO7y85n2z7DZDBxx+g7/L7/gk0LUBSFaUOnNTf0et015i66p3Tng7UfcL649ng5asPLxWYPH46qqs3asSykhXgLPSQSurTQhRbK8ZW71Onab/2taHsGg69jUq/XY1QbT+grdq+grLqMqYOmkhid6Nf1VVVl8+HN9EztSWxk2zQ29Ho9T898Gqfbye//8/tajylq/S10gOHdumHU6xvfTjSchHhCD4ku91iLtNBFx/MCTqMRnUcSerDIL8hn596dJMYn0rNrTwyexl8C/7XhX4Cve9xfx3KPUVpVyuSsyU1ee92BdbWOmY1mfnDzD+o9/9bht/K3lX/js62f8Z0bvkP/zpcrOTb8+2cxGhmYns6O06dxuFyYQ3jrUL/Et+3OgoEmJBJ6jDkGBQUV6foUHctmMKBvYAxTBBa3283Lf34Zp8vJYw885muhuxpOcGcLz7L5yGYyEjKY0H+C3/fJLfXtVZ8Uk9Toef/a+K86x6IjohtM6Iqi8Oydz3LfH+7jf/79P8x7fJ7veCNd7uDbhnT32bNcrKigS0LzS0GHjKiokC77CiGS0HWKjkhTJJXOpveAFqItVekgTrrcA57X6+WVN18h+2A206dO56bJvm2KLy9Zq88nGz9BVVXuGnsXOp3/o5Mllb69uGOsjQ8FLnp2kd+T4i4b23csk7ImsXb/WrYc2cKYPqNRmmjIJFzacrSwsjK8E3pS42+wQkFIjKGDTIwT2qjQqyhIUaNA5vV6+Z+//A9fff0VN0y8gacee6rmMYO9/jaN1+tl4eaF6BQds8fNbtb9LCZfYRWHy9HyoBvx7B3PolN0vPLpK+gNTW9berm0a6NbioYDSejBQ8bRhRbKFQ86RbrcA5XX6+XVv7zK8rXLmTJ+Cj//0c9rWttGnZGGGrfrDqwjtySX8f3Hk5HQyM5d9UiM8k2eK60qbU3oDRrQZQC3j7qd7FPZfLFzSZPnF1dXA5Ac4hPCmiQJPXgkRvg3A1WItlSKC4WmW0mi411O5ivWruD6cdfz3H89V6tojpmGx1Mvj2/fM/6eZt+3d3pvdIqOE/knmh+0n35y208wGUy8+umfcHsaf0N5OC+PjLi4mq73sCUJPXikRLZv0QQh6uMFnCbpcg80l7vZV6xdweQxk3nuiefqVMCzeOqvOV5UUcSqvatIjE5k6uCp9Z7TmFhrLP0692Pf6X3ttqVo58TOPDjpQU7kn2FeI2vMzxQXk1dezsTGthMNB1ar7yPEhcSkOICEiAQMOoNs0iI6XLnJi8moBvUGLaHm/QXvs3ztciIsEXRJ78KHC+tujnL74NsZkjSkzvFFWxbh8ri4Y9QdmAwtG3e+achN/HHJH9l9cjfDM4e36BpN+dEtP2Lh5n9yvKCgwXNWXtqZ7PYhQ9olhqCRnKx1BB0iZBK6oigkW5PJrczVOhQRZgr0XrqYPDhdIfPnFPTyLuYBYLPb+PDT+nc6y7Jm1ZvQW9Pdftk94+/hjc/f4N9b/91uCT0+Kp6np83khU/nN3jOR1u3khIdzR1D668PHzbS/avBH+xCYvvUy7ae20p2frbWYYgwowNGl3amuCx091kONRadhdSLqe16jyfffZLV+1az8ZWNRFna53ejq/JVg5Xijubn0/ell/jl9Om8OH16u9w/aNx1F4TBkr2QGUMHGUcX2vACLotMjAsmFrX99+x+auZT2F123l/9frtc32h0Nlr29deff06n2FieuuGGdrl/0IiICItkDiHU5Q6S0IV2KgxVQGiXlQwlZlf7VwzrnNiZ38/9PYXlhe1yfbOhgoYWWLg8HvqmpjJ3zBgiQ7w6WpPCpLsdQiyhR5oisRqtVLuqtQ5FhJkivZM0I7ikoR4UjLaOqWk+/dr26+o2q8UNPmbU63n+1lvb7d5BJaN5dQSCWUh1uYO00oU2CnFiNEuBmWBgUAzoHcG9fz2AxX1R6xCCgyT04JUWlaZ1CCIMqQo4LE6twxB+sCjtP37e3gwGNwa37F3RpOjokN8y9Uohl9C7xnbVOgQRpiqNMtQTDMye4B9TNhslmfule3etI+hQIZfQ4yxxRJvC5x2ZCByFxjJ0Ulsm4EVURWgdQquZKdU6hODQo4fWEXSokEvoIK10oY0qxQmRMo4eyEw6E3p7KIyf52sdQuCLjIS08BqCDcmE3iW2i9YhiDBVGVmldQiiEZFq8G9QYjQ6MLpKtQ4j8IVZ6xxCNKGnR6dj0IXUijwRJApMRSiK9LsHKost+CfERRqLtA4hOPTsqXUEHS4kE7pBZ6BTVCetwxBhyKZz4ImQDYICkV7RY6pq2WYrgcTqOqN1CIHPaoXU9i3tG4hCMqGDjKML7VRFy2z3QBSpREKQ71xhMjkwusq0DiPw9egBYdhTJgldiDZWaC6UbvcAZHEEf3e71dDwVqniCmG6/3vIJvRoczTxFqmtLTqeQ3HgiZHZ7oFEQcFSGQIJ3XlW6xACX1ISpIRnxdCQTegAvRPD812a0F5pVKnWIYgrROgjUDzB3WtiNtkxusu1DiPw9e+vdQSaCemE3iuhFwrB/UcsglOBoQDFJL97gSLSFfzL1aJ1MhmuSUYj9OqldRSaCemEHmWKolO0zHYXHU9FxRZn0zoMAegUHRHlwV0dzmBwY7Wf1jqMwNerly+ph6mQTugAfRL7aB2CCFO5llykg0h7kUokiju4fxDRxlwUvFqHEfjCuLsdwiCh94jrIUVmhCYcigNnrOzAprXI6uDubtfpVKIcx7QOI/AlJ/smxIWxkE/oRr2RHnHhVwJQBIbCKNmzWksmnQlzZXDvrhZpKUTnlTeGTbrmGq0j0FzIJ3SQ2e5CO+X6StRoqRynlWhv8O+8GOM8qnUIgS8mBjIztY5Cc2GR0DOiM7AarVqHIcJUSUyh1iGErWCfDBcZUYrBXaF1GIFvyJCwrAx3tbBI6IqiyOQ4oZmLhlL0sq1qh4vSRaF3Bu9WqYqiEuc6oHUYgS8yEvrI6zuESUIHyErOQqeEzZcrAkxpfKFMeO9gUY4orUNoleiIixjclVqHEfgGDQKdvLZDGCX0SFMkmfEyxiK0cV5fginOoXUYYcOgGDCXB+9kOL3eQ6xNWudNsljCfqnalcImoQMMSh2kdQgijOXH5GEK35oXHSpajQ7qndViTWfQqS6twwh811wDBlmWfFlYJfREayIZ0RlahyHCVJ5iwxhXqnUYIU+n6IgqC97udqPJQZRNZrY3KSICBg7UOoqAElYJHeCaVFmrKLRzKqKYaJkg165iiEHnCt6XtnjdUZRg7l7oKMOGhXWZ1/oE7299C3WN7UqcJU7rMESYKlKcqDH56MPuL69jKChB3TqPiigmwn5e6zACX2ysjJ3XIyxfVmQsXWjpiKma+BhZW9weYnQxQbtUzWBwE2/fo3UYwWHUKJnZXo+w/I70TuhNhCG4C06I4FWOm2prMRFm2WyjrUWXBW9luET9QZkI54/0dOjeXesoAlJYJnS9Ti9j6UJThwxO4qx5sja9DcXoYtA7grN1HmO9iMWRq3UYgU9RYOxYraMIWGGZ0AEGpgwk0hjcuzCJ4GVTPORGuUmIqdI6lJARUx6jdQgtYjQ5iLPt1TqM4NC/PyQkaB1FwArbhG7QGRiePlzrMEQYO6xUo4+4SEykdLO2VpQuCr09+FrnCpDEPhRVVj40yWqFkSO1jiKghW1CB+ib2FdmvAvNqArstXiJNZ7CYpZlSq0RWxGrdQgtkhBxHJOzSOswgsP48WAyaR1FQAvrhK4oCiPSR2gdhghjpbi4EG0mOeKUFLxqoWhdNAZb8H3zYqz5RNmOaR1GcOjZUybC+SGsEzpAj/gepESmaB2GCGOHdNW4TCopUbnoZJZcs+gUHXGlcVqH0WxWSzlx1dlahxEczGYYN07rKIJC2Cd0gFEZo7QOQYQxL7DPAkbKSI4t1jqcoBKnxqFzBtfLmMnkINGxXarB+WvMGF+ZV9Gk4PpLaCedojvRJaaL1mGIMFakOMmLtGLhIgkxNq3DCQpGnZGo4uCqCqfXe0j27kCnurUOJTh06SJ7nTeDJPRLRnUehSKrgoWG9uttOI1GonWniY6UF/ymJDgSULzB8zerKCrJ+r2yx7m/LBaYOFHrKIKKJPRLEiISGJA8QOswRBjzKCoHrL4EFW84SYRFKsk1xKq3YimzaB2G33Q6lRTTXszOi1qHEjwmT4ZIqRXSHJLQrzAiYwRWo1XrMEQYu4iT01ERKIqHZPNxIiNkfXJ94svjtQ7BbzqdSooxG4sjT+tQgsegQdC1q9ZRBB1J6Fcw6U2M7jxa6zBEmMvR2yiN8CX1JNMxKTxzlTglLmiWqel0XlIMezA78rUOJXikpEgBmRaShH6VXgm9yIjO0DoMEeZ2muw4jCZQVOKNx4mLdmgdUkDQK3piSoKjxGtNMpdudv+ZTDBliuyk1kLyXavHhG4TMOiCowUgQpMblV1WL95LL2yx+pMkxlRrHJX2EtwJKO7Anwin03lJNezG7CzQOpTgMnkyRAfvjnlak4RejxhzDNemX6t1GCLMlePmcNQ3pS6jdGdIji3XMCJtRemisJYE/hwXg8FFmn47Jmeh1qEEl0GDpBpcK0lCb8A1KddIBTmhuTOKndyob5KYVblAalxJ2C2w1Ct64ksCfyKcxVxNmncjRlep1qEEl27dYJQU+GotSegNUBSFid0molPkWyS0tU9XTZXlmyVaFvJJjSsIq2HGZGcyOldgf8FREcWkODai98p8h2ZJSvKNmyvh9ja17QX2X4jGEiISGJkhsy2FtlQFtpucuK7YvcVMEekxZ8Nil7YYXQzmMrPWYTRIUVQSI06QaNuOgtQOaBarFW66CdmZqG1IQm/CoNRBUhZWaM6heNkRCR7dN3t+66ki1XKEuCi7hpG1L6POSFxRnNZhNMhgcJNq3EOU7ajWoQQfgwFuvlmKx7QhSeh+uK7HdVJwRmiuHDd7ovWoVw4DKSqxhlOkxRWi1zf83GCVVJ2E4gnMrtioiCI6edbLsrSWUBS4/npfd7toM5LQ/WAxWLi+x/VS611orhAn+6ONdY6bKSQ9+iRR1tCpAR+vxGOqNDV9YgczGNykWPaTaNuBTpWiPy0ybpzMaG8HktD9lB6dzpC0IVqHIQQXFAeHY+puJ6nDQaLpGKlxJUHfWjfrzEQXBt565KiIYjp5vybCfl7rUILX6NEwQPbNaA+KqqqhP6umjaiqypIjS8irlJrMQnu91Agyy+vfatWrmih1d6WiKvgmG+kVPZ0qOqG3B867EoPBTYIhRxJ5a40YAUOHah1FyJKE3kyVzko+PfgpDo8sTRHa6+uNoHtFw/unO0ikzJaIzRE8nXFpzjTM5YExq12n8xJjuUC07bDsYd5aQ4f6ErpoN5LQW+BU6Sm+PP6l1mEIAcAAr5UuFY2XhXWoiZTak7A7AnseSIKaQHSR9l3tChBtLSDGfhC9N3RXEXSYQYN8Xe2iXUlCb6HdubvZfmG71mEIAUB/r5WuTSR1ADtJlFUnYncGXmKP1kWTcDFB6zCIjCglznUQg7tC61BCQ1aWbxKcaHeS0Fth7am1HCk6onUYQgDQU42gdwNj6lezk0JpdTyOAEnsFp2FlMIUFK828SiANaKMGM9RTM4iTWIIScOGwbWyL0ZHkYTeCl7Vy+dHPie3MlfrUIQAIEO1kFXhRFH9q1hmJ5XSqjgcLu0Su17Rk16Wjs7Z8eP8er2HKPNFoh3H0HtkN7s2oygwdqyvdS46jCT0VrK77XyW8xnljvDdBUsEliRMDKnwoPd6/DpfVcGlJFDtjqOy2oSnA6uXKiik2dM6fL25yWQnWn+OSPspFNW/75Pwk07nKxrTs6fWkYQdSehtoMxexmc5n8nMdxEwYjBwbRUY3c2cma0qOJREql2xVFYb8bbzq0OSJ4nIko4p/ak3uLGaSoh0n5Pqbu3FaIQbb4SMDK0jCUuS0NvIhYoLfHH0C7x+dnUK0d4iVD2j7HrMTmeLnq+qehxKElXOaKqqDbT1C0WcEkdsQWwbX7U2o9FFhLEYq+s8ZldBu94r7EVE+GqzJydrHUnYkoTehg4XHmbd6XVahyFEDaOqMMphJtLRuqVXqmrAriRjc0XicBpwtrLiaYwuhviLbb+/uaKomE02zPoyrK5zmFzFbX4PUY+kJF/LPCpK60jCmiT0NrYnbw/bzm/TOgwhauiAIW4ryVVtN+lLxYSTaJzeCJxuCw6nAZefvftRuigSCxJpiya/weDCbKzCTClmTxFGZ7FsYdrR+vSBCRMI+nrDIUASejvYlbuLHRd2aB2GELV0Uy30qXSi87ZPwlMx4SIapzcSh9uE26PH7QGvR6kZi7fqrSRdTEJR/Z9Vr9e7Meidvg/FjkGtRu+twuQqlqIvWtLpfMViBg7UOhJxiST0drLjwg525e7SOgwhaonBwDCbrsXj6i2lqgYMaiQJ540omFAVI17FV2dewYOien2fvR4U1XPpmBuDp0pmoQciiwWmToX0dK0jEVeQhN6Otp7bSnZ+ttZhCFGLXlUY6rGQWOVfEZq2YFJNpJ52o3NLd3jQS0nxJXMZLw84ktDb2eazm9l3cZ/WYQhRR081gl4VDr+L0LSUUTWResaN3iXJPKgpiq/y29Chvu52EXAkoXeAjWc2cqDggNZhCFFHHEaGVCuYXe3TBS/JPETExPiKxaSkaB2JaIQk9A7y9emvOVR4SOswhKjDgMI17ghS2nAWPIBZNZEi3ezBr18/XxlXg0HrSEQTJKF3oG3nt7Enb4/WYQhRryRMDLTRJhPmLF4Lyacd6Dzy8hK0LBaYOBG6d9c6EuEnSegd7GDBQTae2Yja5nW3hGg9RYV+qpUulTaUFr40WD0Wkk47UNq7bqxoP336+JakWSxaRyKaQRK6Bk6VnmL1ydW4vc2ssy1EB4nCwGCHgSh789Z5R7kiSDhtIzA2ZRXNFhvrKxIjy9GCkiR0jeRX5rPi+ArsbimMIQJXdzWCXlVO9J6m14LH2q3EnZMtSIOSweCbvT5okFR8C2KS0DVUai9l2dFlVDgrtA5FhIBfzPgFRblFtY4ZjAai4qPo1q8b4+8Yz6AJg5p9XRM6BrnNtdatj7vvF5zPL+Lrj1+ma1oSCRURROV33Lp20YZ69vR1r8u68qAn0xY1FGeJY2a/mSw/tpzC6kKtwxEhInNwJildfMuLbJU2zh4+S/b6bLLXZzPlvinM/snsZl3PiZcdBhtxsUb6O/XE2L7pVdKjJ6XQjKU0tJP5qcJCejz3HN0SEzn1299qHU7bSE+HESMgNVXrSEQbkYSuMavRyow+M1h3eh0nSk5oHY4IAeNvH8/YGWNr/u9xe/jkD5+w9pO1rJq/ipE3jaR7VvdmX7cUF5tNLlJMJi6XYk/OU7BEOdooctEhkpN9ibxzZ60jEW1Myv0EAKPeyNSeUxnTeQw6RX4kom3pDXrueuIuLJG+Gct7v97bqutdxIlH72sLGKLiWhue6Cjx8XDDDXDHHZLMQ5S00APINanXkByZzFcnvqLaJZOLRNsxmo2kdE3hzKEzlBeV13n80LZDrF2wlhN7T1BVVoU12krm4ExufOhGMgdl1pynoDAiYwQWw6XlTDfe6JtEtXMnXLxYc97CnTv5+4YN7DxzhjKbjeToaK7r04dfTJvGgCtmUK84cICb//xn+qWlcehXv6o3drfHQ+ef/Yz88nL2PP88g7t0AWDbyZMs3LWLtUeOcKa4mOKqKuKtVkZ2784TU6YwtX//Oteat2kTD7//PnPGjOEv997Lb774ggU7d3K2pIR4q5WbBgzgt7ffTkb8N3u1z503j/c3bwbgdFERymOP1bqm+r//29S3X1vx8TBkCPTq5SvfKkKWJPQAkxaVxqz+s1h1chUXKi5oHY4IIfYq39h3TGJMreML/7SQlR+tRNEpdOvfjd5De1OcV0z2umz2fr2XB557gHG3jcNisHB9j+vpHHNV665LF99HXh7u7Gzuf/55PtmxA7PBwPBu3ciIi+NIfj4fb9vGot27WfS973HzpS03b+jfn87x8eTk5bHlxAlG9+xZJ+5l+/eTX17OsK5da5I5wC8++4w1hw+TlZ7O8K5diTSbOV5QwNJ9+1i6bx9/mj2bJ6ZMqfd7UWazMfa11zhTXMyEXr0YmJ7O5hMn+GDLFtYdPUr2Cy8QGxEBwPhevah0OPh01y4izWbuGjasxT+DDtW5M1xzje9nI8KCJPQAFGGM4Nbet7L9wnapLCfaRO7JXArP+yZeDpr4zUz3r//9NSs/WklKlxQee+0xOvf+Jlkf2XWEvzz5F+a/Mp+xY8dy53V3EmVqZCZ0WhovvfEGn+zYwaisLP4xZw49YmNrHl64cyf3/P3v3PfOO5x4+WXirFZ0Oh1zRo/m5WXLmLd5c70J/b1NmwB4eOzYWsefuuEGPnzkETpdcQ+AzcePc/Of/8wzn37KXcOG1WptX/bZnj3cNGAAXz/9NDGXEndJVRXX//GP7Dl7lrfWruXn06YB8O3x45narx+f7tpFUlQU8+bObfh7oDW93tcSv+YaSEjQOhrRwWTANkApisLIjJHclHkTJr1J63BEkLJV2ji45SBvP/M2Xo+XWx69he4DugPg9XpZ+n9LAfj2b79dK5kD9BnWh+nfno7b5ebIl0caT+ZAcXExf/zjH7FYLHy6YgU9nnwSxo3zFSsB7ho+nMcmTKCkupqPtm6ted7D48YB8M/t27G7XLWuWVBRwdJ9+zAbDNw3cmStx6YNHFgnmQOMyczkh5Mn4/J4+E92/dsXR5rNvDd3bk0yB4iPjORnN90EwFeHgmzfhchI305o990HkyZJMg9T0kIPcN3iunHXgLtYd2od5yvOax2OCALv/+p93v/V+7WO6fQ6HvnvRxg1bVTNsbOHz1JaUEpy52S69e9W5zoJEQl8d9Z3Wfj6QjZfGkNuzJo1a7DZbEyZMoWMjAzfwaws30deHhw9yuSsLN5at45Nx4/zo+uuAyAzOZmJvXuz/uhR/r17N/dekbg/3roVl8fD7OHDSYiMrHPPospKPt+3j/0XLlBSXY3rUgGco5fG8w/n59cb67XdutX7ZqB/p04AnC8tbfLr1ZzB4Kuz3qcPZGTI+LiQhB4MokxR3NrnVg4VHGLr+a04Pe2z1aUIDVeuQ68oqeDYnmPYq+zMf3U+KV1S6DGwB0BNF3zBuQIeu/axBq8HUFBQ0OR9T5zwLbtctWoVShPJpcDt9u2p7fXtxPbIuHGsP3qU9zZtqpXQ37v0RuLq7naAv339NU8uWECVo+Flc+W2+tfHd22gBRtzqXa53R3AZZk7dfIl8Z49wWjUOhoRQCShB5H+yf3pEtuF9afXc678nNbhiAB19Tp0W6WNvz79Vw7vOMzffv43frngl5gsJryXkmlMYgxZY7IAMOgMpEWlYTVaa10zKSmpyftevl6vXr0Yd6kbvSH9+vWDBx+E48fh1Cm+de21PP7Pf7IqJ4dzJSV0jo9n15kz7D13joy4OG4cMKDW83eePs1jH3+MXlH4nzvvZMagQXRNSMBqMqEoCv+3fj2Pffxxg1sg6YKpNasokJYG3bpBjx4QHa11RCJASUIPMlGmKG7pfQs5hTlsObdFWuuiSRFREXznle/w0l0vUZRbxMqPVnLrt28lIdXXSo2KjWLuL+fSK6EX47uOb/GcjS6XZlP37duXefPm+fekAQNgwACsLhez16/nnU8+4f0dO3juhhuYd2ky3JwxY9Dpak/3WbBzJ6qq8vj11/PTS+PeVzp6xRK6oGQy+Wand+vm+2w2ax2RCAIyKS5I9Uvqx7cGfIsuMbIkRTQtOj6aWx65BYCVH62kuqKa7lndiYqLIvdkLr09vbm+x/WtmoA5ZcoUTCYTa9eu5WJzE6rRyCNPPAHA+9nZOKZPZ/7u3QDMnTChzunFVVUAdEtMrPOY3eXi00vPbSsmg6/t4/Zjk5oW0et9rfDBg2H6dHjoIZgyxTdjXZK58JMk9CAWaYpkWu9pTO4+uU4XqRBXm/StSSSkJWCrtLHyo5VYTBZ+9MyPUFWVxx54jA0bNtR5jsfjYfXq1WzZsqXJ66empvL4449TVVXFjBkz2LdvX51zHA4HixcvJicnp85jY8eOpW/fvhw9epRnX3uNotJSxo8fT+9nn/VVNxs7FjIzISqqZvLa+5s3U3HFFq92l4sfzJ/PycK23RshOToak8FAXnl5zZuJVrFYfK3vUaPgtttg7lzf51GjfDXWdfLSLJpPutxDQJ/EPvSI60F2fjZ78/fKPuuiXkaTkenfnc4Hv/6Atf9ay/++/L88/IuH8ZR6+N3vfseECRPIysqiV69eREREkJeXx549eygtLeWvf/0ro0ePbvIer776Krm5ucyfP58hQ4YwePBgevbsicFg4Ny5c+zZs4eqqiqWLVvmG0e/ysMPP8zPfvYzXn/9dQAeeeQRX3JLTvZ9XCpI8/Dkyby+aRO7z56lxwsvMKFfP/QeD18fOYLN5eKJ66/n9dWr2+57p9dz26BBLNy1iyH//d+M79ULq8nXm/H3hx5q+Ik6nW/ZXnz8Nx8JCRAX12axCXGZJPQQYdQbuTb9Wvol9WPb+W0cKz6mdUgiAN1y1y1s+WQLR3KO8Jc//YVXXnmF1157jdtvv5233nqLDRs2sHz5ckwmE506dWLy5MlMnz6dO++806/rGwwGPv74Yx544AH+/ve/s3XrVvbv309kZCSdOnVixowZ3HbbbUycOLHe5z/00EM899xzeDweIiMj+da3vlXveXHp6ezYs4eXXnqJFStWsCw7m8TERG689VZe+sEP2LBuHaxe7dsSNC0NqquhgRnv/vrfBx4gMSqKZfv3s3DXrpolcn//4Q/BavWtBbdafR+Xk3hMjLS2RYeR/dBD1MWqi2w6u4mLVUE+OUi0CYvBwtC0oWSlZIX3BkBu9zfJ3eUCj6f+D6/Xl4h1Ot/4tl7vWyJmNPrWf5vNvsQt49sigEhCD3HHio+x7fw2Kp2VWociNGDWmxmUOoiBKQMx6mXNshChTBJ6GHB73RwqOER2frbs4hYmTHoT16RcwzWp10jpYCHChCT0MOLxesgpzGFP3h6qXG0wU1cEHJPexMCUgQxKHSSJXIgwIwk9DHlVL0eKjpCdl02Zo0zrcEQbMOqMZKVkMTh1MGaDjOsKEY4koYcxVVU5VXqK7PxsmTwXpGLNsWSlZNEnsY+0yIUIc5LQBQC5FbkcKjzEyZKTeNR2qoYl2oSCQpfYLmQlZ9ElVioFCiF8JKGLWuxuO0eLjpJTmEOJvUTrcMQVzHozfZP6MiB5ADHmGK3DEUIEGEnookH5lfnkFOZwvOS4VJ/TUEpkCv2S+tEroRcGndSCEkLUTxK6aJLL4+JY8TFyCnMoqG56X2zReknWJHrG9yQzPpNos2yXKYRomiR00SyVzkpOlZ7iVOkpcityURvccVo0V7wlnsyETDLjM4m1xGodjhAiyEhCFy1md9s5U3aGU6WnOFd+TrrlWyDeEk/3uO5kJmSSEJGgdThCiCAmCV20CbfXzbnyc5wqPcX58vNSuKYBkcZIMmIyyIjOICMmQ7a9FUK0GUnool1UOCrIq8wjtzKXvMo8Su2lWoekCZPeRHp0OunR6XSO6UycJU7rkIQQIUoSuugQdrfdl+ArfAm+yFaEV/VqHVab0it6EiISSLQmkmRNIsmaRLI1GUVRtA5NCBEGJKELTXhVL2X2MkrsJZTYSmo+lznKgiLRm/QmkqxJJEb4kneiNZE4S1x4b00qhNCUJHQRULyql3JHeU2Sr3ZV13zYXDaqXdUdUsnOqDMSaYok0hhJlCmKaHM00aZoYswxRJujZexbCBFwJKGLoOP0OGsleIfHgVf14lW9eLwe32fVU+eYXqfHoDNg1Bkx6o11Pht0Bkx6E5HGSNk7XAgRdCShCyGEECFABvyEEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgBktCFEEKIECAJXQghhAgB/x8cHYhJvKx/fAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbAAAAFgCAYAAADacCwFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABlW0lEQVR4nO3deXxU1fn48c+dfcm+koUECPuOLCqLgCIILqgo7kVr+7PVUmutS6u129cuWlutbdVWRVyo1qVuiArIoiKLIDuyBhKSkJVsk9nn/P4YEg3ZwyR35s55v168lJk7d56EZJ57z3nOcxQhhECSJEmSIoxO7QAkSZIkqTtkApMkSZIikkxgkiRJUkSSCUySJEmKSDKBSZIkSRFJJjBJkiQpIskEJkmSJEUkmcAkSZKkiCQTmCRJkhSRZAKTJEmSIpJMYJIkSVJEkglMkiRJikgygUmSJEkRSSYwSZIkKSLJBCZJkiRFJJnAJEmSpIgkE5gkSZIUkWQCkyRJkiKSTGCSJElSRJIJTJIkSYpIXU5g/fr1Q1GUZn/MZjPZ2dnMnz+f999/vyfiPCONMR89elTtUEJq7dq1KIrCjBkz1A5FkiSp13X7DmzKlCksWrSIRYsWMW/ePAwGA++++y6XXnopP/3pT0MSnFYTjyRJknTmDN194fe+9z1uvvnmpr/7fD7uuusu/v73v/PXv/6V6667jokTJ4YixjO2evVqvF4vWVlZaociSZIkhUjI5sAMBgOPPvoocXFxALz33nuhOvUZy8vLY+jQoRiNRrVDkSRJkkIkpEUcFouFQYMGAVBaWtri+dWrV3PllVeSkZGByWQiLS2NK664gi+++KLZcS+88AKKonDs2DEA+vfv32zObe3atUDzOaCGhgYeeughhg0bhs1mo1+/fk3n62go8o033uCiiy4iNTUVk8lEVlYWN954I3v37m123EcffYSiKAwbNqzN74HP56NPnz4oisKOHTuaPed0Onnsscc455xzSEhIwGKxMGTIEO69914qKyvbPOeLL77IxIkTsdlsJCUlcdFFF/Hpp5+2ebwkSVI06PYQYltqa2sBSE9Pb/b4z372Mx577DF0Oh0TJkxg2rRpFBQU8M477/Dee+/x73//m1tuuQWAgQMHsmjRIt544w0cDgcLFiwgJiam6Vx9+vRpdm6Xy8WMGTPYu3cv5513HmPGjGk3ITTy+XzccMMN/Pe//8VsNjN+/HiysrI4cOAAr7zyCm+99RZvvfUWF110EQAXXngh2dnZfP3112zcuJFzzjmnxTlXrFhBaWkpZ511FmPGjGl6vLi4mIsuuohdu3aRlJTExIkTiY2NZdu2bTz66KO8/vrrrF27ltzc3Gbnu/POO/nb3/6GTqdj6tSpZGZmsnPnTmbMmMHixYs7/BolSZI0S3RRbm6uAMSSJUtaPLd3716h1+sFILZs2dL0+L/+9S8BiIEDB4odO3Y0e826detEbGysMJlM4sCBA62+V35+fquxrFmzRgACEKNHjxYlJSXtxnz6eX7xi18IQJx99tniyJEjzZ57/fXXhV6vF4mJieLkyZNNjz/wwAMCELfddlur73XFFVcIQDz55JNNjwUCATFlyhQBiFtvvVXU1tY2Pef1esXdd98tADFz5sxm53r//fcFIOx2u1i/fn2z537/+983fe3Tp09vNRZJkiQtC0kCq66uFh999JEYOnSoAMSDDz7Y9Jzf7xeZmZkCEF9++WWr53zkkUcEIO6+++5W36szCez0D/iOzlNZWSmsVquwWCzi+PHjrb7u9ttvb5GMDh06JAARHx8vnE5ns+PLysqE0WgUZrNZVFZWNj2+YsUKAYixY8cKr9fb4n38fr8YOXKkAMSuXbuaHp81a5YAxH333ddqfGPHjpUJTJKkqNXtObBbbrmlaU4qISGBOXPmcPDgQV5++WV+97vfNR331VdfUVxcTF5eHuPHj2/1XI3rmDZs2NCtWNLS0pg2bVqXXrNmzRqcTidTpkxpszqxtbjy8vI477zzqKmp4X//+1+z41955RW8Xi/z588nKSmp6fHly5cDsGDBAgyGlqO2Op2O8847r9l7+Xw+PvvsMwBuvPHGVuP7zne+05kvVZIkSZO6PQc2ZcoUBg4cCEB5eTmffvopdXV1/PCHP2TQoEFMmjQJgCNHjgBw+PBhFEVp95zl5eXdiuXbBRud1RjX6tWruxzXd7/7XdavX8+SJUu47rrrmh5fsmQJQNNc3unv9ctf/pJf/vKXnXqvyspKXC4XECxiaU1bj0uSJEWDkK0Dq6mp4YorrmDNmjUsXLiQvXv3YrPZCAQCQLDwYs6cOe2eMyUlpVuxWK3WLr+mMa6BAwcyZcqUdo8dOnRos79fffXVLF68mNWrV3P8+HGys7PZtm0bO3fuJCsri9mzZ7f6XlOnTiUvL6/d9xoxYkRXvxRJks7QwYMHeeKJJ/jkk084duwYfr+flJQUMjIyOPvss5k5cyYLFixQO0zNmjFjBuvWrWPNmjVd6iwUsirE+Ph4XnvtNYYOHcqxY8f4y1/+woMPPkjfvn0BSE5O5oUXXgjV252xxriGDBnS5bhsNhsLFy7kueeeY+nSpTzwwANN51i0aBE6XfOR2cb3mj9/Pj/72c869R7JycmYzWbcbjdHjx5tNbHJDiWSdObeeustrr/+etxuN8nJyUyZMoXU1FROnjzJ9u3b+cc//sGrr74qE1gYCuk6sNTUVB588EEA/vznP1NdXc3EiRNJSUlh79697Nmzp0vnM5lMQHA+KNQuuOACTCYTa9eupaysrMuv/+53vwvA0qVLcbvdLFu2DKDZXWmjuXPnAvD6668jhOjU+Q0GQ9Od4SuvvNLqMS+99FJXw5Yk6VtKS0tZtGgRbrebu+++m+PHj/Pxxx/zyiuv8MEHH1BcXMyXX37J97//fbVDlVoR8m70t99+Ozk5OdTU1PDYY49hNBr51a9+hRCCK664oqkw4dv8fj+ffPIJGzdubPZ4dnY2QJcTX2ekp6ezePFiHA4Hl156Kbt27WpxjNvt5t133+Xrr79u8dzkyZMZMmQIBw8e5L777qOyspKpU6c2LeT+tvnz5zNx4kQ2b97MLbfc0upc38mTJ3n66aebJeuf/OQnADz55JMtClweeeQRtm3b1tUvW5Kkb3n//fepr68nMzOTP//5z1gslhbHjB8/nj/84Q8qRCd1qKtli+2tA2v0/PPPC0DExsY2lZPfc889TSXvI0aMEPPnzxfXXnutmDFjhkhISBCAeOqpp5qd5+9//7sARExMjLjyyivFrbfeKm699Vbx9ddfCyG+KaPvqIy8rXJ8r9crrr/+egEInU4nxo0bJxYsWCCuueYaMWXKFGG32wUgVqxY0ep5//jHPzZ9TYB4/vnn24yhqKioqezdbreLyZMni2uvvVZceeWVYuzYsU3r504vzb/jjjua4psxY4a47rrrxIgRI4ROpxN33nmnLKOXpDPw8MMPC0CMGTOmy6/1er3iueeeExdccIFITk4WJpNJZGVliQsuuED87W9/a3bsokWLmj43d+3aJRYuXCj69OkjdDqd+NWvftXsnP/+97/F9OnTRWJiojCZTKJfv37iBz/4gSgoKGgzlqKiInHXXXeJoUOHCqvVKmJiYsSECRPEk08+2erSnW/Hc+TIEXHjjTeK9PR0YTKZxIABA8QDDzwgXC5Xm+/35Zdfiuuvv1707dtXmEwmkZiYKGbPni2WL1/e5msKCgrELbfcIvr06SPMZrMYOHCg+MUvfiEaGhrE9OnTBSDWrFnT9je8FT2SwHw+nxg+fLgAxP3339/0+Oeffy5uuOEGkZubK8xms4iNjRWDBw8Wl19+uXj22WdFVVVVs/P4/X7xhz/8QYwYMUJYLJamRNH4RZ5pAmv0wQcfiCuvvFJkZWUJo9EoEhISxLBhw8S1114rli1bJhwOR6uvKy4ubko8drtd1NXVtRuHy+USTz/9tJg5c6ZITk4WBoNBpKWlibFjx4o77rhDfPTRR62+7vnnnxfjx48XFotFxMfHi1mzZok1a9Z0+uuPdj6/T/gDfrXDkMLQSy+9JACh1+vFqlWrOv266upqMXXqVAEIo9Eopk+fLq677joxc+ZMkZqaKk6/N2hMGN///veF2WwW/fr1EwsXLhSXXnqp+POf/yyEEKK2tlbMmDGj6aJ9+vTp4qqrrhJDhgwRgEhOThbbtm1rEcu6detEYmKiAES/fv3EZZddJubMmdP02OzZs4XH42k1njvvvFPExcWJ3NxcsXDhQjFr1ixhtVoFIC6//PJWv/bHH39c6HS6prWtV111lZg6daowmUwCEL/5zW9avGbfvn0iLS1NACIjI0NcffXVYt68ecJqtYpzzz1XnHvuud1KYIoQnZyUkSSVBEQAh8eBw+vA4XHg9Dlxep1N/3X73fgDfnwBH34R/K8v4MMf8OMX/qbzKCjoFB16nR69okev0wf/rugx6o1YDVasRmur/7UZbZgNZhW/C1JPqK+vZ+jQoRQVFaEoCtOnT+eCCy7grLPOYuLEiaSmprb6ugULFvDWW28xbtw43nrrrWZLeXw+H8uXL2f+/PlNj918880sXboUgPvvv5+HH364RbHXDTfcwLJly7jkkkt47rnnSEtLa3ru8ccf56677mLQoEHs27cPvV4PwIkTJxg5ciRVVVX84x//4Lbbbms6b2VlJQsXLuSTTz7hN7/5DQ899FCr8TzwwAP85je/aTrn7t27Oeecc3A4HGzYsIFzzz236XUfffQRc+fOJTk5mTfffLNp/SrArl27mDdvHsePH2ft2rVMnz696blJkyaxZcsWFi5cyNKlS5uGagsKCjj//PM5fPgwQJerEGUCk8JGvaeealc11a5qTjpPUu2qpsZdQ4O3Qe3QADDpTSRYEog3xwf/awn+N84ch0EX8raiUi/Zv38/ixYtYtOmTS2eGzt2LLfddhvf//73mz7gd+zYwdixY7FYLBw6dKhT2zQ1JozBgwezd+/epnM12rdvHyNGjCAjI4Ovv/6a2NjYFue4+OKL+eCDD3jvvfe45JJLgGAy/NOf/sSPfvQjnnzyyRavKSoqon///iQkJFBaWtq05rUxnvHjx7Nly5YWa2F/+MMf8vTTT/Pb3/622drVc845h02bNvHGG2+0WpX5+uuvs3DhQhYsWMAbb7wBwOeff87UqVOx2+0cO3aM5OTkZq95++23ueKKK4CuJzD5Wyf1uoAIUNFQQWl9KRUNFZx0naTGVYM34FU7tHZ5/B7KHGWUOVpWrcaYYkiyJpFmT2v6Y9KbVIhS6qohQ4awceNGNm/ezPLly9m0aRPbtm2jvLyc7du388Mf/pA333yT5cuXYzKZ+PDDD4FgQunqHoOXX355i+QF8MEHHyCEYO7cua0mLwiulfrggw/YsGFDUwJr7PJzzTXXtPqarKwsBg0axN69ezl48CCDBw9u9vwll1zSaiOHxh03ioqKmh6rqKhg8+bNWK1WLr300jZjhObdixp3D7noootaJC8IFrnFx8dTU1PT6jnbIxOY1OMavA2U1pdS6ihtSlrfHtrTgnpPPfWeegpqCpoeS7AkkGZPI9WWSpo9jWRbMjol5IW/UohMmjSpqYOQEIKvvvqKRx99lFdffZVVq1bxxBNPcM899zRt83R6g4POaKtrUGO3nueee47nnnuu3XN8u4q58XWdaaVXXl7eIoHl5OS0emzjvo6N3YAA8vPzEULgdDoxm9sfTv92jMePHwfa7hykKAr9+vVrsf1UZ8gEJoWc2+emsLaQwppCSupLqPfUqx2SKhqHQw9UHgDAoDOQGZtJ37i+ZMdlE2+JVzlCqS2KonDWWWfxn//8h4aGBt59913efvtt7rnnnjM6b1tdgxq79YwdO7bZNkytOfvss1u87qqrrsJut7f7utbufk6fh2tP43vFxMSEzaJumcCkkKhoqKCwppCCmgLKHGUI5NTq6XwBHwU1BU13aXHmOLLjsukb15fM2EyMerljeDiaPXs27777LhUVFcA3dy2trQ/trsZuPVOmTOHvf/97l17XuBZ1woQJIYunrfeCYHJ//vnnO538GodZ2+sc1HhX21VyPEPqFn/Az9Hqo6w7uo6Xd77MW/veYkvxFkodpTJ5dVKtu5a95Xv56PBHLN2xlPcPvM/e8r24fK6OXyyFRGdq2AoKghccjY0VGje4bezUEQqN3XrefffdZsN2nX3df//735DE0Z7MzExGjx5NXV1d0zxgZzRWI3744YdUVVW1eP7dd9+lurq6WzHJBCZ1SXFdMeuPreelnS/x8eGP2V+5P2yqBCNZQAQorivms4LPeHnny6w4uIIDlQfw+D1qh6Zp//znP1m0aFGrWzkJIXjrrbea7oiuvfZaIDjMN3/+fJxOJ/Pnz29KcI18Ph/vvvtul+IYN24cCxYsoLCwkCuvvLLVuxWHw8Err7xCaWlp02P33HMPCQkJ/OUvf+Gxxx7D42n585Kfn8/LL7/cpXja8n//939AcMeN9957r8XzQgg2bdrExx9/3PTYtGnTOOuss6ivr+eOO+7A7XY3PVdYWNjp/rCtkUOIUoeqnFUcrDzIoapDOLwOtcPRvIAIBOcQawvRK3py4nPIS8ojNz4Xva5lBZvUfV6vlxdffJEXX3yR1NRUxo0bR0pKCtXV1ezdu7cpkdx4443ceuutTa9bsmQJ8+bNY+PGjQwaNIjJkyeTmZnJiRMn2LVrF+Xl5Z3ue/rtc1ZXV7NixQqGDBnCmDFj6N+/P0IIjh49yo4dO/B4POzbt4/09HQgeFf4zjvvsGDBAn72s5/xyCOPMHLkSDIyMqipqWHfvn0cPnyYs88+u819Bbvi0ksv5YknnuDuu+/msssuY+DAgQwZMoT4+HjKy8vZsWMHZWVl3Hfffc125XjppZeYMWMGr776KuvXr2fq1Kk0NDTwySefMHr0aFJSUvjiiy+6HI9MYFKr3D43+yv3c6DyAFXOlrf9Uu/wCz/51fnkV+dj0psYmDSQ4anDSbImdfxiqUO33nor/fv3Z/Xq1WzatIm9e/dSWlqKwWAgMzOT6667ju985ztNw4aNEhMTWbduHc8//zzLli1j+/btbNiwgbS0NMaOHcvll1/e5VhiY2P5+OOPee2113j55ZfZunUr27dvJy4ujoyMDG644QYuu+yyFlsynXfeeezZs4e///3vLF++nC1btuB2u0lLSyMnJ4cbb7wxpEUXP/7xjzn//PN58sknWbNmDatXr0an09GnTx/GjRvHxRdf3OL9hg8fzpdffsmvfvUrVqxYwdtvv012djaLFy/moYceYt68ed2KRS5klpqpclaxu2w3h6oO4QuEfhcA6cyZ0NHf1odRCYNIrBUojgZwOoN/GhqCf9xuECL4B9r+7+mPmUxgs33zx25v/f9b2VlcknqbTGASARHgaPVR9pTtoaS+RO1wpFPsQk8iRuIDOmL8Aqvfj9HnQ3eqnBlAj54Yp4nYcg96Ty+urWtMdHY7xMRAamrwT3IydKE0W5LOhExgUczlc7GvfB97y/fKuS2VJWEkJWAg1g82fwCL19ssUXVEAWw+K3GVfkx1KhZ+6PWQlARpad8ktYQEaKXbgySdKZnAopDL52Jn6U72lO0J+/ZNWqQISFZMpPr1JPgFdrcXfSB0d0+WgJm4agVrVZiU4xuN3ySz1NRgcouJUTsqSQNkAosiMnGpJwkjGQEjCd4ANo+nS3dX3WUJWEgsD6h7R9YWqxX69oXcXMjODiY5SeoimcCigNvnZmfpTnaX7ZaJq5coAtIVMxk+PYluD0afegUxdp+VhFIvBmeYFuXo9ZCREUxmubny7kzqNJnANMzj97CzdCe7SnfJxNUL9EIhEzN9fArxLk9IhwXPlALEeGzEl7jQe3v+7u+MpKRAXl7wj0xmUjtkAtOo/RX72Vy0GafPqXYompcuzOT4FBKdHhQR3slBh444p4XYEie6QAT86qelBRPZgAHBikdJ+haZwDSmtL6UDYUbKG8o7/hgqdusQk//gIl0tw+TN/LubvXoSagzYS91EjH1gZmZMGIE9OsnqxolQCYwzXB4HGwq2sShqkNqh6JZioC+WMjyQpwzTCr8zpBRGEmo0WOriKCvJzYWRo6EIUOC69GkqCUTWITzB/zsLN3JVye+kp0zeohRKAwSVjIaPBj82vweWwJmkkr8GMO10KM1RiMMHRpMZm3sYixpm0xgEexE/QnWHV1HjbvrW3FLHbOhZ7DfTGqDq1fK3tWmoJBYbyH2RITNmypKcFhx1Cjo00ftaKReJBNYBPL6vWwu2sye8j1qh6JJCRgZ5DWQ2OBCicK9zSwBC8lFXgzu8Kmi7LTU1GAiGzBAtrSKAjKBRZjiumLWHl1Lvade7VA0JwUTgzw6zcxvnQkdOhJrzcSURdjdWCO7PVjwMWKEXCStYTKBRQhfwMfmos3sLtutdiiaEysMDPcaSXBG6Id1D7L5LSQd94T/2rG22GwwcSIMHiwrFzVIJrAIUOYoY03+GjnXFWIWdAz3WUh1yB2l26NHT1K1MbIqFU+XkgKTJ8s5Mo2RCSzM7SrdxaaiTQTCfIFsJDGgMMRvIbPBHRXFGaFi91lJOu5G54vg79mAAXDOObLDh0bIBBamPH4P646uI786X+1QNGWgsNLP4UHvj8AChTCgx0BylSF8Ot13h14Po0fDuHFyY84IJxNYGKpoqGDVkVXUumvVDkUz4oWR0W4dNrdb7VA0Ic5tI6GwIXK6eLTGZoNJk4LzY1JEkgkszOwr38eGwg34hbxDCAUdMMJvI6PeGZUl8T3J5reQfMwdGT0V25OaGpwfS09XOxKpi2QCCxO+gI/1x9bLVlAh1AczwxoCEdmrMFKYhInU4/7IXDN2uoEDYcoUMJvVjkTqJJnAwkCDt4EPD31IRUOF2qFoggkdo71mkhtkWXxv0KMntUyPuTYMN87sqpgYmDkzuD+ZFPZkAlNZlbOKDw99KBcmh0imMDPM4ddsz8JwpaCQXGPBXq6BiwZFCRZ4nHWW7OYR5mQCU1FhTSGrjqySm02GgA4Y47ORJtd0qSrBaSO+SCP/BunpcP75slFwGJMJTCX7yvfxeeHncn1XCCRgZGyDgtmrgSEsDbD7rCQfc6Jo4ZPFZIJp04KbakphRyYwFWw8vpGdpTvVDkMTBgorA+pcKPLHOKyYA2ZSC72R24LqdIMHBws8ZF/FsCITWC8KiACf5H/CkZNH1A4l4hlQGO+xyP6FYcyAgdQSBZNDI0Pk8fFwwQXBtlRSWJAJrJf4A35WHllJQU2B2qFEvBgMTGjQySHDCKBDR2q5EUuNRhaQ63TB5sBjxqgdiYRMYL3CF/Dx8eGPOV57XO1QIl4fzIys86EPaGDdUZTQXBIDyM6GWbOCc2SSamQC62G+gI8PD31IcV2x2qFEvEEBKwPq5JBhJNJkEktMhLlzZWNgFckE1oO8fi8rDq3gRP0JtUOJaIqAcX6b3PYkwmkyidlsMGdOsB2V1OtkAushHr+HFQdXUOooVTuUiGZCx0S3iRhXBHc/l5poMokZDMHijtxctSOJOjKB9QCv38vyg8spc5SpHUpEs6Hn7Aad7GWoMTp0pJUbMWspiSkKnHsujBypdiRRRSawEAuIACsOrqCorkjtUCJarDAw0QFG2RJKk3ToSD+hx1SvsYuTsWODW7RIvUI2+gqxNflrZPI6Q4nCyNkOIZOXhgUIUNYngNeqsQ0lt2+Hzz4DeV/QK2QCC6ENhRs4fPKw2mFEtBRMjK8PyB2To4AfP2VZ4DPr1Q4ltPbuhU8+gYBGupCEMZnAQuSrkq/YXbZb7TAiWh/MjJNrvKKKDx9l2Tr8Ro19FB0+DB99BD45itCTNPZTo46vK75mS/EWtcOIaNnCwuhaDzp51Rp1vIqXshwDAZ2idiihVVgIH3wgk1gPkgnsDB2rPsanxz5VO4yIlomZ4bVu2ZA3inkUD1U5FrXDCL0TJ2DVKjmc2ENkAjsDJ50n+ST/EwTyg7e7UoWJkbVeFPk9jHoOg5PaDJvaYYReQQGsX692FJokE1g3efwePj78sdyM8gwkCiNj6v0ock806ZRqewOueLPaYYTegQOwebPaUWiOTGDdIITgk/xPqHHXqB1KxIoVBsY7ArJgQ2pGABWpPvxaq0yEYIn9rl1qR6EpMoF1w5fFX8ptUc6ADT0TG5Cl8lKr/PgpzzIgNFbTAcAXX8ChQ2pHoRkygXVR/sl8vjrxldphRCyz0DGpQY9RVmZJ7XDr3Jzsq8H5MIC1a+G43FopFGQC64KTzpOsPbpW7TAilg6Y5DbKjSilTqkzNeBIs6odRugFArByJVRUqB1JxJMJrJO8fq8s2jhDZ/ms2NwaauAq9bjKOBeeGA1uGun1wooVUFurdiQRTSawTtpQuEEWbZyBwQEryQ65GaXUNQJBeXpAe506AJzO4EJnp/y96C4N/lSEXv7JfPZX7lc7jIiVIcz0lzspS93kU3xUZmvwLgyCd2ArVoAsaOoWmcA60OBt4NMC2Wmju2KFgRH1smBDOjNOvYuaLI0WdVRUBKsTpS6TCawDa4+uxeWTuwF3hwGF8U6dXOslhUSNtQFPjFHtMHrG3r2Qn692FBFHJrB27CrdxfFaWe7aXRM8FllxKIWMAKrSNdx0bP16qK9XO4qIIhNYG6qcVWwukq1fumtQwEq8nJyWQsyteKjP0GBpPYDbHdxHTDa17jSZwFoREAHW5K/BL+TQV3fECyP96+Wwq9Qzqu1u7W2C2ejECdi6Ve0oIoZMYK3YXbabSmel2mFEJB0w1qXIrVGkHhMgQFWmRufCAL76CkpK1I4iIsgEdpp6Tz1bi+UVUHeN8lmxeOS8l9SznHoXDSka3D8MgkOIn3wSHFKU2mVQO4Bws6Fwg+y20U2ZmOkjFyt/Qyj4FTMBxYQQRgIYCAgDAaFDnLp2/OZGVUGgoIgAOl0ABT86JYBO8Qf/Hw96nCjIYe1GVQleLNU6dD4NbsfjcAR7Js6Zo3YkYU0msG85Vn2Mo9VH1Q4jIlnQMaw++j5chYCAYseLFX/AgtdvxBsw4PXq8fbA8je9DowGgcHgx6D3Y9R5MCgOjNRFXXLz46c6y0rSMY1eNB07Bnv2wIgRakcStmQCO8UX8PF54edqhxGxxnjMGPwa/SD5lgBWPCIGb8COy2fC5dZ1arf4B/74AAXFBbzw1xfQ67tfgOAPgN+jgMdA8NfXDMQCfTAawWL0YTJ4+Pkff8/WXXv4ZOlvsZi0O19UZ3RiizdjqdHocNvGjZCRAUlJakcSlmQCO2Vr8VbqPXINRndkCQsJYVwy/7+VG9m86xC7Dx5jf34xHq+PR+9ZxNUXTW71+L8ufY8nXny/zfO9+tSrZKQld/r9t+/ZzmdbPuOhux5qlryWvLaEF/77QpuvmzpxKg/f/zAAX+3+ip/86ifNnjcZTSQnJnPW6LO4acFNZKRl4PUGE9s1l9/GB2sW8eSrO/jBjQsw6+owUo2C9obbqlICZNSCosW6Ib8fVq+GBQtAJ0sWTicTGMFtUnaVyZ1Su0MvFIY0hHerqD8veZei0kqS4mNITYqnqLT9CtMAweKAiy+YQ2pyRovnY+wxXXr/5/7zHH1S+zBz8sxWn59+znT65/Rv8XhOVk6Lx4bkDeHc8ecCUO+oZ/ue7SxftZz1G9fz9B+eJjszG4C+mX2ZMnEKS1//D5fOXoDVkoZOScNq9WMz1mNVKlHQRrGNV/FSm2kjvqhB7VB6xsmTsHs3jB6tdiRhRyYwYFPRJgJCe1emvWGk34rRF94fHH+6+yb6ZaeRnZ7MP//zIY88+78Wx/iIpcGfRL3TSp0r2JfuwvPmMm7kuDN67/yCfHbu28lNC25C18YV9PRzp3PB1As6db4heUO45Zpbmv4uhOD3T/6ej9d9zEtvvsTPF/+86bkLp1/I+k3r+eSzT7h41sUEBDga9DiIRyEeq9WP3eTAolSiI7KH4GqsTmxWA0ZneF9Mddu2bTBoEFg1uoi7m6L+nrS4rpiCmgK1w4hIicJIH0d4Jy+AqeOHkZ3ecsgvgJW6QDYnGoZQVJ3FyTpryAsvVqxZAcCMyTNCe+JTFEXhiouuAODrQ183e+7cs87FYrawYu2KFq8TQINTT3lNHIXV/anwDMJNKgilR+LsaQJBTbp25/rweGDLFrWjCDtRfwe28fhGtUOIWKPckXv9U+9Po7A6t91jduzdwb6D+1AUheyMbMaPHo/N2rWO6Ft3bcVqsdK/b8shwlBTlObJx2g0MnjAYPYc2IPT5cRqafvq3dGgx9GQjMGQTJzNiV1Xho7wnddsjcPgJF7Ld2H798Pw4ZCSonYkYSOqE9jhqsNUNMhtvbtjUMCK1R05H3BC6HGSTr0n+MvvdndcCbjktSXN/h5jj2Hxdxdz0YyLOvWeDc4Gjhw7wojBI9qtPFz3xToKilqOAlx/xfWYTeZ230MIwTsfvQPA0EFDWzw/JG8IO/ftZN/BfZw16qwOY/b5oKrWShW5xNj8xJoqMVHV4evCRU26kZSjGk1gQsCGDXDZZWpHEjaiNoEFRIAvi79UO4yIZEJHriMyCgCEMFAvMqmpt+EPgM/f8RBZXr887rvjPsaNGEdSYhJV1VV88eUXPP/a8/zx738k1h7LlIlTOjxP5clKAoEAifGJ7R63buM61m1c1+Lxqy65qkUC2394f1NidTQ4+GrPVxzKP0RcTBw3LbipxTmSEoLl1+WV5R3Ge7r6Bj31DWlYzKnEW09ioazL5+htmr8LO3ECDh+GvDy1IwkLUZvADlQeoMZdo3YYEWmYz4w+EN53X0IYcIhMqk8lrq447+zzmv09Iy2DK+ddSW52Lnf/9m6eXfZspxJYTV3w56ujqsWH7nqo00Uc+w/vZ//h4O7gRoORlKQULpl1CTctuIk+aX1aHB8bE9sslu5wuRVc7iTMpkQSbNWYRSlKGE+VafouDGDTJsjNBUPUfnw3icrvgD/gZ1vJNrXDiEg29KQ3hG+neYEehz+Taoe9y4mrI+NHjyczPZMjBUdwNDiw2+ztHt949+QJ4Z5ol82+jLtvu7vTx7s9wepCi/nM+wa6PQqlnkTMpgSSbOVhO7To0DcQbzVq9y6svh527IDx49WORHWROwt/BvZX7peLlrtpuNcUtp3mnaIPxXWDqKwLffJqFB8XD4DL3XEST4hLAKC2vrZngumEuvq6ZrGEgtujUFKdRqVnEH7aT+KqUBRtVyRCMIHJzS+jL4EJIdhZulPtMCJSAkaSG8Jv6NBLPGXOwZTVJODrwXaATpeTo4VHsVqsTYmsPSlJKcTHxlNYVNhzQXWgsThkQM6AkJ+7vkFPUXVfanz9EZhCfv4z4dA34LVqeIDJ5wu2mYpyUZfAjlYfpdat3hVxJBvWicq93iQwUe0bQHF1Bs4QlfQ3OBsoLG6ZcNxuN48+9SgNzgZmTJ6BQd/xh6OiKIwaNoqSshKqa6pDEl9X7Tu4j+TE5KYOHaEmgOp6M0W1A2gQLefgVBMNd2FHjkT9vmEavkRp3Y7SHWqHEJHSMBHnCp+5L6foQ2VdQqeGCt9f9T679gVbhR0pOALA8tXL2b5nOwCjho3iklmXAMFih5t+fBNDBw4lNyuXpIQkqmqq2LpzK+WV5QzIGcAPv/PDTsc57expfLb5M7bs3MKF0y7s2hd5hopOFFFSVsL8OfN7/L38ASivScBujSXJfDws1pBpfi4MggUdl1+udhSqiaoEVlJXQpkj/EuBw9EQV3iUnQUwc9KTQ31D5+8Gd+3bxYdrP2z+2Ne72PX1N/0vGxNYXEwcl8+5nH2H9rFx20bqHHWYTWZys3NZMG8BV869ErO5/bVZ3zZz8kz+seQfrFy3stcT2Mr1K4Fg4UdvcTj1ON25JMXWYaNI3WpFRaEmzUDKMQ0nsLKy4F1YRsuendFAESJMZ+R7wIeHPpRto7ohDRPjatRf9+UijYq6JPwRtu3Vc/95jlf+9wrL/r6s1VL3nuDz+7jhRzeQkZbB4795vFfe83RWS4AUS6Gqd2MKkFGk4XVhAH37wty5akehiqiZAzvpPCmTVzfledT9MRFCz0lfHqXVkZe8AK67/DriYuJ46c2Xeu09P1rzEaXlpdy+6PZee8/TOV06imtzgz0WVSKAujSNz4UVFkJVeC5p6GlRk8Bk5WH3JAojcU715r782Ch1DqS2PnI/hGxWGw/8+AH6pPXB31sZWIF7fnAPgwcM7p33a4M/ACeqk6nx91OtUbDD6CagC48h8B6zfbvaEagiKoYQPX4PL+14Cb+IwMt3lU30WklSqXTeTSrltck9tqZL6l1Wc4BkayF6FYYUk+usxJSqX1jSYxQFrr0WYmPVjqRXRcUd2MHKgzJ5dYMNvSrJSwioC+RyolomLy1xunWU1OXipf3ekD2hPkbjv/9CwK7o25Q3KhLY/sr9aocQkQb6O19tFyoCPZXeQVTVyo37tMjvh5LqdJy9vGbMrXjwxETuMHSn7N8PXq/aUfQqzSewioYKuWVKN+iFQlpD7+7SGxAmypwDcXShRF6KPAIoq0mgLtD+fmyhVp+k8VVDXi8cOKB2FL1K8wlsf4W8++qO/ljQB3pv2MWPndKGAbjcGp9sl5pU1Vo56c1D9NLHkMMUBcUce/aoHUGv0nQC8wf8HKo6pHYYESnD3XuTT17iKanriye6Rj8koNZhpMKdhxA9f9cdIIAz+cy78oe16mo4flztKHqNphNYfnU+bn/vDoNpQRwGbO7e+b65SeZETUZEru+SQqPBqQ8mMXo+iTliNF90Dbt3qx1Br9F0ApPDh93Tz9c7k90ekiirSSUQBZ8pUvsaXDrKXXkI0bPzVC69C79R0x97wYXNdXVqR9ErNPsv6fQ6Ka4rVjuMiKMISHX1/FiehyRKa9Jk8pKaOF06ylx5BOi5CygBNCT3fnVtrxIC8vPVjqJXaDaBHas5hkB+OnZVJhYM/p7tG+chQSYvqVUut0K5c0CPDic67FGwuPDIEbUj6BWaTWBHq4+qHUJE6uvt2SotLwmU1vaRyUtqk8utUOEe0GOtp9yKW9ubXUKwS73DoXYUPU6TCczr91JUW6R2GBHHLHQ92vfQjz2YvKLgAlg6Mw1OPVW+vJ47f1J47SDdI6JgGFGTCaygpkC2juqGvphRemjYVWCi1NFXtoaSOq3OYaDG379Hzu0yR8HnQxQMI2oygcnhw+5J9fZU8tJR7uoXbV1upBCorjNTL/qG/LxuXRQsai4tBaeGGxijwQTmD/jlvl/doBcKsa6eWft10jsAp0tzP2pSL6msseMmJaTnFIA7QVYjRjrNfaoU1RXhDchL/a7KxIzSAzvr1AVyqXNofMJc6nHltSn4sYX0nC675j7+WpIJLLIcr42eNiqh1McX+uEUN8myq7wUEv4AlDf0DWnLKZepZ5eLhIWSEnCptyFtT9NcApOLl7tOERDv8oT0nH4slNept5W8pD1uj8JJX+iKOjyKR/tdOQIBOHpU7Sh6jKb+9Vw+F1XOKrXDiDhpmEPbeV4oVLpyZH9DKeTqHAYcgeyQnc+l9Xkw0PQwoqYSWEldidohRKSMQGi7HtQFcmXRhtRjKmtj8CpxITmXK7TTauGpqAg8oR1hCRea+pSRw4fdE+8J3VyAm2Sq6jS+ZYWkKgFU1meEZD7MZYyCgq9AINjgV4M0lcBK6uUdWFcZhYIlRFdnAj2V9XLeS+p5bq9CbSDnjM/jw6f9tlIQXBOmQZpJYHL+q3vSCN0cQK0vB28UFHZJ4aG6zoyHpDM+jyuud7YPUlV5udoR9AjNJDA5fNg9yYHQ/Ah4SKK6PgomxKWwUulIO+OhxKiYB6uoQItNSDWTwErrtXmL3NPivWdeKiiEngpHWgiikaSu8Xih7gyrEt16j/Y3XvL7oUp7I1SaSWCVzkq1Q4g4OsAagvmvukBf2edQUk11nRU/9m6/3o8fnz0KhhHLytSOIOS0k8AaZALrqlRx5u2j/NipllWHkooEcNKdeUbn8Fp7bgPNsCETWHiq99Tj9vdMI1otSwnB+q9qT4b2h1+ksOdw6nHR/WFsr0njnelBJrBwJe++uifWf2apx0MS9Q1RUIIsRYSq+qRuF3R4o2B/S6qrNbegWRsJTM5/dYvV1/2adyGgqkGu+ZLCh9cHDro3lOjVR0nfM43dhWkigVU0VKgdQsTRCwXTGVReOMnA7YmCYRcpolTX2RGi66MCPiVKFjBqbD2YJhKYXMDcdfGcwdCfUKh2xIcuGEkKEX8AHKLrd2EBAvjMspAj0kR8AvP6vdS6a9UOI+IknEECayBDdtyQwlZ1vY0AXS+L99qiYD5XJrDwUuepUzuEiBQX6Obwn1CodoSmE3g0c7ldLPj+Ah596tFee8+CogLOv/p8/vfh/3rtPdXgD4Aj0PW7MK8lCu7AnE6or1c7ipCJ+EuOOrdMYN1h62YHDqfSR5N3X8v+t4xnXn4GgH/+4Z+MGDyixTGOBgdLXlvC+o3rqaquIjkxmRnnzmDRwkXYrF3rR/TqO69SU1vDd676DgB/ePIPfLj2w06//v477mfu+XO586E72b5ne9PjiqJgt9npn9Ofi8+/mItmXoSiBC9WcrJyOH/q+Sz971LmTJ/T5ZgjSXW9FXucER2dn+f1maJkQUh9PcTEqB1FSER+ApN3YN1i9XU9gQkBNQ3au/s6UnCEJa8twWqx4nQ5Wz3G6XLy44d+zKH8Q0wcM5ELpl7AwfyDvPruq2zfu52//e5vmE2d6wXpaHDw6juvMnPKTNJT0wGYOmkqfdL6NDtu+57tbN+znakTpzKw/8Bmz53+92suuwarxUogEKC4tJj1m9aza98u9h/Zz0++95Om4667/DpWrl/Jm8vf5KarbupUvJEoEABnIB277ninX+PVa69XYKscDrUjCJmIT2D1Hu3cDvcWHWDwd/02yqOk4vZqq/LQ5/Pxhyf/wMB+A8nKyGLl+pWtHveft//DofxDXH/59dx2021Njz/z0jMse3sZr7//OjdeeWOn3vPjdR/jdDmZM31O02PTzp7GtLOnNTtuyWtLggls0lTmnj+33XNec9k1JCcmN/398LHD/PD+H/L2h29zzaXXkJGeAUBebh55uXm8v+p9brjyBnS6iJ9FaFNNQww2Oyid/JH16jQ4tNAaDSWwiP/plUOIXWfr5nVLnTshtIGEgZfefImjhUe574770OtanwMRQrB89XKsFivfufo7zZ77ztXfwWqxsnzV8k6/54o1K4iLieOsUWedUeztycvNY8yIMQgh+Prw182emzl5JifKT/DV7q967P3DgdcXvOjqLD9+AoaI/0jsmExg4UMOIXadvRvdCvxYcTi1Ncl94MgBXnrzJRYtXES/vv3aPO54yXEqqioYNXQUVou12XNWi5VRQ0dRXFpMWUXHFV519XUczD/I0EFDe+3uRzntFmTEkOD83tZdW3vl/dVU60rs0vFRUYkoE1j4kEOIXWftxj+7w6+t7VI8Xg+//9vvGdhvINfNv67dY4+XBOdRsjKyWn2+8fHG49qzZ/8eAoEAQwYM6WLEXZNfkM+OPTtQFIUhec3fq/Hvu7/e3aMxhIMGlw4fsZ0+PirWgjU0qB1ByET05YbX78Xlc6kdRsSxii7OYwmFugZrx8dFkOdffZ7jJcf516P/Qq9v/0OrviF4kRRja71yy261NzuuPWWVwbu0xISu3Rl05LV3X2sq4igpK2H9xvW4PW4WzFtARlpG83htdkwmE+WV2urK0BaHL4V4Q+dGakTEX9J3gobuwCI6gTl9rVeMSe2zdLHYyqWk0Y2ixbC1e/9uXnv3NW5eeDMDcgb06nvX1gcX3beVDLvrtXdfA06V0VvtDMkbwsUXXMycGXNaPT4uJo6aupqQxhCu6pxm4mMUUDoukw/otVWk1CqZwMKD2ye3UOkOc6Br613qPdopnff5g1WHA3IHcMMVN3TqNY3Jpq07LIfT0ey49jSW2nu8oe0K/tazbzWrQuyI2+PudNl/pPP7wa0kY6bjnqmBTiS5iBcIgMsFlsjfxy+iE5jHr52tAYQQ/P6m32OPt/OTf/yky6/ft2kfj9/xOD96/EeMmjqq3WNNgc7fggUw0tCgnXkBp8vZNFd1wTUXtHrM7T+/HYD/u/f/mHb2NLIzglvWF5UUtXp84+ONx7UnIS4BCBZzqCUQCOBocNAvu59qMfQ2hzcBs7HjBCai4Q4MgndhMoGpK1QJzOv28r9//I9j+45RXliOo9aBLdZGalYqUy6fwjnzzkFv+OZD3O/zs2PdDnas38HRPUc5WXoSFMjon8HkSycz7Ypp6PRdG0zfuHwjBV8XcN+S+5o9fmDrAT7936cc/OogdVV1xKfGM+zsYVx222XEp3zTUHfY2cMYOHYgb/3tLUacO6Ld9zf4O5/A3CJZUxtWmgwmLr7g4laf27F3B8dLjjNl4hQS4hKaFhZnZ2STkpTCrq934XQ5m1UiOl1Odn29i4y0DNJSOi50aRyyLCguCMFX0z3HS44TCAQYkNu7w6dqcjQYSIzXo9D+WHhP34E1eDwM+uUvuXjkSP51U+8vJF+1bx8XPv44y00m5t18c6+/f6hFdAIL1S7Mbqeb9W+up9+IfoyaOoqYxBgaahvYs2EPL/72Rb78+EsW/21xU9lz+fFynrnvGcw2M0MnDmX0eaNx1bvY+elOlv1xGbs+38Udf7mjRflyWwKBAO/96z0GjhvIgFHffKg46508dttjxCXHMXLySGISYziy8wif/e8z9m3cxy//80usMd98mM7+zmz++dN/suXjLZw99+w2308nOp/AGrydr+CKBGazmXtvv7fV5/7w5B84XnKcG668oVkrKUVRuPiCi1n6+lJefP3FZguZX3z9RZwuJzcu6Nwi5gG5A4iLiWPfgX1n9oWcgb0H9wIwZvgY1WLobQEBx8p1rF7/EWs37eZw4QnKq2qJj7UzYWQet10zh3HD+reawGqdTn793nu8+dVXnKitJSM+nqvPOotfXXIJMV28i3n0o4+oqK/nwYuDF1E3v/ACS7/4otOvX7JoETdPnsyMxx5j3YEDmA0G9v/2t+Qmtxw+HvrQQ+wvLUU880zTY7OGDWPqwIHc+/DDzLnppg4LmMJdRCewUN2B2eJsPL72cQzG5t8Ov8/P43c8zt6Ne9mzYU/T0JzZZua6+67j3EvOxWz9Zh7hKudVPHbbY+z6dBfbVm9j/KzxnXr/PRv2UFlcybzvzmv2uKJTuPquq5l+9XSMpm+6a//r5/9i68qtbFu9jSnzpzQ9PnLySGISYlj/5vp2E5giOneVKYQBh4aGD8/EdZdfx+dbPmfZ28s4mH+QwQMGc+DIAbbs2MLQgUO5+uKrO3UeRVGYMmkKH675kLLKMtKSe395wpc7vkSv1zN5wuRef281vfDmx7zw+uvkZqYybfxwkhJiOXq8jI8/387Hn2/niV/cytXTm3dDcbjdTH/sMbYXFjJ7+HCumziRrwoL+fPKlaw7eJD1P/sZFmPnOt/XOp38eeVKrpkwgZykJAAuHzuWfqcln7UHDrDuwAHmjxnD2L59mz13+t/dPh8PvvMOL333u53+Ptw7ezaX/fOfvPrqq9xwQ+fmgcOVTGCATqdrdVGp3qBn3MxxHNh6gLLCbxapJqYlMuPqGS2ON1vNzLp+Fs89+BwHth7odALb8O4GFEVh3Pnjmj1usVmYdcOsFscPGjeIrSu3Uney+TyK3qBn7IyxfPb2Z5QVlpHWt/UPR10n58BcpGhq+PBMWC1WnvjtE7zw3xdYt3EdX+35iuSEZK657BpuXngzZnPnCyLmz57Pik9WsOrTVVx/+fU9GHVLLreLzzZ/xrnjzyUlKaVX31ttg/qP4NXH7uacsYObPb5550FuuOevPPjEMi6dPKnZc4989BHbCwu5b84c/njllU2P3//WW/zpo4/466pV/Hxu+22+Gr20cSP1bjffOeecpscuHzuWy8eObXbcr997j3UHDnD52LHcPLn9i4y81FSWbd7MPbNnMzq74zlYgItGjiQlIYGnn3464hNYRK966OkqxEAgwJ4NewDIzOvc9gyNc2W6TrakEUKwf+t+0nPTscfZOzze5/Wx6YNNKIrC8HOGt3i+cQjy6y1ft3gOuvYP7vRpo2N1Z/188c9Z9+a6VjvRA8TYY/jRLT/i9WdeZ/Vrq/nvM//l9kW3d7mr+7BBwxg1bBQrVq8g0M7FxC3X3MK6N9e12wfxid8+wbo313W6AnHVp6twupwsvGRhl2LWgqlnn8f4sS1HJiaNHsQ5Y4dQU9fA3vzCpseFEDz7+efEmM388uLm86a/vPhiYsxmnv3ss06//5ING0iy2zl/SOgWsf/f/PkEhOC+t97q9GuMej2XT53KZ599xqFDh0IWixoiOoF5A53fKqEzfF4f7z3zHu8+8y7/+dN/+M3C37B7w24mXzqZYZOGdeocG97dAMDws1sml9aU5JfgqHGQMyynw2O9Hi9P3/s0+bvzmXfrPHKGtnxN7vBcAA7vONzqOQxd+Cd3uiL6Bj2s/fA7P6SguIDVn6/utff0+X28/ObLTJk4hTEjomf+69uc/taXhBhPzQXpvlWFeLCsjOLqaqbk5WE/7Q7bbjYzJS+PIxUVFFZ1vCP8SYeDrwoLmdSvX0hbiM0YPJi5I0fy4Z49rNm/v9OvO3d48PPpk08+CVksaojoBBboQjFCZ/i8Pt7/9/ss//dy1r6+ltJjpVx404Xc+EDnJujXv7We3Rt2M2TikA5L2RtVl1UDEJfU/lorv8/PM/c+w65Pd3HhTRdy2Q8ua/W4xvOcLDvZ6vMGOldY4iNWU4uXw82IwSO4+7a7CXShIvRMlZWXMWfGHO64+Y5ee89w0+BuOdRbVFrFZ9v2kZYcz5AB37QLO3hq9+JBaa0PxTc+frATuxx/ceQIASEYn9PxhWpX/fGKK9ApCve99Raik/PbEwYHh1E///zzkMfTmyL6Eruz/1idZbFZeObLZwgEAtSU17Dz0528/Y+3ObLzCIufWNys4u90Oz/dyauPvEpyRjK3/u7WTr9nfXVwcawttv1hqPVvrmfXZ7uYevlUrrrzqjaPs8cHhyEd1a2vtjd0so2UR8R3fJB0Ri6b3fpFSE/J7JPJLdfc0qvvGW7cHh0BmxkdwekHr8/PXX98Ho/Xx/3fvxJFDwJQgBpnsNNPvLX13/u4U483Htee4yeDF5TpcaFvCjA6O5sbzz6bFzdu5PWtW1k4YUKHr0mPD/5+Hz/e+f3SwlFE34GJHiox0Ol0JKYnMv2q6dz4wI0c3nGYD57/oM3jd322i2fue4a45DjueuquZuuzOmKymIDgWrT27P8yODxw/rXnt3ucxxUsbDFaWq+MMnSytN/p1VbvQ0lq5CH4+xkIBPjZIy+weedBrrt4KldeGCyuED2wpUrlqfZNCbae2QX7d/PnYzYYePCdd/D5Ox46SYoNLo+pqOh4cXc4i+wEFuI7sNY0Fkoc2Hqg1ed3fbaLp+99mpiEGH769E9Jze78/kMAMQnBQglHbfv9ydyu4BWj2dZ+tVvjeWITur9+SwhwumT5vKRNHr+NQCDAPY++yDurN3PFrLN5+CffVOMJXfAiL76DO6zaDu7Qvs16qtTe5Q3tvH2jnKQk7pgxg4NlZfzr0087PN55KnZbDyXU3hLRCaw3VJdXAzTrxNGoMXnZ4+z89Omftlm23p7MvEwUnULpsdJ2j5ty2RSu+NEVHVYqNp4na2DrW3/4O5H0/UoMvTg1I0m9qsFt4GePLOXNj7/gsvMn8ud7b25WWKGc6hXa0RxXR3Nk35Z66o6nqgcb6T4wbx4JNhu/Xb6celf7u3RU1QenLlJTu3bBHW4iOoF1ttNFR4qPFDcNvX2bx+Xh9b++DsDIKSObPbf78908fe/T2GJt/PTpn5Kek96t97bF2sgemM2xfcfaLakeMGoAY2eMbRpybEv+7nwABo8f3OrznclLXhFd5fNS9AgEAvz6sUd4a+VGLpkxgb/e/130p7Vd+3YCy0xI4PPDh3G4my/ZcbjdfH74MP1TUuh7alFye0ZlBS8o95e2f6F6JpLsdu6bM4fS2loeW7my3WP3n5r7GjWqc8Vm4SqiiziUTlbUdWTryq2sWraKgWMGkpyZjMVuobqsmt0bduOocTBw3EBmXf/NguITR0/w1D1P4fP4GDJ+CFs+2tLinMmZyUy+tHOdDsbOHMt7z7xH/q588sbktXrMkoeWcGDbAR5+92FSMttegLpv0z5scTYGjRvU6vP+TqQwTyDym3xK0ukCgQB/+sef+HDtR8ydfjaP/2JRi+QF3yQwRVH43pQp/Hb5cn63fHmzhcy/W76cerebX3RyEfOorCyS7HY25eeH5otpw53nn8/f16zhsVWrsJnavtjddKrkfvr06T0aT0+L7AQWojuwUdNGUV1RzZEdRziy6wiuBhfWGCvZg7KZMHsCUy6b0mwIsaaiBp/HB8CWj1smL4DBZw3udAKbevlUlj+7nE0rNrWZwDqjoriCwzsOc/6152M0t17EEejEt8zjbf8uT5Ii0dLXl/Lh2g+xWqzkZOXy5MstC7PmTB5LrvJNu6Z758zhnR07+NNHH/FVYSFn5eSwraCAj/fuZWK/fvzkgtZ3NDidoijMHzOGF774guMnT5KdGNoNTRtZTSZ+femlfP+ll6hrZxhx5datJCYmct555/VIHL0lohOYXglNoUG/4f3oN7xfp48fMmEIz3z5TMcHdlJCagITZ09ky8dbuHLxlVjsLe+A7v7X3R2e57O3P0On1zHzmpltHtOZIUS3J6JHliWpVSfKTgDBHQSeWfbfVo/JSU9h7sBvEpjdbGbdz34WbOa7bRtr9u8nIz6euy+8kF9dcgnWdu5yTveD885jyYYNLNu8mXvntL7RaCjcMnkyf1m1in0lJa0+f7Sigs937+bOO+/EEuFbqiiiN0r5esgXhV+wq2yX2mGERGVJJb+66lfMu3Vei6a+neGodfCLS3/B5Esnc83PrmnzOL2AWbVtn8ePjePVoV9sKUnhxG71k2I+2OJxvdCRfbjnKpimPfoo5XV17P31r0PakaMrHnz7bR5ZuZJ9X39NXl73R3zCQURfapsN2tlRNjkjmZt/fTMWW/euiCqLK5l1/Swu/l7re1016uhX049c/yVpn8fb+kefovTsR+KjCxawv7SUV7/8skffpy0nHQ6eXLOGH15xRcQnL4jwIUSTXltzNRMu7HgFfVtyhua02hvxdEIBoShtbqniE9q5KJCktnh9CgIdymmXdDrRswnsnAEDeOaGG/B3YVf0UMqvrOSuWbNYfPvtqrx/qMkEFoX8Oj0Gv6/V53z+zu1tJEmRzo8NA/XNHtP1woTK/1OxcOKsnBzOysmBzM7trhHuInsIUS/vFrrD30rpcCNfIKKvaSSp01obLu/pO7CwEaONtZ4R/a8l78C6x9vO8gOfX7aQkqKDT7T8/FA62ew64kV4C6lGMoFFIV871U9eX0T/SEhSp/kDLS/W9J1ZKKkF9o43z40EEf1ppaUqxN7k0bV3B9aLgUiSigKi5XC54o/YVUWdZzKBURtz3RGdwGxGW8jaSUUTl9L6L6lo5RdakrSq1TuwaLiA08jwIUR4AtMpOuwmbdwK9yZXG//qQtHGVZkkdYa/leFCgzsKtmHQyPAhRHgCA4gxaaOapjc5ldZ/SQORvapCkrok4G+ZwIzOKLgFkwksfMSaur9xY7SqFq1vqhdAViBK0SNwWsWhgoLB1fr6SE2RCSx8xJplAusqlxIg0EolopB3YFIUOb0NrEFEyRC6TGDhQw4hdo/b0DJZicj/cZCkLmh+B2b0R8nPvyziCB9yCLF7XIaWw4UKUVBCLEmnnN4O1OCLkopmeQcWPuQQYvc49FHyyypJbTj9cs3oUSWM3hcfr3YEIRPxCSzGFBOyjS2jSX0rlYhC3oFJ0eS0H3ejKwoqEBMSgguZNSLiE5hO0ZFo7ZntubWsRmn5yyoXhUtR5bQfd0ND69W5mpKWpnYEIRXxCQwgyZqkdggRp5bWflnlHZgUPb5diKsTOvS+KPj5lwks/KTYUtQOIeIEAFeLoYQo+AWWpFOUb7VUM0ZLG7XUVLUjCClNJLBka7LaIUSkOmPzX1odUbCIU5JO+fa2eIZW+iJqjl4Pydr6rNRGArNp6x+lt1Trm99xyQQmRROFbwqZoqICMSWl+bipBmjiqzHpTXI9WDeUnzYPprTRYkqStEj/rZsuozsKhs81NnwIGklgIO/CuqNO8eHXffNbrBANl6GSFKT71lISS00U/OxrrIADNJTAZCFH9zjM3/R/U5Tm8wKSpGV6XXDI3CRM6HxRsI2KTGDhKyMmQ+0QIlKtofmPgMEQBUMpkgQYdMEhc4snCgo4LBaIi1M7ipDTTAJLs6fJjhzdUKlrvqDZaIiCK1FJAvSnhswtjii4aNPg3RdoKIHpdXrSY9LVDiPilOFGKN/8GBj0spBDig56xY0CmGvcaofS8zRYwAEaSmAghxG7IwDUWb5Z0GzUyQQmRQcdLkwBEzp/FNyBZWjzs1FTCSwzNlPtECJSheGbpnAGxaViJJLUO/Q60OHF4omCDhwWi0xgkUDOg3VPke6bEmI9MoFJ2mcyBed6zY4o6EDfr1+wxFiDNJXA5DxY9zTgx20MDiPqRYPsSS9pntngPTX/FQXrv/r3VzuCHqOpBAZyGLG7TppODaUoArM5CuYEpKhm1LsxB8zoAhr/WTeZICtL7Sh6jOYSWE58jtohRKQT+m/6IJqNspBD0jaj4sAcDeu/cnM11//w2zT3laXYUmRfxG4ow9PUVsqkd6ocjST1LAP1WOqjYP5rwAC1I+hRmktgAP0S+qkdQsQRClSfKqc3KfUqRyNJPcdkFOgIYK7V+PyX0QjZ2WpH0aM0mcD6J2p30rInHTcGK7MMoh6drOSQNMpi9mD1mVG0Pv+Vk9O85b4GaTKBpdvTsRqsaocRcU7gxqc3nCrkkC2lJG2yGhqwR8Mgg4arDxtpMoEpikJuQq7aYUSkcmtwGNFqkuvBJG2yUIu1UuM/3wZD8A5M4zSZwEDOg3XXscYO3bpalSORpNAzGgWxHlA0PnpI377BJKZxmk1gWbFZGHXGjg+UmqlRvDhNJoyiRu4NJmmOzezFftLX8YGRLgqGD0HDCUyv0zMgUdslpD2lzBycB7NaouAXXYoqMUoD5jqNr3M0mYLrv6KAZhMYwNCUoWqHEJGO6oLzA1Zjg8qRSFLo6BRIckdB9cbgwcES+iig6QSWHpNOgiVB7TAijosAJ60WLMpJtUORpJCxWf3EVmm8eANgxAi1I+g1mk5gIO/CuuuIKYAOJ1ZZTi9pRILSgMGl8WHxvn0hPl7tKHqN5hPYoKRB6BTNf5khV4EHp9mM3exQOxRJComUhhq1Q+h5UXT3BVGQwKxGq2zw200FZgWrrkLtMCTpjNktAWKqNH4xFhcXFWu/vk3zCQzkMGJ3HcNFQO/HZomCpqeSpiXTgM6v8cVfI0eqHUGvi4oE1jeuLzajTe0wIo5QoMRmwmaOgsotSbN0CqTVV6odRs+yWGBo9F2oR0UCUxSFkWnRd3USCod0Liz6KtncV4pYCRYv1mqNbxE0YkRUdN44XVQkMIDhqcNlZ45u8BCg0qon1q7xrSckzcpw16Dp6y+DIeqKNxpFTQIz6U1yLqyb9hlc2E0aH4KRNMlqhMQKjf/sDh0aHEKMQlGTwABGpY+SJfXd4CJAlc2L1SLXhEmRpY/PgU7L+37pdDB6tNpRqCaqPs1jTDGyP2I37dO5sFtlh3opcugV6FNVonYYPWvQIIiJUTsK1URVAgMYkz5G7RAiklsJUG2rxaDtDV4lDemDC6Nfw503DAaYOFHtKFQVdQks2ZZMdly22mFEpAMGF7ExGl8MKmmCDsiuOaF2GD1r3DiwRffyoKhLYABj+4xVO4SI5CFATexJ9PIuTApzaToPZo+GG/fGxkb13FejqExgmbGZZMZmqh1GRDqgdxATK7dZkcKXTlHIqStVO4yedc45yCvJKE1gABMzo3vsuLt8CEpjKuXvjhS2+ui9WJwaHurOzIyaHZc7ErUJLD0mXTb57aajegfGBI13NpAikl6no291sdph9BxFgcmT1Y4ibERtAgOYlDUJRdtr9HvMUfsJ9NHXuUYKc5mKC7NLw0Pcw4ZBUpLaUYSNqE5gSdYkBiUPUjuMiFStuPEm1akdhiQ1MeoUsiuPqx1GzzGbo75s/nRRncAAJmROQK/ICZ3uOGotwWCV3Tmk8JDjr8Po03DPzgkTgklMahL1CSzGFCM71XeThwAnE+WGl5L6bDpBRlWR2mH0nKQkGD5c7SjCTtQnMIBxGePkfmHdVGSsQon3qh2GFOXy3BXohIZ7Hp57brCAQ2pGJjCCnerPzT5X7TAiVkFcEbJHsqSWFKOPxBoNd5wfOBCystSOIizJj51T8pLyZIupbnLoXNSm1qgdhhSF9DoYUK3hocO4OJg6Ve0owpZMYN8yNWeqLOjopiJLCf4Yv9phSFGmr64ei1OjaxJ1Ojj/fDCZ1I4kbMkE9i1x5jjGZYxTO4yIdTzxOIpejtNLvSPe7CO7TMN3XxMnQlqa2lGENZnATjO2z1gSLAlqhxGRnDonNWlyKFHqeSYjDKov1u5mldnZMEZu/dQRmcBOo1N0TM2RY87dVWwqxp8ghxKlnqMA/XUV2Go02nHDaoUZM9SOIiLIBNaKzNhMhiQPUTuMiJUfl49ilkOJUs9Ij3GQVqLhqsOZM6N+n6/OkgmsDZP7TibWFKt2GBHJp/goSStB0ckkJoVWrM1HblUpilaHDkePDg4fSp0iE1gbjHoj5/c/Xzb77aZqXTV1abJXohQ6RiPk+Esx12q0XVRqKkyapHYUEUUmsHakx6TLqsQzcNx8HF+iT+0wJA3QKZBpKyOpWKMXRUYjXHBBsHRe6jT53erAWRlnkWaXpazddST2CAGbbPgrnZnkuFoyyupRNDpyyLRpwUXLUpfIBNYBnaJjZr+ZGHRy86vuCCgBjiUfRZFrMaVuirX7yHBXY6rT6NDhyJHBdlFSl8kE1gnxlngm95W7oHaXW+ehJK1IFnVIXWa3+emjP058kUZL5gcMCDbqlbpFJrBOGpoylAGJA9QOI2JV6+uoTq9AkR21pU6yWQOkGY+RVuTV5tBhZmawVZT8neg2mcC6YEa/GSRZ5Xbe3VViKseTVqt2GFIEsFoCpJqPkFYOercGF8YnJ8Ps2bJo4wzJ714XGHQGZufNxqyXu6J21xFzEXqZxKR2WMyCVPMRkuuM2iyZj4uDuXNlk94QkAmsi+LMcVww4AK5PqybhAJ7LcUY02rld1BqwWwSpFqPEOc1ElOqwS7zVivMmyc7bYSITGDdkB2XzcSsiWqHEbGEAnusxZhSq2USk5qYjYI0az62gI7EQg0mL6MxeOcly+VDRiawbhrbZ6ws6jgDAWCX7QS29GpkcaJkNEKq/RgmJUDKcZ/2ijZ0OrjwQkhJUTsSTZEJ7AzIoo4zEwC2m0uxpVXJuewoZjRAuv0YBtyklurQezRYtDFjhuxx2APkx8YZMOgMXDTwIuxGu9qhRCy/IthuKScmpQK9/GmMOlZzgD4x+ehxklRr0eZi5XPPlQuVe4j8yDhDMaYY5g2aJysTz4APwTZrJbEp5ej1akcj9Za4GC+p1kPocBPnthFTpsF5rwkTYNQotaPQLEUIobXRZlWcqD/BBwc/wBeQzWu7yygUJjoTqK9Ow+OVE2Nalhpfj005DkCsx0pSgQaT15QpMGKE2lFomkxgIXSs+hgrj6wkIGTz2u4yoeMcpwWnI5M6h+w/qTV6HaTGlWMmuCFljNdK8jGNJS+dLjjnJYcNe5xMYCG2v2I/646tUzuMiGYWOia5jeBKprI2Dq3uXRhtzCZBqq0QPcG+hja/lZR8p7aWUhgMMGsW5OSoHUlUkAmsB2w/sZ3NRZvVDiOi6YDxXitxDQbKHZlySDHCxdj8JJmOoBCsMLT6LaQedWmrXN5kgosugj591I4kasgE1kM2F21m+4ntaocR8Ub4bWTVezjp7SeHFCNUUpyTGOVYU89aS8BM2lEPipZurWNigskrSS6r6U0ygfWgL4u/ZFvJNrXDiHi5wsKQOjfOQIYcUowgRgOkxJRjOjXfBWAWJtKOetH5NfSPmJISTF6yPVSvkwmsh20r2caXxV+qHUbES8HEmPoA+K2UObLwetWOSGpPfIyHeH0BivJNVa5JmEg/5kPn01CRU24uXHBBcO5L6nUygfWCHSd2sKlok9phRDwbeiY59Zg8fur8famusyB/eMOLyQjJ9jJMVDV73CiMpBf40Xs1lLxGjgwuUpb7ealGJrBesqt0F18c/0LtMCKeAYWJbjNxLhd+7FR7MqlvkKuf1aYACbEuYvWFTYUajczCRGqBTzvJS68PJq7hw9WOJOrJBNaL9pbv5bOCz9QOQxPG+m2k1wfLsT0kU+VIwS0rFVVhswRItJRgoK7Fc9aAhZSjbnRambhMSgruoiyLNcKCTGC97GDlQdYdWycXO4fAAGElr96NLhAAoeAQmZysj8Uvv7W9wqCHpJhqrMqJVp+3+4KLlDVTKj98ePDOS/Y7CxsygamguK6YlYdX4va71Q4l4sVgYJxLj80d/F4KTNR4s6hxyN6UPUWvhwS7A7tS0qxI49tiPTaSChp6ObIeYrHA9OnBgg0prMgEppJqVzUfHvqQWnet2qFowtCAlZw6F8qpsg4fsZx0Z9DglP2qQ8VogAR7HVZOoChtb3mS6LARV6KR5JWZGRwylCXyYUkmMBW5fC4+PPQhZY4ytUPRhASMjHEqWDzfbMnhJYE6Twp1DbLMubvMRkG8rQYrpbQ3HqigkFJtxlbh6sXoeohOF+wkP2aMrDIMYzKBqcwf8LPm6BqOnDyidiiaoANG+m1k1De/A/Bjo96fTm29WS6E7iSLWZBgrcJMeYfH6tGTWqbHXKuB/bzi4oJ3XWlpakcidUAmsDAhW0+FVqowMdIpMJ224jmAEWcgnTqnXVYttsFm9RNvrmyxlqstJmEitSiAwaWBrYQGDYKpU8FoVDsSqRNkAgsjR04eYf2x9Xj8GriKDQMGFEb7rKQ6Wp+P8ZBEvTeJeoch6hdEm02CGHMDVn0Vehydfl2sx0pioQYqDWNi4JxzYMAAtSORukAmsDBT665l5eGVVDorOz5Y6pRMzAx1+DH6Wr9DEJhwilQaPFacLkPUDDF2N2kB6NCRXG2K/PkugwHGjYPRo2V5fASSCSwM+QN+Pi/8nK8rvlY7FM3QAUMDNrIcruC6sTYIocejJOLyxdLgNuPRWM/FM0lajUzCREpxAKMzwocMBw+GSZNkhWEEkwksjB2sPMinBZ/iC0T4B0UYsQo9I3wmkhs6twuwHzuuQOKpuzN9xA016nVgNvmxmlxYdd1PWo1iPVYSj7sieyuU9HSYPBlSU9WORDpDMoGFuZPOk6w8spJqV7XaoWhKsjAx3K00LYDuDIEeD0m4/Ha8PiMerx5vGF1b6HRgMfkxG72Y9E6M1DXtfnzG50ZHUo0Ze3nnEn9YiokJ3nENHKh2JFKIyAQWAXwBH5uLNrO7bLfaoWhOlrAwyBXA7Ole4YxAj1fY8WPD6zfj8Rvx+Aw9vt2LXg8mYyCYrHROTErdGd9dtSXihwwNBhg7NjjPJbc90RSZwCJIcV0xa4+upd5Tr3YomtNPWBnQ4MPoC03mEejwY8eHFSEUBHqE0CHQERAKAl3wcaEQOPVHCAWdItDpBDol8K0/fnSKFz1u9HjQCVe7C4pDRQFi3TYSipyRO2Q4cCCcfTbY7WpHIvUAmcAijMfvYePxjbLAowfogLyAlZwGLwZ/hN5thIhZmEgqFZjqI7CKxWAIrucaNQoSEtSORupBMoFFqOK6YtYfWy97KfYARUAOFnLcoktzZFqgQ0eCw0xMiZOIW+Zts8GIETBsWLABr6R5MoFFMF/Ax5aiLewu242IuPq4yJCEkTyvgUSnC0Xjvyo2v4WkYi96d9uNesNSSkrwbisvL1jJIkUNmcA0oMpZxYbCDRTXFasdimaZhY6BAQt9nB7NDS8aMJB40oCtMoIWJStKcHuTUaMgI0PtaCSVyASmIfkn89l4fCN1npY740qhobXhxVhPsEhD54+QjwGjEYYMgZEjg013pagmE5jG+AN+dpTuYPuJ7XIBdA9rHF5McLnb7e4RjiwBMwkVInK6x6elBSsKBw8Gk0ntaKQwIROYRjk8DjYVbeJQ1SG1Q9E8HZAhLGT4FRJcHvT+8J1DMgsT8VUK1pMRcPeYkhJsrpuXB7GxakcjhSGZwDSuoqGCL4u/pKCmQO1Qoka6MJPh15Pk9rTZQLi3mYSJhGod1nCf50pJgX79gkkrPl7taKQwJxNYlCh3lLO1ZKtMZL0sCSNZfiPJbh9mb+8P1xmFkfhaQ/i2gNLrISsrWJCRkyMXHEtdIhNYlNFyIvvFpb+gsqT5NjQGo4GYxBhyh+Yy9YqpjJ42OqTv9fC7D5OSmdKp18QKA9nCRKI3gM3jRR/ouaFGAwbia43Yy8JsPZdOB0lJwTmt7OzgH9neSeom+ZMTZVLtqVw08CJNJ7K8MXmk9Q1uB++sd1K4v5Ad63ewY/0OLrj+Ahb+dKEqcdUpPvYpPjADZojBQErAQEJAR6zPj9XjOeO1ZgYMxNUbiSl1oojQDV8eraig/wMPkJuczNHf/77zL0xICHZ9b/yTkiL33ZJCRiawKNWYyKpd1ewu283ByoN4AxHYNqgVUy+fyuRLJzf93e/z89+//Je1/13L6mWrmTRnEv1G9FMvwFPq8VGv8wWrQAygmCFRMZIU0JPgV7B7fVg62RXYHDATV6tgrXChoNK8m90evLP6dsKSFYNSD5IJLMolWBKYmjOVSVmTOFB5gN1luzXXnkpv0HPVnVexcflGXA4XOz/dGRYJ7HRCgSq8VOm8waRmBAMKSRixCz02oWD1CywBgcnnx+jzY/eZiav0Y6rrpapCnS447Ge3B1s32e3BbUpSUuTGkFKvkwlMAsCkNzEybSQj00ZSUFPAnrI9FNYWqh1WyBjNRtJy0ijYV0BtZcsEvW/zPta+vpYjO4/gqHFgi7WRNyaP2d+ZTd7ovC6/39ZVW/nsnc8o2FeAs95JbGIsgycMZu4tc8kckNl03J4v9vC3xX+jT78+/OaN37Q4jw9Bic/J/RffT21lLQ8ue5ARo0YwLGUYtUdqee+N/7F23ToKCgupqq4mMTaWScOGceeVVzLrrLOCJ2kclhSCFz78kFsefZRFs2fzjx/9iP9btozX16+nsLycxJgY5kycyO8XLyarf/9gQrLZuPmnP2XpsmUAHCsvR7n44mYxyml0SS0ygUkt5MTnkBOfQ627lgOVBzhUdUgTd2UuR7CEPC65eQeHNx5/g5Uvr0TRKeQOy2XQuEFUnahix7od7Px0Jzc+cCNTLpvSqffw+/w898vn2LpyKwaTgdyhuSSkJVBaUMrmFZv56pOv+MGjP2Dk5JEADDt7GInpiZw4eoIju44wYNSAFufcvWE3tZW1DBoxiNsuvo3suGwAZt04izVr1jBixAjGT5yI3W7n8OHDvP/557z/+ec8/vjj3Hnnnc1PVhCc86yx2Zj84IMUFBQwbdo0Rup0fPHFF7z48ces27+fHTt2EH+qjH3qzJnUu928+eab2O12rrrqqs5/0yWpB8kEJrUpzhzHhMwJTMicQJmjjENVhzhcdRinL0xLsttRkl9CRVEFAKPP+6YS8dP/fcrKl1eS1jeN2x65jexB2U3PHdh2gH/c9Q+W/WEZA8cOJD0nvcP3ee+Z99i6civ9R/bnew9/j5SsbyoUt67ayrMPPMtzDz7Hw+88jC3Whk6n45yLz2HF8yv44r0vmiUwnaIjOy6bN1a/AcCPf/DjpuQFcPfdd/PSSy+RcVovwC+++IKLLrqIe+65h6uuuoqsrKwWcb799tvMmTOHTz/9lLhTLZlOnjzJ+eefz/bt2/nnP//Jz3/+cwC+973vMWvWLN58801SUlJ44YUXOvw+SFJvkK2bpU5Js6cxue9kbhh9A3MHzmVg0kAMuvC//nHWO9m7cS9P3/M0AX+AebfOo9/wfgAEAgHe/9f7AHzv999rlrwABp81mHm3zsPn9fHpW592+F6OGgerlq3CaDZy2yO3NUteAONnjWfaldNoqG1g0webmh5vvLvb8vEWfG4fmbGZTMuZxo2jb2R8/HjWfrwWs9nM9ddf3+x8c+fObZG8AM4991zuuOMOvF4v77zzTqux2u12lixZ0pS8ABITE7n//vsBWLVqVYdfrySpLfw/gaSwolN09I3vS9/4vvgCPorriimoKaCwpjBsmggv/c1Slv5mabPHdHod3/3ddzl77tlNjxXuL6S6vJrU7FRyh+W2eq4h44cAcHjn4Q7fd/+X+/G6vQydNJTEtMQ2z7fu9XUc3nmYmdfMBCAjJ4ORE0eye8tubIdtXDL5kqbjn37labxeLwsXLiQpKanF+SorK1m+fDm7d+/m5MmTeE9VLR48eDAY0/79rcYxYcKEVpPfsGHDACgqKurw65UktckEJnWbQWdomi8DqHZVU1BTQEFNASfqTxAQ6jS4/fY6sLqTdRzafgiXw8WyPy4jrW8a/Uf2B2gaUiw/Xs5tE25r95z1J+s7fN/yonIAvt78dYfnc9W4GJ0+muy4bDJiMtDfoefmm29m2UvLuPmmm5uOW7JkCQC33HJLi3P8+9//5q677sLhcLT5PrW1rc9d5uTktPp44x2ZyxXmLackCZnApBBKsCSQYElgdPpovH4vRXVFlNSVUOoopaKhotcS2unrwJz1Tp762VPs/3I///75v/n167/GZDERONVBPi45jhHnjmj3nDEJMR2+rwgEq/HS+qaRN6Z55aJBZ8BisGA32rGb7IwcPpJzss9pev7qq69m8eLFrF69muPHj5Odnc22bdvYuXMnWVlZzJ49u9n5tm7dym233YZer+dPf/oTl156KTk5OdhsNhRF4V//+he33XZbmxWCOrnxo6QBMoFJPcKoN9IvoR/9EvoBwW1eKhoqKHWUUlpfSqmjlAZvQ6/EYo2x8v0/fJ9fXfUrKksqWfnySi7+3sUkpQeH5GLiY7j51zef8fsk9Qmer09uH37x2C9Is6c1/bEZ218jZbPZWLhwIc899xxLly7lgQceaCqWWLRoUYuE8/rrryOEYPHixdx7770tztc4hChJWiYvw6ReodfpSY9JZ3T6aC7Mu5AbR9/I9aOuZ3bebCZmTmRQ0iBSbakYdcYeef/YxFjmfXceACtfXklDXQP9RvQjJiGGkvwSig93bzfrWFMsfeP6MjJtJLdccQsmk4nDXx1mUsIkJmVNol9Cvw6TV6Pvfve7ACxduhS3282yU2uvbr755hbHVlVVAZCb23LuzuVy8eabb3br62mL6VRHDV+YdNeXJJAJTFJRjCmGfgn9GJcxjpn9Z3LFsCu4ZdwtXD/qeuYNmsfkvpMZnjqcnPgckq3JWAyWM3q/6VdPJ6lPEs56JytfXoneoOeS71+CEIKn7nmKQ9tb7p1mwMDxHcepOVTDkOQhTMqaxOy82U1J6dIhlzJ30Fwm953MlOFTWLx4MQ6Hg0svvZRdu3a1OJ/b7ebdd9/l66+/bvHc5MmTGTJkCAcPHuS+++6jsrKSqVOnMmjQoBbHNhZbLF26lLq6b4pnXC4Xt99+O/n5+d3+PrUmNTUVk8nEiRMnmpKnJKlNDiFKYSfGFEOMKabZmqdG/oAfp8+J0+vE6XPi8rlw+9z4hR+TPniXkBWbxZDkIfiFH1/Ahz/gR1EU9IqeW++6lUfve5S1r61l8Z2L+cmPf4Kx3shLT73Eo997lMHDBpOXl4fdZqeirILt27dTXV3NU089xfR+05vi0CmtX/v98Y9/pKSkhGXLljF27FjGjBnDgAEDMBgMHD9+nO3bt+NwOFixYgVDhw5t8fpbbrmF+++/nyeeeAL45q6steOeeOIJvvrqK/r378+0adPQ6/V8+umnOJ1O7rzzzqZzhILRaOSyyy7jjTfeYOzYsUydOhXbqdZRzz77bMjeR5K6REiSRuTm5gpALFmypM1jfD6fGD58uADE/fff3/T4559/Lm644QaRm5srzGaziI2NFYMHDxaXX365ePbZZ0VVVVWr75Wfn9/q+3zwwQfiyiuvFFlZWcJoNIqEhAQxbNgwce2114ply5YJh8PR6uuKi4uFXq8XgLDb7aKurq7Nr6W8vFzcfvvtIi8vT5jNZpGZmSluvPFGcfDgQbFkyRIBiEWLFjV7TVuPN8rPzxeAyM3NbfFcZWWluO2220ROTo4wGo0CEPIjRFKT3A9MkiRJikhyDkySJEmKSDKBSZIkSRFJJjBJkiQpIskEJkmSJEUkmcAkSZKkiCQTmCRJkhSRZAKTJEmSIpJMYJIkSVJEkglMkiRJikgygUmSJEkRSSYwSZIkKSLJBCZJkiRFJJnAJEmSpIgkE5gkSZIUkWQCkyRJkiKSTGCSJElSRJIJTJIkSYpIMoFJkiRJEUkmMEmSJCkiyQQmSZIkRSSZwCRJkqSIJBOYJEmSFJFkApMkSZIikkxgkiRJUkSSCUySJEmKSDKBSZIkSRFJJjBJkiQpIskEJkmSJEUkmcAkSZKkiCQTmCRJkhSRZAKTJEmSIpJMYJIkSVJEkglMkiRJikgygUmSJEkRSSYwSZIkKSL9fwqPuHhbM8swAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    for subset_screened_desc, subset_retrieved_desc in zip(\n",
    "        [\"All\", \"Only RCTs\"], [\"All\", \"Is RCT\"]\n",
    "    ):\n",
    "        fig, ax = draw_venn_rsi(\n",
    "            title=None,\n",
    "            ids_retrieved=metrics_w_manual[subset_screened_desc][subset_retrieved_desc][\n",
    "                \"ids_retrieved\"\n",
    "            ],\n",
    "            ids_screened=metrics_w_manual[subset_screened_desc][subset_retrieved_desc][\n",
    "                \"ids_screened\"\n",
    "            ],\n",
    "            ids_screened_relevant=metrics_w_manual[subset_screened_desc][\n",
    "                subset_retrieved_desc\n",
    "            ][\"ids_screened_relevant\"],\n",
    "            add_pn_labels=True,\n",
    "            circle_label_size=16,\n",
    "            label_size=14,\n",
    "        )\n",
    "        fig.savefig(\n",
    "            PATH_LATEX_PLOT_OUTPUT\n",
    "            / f\"venn3_screened_found_included_{subset_screened_desc}.pdf\",\n",
    "        )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Screening Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAFgCAYAAAChYpM4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABRwUlEQVR4nO3deVwV1f8/8NcF5MpOIKsg4IKKgoqV4obmgkvmlrmLfkytcMVMKfcN1zTMNFPBSnNLK83cSDD3RHElUETBBDEXEBRQ7vn94Y/5emXxDt4LXH09H495PLhnZs55vy8XeDNzZkYhhBAgIiIi0kMG5R0AERERUWmxkCEiIiK9xUKGiIiI9BYLGSIiItJbLGSIiIhIb7GQISIiIr3FQoaIiIj0FgsZIiIi0ltG5R2ArqlUKty8eRMWFhZQKBTlHQ4RERFpQAiBBw8ewNnZGQYGxR93eeULmZs3b8LV1bW8wyAiIqJSSElJgYuLS7HrX/lCxsLCAsDTN8LS0rKcoyEiIiJNZGZmwtXVVfo7XpxXvpApOJ1kaWnJQoaIiEjPvGhaCCf7EhERkd5iIUNERER6i4UMERER6a1Xfo6MJoQQePLkCfLz88s7FKqgDA0NYWRkxEv4iYgqmNe+kMnLy0NqaioePnxY3qFQBWdqagonJycYGxuXdyhERPT/vdaFjEqlQlJSEgwNDeHs7AxjY2P+x02FCCGQl5eH27dvIykpCbVq1Srx5kxERFR2XutCJi8vDyqVCq6urjA1NS3vcKgCMzExQaVKlXD9+nXk5eWhcuXK5R0SERGBk30BgP9dk0b4OSEiqnj4m5mIiIj0FgsZIiIi0luv9RyZ4rhP/r1Mx7s2v0uZjldeZsyYgV9++QWxsbHlHQrR622GVTmOnVF+Y9MriUdk9NCQIUOgUCgwf/58tfZffvlF9lVX7u7uWLZsmUbbKRQKKBQKmJqawtvbG2vWrJE1VkU1Y8YMNGzYsLzDICKiUmAho6cqV66MBQsW4N69e2U25qxZs5CamooLFy5g4MCBGD58OP74448yG5+IiOh5LGT0VLt27eDo6IjQ0NASt/v5559Rr149KJVKuLu7Y8mSJdK61q1b4/r16xg/frx0tKUkFhYWcHR0RPXq1TFp0iTY2Nhg//790vr79+/jww8/hJ2dHSwtLfHOO+/g7NmzJfa5Zs0a1K1bF5UrV0adOnXwzTffSOuaNWuGSZMmqW1/+/ZtVKpUCYcOHQIA/PDDD3jzzTel2Pr374/09HRp+6ioKCgUCkRGRuLNN9+EqakpmjVrhvj4eABAREQEZs6cibNnz0rvQURERIkxExFRxcFCRk8ZGhpi3rx5WL58OW7cuFHkNjExMfjggw/Qt29fnD9/HjNmzMDUqVOlP9Tbt2+Hi4uLdKQlNTVVo7FVKhV+/vln3Lt3T+0ut71790Z6ejr++OMPxMTEwNfXF23btsXdu3eL7GfDhg2YNm0a5s6di7i4OMybNw9Tp07F+vXrAQADBgzApk2bIISQ9tm8eTOcnZ3RsmVLAMDjx48xe/ZsnD17Fr/88guuXbuGIUOGFBrriy++wJIlS3Dq1CkYGRnhf//7HwCgT58+mDBhAurVqye9B3369NHofSAiovLHyb56rEePHmjYsCGmT5+OtWvXFlr/5Zdfom3btpg6dSoAwNPTE5cuXcKiRYswZMgQ2NjYwNDQUDqa8SKTJk3ClClTkJubiydPnsDGxgYffvghAODw4cM4efIk0tPToVQqAQCLFy/GL7/8gm3btmHEiBGF+ps+fTqWLFmCnj17AgA8PDxw6dIlfPvttwgMDMQHH3yAcePG4fDhw1LhsnHjRvTr1086elRQkABA9erVERYWhrfeegtZWVkwNzeX1s2dOxf+/v4AgMmTJ6NLly7IycmBiYkJzM3NYWRkpNF7QEREFQuPyOi5BQsWYP369YiLiyu0Li4uDs2bN1dra968OS5fvlyqB2ROnDgRsbGx+PPPP9GkSRMsXboUNWvWBACcPXsWWVlZsLW1hbm5ubQkJSUhMTGxUF/Z2dlITEzEsGHD1LafM2eOtL2dnR06dOiADRs2AACSkpJw7NgxDBgwQOonJiYGXbt2RbVq1WBhYSEVK8nJyWrj+fj4SF87OTkBgNopKCIi0k88IqPnWrVqhYCAAISEhBR5SkWbqlSpgpo1a6JmzZrYunUrvL298eabb8LLywtZWVlwcnJCVFRUof2sra0LtWVlZQEAvvvuOzRp0kRtnaGhofT1gAEDMGbMGCxfvhwbN26Et7c3vL29ATwthgICAhAQEIANGzbAzs4OycnJCAgIQF5enlqflSpVkr4uOJqjUqlK9T7QK4SXIRPpPRYyr4D58+ejYcOGqF27tlp73bp1ceTIEbW2I0eOwNPTUyoWjI2NS3V0xtXVFX369EFISAh+/fVX+Pr6Ii0tDUZGRnB3d3/h/g4ODnB2dsbVq1fVjrA8r1u3bhgxYgT27NmDjRs3YvDgwdK6f/75B3fu3MH8+fPh6uoKADh16pTsXEr7HhARUfmrMKeW5s+fD4VCgXHjxkltOTk5CAoKkk5X9OrVC7du3Sq/ICsob29vDBgwAGFhYWrtEyZMQGRkJGbPno2EhASsX78eX3/9NT799FNpG3d3dxw6dAj//vsv/vvvP1njjh07Fjt37sSpU6fQrl07+Pn5oXv37ti3bx+uXbuGo0eP4osvvii2uJg5cyZCQ0MRFhaGhIQEnD9/HuHh4fjyyy+lbczMzNC9e3dMnToVcXFx6Nevn7SuWrVqMDY2xvLly3H16lX89ttvmD17tqwcCt6DpKQkxMbG4r///kNubq7sPoiIqHxUiCMyf//9N7799lu1eQwAMH78ePz+++/YunUrrKysMGrUKPTs2bPQUQZt08c77c6aNQubN29Wa/P19cWWLVswbdo0zJ49G05OTpg1a5baKahZs2Zh5MiRqFGjBnJzc9WuEHoRLy8vdOjQAdOmTcPu3buxe/dufPHFFxg6dChu374NR0dHtGrVCg4ODkXu/+GHH8LU1BSLFi3CxIkTYWZmBm9vb7ViFnh6eqlz585o1aoVqlWrJrXb2dkhIiICn3/+OcLCwuDr64vFixfjvffe0zgHAOjVqxe2b9+ONm3a4P79+wgPD9f5aToiItIOhZDzl0sHsrKy4Ovri2+++QZz5sxBw4YNsWzZMmRkZMDOzg4bN27E+++/D+DpqYS6devi2LFjaNq0qUb9Z2ZmwsrKChkZGbC0tFRbl5OTg6SkJHh4eKBy5cpaz41eLfy8vIJe1zkyr2vepFdK+vv9rHI/tRQUFIQuXbqgXbt2au0xMTF4/PixWnudOnVQrVo1HDt2rKzDJCIiogqoXE8tbdq0CadPn8bff/9daF1aWhqMjY0LXfHi4OCAtLS0YvvMzc1Vm+OQmZmptXiJiIioYim3IzIpKSkYO3YsNmzYoNXD9KGhobCyspKWgqtZiIiI6NVTboVMTEwM0tPT4evrCyMjIxgZGSE6OhphYWEwMjKCg4MD8vLycP/+fbX9bt26VeIdWENCQpCRkSEtKSkpOs6EiIiIyku5nVpq27Ytzp8/r9Y2dOhQ1KlTB5MmTYKrqysqVaqEyMhI9OrVCwAQHx+P5ORk+Pn5FduvUqmUbpFPREREr7aXLmTy8/Nx/vx5uLm54Y033tB4PwsLC9SvX1+tzczMDLa2tlL7sGHDEBwcDBsbG1haWmL06NHw8/PT+IolIiIierXJPrU0btw46QGF+fn58Pf3h6+vL1xdXYu8Pf3LWLp0Kd5991306tULrVq1gqOjI7Zv367VMYiIiEh/yT4is23bNgwcOBAAsHPnTiQlJeGff/7BDz/8gC+++OKlblb3fCFUuXJlrFixAitWrCh1n0RERPTqkn1E5r///pMm2+7evRu9e/eGp6cn/ve//xWa80IVV1RUFBQKRaHJ1ERERPpE9hEZBwcHXLp0CU5OTtizZw9WrlwJAHj48KHaU4v1Wlnf9bIUd7ocMmQI1q9fDwAwMjKCi4sLevfujVmzZun1XWejoqLQpk0b3Lt3r8inZhMRET1LdiEzdOhQfPDBB3BycoJCoZDuvHvixAnUqVNH6wFS8Tp27Ijw8HA8fvwYMTExCAwMhEKhwIIFC8o7NCIiojIh+9TSjBkzsGbNGowYMQJHjhyRLnU2NDTE5MmTtR4gFU+pVMLR0RGurq7o3r072rVrh/379wMAVCoVQkND4eHhARMTEzRo0ADbtm0rsb/Dhw+jZcuWMDExgaurK8aMGYPs7GwAwOeff44mTZoU2qdBgwaYNWsWgKcP/2zfvj2qVKkCKysr+Pv74/Tp02rbKxQKrFmzBj169ICpqSlq1aqF3377DQBw7do1tGnTBgDwxhtvQKFQ8OGNRERUolLdEO/999/H+PHj4eLiIrUFBgaiW7duWguM5Llw4QKOHj0KY2NjAE/vcPz9999j1apVuHjxIsaPH4+BAwciOjq6yP0TExPRsWNH9OrVC+fOncPmzZtx+PBhjBo1CsDTJ1CfPHkSiYmJ0j4XL17EuXPn0L9/fwDAgwcPEBgYiMOHD+P48eOoVasWOnfujAcPHqiNNXPmTHzwwQc4d+4cOnfujAEDBuDu3btwdXXFzz//DODpPYNSU1Px1Vdfaf29IiKiV0ep7iMTGRmJyMhIpKenQ6VSqa1bt26dVgKjF9u1axfMzc3x5MkT5ObmwsDAAF9//TVyc3Mxb948HDhwQLp5YPXq1XH48GF8++238Pf3L9RXaGgoBgwYgHHjxgEAatWqhbCwMPj7+2PlypWoV68eGjRogI0bN2Lq1KkAgA0bNqBJkyaoWbMmAOCdd95R63P16tWwtrZGdHQ03n33Xal9yJAh6NevHwBg3rx5CAsLw8mTJ9GxY0fY2NgAAOzt7TlHhoiIXkh2ITNz5kzMmjULb775pjRPhspHmzZtsHLlSmRnZ2Pp0qUwMjJCr169cPHiRTx8+BDt27dX2z4vLw+NGjUqsq+zZ8/i3Llz2LBhg9QmhIBKpUJSUhLq1q2LAQMGYN26dZg6dSqEEPjpp58QHBwsbX/r1i1MmTIFUVFRSE9PR35+Ph4+fIjk5GS1sXx8fKSvzczMYGlpifT0dG28JURE9JqRXcisWrUKERERGDRokC7iIRnMzMykoyHr1q1DgwYNsHbtWunOyL///juqVq2qtk9xj2/IysrCyJEjMWbMmELrqlWrBgDo168fJk2ahNOnT+PRo0dISUlBnz59pO0CAwNx584dfPXVV3Bzc4NSqYSfnx/y8vLU+qtUqZLaa4VCUejIHpWxsr5ST21s+VftEREVkF3I5OXloVmzZrqIhV6CgYEBPv/8cwQHByMhIQFKpRLJyclFnkYqiq+vLy5duiQVRkVxcXGBv78/NmzYgEePHqF9+/awt7eX1h85cgTffPMNOnfuDODpE87/++8/WXkUzPHJz8+XtR8REb2eZE/2/fDDD7Fx40ZdxEIvqXfv3jA0NMS3336LTz/9FOPHj8f69euRmJiI06dPY/ny5dK9Z543adIkHD16FKNGjUJsbCwuX76MX3/9VZrsW2DAgAHYtGkTtm7digEDBqitq1WrFn744QfExcXhxIkTGDBgAExMTGTl4ObmBoVCgV27duH27dvIysqS9yYQEdFrRfYRmZycHKxevRoHDhyAj49PodMEX375pdaCI3mMjIwwatQoLFy4EElJSbCzs0NoaCiuXr0Ka2tr+Pr64vPPPy9yXx8fH0RHR+OLL75Ay5YtIYRAjRo11E4dAU+vWBs1ahQMDQ3RvXt3tXVr167FiBEjpGdvzZs3D59++qmsHKpWrYqZM2di8uTJGDp0KAYPHoyIiAhZfRAR0etDIYQQcnYouM9HkZ0pFPjzzz9fOihtyszMhJWVFTIyMmBpaam2LicnB0lJSfDw8NDru+FS2XilPy+v6xwZ5l0OY3NOFGmmpL/fz5J9RObgwYMvFRgRERGRtpTqhngFbty4gRs3bmgrFiIiIiJZZBcyKpUKs2bNgpWVFdzc3ODm5gZra2vMnj2bl9ASERFRmZJ9aumLL77A2rVrMX/+fDRv3hzA02f0zJgxAzk5OZg7d67WgyQiIiIqiuxCZv369VizZg3ee+89qc3HxwdVq1bFJ598wkKGiIiIyozsU0t3795FnTp1CrXXqVMHd+/e1UpQZU3mhVv0muLnhIio4pFdyDRo0ABff/11ofavv/4aDRo00EpQZaXgHjgPHz4s50hIHxR8Tp6/dxIREZUf2aeWFi5ciC5duqg9WfnYsWNISUnB7t27tR6gLhkaGsLa2lp6YKGpqSkfgkmFCCHw8OFDpKenw9raGoaGhuUdEhER/X+yCxl/f38kJCRgxYoV+OeffwAAPXv2xCeffAJnZ2etB6hrjo6OAMCnL9MLWVtbS58XIiKqGGQXMgDg7Oz8ykzqVSgUcHJygr29PR4/flze4VAFValSJR6JISKqgDQqZM6dO4f69evDwMAA586dK3FbHx8fjQdfuXIlVq5ciWvXrgEA6tWrh2nTpqFTp04AgNatWyM6Olptn5EjR2LVqlUaj6EpQ0ND/qEiIiLSMxoVMg0bNkRaWhrs7e3RsGFDKBSKIq/gUCgUyM/P13hwFxcXzJ8/H7Vq1YIQAuvXr0e3bt1w5swZ1KtXDwAwfPhwzJo1S9rH1NRU4/6JiIjo1aZRIVPwJOWCr7Wla9euaq/nzp2LlStX4vjx41IhY2pqynkJREREVCSNLr92c3OTrua5fv06qlatKj2eoGCpWrUqrl+/XupA8vPzsWnTJmRnZ0tXQwHAhg0bUKVKFdSvXx8hISEvvFQ6NzcXmZmZagsRERG9mmRP9m3Tpg1SU1Nhb2+v1p6RkYE2bdrIOrUEAOfPn4efnx9ycnJgbm6OHTt2wMvLCwDQv39/uLm5wdnZGefOncOkSZMQHx+P7du3F9tfaGgoZs6cKTctIiIi0kOyCxkhRJH3Wrlz5w7MzMxkB1C7dm3ExsYiIyMD27ZtQ2BgIKKjo+Hl5YURI0ZI23l7e8PJyQlt27ZFYmIiatSoUWR/ISEhCA4Oll5nZmbC1dVVdlxERERU8WlcyPTs2RPA0wm9Q4YMgVKplNbl5+fj3LlzaNasmewAjI2NUbNmTQBA48aN8ffff+Orr77Ct99+W2jbJk2aAACuXLlSbCGjVCrVYiMiIqJXl8aFjJWVFYCnR2QsLCxgYmIirTM2NkbTpk0xfPjwlw5IpVIhNze3yHWxsbEAACcnp5ceh4iIiPSfxoVMeHg4AMDd3R2ffvppqU4jPS8kJASdOnVCtWrV8ODBA2zcuBFRUVHYu3cvEhMTsXHjRnTu3Bm2trY4d+4cxo8fj1atWsm6Vw0RERG9umTPkZk+fbrWBk9PT8fgwYORmpoKKysr+Pj4YO/evWjfvj1SUlJw4MABLFu2DNnZ2XB1dUWvXr0wZcoUrY1PRERE+q1UjyjYtm0btmzZguTkZOTl5amtO336tMb9rF27tth1rq6uhe7qS0RERPQsje4j86ywsDAMHToUDg4OOHPmDN5++23Y2tri6tWr0qMFiIiIiMqC7ELmm2++werVq7F8+XIYGxvjs88+w/79+zFmzBhkZGToIkYiIiKiIskuZJKTk6XLrE1MTPDgwQMAwKBBg/DTTz9pNzoiIiKiEsguZBwdHXH37l0AQLVq1XD8+HEAT5/BVNSDJImIiIh0RXYh88477+C3334DAAwdOhTjx49H+/bt0adPH/To0UPrARIREREVR/ZVS6tXr4ZKpQIABAUFwdbWFkePHsV7772HkSNHaj1AIiIiouLILmQMDAxgYPB/B3L69u2Lvn37ajUoIiIiIk3IPrUUHh6OrVu3FmrfunUr1q9fr5WgiIiIiDQhu5AJDQ1FlSpVCrXb29tj3rx5WgmKiIiISBOluvzaw8OjULubmxuSk5O1EhQRERGRJmQXMvb29jh37lyh9rNnz8LW1lYrQRERERFpQnYh069fP4wZMwYHDx5Efn4+8vPz8eeff2Ls2LGc9EtERERlSvZVS7Nnz8a1a9fQtm1bGBk93V2lUmHw4MGcI0NERERlSnYhY2xsjM2bN2P27Nk4e/YsTExM4O3tDTc3N13ER0RERFQs2YVMAU9PT3h6emozFiIiIiJZNCpkgoODMXv2bJiZmSE4OLjEbb/88kutBEZERET0IhoVMmfOnMHjx4+lr4ujUCi0ExURERGRBjQqZA4ePFjk10RERETlSfbl10REREQVhUZHZHr27Klxh9u3by91MERERERyaFTIWFlZ6ToOIiIiItk0KmTCw8N1HQcRERGRbKWeI5Oeno6//voLf/31F9LT00vVx8qVK+Hj4wNLS0tYWlrCz88Pf/zxh7Q+JycHQUFBsLW1hbm5OXr16oVbt26VNmQiIiJ6xcguZDIzMzFo0CBUrVoV/v7+8Pf3R9WqVTFw4EBkZGTI6svFxQXz589HTEwMTp06hXfeeQfdunXDxYsXAQDjx4/Hzp07sXXrVkRHR+PmzZuy5usQERHRq012ITN8+HCcOHECu3btwv3793H//n3s2rULp06dwsiRI2X11bVrV3Tu3Bm1atWCp6cn5s6dC3Nzcxw/fhwZGRlYu3YtvvzyS7zzzjto3LgxwsPDcfToURw/flxu2ERERPQKkv2Igl27dmHv3r1o0aKF1BYQEIDvvvsOHTt2LHUg+fn52Lp1K7Kzs+Hn54eYmBg8fvwY7dq1k7apU6cOqlWrhmPHjqFp06ZF9pObm4vc3FzpdWZmZqljIiIioopNdiFja2tb5FVMVlZWeOONN2QHcP78efj5+SEnJwfm5ubYsWMHvLy8EBsbC2NjY1hbW6tt7+DggLS0tGL7Cw0NxcyZM2XHQUREr7gZ5XgF7gx5Uy9Ic7JPLU2ZMgXBwcFqxURaWhomTpyIqVOnyg6gdu3aiI2NxYkTJ/Dxxx8jMDAQly5dkt1PgZCQEGRkZEhLSkpKqfsiIiKiik2jIzKNGjVSe47S5cuXUa1aNVSrVg0AkJycDKVSidu3b8ueJ2NsbIyaNWsCABo3boy///4bX331Ffr06YO8vDzcv39f7ajMrVu34OjoWGx/SqUSSqVSVgxERESknzQqZLp3767jMP6PSqVCbm4uGjdujEqVKiEyMhK9evUCAMTHxyM5ORl+fn5lFg8RERFVXBoVMtOnT9fJ4CEhIejUqROqVauGBw8eYOPGjYiKisLevXthZWWFYcOGITg4GDY2NrC0tMTo0aPh5+dX7ERfIiIier3InuyrTenp6Rg8eDBSU1NhZWUFHx8f7N27F+3btwcALF26FAYGBujVqxdyc3MREBCAb775pjxDJiIiogpEdiFjYGCgNl/mefn5+Rr3tXbt2hLXV65cGStWrMCKFSs07pOIiIheH7ILmR07dqi9fvz4Mc6cOYP169fzsmciIiIqU7ILmW7duhVqe//991GvXj1s3rwZw4YN00pgRERERC9S6odGPq9p06aIjIzUVndEREREL6SVQubRo0cICwtD1apVtdEdERERkUZkn1p644031Cb7CiHw4MEDmJqa4scff9RqcEREREQlkV3ILFu2TO21gYEB7Ozs0KRJk1I9a4mIiIiotGQVMkIINGvWDHl5eahduzaMjMr1NjRERFQK7jkby23sa+U2Mr2qNJ4jk5SUBB8fH9SpUwc+Pj6oUaMGTp06pcvYiIiIiEqkcSEzceJEPHnyBD/++CO2bdsGFxcX2Q+IJCIiItImjc8NHT58GNu2bUOLFi0APL3c2sXFBdnZ2TAzM9NZgERERETF0fiITHp6OmrVqiW9dnJygomJCdLT03USGBEREdGLaHxERqFQICsrCyYmJlKbgYEBHjx4gMzMTKnN0tJSuxESERERFUPjQkYIAU9Pz0JtjRo1kr5WKBSyHhpJRERE9DI0LmQOHjyoyziIiMocL0Mm0n8aFzL+/v66jIOIiIhINq09NJKIiIiorLGQISIiIr3FQoaIiIj0FgsZIiIi0lssZIiIiEhvaXTVUs+ePTXucPv27aUOhoiIiEgOjY7IWFlZSYulpSUiIyPVnnwdExODyMhIWFlZ6SxQIiIioudpVMiEh4dLi4ODAz744AMkJSVh+/bt2L59O65evYq+ffuiSpUqsgYPDQ3FW2+9BQsLC9jb26N79+6Ij49X26Z169ZQKBRqy0cffSRrHCIiIno1yZ4js27dOnz66acwNDSU2gwNDREcHIx169bJ6is6OhpBQUE4fvw49u/fj8ePH6NDhw7Izs5W22748OFITU2VloULF8oNm4iIiF5BGt/Zt8CTJ0/wzz//oHbt2mrt//zzD1Qqlay+9uzZo/Y6IiIC9vb2iImJQatWraR2U1NTODo6yg2ViIiIXnGyC5mhQ4di2LBhSExMxNtvvw0AOHHiBObPn4+hQ4e+VDAZGRkAABsbG7X2DRs24Mcff4SjoyO6du2KqVOnwtTUtMg+cnNzkZubK71+9sncRERE9GqRXcgsXrwYjo6OWLJkCVJTUwEATk5OmDhxIiZMmFDqQFQqFcaNG4fmzZujfv36Unv//v3h5uYGZ2dnnDt3DpMmTUJ8fHyxV0eFhoZi5syZpY6DiIiI9IfsQsbAwACfffYZPvvsM+loh6Wl5UsHEhQUhAsXLuDw4cNq7SNGjJC+9vb2hpOTE9q2bYvExETUqFGjUD8hISEIDg6WXmdmZsLV1fWl4yMiIqKKp1Q3xHvy5AkOHDiAn376CQqFAgBw8+ZNZGVllSqIUaNGYdeuXTh48CBcXFxK3LZJkyYAgCtXrhS5XqlUwtLSUm0hIiKiV5PsIzLXr19Hx44dkZycjNzcXLRv3x4WFhZYsGABcnNzsWrVKo37EkJg9OjR2LFjB6KiouDh4fHCfWJjYwE8PZ1FRERErzfZR2TGjh2LN998E/fu3YOJiYnU3qNHD0RGRsrqKygoCD/++CM2btwICwsLpKWlIS0tDY8ePQIAJCYmYvbs2YiJicG1a9fw22+/YfDgwWjVqhV8fHzkhk5ERESvGNlHZP766y8cPXoUxsbGau3u7u74999/ZfW1cuVKAE9veves8PBwDBkyBMbGxjhw4ACWLVuG7OxsuLq6olevXpgyZYrcsImIiOgVJLuQUalUyM/PL9R+48YNWFhYyOpLCFHieldXV0RHR8vqk4jkc8/ZWG5jXyu3kYnoVSD71FKHDh2wbNky6bVCoUBWVhamT5+Ozp07azM2IiIiohLJPiKzZMkSBAQEwMvLCzk5Oejfvz8uX76MKlWq4KefftJFjERERERFkl3IuLi44OzZs9i0aRPOnTuHrKwsDBs2DAMGDFCb/EtERESka7ILGQAwMjLCwIEDtR0LERERkSylKmQuX76MgwcPIj09vdCDIqdNm6aVwIiIiIheRHYh89133+Hjjz9GlSpV4OjoKN3ZF3g68ZeFDBEREZUV2YXMnDlzMHfuXEyaNEkX8RAREekEbzPwapJ9+fW9e/fQu3dvXcRCREREJIvsQqZ3797Yt2+fLmIhIiIikkX2qaWaNWti6tSpOH78OLy9vVGpUiW19WPGjNFacEREREQlkV3IrF69Gubm5oiOji70+ACFQsFChoiIiMqM7EImKSlJF3EQERERySZ7jkyBvLw8xMfH48mTJ9qMh4iIiEhjsguZhw8fYtiwYTA1NUW9evWQnJwMABg9ejTmz5+v9QCJiIiIiiO7kAkJCcHZs2cRFRWFypUrS+3t2rXD5s2btRocERERUUlkz5H55ZdfsHnzZjRt2lTtrr716tVDYmKiVoMjIiIiKonsIzK3b9+Gvb19ofbs7Gy1woaIiIhI12QXMm+++SZ+//136XVB8bJmzRr4+flpLzIiIiKiF5B9amnevHno1KkTLl26hCdPnuCrr77CpUuXcPTo0UL3lSEiIiLSJdlHZFq0aIHY2Fg8efIE3t7e2LdvH+zt7XHs2DE0btxYFzESERERFUn2ERkAqFGjBr777jttx0JEREQkS6luiJeYmIgpU6agf//+SE9PBwD88ccfuHjxoqx+QkND8dZbb8HCwgL29vbo3r074uPj1bbJyclBUFAQbG1tYW5ujl69euHWrVulCZuIiIheMbILmejoaHh7e+PEiRP4+eefkZWVBQA4e/Yspk+fLruvoKAgHD9+HPv378fjx4/RoUMHZGdnS9uMHz8eO3fuxNatWxEdHY2bN2+iZ8+ecsMmIiKiV5DsU0uTJ0/GnDlzEBwcDAsLC6n9nXfewddffy2rrz179qi9joiIgL29PWJiYtCqVStkZGRg7dq12LhxI9555x0AQHh4OOrWrYvjx4+jadOmcsMnIiKiV4jsIzLnz59Hjx49CrXb29vjv//+e6lgMjIyAAA2NjYAgJiYGDx+/Bjt2rWTtqlTpw6qVauGY8eOvdRYREREpP9kH5GxtrZGamoqPDw81NrPnDmDqlWrljoQlUqFcePGoXnz5qhfvz4AIC0tDcbGxrC2tlbb1sHBAWlpaUX2k5ubi9zcXOl1ZmZmqWMiIiKiik32EZm+ffti0qRJSEtLg0KhgEqlwpEjR/Dpp59i8ODBpQ4kKCgIFy5cwKZNm0rdB/B0ArGVlZW0uLq6vlR/REREVHHJLmTmzZuHOnXqwNXVFVlZWfDy8kKrVq3QrFkzTJkypVRBjBo1Crt27cLBgwfh4uIitTs6OiIvLw/3799X2/7WrVtwdHQssq+QkBBkZGRIS0pKSqliIiIioopP1qklIQTS0tIQFhaGadOm4fz588jKykKjRo1Qq1Yt2YMLITB69Gjs2LEDUVFRhU5XNW7cGJUqVUJkZCR69eoFAIiPj0dycnKxj0NQKpVQKpWyYyEiIiL9I7uQqVmzJi5evIhatWq99GmboKAgbNy4Eb/++issLCykeS9WVlYwMTGBlZUVhg0bhuDgYNjY2MDS0hKjR4+Gn58fr1giIiIieYWMgYEBatWqhTt37pTqCMzzVq5cCQBo3bq1Wnt4eDiGDBkCAFi6dCkMDAzQq1cv5ObmIiAgAN98881Lj01ERET6T/ZVS/Pnz8fEiROxcuVK6eqi0hJCvHCbypUrY8WKFVixYsVLjUVERESvHtmFzODBg/Hw4UM0aNAAxsbGMDExUVt/9+5drQVHREREVBLZhcyyZct0EAYRERGRfLILmcDAQF3EQURERCSb7PvI7N69G3v37i3Uvm/fPvzxxx9aCYqIiIhIE7ILmcmTJyM/P79Qu0qlwuTJk7USFBEREZEmZBcyly9fhpeXV6H2OnXq4MqVK1oJioiIiEgTsgsZKysrXL16tVD7lStXYGZmppWgiIiIiDQhu5Dp1q0bxo0bh8TERKntypUrmDBhAt577z2tBkdERERUEtmFzMKFC2FmZoY6derAw8MDHh4eqFu3LmxtbbF48WJdxEhERERUJNmXX1tZWeHo0aPYv38/zp49CxMTE/j4+KBVq1a6iI+IiIioWLILGQBQKBTo0KEDOnTooO14iIiIiDSm8amlY8eOYdeuXWpt33//PTw8PGBvb48RI0YgNzdX6wESERERFUfjQmbWrFm4ePGi9Pr8+fMYNmwY2rVrh8mTJ2Pnzp0IDQ3VSZBERERERdG4kImNjUXbtm2l15s2bUKTJk3w3XffITg4GGFhYdiyZYtOgiQiIiIqisaFzL179+Dg4CC9jo6ORqdOnaTXb731FlJSUrQbHREREVEJNC5kHBwckJSUBADIy8vD6dOn0bRpU2n9gwcPUKlSJe1HSERERFQMjQuZzp07Y/Lkyfjrr78QEhICU1NTtGzZUlp/7tw51KhRQydBEhERERVF48uvZ8+ejZ49e8Lf3x/m5uZYv349jI2NpfXr1q3j5dhERERUpjQuZKpUqYJDhw4hIyMD5ubmMDQ0VFu/detWmJubaz1AIiIiouKU6s6+RbGxsXnpYIiIiIjkKNWdfYleWTOKLtTLZuyM8hubiEhPyX5oJBEREVFFUa6FzKFDh9C1a1c4OztDoVDgl19+UVs/ZMgQKBQKtaVjx47lEywRERFVOBoVMr6+vrh37x6Ap48qePjwoVYGz87ORoMGDbBixYpit+nYsSNSU1Ol5aefftLK2ERERKT/NJojExcXh+zsbLzxxhuYOXMmPvroI5iamr704J06dVK7O3BRlEolHB0dX3osIiIievVoVMg0bNgQQ4cORYsWLSCEwOLFi4u91HratGlaDTAqKgr29vZ444038M4772DOnDmwtbUtdvvc3Fy1p3BnZmZqNR4iIiKqODQqZCIiIjB9+nTs2rULCoUCf/zxB4yMCu+qUCi0Wsh07NgRPXv2hIeHBxITE/H555+jU6dOOHbsWKH72BQIDQ3FzJkztRYDERERVVwaFTK1a9fGpk2bAAAGBgaIjIyEvb29TgMDgL59+0pfe3t7w8fHBzVq1EBUVJTak7ifFRISguDgYOl1ZmYmXF1ddR4rERERlT3Z95FRqVS6iEMj1atXR5UqVXDlypViCxmlUgmlUlnGkREREVF5KNUN8RITE7Fs2TLExcUBALy8vDB27FidPzTyxo0buHPnDpycnHQ6DhEREekH2feR2bt3L7y8vHDy5En4+PjAx8cHJ06cQL169bB//35ZfWVlZSE2NhaxsbEAgKSkJMTGxiI5ORlZWVmYOHEijh8/jmvXriEyMhLdunVDzZo1ERAQIDdsIiIiegXJPiIzefJkjB8/HvPnzy/UPmnSJLRv317jvk6dOoU2bdpIrwvmtgQGBmLlypU4d+4c1q9fj/v378PZ2RkdOnTA7NmzeeqIiIiIAJSikImLi8OWLVsKtf/vf//DsmXLZPXVunVrCCGKXb9371654REREdFrRPapJTs7O+lU0LNiY2PL5EomIiIiogKyj8gMHz4cI0aMwNWrV9GsWTMAwJEjR7BgwQK1y56JiIiIdE12ITN16lRYWFhgyZIlCAkJAQA4OztjxowZGDNmjNYDJCIiIiqO7EJGoVBg/PjxGD9+PB48eAAAsLCw0HpgRERERC9SqvvIFGABQ0REROVJ9mRfIiIiooqChQwRERHpLRYyREREpLdkFTKPHz9G27ZtcfnyZV3FQ0RERKQxWYVMpUqVcO7cOV3FQkRERCSL7FNLAwcOxNq1a3URCxEREZEssi+/fvLkCdatW4cDBw6gcePGMDMzU1v/5Zdfai04IiIiopLILmQuXLgAX19fAEBCQoLaOoVCoZ2oiIiIiDQgu5A5ePCgLuIgIiIikq3Ul19fuXIFe/fuxaNHjwAAQgitBUVERESkCdmFzJ07d9C2bVt4enqic+fOSE1NBQAMGzYMEyZM0HqARERERMWRXciMHz8elSpVQnJyMkxNTaX2Pn36YM+ePVoNjoiIiKgksufI7Nu3D3v37oWLi4tae61atXD9+nWtBUZERET0IrKPyGRnZ6sdiSlw9+5dKJVKrQRFREREpAnZhUzLli3x/fffS68VCgVUKhUWLlyINm3aaDU4IiIiopLIPrW0cOFCtG3bFqdOnUJeXh4+++wzXLx4EXfv3sWRI0d0ESMRERFRkWQfkalfvz4SEhLQokULdOvWDdnZ2ejZsyfOnDmDGjVq6CJGIiIioiKV6j4yVlZW+OKLL7Blyxbs3r0bc+bMgZOTk+x+Dh06hK5du8LZ2RkKhQK//PKL2nohBKZNmwYnJyeYmJigXbt2fPI2ERERSUpVyNy7dw+LFy/GsGHDMGzYMCxZsgR3796V3U92djYaNGiAFStWFLl+4cKFCAsLw6pVq3DixAmYmZkhICAAOTk5pQmbiIiIXjGyC5lDhw7B3d0dYWFhuHfvHu7du4ewsDB4eHjg0KFDsvrq1KkT5syZgx49ehRaJ4TAsmXLMGXKFHTr1g0+Pj74/vvvcfPmzUJHboiIiOj1JHuyb1BQEPr06YOVK1fC0NAQAJCfn49PPvkEQUFBOH/+vFYCS0pKQlpaGtq1aye1WVlZoUmTJjh27Bj69u1b5H65ubnIzc2VXmdmZmolHiIiIqp4ZB+RuXLlCiZMmCAVMQBgaGiI4OBgXLlyRWuBpaWlAQAcHBzU2h0cHKR1RQkNDYWVlZW0uLq6ai0mIiIiqlhkFzK+vr6Ii4sr1B4XF4cGDRpoJaiXERISgoyMDGlJSUkp75CIiIhIRzQ6tXTu3Dnp6zFjxmDs2LG4cuUKmjZtCgA4fvw4VqxYgfnz52stMEdHRwDArVu31K6IunXrFho2bFjsfkqlkncYJiIiek1oVMg0bNgQCoUCQgip7bPPPiu0Xf/+/dGnTx+tBObh4QFHR0dERkZKhUtmZiZOnDiBjz/+WCtjEBERkX7TqJBJSkrSyeBZWVlq82qSkpIQGxsLGxsbVKtWDePGjcOcOXNQq1YteHh4YOrUqXB2dkb37t11Eg8RERHpF40KGTc3N50MfurUKbXnMwUHBwMAAgMDERERgc8++wzZ2dkYMWIE7t+/jxYtWmDPnj2oXLmyTuIhIiIi/SL78msAuHnzJg4fPoz09HSoVCq1dWPGjNG4n9atW6udrnqeQqHArFmzMGvWrNKESURERK842YVMREQERo4cCWNjY9ja2kKhUEjrFAqFrEKGqKJxz9lYbmNfK7eRiYj0l+xCZurUqZg2bRpCQkJgYFCqJxwQERERaYXsSuThw4fo27cvixgiIiIqd7KrkWHDhmHr1q26iIWIiIhIFtmnlkJDQ/Huu+9iz5498Pb2RqVKldTWf/nll1oLjoiIiKgkpSpk9u7di9q1awNAocm+RERERGVFdiGzZMkSrFu3DkOGDNFBOERERESakz1HRqlUonnz5rqIhYiIiEgW2YXM2LFjsXz5cl3EQkRERCSL7FNLJ0+exJ9//oldu3ahXr16hSb7bt++XWvBEREREZVEdiFjbW2Nnj176iIWIiIiIllkFzLh4eG6iIOIiIhINt6el4iIiPSW7CMyHh4eJd4v5urVqy8VEBEREZGmZBcy48aNU3v9+PFjnDlzBnv27MHEiRO1FRcRERHRC8kuZMaOHVtk+4oVK3Dq1KmXDoiIiIhIU1qbI9OpUyf8/PPP2uqOiIiI6IW0Vshs27YNNjY22uqOiIiI6IVkn1pq1KiR2mRfIQTS0tJw+/ZtfPPNN1oNjoiIiKgksguZ7t27q702MDCAnZ0dWrdujTp16mgrLiIiIqIXkl3ITJ8+XRdxEBEREcnGG+IRERGR3tK4kDEwMIChoWGJi5GR7AM8JZoxYwYUCoXawtNXREREVEDjymPHjh3Frjt27BjCwsKgUqm0EtSz6tWrhwMHDkivtV0sERERkf7SuCro1q1bobb4+HhMnjwZO3fuxIABAzBr1iytBgc8LVwcHR213i8RERHpv1LNkbl58yaGDx8Ob29vPHnyBLGxsVi/fj3c3Ny0HR8uX74MZ2dnVK9eHQMGDEBycnKJ2+fm5iIzM1NtISIioleTrEImIyMDkyZNQs2aNXHx4kVERkZi586dqF+/vk6Ca9KkCSIiIrBnzx6sXLkSSUlJaNmyJR48eFDsPqGhobCyspIWV1dXncRGRERE5U/jQmbhwoWoXr06du3ahZ9++glHjx5Fy5YtdRkbOnXqhN69e8PHxwcBAQHYvXs37t+/jy1bthS7T0hICDIyMqQlJSVFpzESERFR+dF4jszkyZNhYmKCmjVrYv369Vi/fn2R223fvl1rwT3P2toanp6euHLlSrHbKJVKKJVKncVAREREFYfGhczgwYPVHk1QHrKyspCYmIhBgwaVaxxERERUMWhcyEREROgwjKJ9+umn6Nq1K9zc3HDz5k1Mnz4dhoaG6NevX5nHQkRERBVPhb4py40bN9CvXz/cuXMHdnZ2aNGiBY4fPw47O7vyDo2IiIgqgApdyGzatKm8QyAiIqIKrEIXMkRERPSSZliV49gZOh+CD40kIiIivcVChoiIiPQWCxkiIiLSWyxkiIiISG+xkCEiIiK9xauWXuQVn+1d/Nivad5ERKRXeESGiIiI9BYLGSIiItJbLGSIiIhIb7GQISIiIr3FQoaIiIj0FgsZIiIi0lssZIiIiEhvsZAhIiIivcVChoiIiPQWCxkiIiLSWyxkiIiISG+xkCEiIiK9xYdGvoB7zsZyG/tauY1MRESvilf97xiPyBAREZHeYiFDREREeksvCpkVK1bA3d0dlStXRpMmTXDy5MnyDomIiIgqgApfyGzevBnBwcGYPn06Tp8+jQYNGiAgIADp6enlHRoRERGVswpfyHz55ZcYPnw4hg4dCi8vL6xatQqmpqZYt25deYdGRERE5axCX7WUl5eHmJgYhISESG0GBgZo164djh07VuQ+ubm5yM3NlV5nZGQAADIzM0sVgyr3Yan204bSxqwVuaL8xi7HvF/X7zfzLnvMu+wx77L3MnkX7CvEC/4eiQrs33//FQDE0aNH1donTpwo3n777SL3mT59ugDAhQsXLly4cHkFlpSUlBJrhQp9RKY0QkJCEBwcLL1WqVS4e/cubG1toVAoyjSWzMxMuLq6IiUlBZaWlmU6dnli3sz7dcC8mffroDzzFkLgwYMHcHZ2LnG7Cl3IVKlSBYaGhrh165Za+61bt+Do6FjkPkqlEkqlUq3N2tpaVyFqxNLS8rX64Bdg3q8X5v16Yd6vl/LK28rK6oXbVOjJvsbGxmjcuDEiIyOlNpVKhcjISPj5+ZVjZERERFQRVOgjMgAQHByMwMBAvPnmm3j77bexbNkyZGdnY+jQoeUdGhEREZWzCl/I9OnTB7dv38a0adOQlpaGhg0bYs+ePXBwcCjv0F5IqVRi+vTphU51veqYN/N+HTBv5v060Ie8FUK86LomIiIiooqpQs+RISIiIioJCxkiIiLSWyxkiIiISG+xkCEiIiK9xUKmBKGhoXjrrbdgYWEBe3t7dO/eHfHx8Wrb5OTkICgoCLa2tjA3N0evXr0K3cBvzJgxaNy4MZRKJRo2bFjkWEIILF68GJ6enlAqlahatSrmzp2rq9RKVFZ5z5gxAwqFotBiZmamy/SKVZbf771796Jp06awsLCAnZ0devXqhWvXrukos5KVZd5btmxBw4YNYWpqCjc3NyxatEhXab2QNvI+e/Ys+vXrB1dXV5iYmKBu3br46quvCo0VFRUFX19fKJVK1KxZExEREbpOr1hllXdqair69+8PT09PGBgYYNy4cWWRXrHKKu/t27ejffv2sLOzg6WlJfz8/LB3794yybEoZZX34cOH0bx5c9ja2sLExAR16tTB0qVLyyTHCv2spfIWEBAgwsPDxYULF0RsbKzo3LmzqFatmsjKypK2+eijj4Srq6uIjIwUp06dEk2bNhXNmjVT62f06NHi66+/FoMGDRINGjQocqzRo0eL2rVri19//VVcvXpVnDp1Suzbt0+X6RWrrPJ+8OCBSE1NVVu8vLxEYGCgjjMsWlnlffXqVaFUKkVISIi4cuWKiImJEa1atRKNGjXSdYpFKqu8d+/eLYyMjMTKlStFYmKi2LVrl3BychLLly/XdYpF0kbea9euFWPGjBFRUVEiMTFR/PDDD8LExEQtp6tXrwpTU1MRHBwsLl26JJYvXy4MDQ3Fnj17yjTfAmWVd1JSkhgzZoxYv369aNiwoRg7dmxZpllIWeU9duxYsWDBAnHy5EmRkJAgQkJCRKVKlcTp06fLNN8CZZX36dOnxcaNG8WFCxdEUlKS+OGHH4Spqan49ttvdZ4jCxkZ0tPTBQARHR0thBDi/v37olKlSmLr1q3SNnFxcQKAOHbsWKH9p0+fXuQv+EuXLgkjIyPxzz//6Cz2l6GrvJ8XGxsrAIhDhw5pLfaXoau8t27dKoyMjER+fr7U9ttvvwmFQiHy8vK0n4hMusq7X79+4v3331drCwsLEy4uLkKlUmk3iVJ42bwLfPLJJ6JNmzbS688++0zUq1dPbZs+ffqIgIAALWdQOrrK+1n+/v7lXsg8ryzyLuDl5SVmzpypncBfUlnm3aNHDzFw4EDtBF4CnlqSISMjAwBgY2MDAIiJicHjx4/Rrl07aZs6deqgWrVqOHbsmMb97ty5E9WrV8euXbvg4eEBd3d3fPjhh7h79652EyglXeX9vDVr1sDT0xMtW7Z8uYC1RFd5N27cGAYGBggPD0d+fj4yMjLwww8/oF27dqhUqZJ2kygFXeWdm5uLypUrq7WZmJjgxo0buH79uhYifznayjsjI0PqAwCOHTum1gcABAQEvNTPijbpKu+KrqzyVqlUePDgQYV5b8oq7zNnzuDo0aPw9/fXUuTFYyGjIZVKhXHjxqF58+aoX78+ACAtLQ3GxsaFHkrp4OCAtLQ0jfu+evUqrl+/jq1bt+L7779HREQEYmJi8P7772szhVLRZd7PysnJwYYNGzBs2LCXDVkrdJm3h4cH9u3bh88//xxKpRLW1ta4ceMGtmzZos0USkWXeQcEBGD79u2IjIyESqVCQkIClixZAuDpfIrypK28jx49is2bN2PEiBFSW1paWqE7kTs4OCAzMxOPHj3SbiIy6TLviqws8168eDGysrLwwQcfaC3+0iqLvF1cXKBUKvHmm28iKCgIH374odbzeF6Ff0RBRREUFIQLFy7g8OHDWu9bpVIhNzcX33//PTw9PQEAa9euRePGjREfH4/atWtrfUxN6TLvZ+3YsQMPHjxAYGCgTsfRlC7zTktLw/DhwxEYGIh+/frhwYMHmDZtGt5//33s378fCoVC62NqSpd5Dx8+HImJiXj33Xfx+PFjWFpaYuzYsZgxYwYMDMr3fypt5H3hwgV069YN06dPR4cOHbQYne4wb93mvXHjRsycORO//vor7O3tSz2WtpRF3n/99ReysrJw/PhxTJ48GTVr1kS/fv1eJuwX4hEZDYwaNQq7du3CwYMH4eLiIrU7OjoiLy8P9+/fV9v+1q1bcHR01Lh/JycnGBkZSUUMANStWxcAkJyc/HLBvwRd5/2sNWvW4N13360Qz9DSdd4rVqyAlZUVFi5ciEaNGqFVq1b48ccfERkZiRMnTmgrDdl0nbdCocCCBQuQlZWF69evIy0tDW+//TYAoHr16lrJoTS0kfelS5fQtm1bjBgxAlOmTFFb5+joWOgKr1u3bsHS0hImJibaTUYGXeddUZVV3ps2bcKHH36ILVu2FDq1WB7KKm8PDw94e3tj+PDhGD9+PGbMmKHtVArT+SwcPaZSqURQUJBwdnYWCQkJhdYXTJLatm2b1PbPP//IngS5d+9eAUBcuXJFaiuY+BofH6+dZGQoq7wLXL16VSgUCrFz506txF9aZZV3cHCwePvtt9Xabt68KQCII0eOvHwiMpX19/tZgwYNEn5+fqWO/WVoK+8LFy4Ie3t7MXHixCLH+eyzz0T9+vXV2vr161duk33LKu9nVYTJvmWZ98aNG0XlypXFL7/8ot0kSqE8vt8FZs6cKdzc3F4qfk2wkCnBxx9/LKysrERUVJTaJcIPHz6Utvnoo49EtWrVxJ9//ilOnTol/Pz8Cv1ivnz5sjhz5owYOXKk8PT0FGfOnBFnzpwRubm5Qggh8vPzha+vr2jVqpU4ffq0OHXqlGjSpIlo3759meZboKzyLjBlyhTh7Owsnjx5Uib5Faes8o6MjBQKhULMnDlTJCQkiJiYGBEQECDc3NzUxiorZZX37du3xcqVK0VcXJw4c+aMGDNmjKhcubI4ceJEmeZbQBt5nz9/XtjZ2YmBAweq9ZGeni5tU3D59cSJE0VcXJxYsWJFuV5+XVZ5CyGkz0Djxo1F//79xZkzZ8TFixfLLNdnlVXeGzZsEEZGRmLFihVq29y/f79M8y1QVnl//fXX4rfffhMJCQkiISFBrFmzRlhYWIgvvvhC5zmykCkBgCKX8PBwaZtHjx6JTz75RLzxxhvC1NRU9OjRQ6Smpqr14+/vX2Q/SUlJ0jb//vuv6NmzpzA3NxcODg5iyJAh4s6dO2WUqbqyzDs/P1+4uLiIzz//vIyyK15Z5v3TTz+JRo0aCTMzM2FnZyfee+89ERcXV0aZqiurvG/fvi2aNm0qzMzMhKmpqWjbtq04fvx4GWaqTht5T58+vcg+nv8v9ODBg6Jhw4bC2NhYVK9eXW2MslaWeWuyTVkpq7yL+zkor/tjlVXeYWFhol69esLU1FRYWlqKRo0aiW+++UbtNhO6ovj/iRIRERHpHU72JSIiIr3FQoaIiIj0FgsZIiIi0lssZIiIiEhvsZAhIiIivcVChoiIiPQWCxkiIiLSWyxkiF5g9erVcHV1hYGBAZYtW6bxfkOGDEH37t11FtfrLCoqCgqFotDzYSqaivYZiIiIKPSUYyJ9x0KGypwQAu3atUNAQEChdd988w2sra1x48aNcoissMzMTIwaNQqTJk3Cv//+W+Rj669duwaFQoHY2Ngyiens2bN47733YG9vj8qVK8Pd3R19+vRBenq6Vvp3d3eXVbDpWuvWrTFu3Di1tmbNmiE1NRVWVlal6rOgECpY7Ozs0LlzZ5w/f14LEf+fr776ChEREVrtszitW7dWy+n5pXXr1ujTpw8SEhLKJJ7nFRVTixYtyiUWerUYlXcA9PpRKBQIDw+Ht7c3vv32W4wcORIAkJSUhM8++wwrV65UezqrNjx+/BiVKlWSvV9ycjIeP36MLl26wMnJSasxlcbt27fRtm1bvPvuu9i7dy+sra1x7do1/Pbbb8jOzi7v8MqMsbFxqZ+0/qz4+HhYWlri5s2bmDhxIrp06YIrV67A2NhYC1Gi1IVWaWzfvh15eXkAgJSUFLz99ts4cOAA6tWrB+Dpe2ZiYlKuT9wODw9Hx44dpdfFvc+l/Xml15TOH4JAVIyIiAhhbm4url69KlQqlWjTpo3o0aOHOH/+vOjYsaMwMzMT9vb2YuDAgeL27dvSfn/88Ydo3ry5sLKyEjY2NqJLly5qTw5PSkoSAMSmTZtEq1athFKpLPbZNtevXxfvvfeeMDMzExYWFqJ3794iLS1NCCFEeHh4ic9LKvD8Nv7+/kIIIQIDA0W3bt3EokWLhKOjo7CxsRGffPKJyMvLk/bNyckREyZMEM7OzsLU1FS8/fbb4uDBg8W+Zzt27BBGRkbi8ePHRa5XqVSiRo0aYtGiRWrtZ86cEQDE5cuXhUqlEtOnTxeurq7C2NhYODk5idGjRwshin5OTIG//vpLtGjRQlSuXFm4uLiI0aNHi6ysLGm9m5ubmD17thg0aJAwMzMT1apVE7/++qtIT0+X3mNvb2/x999/S/v8999/om/fvsLZ2VmYmJiI+vXri40bN0rrAwMDi/weHDx4UAAQ9+7dk7Y9fPiw8Pf3FyYmJsLa2lp06NBB3L17t8j3qaj9f/vtNwFAnD17VqOcQ0JCCj3FXAghfHx8xMyZM6X4u3XrJq3Lz88X8+bNE+7u7qJy5crCx8dHbN26VVrfuHFjte9dt27dhJGRkXjw4IEQQoiUlBTp+1iSgp+BM2fOqLWHh4cLKysr6XXBk8rXrl0rXF1dhZmZmfj444/FkydPxIIFC4SDg4Ows7MTc+bMUevn3r17YtiwYaJKlSrCwsJCtGnTRsTGxpYYEwCxY8eOYmN9/uf1RZ8NIZ5+XkeNGiXGjh0rrK2thb29vVi9erXIysoSQ4YMEebm5qJGjRpi9+7davu96HcM6RcWMlSuunXrJlq3bi3CwsKEnZ2dSE9PF3Z2diIkJETExcWJ06dPi/bt24s2bdpI+2zbtk38/PPP0tOWu3btKry9vaWHkxX8YnR3dxc///yzuHr1qrh582ahsfPz80XDhg1FixYtxKlTp8Tx48dF48aNpULk4cOH4sCBAwKAOHnypEhNTS3yCd0nT54UAMSBAwdEamqq9LDPwMBAYWlpKT766CMRFxcndu7cKUxNTcXq1aulfT/88EPRrFkzcejQIXHlyhWxaNEioVQqRUJCQpHv17FjxwQAsWXLFqFSqYrcZu7cucLLy0utbcyYMaJVq1ZCCCG2bt0qLC0txe7du8X169fFiRMnpJju3LkjXFxcxKxZs6Qn3AohxJUrV4SZmZlYunSpSEhIEEeOHBGNGjUSQ4YMkcZwc3MTNjY2YtWqVSIhIUF8/PHHwtLSUnTs2FFs2bJFxMfHi+7du4u6detKsd+4cUMsWrRInDlzRiQmJoqwsDBhaGgoPRH7/v37ws/PTwwfPlyK58mTJ4UKkTNnzgilUik+/vhjERsbKy5cuCCWL19e7B+n5/e/f/++6N+/vwAgPbzzRTlfuHBBAFArogvaCgqN5wuZOXPmiDp16og9e/aIxMREER4eLpRKpYiKihJCCBEcHCy6dOkihHhalNrY2IgqVaqIP/74QwghxI8//iiqVq1aZE7PklPImJubi/fff19cvHhR/Pbbb8LY2FgEBASI0aNHi3/++UesW7dOAFB7wGe7du1E165dxd9//y0SEhLEhAkThK2tbYkPun1RIfP8z+uLPhtCPC1kLCwsxOzZs0VCQoKYPXu2MDQ0FJ06dRKrV6+WPoe2trYiOztbCPG0CHvR7xjSLyxkqFzdunVLVKlSRRgYGIgdO3aI2bNniw4dOqhtU/BfaHx8fJF93L59WwAQ58+fF0L83y/GZcuWlTj2vn37hKGhoUhOTpbaLl68KBUuQvzfkYyijsQUKO6PRmBgoHBzc1Mrfnr37i369OkjhHh6NMjQ0FD8+++/avu1bdtWhISEFDve559/LoyMjISNjY3o2LGjWLhwoXQUSYinT1J/9hd+Xl6eqFKlioiIiBBCCLFkyRLh6empdmToWW5ubmLp0qVqbcOGDRMjRoxQa/vrr7+EgYGBePTokbTfwIEDpfWpqakCgJg6darUVlCIPf/k7Gd16dJFTJgwQXrt7+8vxo4dq7bN84VIv379RPPmzYvt83kF+5uZmQkzMzPpaM97770nK+cGDRqIWbNmSetDQkJEkyZNpNfPFjI5OTnC1NRUHD16VK3PYcOGiX79+gkhnh4VsrKyEk+ePBGxsbHC0dFRjB07VkyaNEkI8bTw7d+//wvzk1PImJqaiszMTKktICBAuLu7qz21uHbt2iI0NFR6DywtLUVOTo5a3zVq1BDffvttsTEBEJUrV5beczMzM7Fjxw6Nf16FKPqz0aJFC+n1kydPhJmZmRg0aJDUVvA5PHbsmBBClOp3DFVsnOxL5cre3h4jR45E3bp10b17d5w9exYHDx6Eubm5tNSpUwcAkJiYCAC4fPky+vXrh+rVq8PS0hLu7u4Ans5nedabb75Z4thxcXFwdXWFq6ur1Obl5QVra2vExcVpJb969erB0NBQeu3k5CRNyj1//jzy8/Ph6emplm90dLSUa1Hmzp2LtLQ0rFq1CvXq1cOqVatQp04daaKqs7MzunTpgnXr1gEAdu7cidzcXPTu3RsA0Lt3bzx69AjVq1fH8OHDsWPHDjx58qTEPM6ePYuIiAi1OAMCAqBSqZCUlCRt5+PjI33t4OAAAPD29i7UVvAe5OfnY/bs2fD29oaNjQ3Mzc2xd+/eQt/LF4mNjUXbtm1l7QMAf/31F2JiYhAREQFPT0+sWrVKVs4DBgzAxo0bATydxP7TTz9hwIABRY515coVPHz4EO3bt1fr8/vvv5e+3y1btsSDBw9w5swZREdHw9/fH61bt0ZUVBQAIDo6Gq1bt5adZ0nc3d1hYWEhvXZwcICXlxcMDAzU2gq+Z2fPnkVWVhZsbW3V8khKSirxcwsAS5cuRWxsrLS0b99eWvf8z6umn41nP3OGhoawtbUt8TOnye8Y0i+c7EvlzsjICEZGTz+KWVlZ6Nq1KxYsWFBou4LJtl27doWbmxu+++47ODs7Q6VSoX79+tJExwJmZma6D/4Fnp+wqFAooFKpADzN1dDQEDExMWrFDgCYm5uX2K+trS169+6N3r17Y968eWjUqBEWL16M9evXAwA+/PBDDBo0CEuXLkV4eDj69OkDU1NTAICrqyvi4+Nx4MAB7N+/H5988gkWLVqE6OjoYidYZmVlYeTIkRgzZkyhddWqVSsyX4VCUWxbwXuwaNEifPXVV1i2bBm8vb1hZmaGcePGFfpevkhpJ7B6eHjA2toatWvXRnp6Ovr06YNDhw4B0Cznfv36YdKkSTh9+jQePXqElJQU9OnTp8ixsrKyAAC///47qlatqrZOqVQCAKytrdGgQQNERUXh2LFjaN++PVq1aiVdbXT58mX4+/uXKtfiFPUZfdHn1snJSSqunvWiS7sdHR1Rs2ZNtbbbt28DKPzzquln40XxP/+Z0+R3DOkXFjJUofj6+uLnn3+Gu7u7VNw8686dO4iPj8d3332Hli1bAgAOHz5cqrHq1q2LlJQUpKSkSEdlLl26hPv378PLy0vjfgquvMjPz5c1fqNGjZCfn4/09HQpl9IwNjZGjRo11K5a6ty5M8zMzLBy5Urs2bNH+uNcwMTEBF27dkXXrl0RFBQkHdHx9fWFsbFxoVx8fX1x6dKlQn+EXtaRI0fQrVs3DBw4EMDTPzYJCQlq739R8TzPx8cHkZGRmDlzZqljCQoKQmhoKHbs2IEePXpolLOLiwv8/f2xYcMGPHr0CO3bt4e9vX2R23p5eUGpVCI5ObnEYsTf3x8HDx7EyZMnMXfuXNjY2KBu3bqYO3cunJyc4OnpWeoctcHX1xdpaWkwMjKSjobqgiafjdJ40e8Y0j88tUQVSlBQEO7evYt+/frh77//RmJiIvbu3YuhQ4ciPz8fb7zxBmxtbbF69WpcuXIFf/75J4KDg0s1Vrt27eDt7Y0BAwbg9OnTOHnyJAYPHgx/f/8XnpZ6lr29PUxMTLBnzx7cunULGRkZGu3n6emJAQMGYPDgwdi+fTuSkpJw8uRJhIaG4vfffy9yn127dmHgwIHYtWsXEhISEB8fj8WLF2P37t3o1q2btJ2hoSGGDBmCkJAQ1KpVC35+ftK6iIgIrF27FhcuXMDVq1fx448/wsTEBG5ubgCenmo4dOgQ/v33X/z3338AgEmTJuHo0aMYNWoUYmNjcfnyZfz6668YNWqUxu9TUWrVqoX9+/fj6NGjiIuLw8iRI3Hr1i21bdzd3XHixAlcu3YN//33n/Sf9bNCQkLw999/45NPPsG5c+fwzz//YOXKlVL8mjA1NcXw4cMxffp0CCE0znnAgAHYtGkTtm7dWuxpJQCwsLDAp59+ivHjx2P9+vVITEzE6dOnsXz5culIGvD0fjB79+6FkZGRdMqjdevW2LBhg9aPxpRGu3bt4Ofnh+7du2Pfvn24du0ajh49ii+++AKnTp3S2jiafDZK40W/Y0j/sJChCsXZ2RlHjhxBfn4+OnToAG9vb4wbNw7W1tYwMDCAgYEBNm3ahJiYGNSvXx/jx4/HokWLSjWWQqHAr7/+ijfeeAOtWrVCu3btUL16dWzevFlWP0ZGRggLC8O3334LZ2dntYLiRcLDwzF48GBMmDABtWvXRvfu3fH333+rna55lpeXF0xNTTFhwgQ0bNgQTZs2xZYtW7BmzRoMGjRIbdthw4YhLy8PQ4cOVWu3trbGd999h+bNm8PHxwcHDhzAzp07YWtrCwCYNWsWrl27hho1asDOzg7A0yMe0dHRSEhIQMuWLdGoUSNMmzYNzs7Oct6qQqZMmQJfX18EBASgdevWcHR0LHQn3E8//RSGhobw8vKCnZ1dkfNnPD09sW/fPpw9exZvv/02/Pz88Ouvv8r+j3vUqFGIi4vD1q1bNc75/fffx507d/Dw4cMX3sV39uzZmDp1KkJDQ1G3bl107NgRv//+Ozw8PKRtWrZsCZVKpVa0tG7dGvn5+VqfH1MaCoUCu3fvRqtWrTB06FB4enqib9++uH79ujQfRRs0+WyUxot+x5D+UQghRHkHQUTa99dff6Ft27ZISUnR6h8YIqKKhIUM0SsmNzcXt2/fRmBgIBwdHbFhw4byDomISGd4HI3oFfPTTz/Bzc0N9+/fx8KFC8s7HCIineIRGSIiItJbPCJDREREeouFDBEREektFjJERESkt1jIEBERkd5iIUNERER6i4UMERER6S0WMkRERKS3WMgQERGR3mIhQ0RERHrr/wF8gMvE5ffjogAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from evaluation.plotting import draw_docs_by_publication_date\n",
    "\n",
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    draw_docs_by_publication_date(\n",
    "        df_screened=df_s,\n",
    "        col_date=\"publication_date\",\n",
    "        col_relevant=\"is_relevant\",\n",
    "        output_dir=PATH_LATEX_PLOT_OUTPUT,\n",
    "        file_name=f\"screened_by_publication_year.pdf\",\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from evaluation.plotting import draw_docs_publication_types\n",
    "\n",
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    draw_docs_publication_types(\n",
    "        df_screened=df_s,\n",
    "        n_most_common=10,\n",
    "        col_ptypes=\"publication_types\",\n",
    "        col_relevant=\"is_relevant\",\n",
    "        output_dir=PATH_LATEX_PLOT_OUTPUT,\n",
    "        file_name=f\"publication_types_included.pdf\",\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Retrieved Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from evaluation.plotting import set_size\n",
    "from matplotlib.dates import AutoDateLocator\n",
    "\n",
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    monthly_retrievals = (\n",
    "        df_query.query(\"source != 'Trialstreamer'\")\n",
    "        .groupby([pd.Grouper(key=\"publication_date_combined\", freq=\"3M\"), \"source\"])\n",
    "        .size()\n",
    "        .unstack()\n",
    "        .rename(index=lambda d: f\"Q{pd.Timestamp(d).quarter} {d.year}\")\n",
    "        .rename_axis(\"Source\", axis=1)\n",
    "    )[[\"Pubmed\", \"ClinicalTrials\", \"Civic\"]]\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=set_size(\"thesis\"), layout=\"constrained\")\n",
    "    monthly_retrievals.plot(\n",
    "        kind=\"bar\",\n",
    "        stacked=True,\n",
    "        ax=ax,\n",
    "        xlabel=\"Month of the Systematic Review Time Frame\",\n",
    "        ylabel=\"Number of Retrieved Publications\",\n",
    "        rot=0,\n",
    "    )\n",
    "    locator = AutoDateLocator()\n",
    "    locator.intervald[\"QUARTER\"] = 1\n",
    "    ax.xaxis.set_major_locator(locator)\n",
    "    fig.savefig(PATH_LATEX_PLOT_OUTPUT / f\"monthly_retrievals_by_source.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "monhtly_retrieved_vs_screened_pm = (\n",
    "    df_query.query(\"source == 'Pubmed' & is_rct_pt == True & is_in_ggponc == False\")\n",
    "    .merge(df_s[[\"pm_id\", \"is_included\"]], on=\"pm_id\", how=\"left\", indicator=True)\n",
    "    .groupby([pd.Grouper(key=\"publication_date_combined\", freq=\"3M\"), \"_merge\"])\n",
    "    .size()\n",
    "    .unstack()[[\"both\", \"left_only\"]]\n",
    "    .rename(\n",
    "        index=lambda d: f\"Q{pd.Timestamp(d).quarter} {d.year}\",\n",
    "        columns={\"left_only\": \"Unscreened\", \"both\": \"Screened\"},\n",
    "    )\n",
    "    .rename_axis(None, axis=1)\n",
    ").astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib.ticker import MaxNLocator\n",
    "\n",
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    fig, ax = plt.subplots(figsize=set_size(\"thesis\"), layout=\"constrained\")\n",
    "\n",
    "    monhtly_retrieved_vs_screened_pm = (\n",
    "        df_query.query(\"source == 'Pubmed' & is_rct_pt == True & is_in_ggponc == False\")\n",
    "        .merge(df_s[[\"pm_id\", \"is_included\"]], on=\"pm_id\", how=\"left\", indicator=True)\n",
    "        .groupby([pd.Grouper(key=\"publication_date_combined\", freq=\"3M\"), \"_merge\"])\n",
    "        .size()\n",
    "        .unstack()[[\"both\", \"left_only\"]]\n",
    "        .rename(\n",
    "            index=lambda d: f\"Q{pd.Timestamp(d).quarter} {d.year}\",\n",
    "            columns={\"left_only\": \"Unscreened\", \"both\": \"Screened\"},\n",
    "        )\n",
    "        .rename_axis(None, axis=1)\n",
    "    ).astype(int)\n",
    "\n",
    "    monhtly_retrieved_vs_screened_pm.plot(\n",
    "        kind=\"bar\",\n",
    "        stacked=True,\n",
    "        ylabel=\"Number of RCTs retrieved from Pubmed\",\n",
    "        xlabel=\"Month of the Systematic Review Time Frame\",\n",
    "        ax=ax,\n",
    "        rot=0,\n",
    "    )\n",
    "\n",
    "    locator = AutoDateLocator()\n",
    "    locator.intervald[\"QUARTER\"] = 1\n",
    "    ax.xaxis.set_major_locator(locator)\n",
    "    ax.yaxis.set_major_locator(MaxNLocator(integer=True))\n",
    "\n",
    "    fig.savefig(PATH_LATEX_PLOT_OUTPUT / f\"monthly_retrieved_vs_screened_rcts.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    df_query.loc[df_query.query(FILTER_HQ).index, \"is_hq\"] = True\n",
    "    df_query[\"is_hq\"] = df_query[\"is_hq\"].fillna(False)\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=set_size(\"thesis\"), layout=\"constrained\")\n",
    "    df_query.query(\"source != 'Trialstreamer'\").groupby(\n",
    "        [pd.Grouper(key=\"publication_date_combined\", freq=\"3M\"), \"is_hq\"]\n",
    "    ).size().unstack().rename(\n",
    "        index=lambda d: f\"Q{pd.Timestamp(d).quarter} {d.year}\",\n",
    "        columns={True: \"High\", False: \"Normal\"},\n",
    "    ).rename_axis(\n",
    "        \"Evidence Quality\", axis=1\n",
    "    ).plot(\n",
    "        kind=\"bar\",\n",
    "        stacked=True,\n",
    "        ax=ax,\n",
    "        xlabel=\"Month of the Systematic Review Time Frame\",\n",
    "        ylabel=\"Number of Retrieved Publications\",\n",
    "        rot=0,\n",
    "    )\n",
    "    locator = AutoDateLocator()\n",
    "    locator.intervald[\"QUARTER\"] = 1\n",
    "    ax.xaxis.set_major_locator(locator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 551.181x340.649 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if OVERWRITE_LATEX_OUTPUT:\n",
    "    df_merged = (\n",
    "        df_query.query(\"source == 'Pubmed' & is_rct_pt == True\")\n",
    "        .dropna(subset=\"pm_id\")\n",
    "        .merge(\n",
    "            df_s.query(\"is_rct_api == True\")[\n",
    "                [\"pm_id\", \"publication_date\", \"is_included\"]\n",
    "            ].dropna(subset=\"pm_id\"),\n",
    "            on=\"pm_id\",\n",
    "            how=\"outer\",\n",
    "            indicator=True,\n",
    "        )\n",
    "    )\n",
    "    df_merged[\"common_publication_date\"] = df_merged[\n",
    "        \"publication_date_combined\"\n",
    "    ].fillna(df_merged[\"publication_date_y\"])\n",
    "\n",
    "    montly_retrieved_vs_screened = (\n",
    "        df_merged.groupby(\n",
    "            [pd.Grouper(key=\"common_publication_date\", freq=\"3M\"), \"_merge\"]\n",
    "        )\n",
    "        .size()\n",
    "        .unstack()\n",
    "        .rename(\n",
    "            index=lambda d: f\"Q{pd.Timestamp(d).quarter} {d.year}\",\n",
    "            columns={\n",
    "                \"left_only\": \"Only Retrieved\",\n",
    "                \"both\": \"Screened and Retrieved\",\n",
    "                \"right_only\": \"Only Screened\",\n",
    "            },\n",
    "        )\n",
    "        .rename_axis(None, axis=1)\n",
    "    )[[\"Only Screened\", \"Screened and Retrieved\", \"Only Retrieved\"]]\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=set_size(\"thesis\"), layout=\"constrained\")\n",
    "    montly_retrieved_vs_screened.plot(\n",
    "        kind=\"bar\",\n",
    "        stacked=True,\n",
    "        ax=ax,\n",
    "        xlabel=\"Year of the Systematic Review Time Frame\",\n",
    "        ylabel=\"Number of RCTs\",\n",
    "        rot=0,\n",
    "    )\n",
    "    locator = AutoDateLocator()\n",
    "    locator.intervald[\"QUARTER\"] = 1\n",
    "    ax.xaxis.set_major_locator(locator)\n",
    "\n",
    "    fig.savefig(PATH_LATEX_PLOT_OUTPUT / f\"yearly_retrieved_vs_screened_rcts.pdf\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:nge]",
   "language": "python",
   "name": "conda-env-nge-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}