{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RendererRegistry.enable('mimetype')"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from splink.duckdb.duckdb_linker import DuckDBLinker\n",
    "import altair as alt\n",
    "alt.renderers.enable('mimetype')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>unique_id</th>\n",
       "      <th>cluster</th>\n",
       "      <th>full_name</th>\n",
       "      <th>first_and_surname</th>\n",
       "      <th>first_name</th>\n",
       "      <th>surname</th>\n",
       "      <th>dob</th>\n",
       "      <th>birth_place</th>\n",
       "      <th>postcode_fake</th>\n",
       "      <th>gender</th>\n",
       "      <th>occupation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Q2296770-1</td>\n",
       "      <td>Q2296770</td>\n",
       "      <td>thomas clifford, 1st baron clifford of chudleigh</td>\n",
       "      <td>thomas chudleigh</td>\n",
       "      <td>thomas</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>1630-08-01</td>\n",
       "      <td>devon</td>\n",
       "      <td>tq13 8df</td>\n",
       "      <td>male</td>\n",
       "      <td>politician</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Q2296770-2</td>\n",
       "      <td>Q2296770</td>\n",
       "      <td>thomas of chudleigh</td>\n",
       "      <td>thomas chudleigh</td>\n",
       "      <td>thomas</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>1630-08-01</td>\n",
       "      <td>devon</td>\n",
       "      <td>tq13 8df</td>\n",
       "      <td>male</td>\n",
       "      <td>politician</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Q2296770-3</td>\n",
       "      <td>Q2296770</td>\n",
       "      <td>tom 1st baron clifford of chudleigh</td>\n",
       "      <td>tom chudleigh</td>\n",
       "      <td>tom</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>1630-08-01</td>\n",
       "      <td>devon</td>\n",
       "      <td>tq13 8df</td>\n",
       "      <td>male</td>\n",
       "      <td>politician</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Q2296770-4</td>\n",
       "      <td>Q2296770</td>\n",
       "      <td>thomas 1st chudleigh</td>\n",
       "      <td>thomas chudleigh</td>\n",
       "      <td>thomas</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>1630-08-01</td>\n",
       "      <td>devon</td>\n",
       "      <td>tq13 8hu</td>\n",
       "      <td>None</td>\n",
       "      <td>politician</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Q2296770-5</td>\n",
       "      <td>Q2296770</td>\n",
       "      <td>thomas clifford, 1st baron chudleigh</td>\n",
       "      <td>thomas chudleigh</td>\n",
       "      <td>thomas</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>1630-08-01</td>\n",
       "      <td>devon</td>\n",
       "      <td>tq13 8df</td>\n",
       "      <td>None</td>\n",
       "      <td>politician</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    unique_id   cluster                                         full_name  \\\n",
       "0  Q2296770-1  Q2296770  thomas clifford, 1st baron clifford of chudleigh   \n",
       "1  Q2296770-2  Q2296770                               thomas of chudleigh   \n",
       "2  Q2296770-3  Q2296770               tom 1st baron clifford of chudleigh   \n",
       "3  Q2296770-4  Q2296770                              thomas 1st chudleigh   \n",
       "4  Q2296770-5  Q2296770              thomas clifford, 1st baron chudleigh   \n",
       "\n",
       "  first_and_surname first_name    surname         dob birth_place  \\\n",
       "0  thomas chudleigh     thomas  chudleigh  1630-08-01       devon   \n",
       "1  thomas chudleigh     thomas  chudleigh  1630-08-01       devon   \n",
       "2     tom chudleigh        tom  chudleigh  1630-08-01       devon   \n",
       "3  thomas chudleigh     thomas  chudleigh  1630-08-01       devon   \n",
       "4  thomas chudleigh     thomas  chudleigh  1630-08-01       devon   \n",
       "\n",
       "  postcode_fake gender  occupation  \n",
       "0      tq13 8df   male  politician  \n",
       "1      tq13 8df   male  politician  \n",
       "2      tq13 8df   male  politician  \n",
       "3      tq13 8hu   None  politician  \n",
       "4      tq13 8df   None  politician  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd \n",
    "pd.options.display.max_rows = 1000\n",
    "df = pd.read_parquet(\"./data/historical_figures_with_errors_50k.parquet\")\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v4+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v4.8.1.json",
       "config": {
        "view": {
         "continuousHeight": 300,
         "continuousWidth": 400
        }
       },
       "vconcat": [
        {
         "hconcat": [
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.9449625015258789,
              "percentile_inc_nulls": 0.9450353980064392,
              "sum_tokens_in_value_count_group": 2780,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 2780
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.8907960653305054,
              "percentile_inc_nulls": 0.8909407258033752,
              "sum_tokens_in_value_count_group": 2736,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 2736
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.8621290326118469,
              "percentile_inc_nulls": 0.8623116612434387,
              "sum_tokens_in_value_count_group": 1448,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 1448
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.8341153264045715,
              "percentile_inc_nulls": 0.8343350887298584,
              "sum_tokens_in_value_count_group": 1415,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 1415
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.8082596063613892,
              "percentile_inc_nulls": 0.8085135817527771,
              "sum_tokens_in_value_count_group": 1306,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 1306
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.7832155227661133,
              "percentile_inc_nulls": 0.7835026979446411,
              "sum_tokens_in_value_count_group": 1265,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 1265
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.7582308650016785,
              "percentile_inc_nulls": 0.7585511207580566,
              "sum_tokens_in_value_count_group": 1262,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 1262
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.7341569066047668,
              "percentile_inc_nulls": 0.7345091104507446,
              "sum_tokens_in_value_count_group": 1216,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 1216
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.7161211967468262,
              "percentile_inc_nulls": 0.7164973020553589,
              "sum_tokens_in_value_count_group": 911,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 911
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.698620080947876,
              "percentile_inc_nulls": 0.6990193128585815,
              "sum_tokens_in_value_count_group": 884,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 884
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6826632022857666,
              "percentile_inc_nulls": 0.6830835342407227,
              "sum_tokens_in_value_count_group": 806,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 806
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6697155237197876,
              "percentile_inc_nulls": 0.670153021812439,
              "sum_tokens_in_value_count_group": 654,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 654
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6573221683502197,
              "percentile_inc_nulls": 0.6577761173248291,
              "sum_tokens_in_value_count_group": 626,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 626
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6471858024597168,
              "percentile_inc_nulls": 0.6476531028747559,
              "sum_tokens_in_value_count_group": 512,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 512
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6370691657066345,
              "percentile_inc_nulls": 0.6375499367713928,
              "sum_tokens_in_value_count_group": 511,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 511
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6269921064376831,
              "percentile_inc_nulls": 0.6274862289428711,
              "sum_tokens_in_value_count_group": 509,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 509
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6171329021453857,
              "percentile_inc_nulls": 0.6176400780677795,
              "sum_tokens_in_value_count_group": 498,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 498
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.6079269647598267,
              "percentile_inc_nulls": 0.6084463596343994,
              "sum_tokens_in_value_count_group": 465,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 465
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5987408757209778,
              "percentile_inc_nulls": 0.5992723703384399,
              "sum_tokens_in_value_count_group": 464,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 464
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5904456377029419,
              "percentile_inc_nulls": 0.5909881591796875,
              "sum_tokens_in_value_count_group": 419,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 419
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5831403136253357,
              "percentile_inc_nulls": 0.5836925506591797,
              "sum_tokens_in_value_count_group": 369,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 369
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5762308835983276,
              "percentile_inc_nulls": 0.5767922401428223,
              "sum_tokens_in_value_count_group": 349,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 349
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5698164701461792,
              "percentile_inc_nulls": 0.5703862905502319,
              "sum_tokens_in_value_count_group": 324,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 324
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5634812116622925,
              "percentile_inc_nulls": 0.5640594959259033,
              "sum_tokens_in_value_count_group": 320,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 320
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5573043823242188,
              "percentile_inc_nulls": 0.557890772819519,
              "sum_tokens_in_value_count_group": 312,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 312
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5513848066329956,
              "percentile_inc_nulls": 0.551979124546051,
              "sum_tokens_in_value_count_group": 299,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 299
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5457029342651367,
              "percentile_inc_nulls": 0.5463047027587891,
              "sum_tokens_in_value_count_group": 287,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 287
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5401991605758667,
              "percentile_inc_nulls": 0.5408082604408264,
              "sum_tokens_in_value_count_group": 278,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 278
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5348538160324097,
              "percentile_inc_nulls": 0.5354700088500977,
              "sum_tokens_in_value_count_group": 270,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 270
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5298252105712891,
              "percentile_inc_nulls": 0.5304480195045471,
              "sum_tokens_in_value_count_group": 254,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 254
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5250935554504395,
              "percentile_inc_nulls": 0.5257226228713989,
              "sum_tokens_in_value_count_group": 239,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 239
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5204806923866272,
              "percentile_inc_nulls": 0.5211158990859985,
              "sum_tokens_in_value_count_group": 233,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 233
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5164815187454224,
              "percentile_inc_nulls": 0.5171220302581787,
              "sum_tokens_in_value_count_group": 202,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 202
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.51287841796875,
              "percentile_inc_nulls": 0.5135236978530884,
              "sum_tokens_in_value_count_group": 182,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 182
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5092949867248535,
              "percentile_inc_nulls": 0.5099450349807739,
              "sum_tokens_in_value_count_group": 181,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 181
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.5058304071426392,
              "percentile_inc_nulls": 0.5064850449562073,
              "sum_tokens_in_value_count_group": 175,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 175
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.49898040294647217,
              "percentile_inc_nulls": 0.49964410066604614,
              "sum_tokens_in_value_count_group": 346,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 173
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.49563461542129517,
              "percentile_inc_nulls": 0.4963027238845825,
              "sum_tokens_in_value_count_group": 169,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 169
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4923086166381836,
              "percentile_inc_nulls": 0.4929811358451843,
              "sum_tokens_in_value_count_group": 168,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 168
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4890221953392029,
              "percentile_inc_nulls": 0.4896990656852722,
              "sum_tokens_in_value_count_group": 166,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 166
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.48587435483932495,
              "percentile_inc_nulls": 0.48655539751052856,
              "sum_tokens_in_value_count_group": 159,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 159
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4827859401702881,
              "percentile_inc_nulls": 0.4834710955619812,
              "sum_tokens_in_value_count_group": 156,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 156
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.47664862871170044,
              "percentile_inc_nulls": 0.4773419499397278,
              "sum_tokens_in_value_count_group": 310,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 155
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4705509543418884,
              "percentile_inc_nulls": 0.47125232219696045,
              "sum_tokens_in_value_count_group": 308,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 154
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4675813317298889,
              "percentile_inc_nulls": 0.4682866334915161,
              "sum_tokens_in_value_count_group": 150,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 150
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.46471065282821655,
              "percentile_inc_nulls": 0.4654197692871094,
              "sum_tokens_in_value_count_group": 145,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 145
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.46195876598358154,
              "percentile_inc_nulls": 0.46267151832580566,
              "sum_tokens_in_value_count_group": 139,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 139
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.459286093711853,
              "percentile_inc_nulls": 0.4600023627281189,
              "sum_tokens_in_value_count_group": 135,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 135
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.45405954122543335,
              "percentile_inc_nulls": 0.45478272438049316,
              "sum_tokens_in_value_count_group": 264,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 132
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.451485812664032,
              "percentile_inc_nulls": 0.45221245288848877,
              "sum_tokens_in_value_count_group": 130,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 130
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4464572072029114,
              "percentile_inc_nulls": 0.44719046354293823,
              "sum_tokens_in_value_count_group": 254,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 127
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.44402211904525757,
              "percentile_inc_nulls": 0.4447585940361023,
              "sum_tokens_in_value_count_group": 123,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 123
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.44164639711380005,
              "percentile_inc_nulls": 0.44238603115081787,
              "sum_tokens_in_value_count_group": 120,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 120
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4393102526664734,
              "percentile_inc_nulls": 0.44005298614501953,
              "sum_tokens_in_value_count_group": 118,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 118
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4370335340499878,
              "percentile_inc_nulls": 0.4377792477607727,
              "sum_tokens_in_value_count_group": 115,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 115
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4347766041755676,
              "percentile_inc_nulls": 0.4355252981185913,
              "sum_tokens_in_value_count_group": 114,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 114
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4325592517852783,
              "percentile_inc_nulls": 0.43331092596054077,
              "sum_tokens_in_value_count_group": 112,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 112
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4304013252258301,
              "percentile_inc_nulls": 0.43115586042404175,
              "sum_tokens_in_value_count_group": 109,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 109
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4283621311187744,
              "percentile_inc_nulls": 0.42911940813064575,
              "sum_tokens_in_value_count_group": 103,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 103
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.42634278535842896,
              "percentile_inc_nulls": 0.4271026849746704,
              "sum_tokens_in_value_count_group": 102,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 102
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4223436713218689,
              "percentile_inc_nulls": 0.42310887575149536,
              "sum_tokens_in_value_count_group": 202,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 101
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4203639030456543,
              "percentile_inc_nulls": 0.4211317300796509,
              "sum_tokens_in_value_count_group": 100,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 100
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.418443500995636,
              "percentile_inc_nulls": 0.4192138910293579,
              "sum_tokens_in_value_count_group": 97,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 97
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4165429472923279,
              "percentile_inc_nulls": 0.41731584072113037,
              "sum_tokens_in_value_count_group": 96,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 96
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4146621823310852,
              "percentile_inc_nulls": 0.4154375195503235,
              "sum_tokens_in_value_count_group": 95,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 95
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4128011465072632,
              "percentile_inc_nulls": 0.4135790467262268,
              "sum_tokens_in_value_count_group": 94,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 94
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4055156111717224,
              "percentile_inc_nulls": 0.4063031077384949,
              "sum_tokens_in_value_count_group": 368,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 92
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.4001108407974243,
              "percentile_inc_nulls": 0.4009055495262146,
              "sum_tokens_in_value_count_group": 273,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 91
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.39832907915115356,
              "percentile_inc_nulls": 0.3991261124610901,
              "sum_tokens_in_value_count_group": 90,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 90
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3948447108268738,
              "percentile_inc_nulls": 0.395646333694458,
              "sum_tokens_in_value_count_group": 176,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 88
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.39139991998672485,
              "percentile_inc_nulls": 0.392206072807312,
              "sum_tokens_in_value_count_group": 174,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 87
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.38971710205078125,
              "percentile_inc_nulls": 0.3905255198478699,
              "sum_tokens_in_value_count_group": 85,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 85
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3864702582359314,
              "percentile_inc_nulls": 0.3872830271720886,
              "sum_tokens_in_value_count_group": 164,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 82
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3848666548728943,
              "percentile_inc_nulls": 0.38568150997161865,
              "sum_tokens_in_value_count_group": 81,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 81
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3816990256309509,
              "percentile_inc_nulls": 0.38251811265945435,
              "sum_tokens_in_value_count_group": 160,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 80
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3801349997520447,
              "percentile_inc_nulls": 0.38095617294311523,
              "sum_tokens_in_value_count_group": 79,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 79
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.37859082221984863,
              "percentile_inc_nulls": 0.3794139623641968,
              "sum_tokens_in_value_count_group": 78,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 78
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.37706637382507324,
              "percentile_inc_nulls": 0.3778916001319885,
              "sum_tokens_in_value_count_group": 77,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 77
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.37556177377700806,
              "percentile_inc_nulls": 0.3763889670372009,
              "sum_tokens_in_value_count_group": 76,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 76
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3740769624710083,
              "percentile_inc_nulls": 0.374906063079834,
              "sum_tokens_in_value_count_group": 75,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 75
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3726118803024292,
              "percentile_inc_nulls": 0.37344300746917725,
              "sum_tokens_in_value_count_group": 74,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 74
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3697214722633362,
              "percentile_inc_nulls": 0.3705563545227051,
              "sum_tokens_in_value_count_group": 146,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 73
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.36687058210372925,
              "percentile_inc_nulls": 0.36770927906036377,
              "sum_tokens_in_value_count_group": 144,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 72
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3654649257659912,
              "percentile_inc_nulls": 0.36630553007125854,
              "sum_tokens_in_value_count_group": 71,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 71
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3640791177749634,
              "percentile_inc_nulls": 0.364921510219574,
              "sum_tokens_in_value_count_group": 70,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 70
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.362713098526001,
              "percentile_inc_nulls": 0.36355727910995483,
              "sum_tokens_in_value_count_group": 69,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 69
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3586743474006653,
              "percentile_inc_nulls": 0.3595238924026489,
              "sum_tokens_in_value_count_group": 204,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 68
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.35734790563583374,
              "percentile_inc_nulls": 0.35819923877716064,
              "sum_tokens_in_value_count_group": 67,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 67
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.35606104135513306,
              "percentile_inc_nulls": 0.35691410303115845,
              "sum_tokens_in_value_count_group": 65,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 65
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3547940254211426,
              "percentile_inc_nulls": 0.3556486964225769,
              "sum_tokens_in_value_count_group": 64,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 64
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3522995114326477,
              "percentile_inc_nulls": 0.35315752029418945,
              "sum_tokens_in_value_count_group": 126,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 63
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3510720729827881,
              "percentile_inc_nulls": 0.35193169116973877,
              "sum_tokens_in_value_count_group": 62,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 62
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.34744906425476074,
              "percentile_inc_nulls": 0.34831351041793823,
              "sum_tokens_in_value_count_group": 183,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 61
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3427768349647522,
              "percentile_inc_nulls": 0.34364742040634155,
              "sum_tokens_in_value_count_group": 236,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 59
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.33933204412460327,
              "percentile_inc_nulls": 0.34020721912384033,
              "sum_tokens_in_value_count_group": 174,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 58
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3382035493850708,
              "percentile_inc_nulls": 0.33908021450042725,
              "sum_tokens_in_value_count_group": 57,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 57
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.33709490299224854,
              "percentile_inc_nulls": 0.33797305822372437,
              "sum_tokens_in_value_count_group": 56,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 56
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.33602583408355713,
              "percentile_inc_nulls": 0.3369053602218628,
              "sum_tokens_in_value_count_group": 54,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 54
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3328779935836792,
              "percentile_inc_nulls": 0.33376169204711914,
              "sum_tokens_in_value_count_group": 159,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 53
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3287600874900818,
              "percentile_inc_nulls": 0.32964926958084106,
              "sum_tokens_in_value_count_group": 208,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 52
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.32674068212509155,
              "percentile_inc_nulls": 0.3276325464248657,
              "sum_tokens_in_value_count_group": 102,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 51
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.32575082778930664,
              "percentile_inc_nulls": 0.32664400339126587,
              "sum_tokens_in_value_count_group": 50,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 50
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3238106369972229,
              "percentile_inc_nulls": 0.32470637559890747,
              "sum_tokens_in_value_count_group": 98,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 49
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.32095980644226074,
              "percentile_inc_nulls": 0.32185930013656616,
              "sum_tokens_in_value_count_group": 144,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 48
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3200293183326721,
              "percentile_inc_nulls": 0.3209300637245178,
              "sum_tokens_in_value_count_group": 47,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 47
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.31729722023010254,
              "percentile_inc_nulls": 0.31820160150527954,
              "sum_tokens_in_value_count_group": 138,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 46
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.31373363733291626,
              "percentile_inc_nulls": 0.3146427273750305,
              "sum_tokens_in_value_count_group": 180,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 45
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.30937814712524414,
              "percentile_inc_nulls": 0.3102930188179016,
              "sum_tokens_in_value_count_group": 220,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 44
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.3076755404472351,
              "percentile_inc_nulls": 0.30859267711639404,
              "sum_tokens_in_value_count_group": 86,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 43
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.30434954166412354,
              "percentile_inc_nulls": 0.30527108907699585,
              "sum_tokens_in_value_count_group": 168,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 42
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.30353784561157227,
              "percentile_inc_nulls": 0.30446046590805054,
              "sum_tokens_in_value_count_group": 41,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 41
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2987864017486572,
              "percentile_inc_nulls": 0.2997152805328369,
              "sum_tokens_in_value_count_group": 240,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 40
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.29569798707962036,
              "percentile_inc_nulls": 0.2966309189796448,
              "sum_tokens_in_value_count_group": 156,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 39
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2934410572052002,
              "percentile_inc_nulls": 0.29437702894210815,
              "sum_tokens_in_value_count_group": 114,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 38
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2912434935569763,
              "percentile_inc_nulls": 0.29218238592147827,
              "sum_tokens_in_value_count_group": 111,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 37
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2891647219657898,
              "percentile_inc_nulls": 0.2901063561439514,
              "sum_tokens_in_value_count_group": 105,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 35
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.28579914569854736,
              "percentile_inc_nulls": 0.28674525022506714,
              "sum_tokens_in_value_count_group": 170,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 34
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.281879186630249,
              "percentile_inc_nulls": 0.28283047676086426,
              "sum_tokens_in_value_count_group": 198,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 33
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2799786329269409,
              "percentile_inc_nulls": 0.2809324264526367,
              "sum_tokens_in_value_count_group": 96,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 32
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2775236964225769,
              "percentile_inc_nulls": 0.27848076820373535,
              "sum_tokens_in_value_count_group": 124,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 31
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.27336621284484863,
              "percentile_inc_nulls": 0.2743287682533264,
              "sum_tokens_in_value_count_group": 210,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 30
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.27049553394317627,
              "percentile_inc_nulls": 0.2714619040489197,
              "sum_tokens_in_value_count_group": 145,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 29
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.26772385835647583,
              "percentile_inc_nulls": 0.2686939239501953,
              "sum_tokens_in_value_count_group": 140,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 28
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2639821171760559,
              "percentile_inc_nulls": 0.264957070350647,
              "sum_tokens_in_value_count_group": 189,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 27
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.26037895679473877,
              "percentile_inc_nulls": 0.26135867834091187,
              "sum_tokens_in_value_count_group": 182,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 26
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2579042315483093,
              "percentile_inc_nulls": 0.25888729095458984,
              "sum_tokens_in_value_count_group": 125,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 25
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.25410306453704834,
              "percentile_inc_nulls": 0.25509113073349,
              "sum_tokens_in_value_count_group": 192,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 24
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.25091564655303955,
              "percentile_inc_nulls": 0.25190794467926025,
              "sum_tokens_in_value_count_group": 161,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 23
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.24612462520599365,
              "percentile_inc_nulls": 0.24712324142456055,
              "sum_tokens_in_value_count_group": 242,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 22
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2411355972290039,
              "percentile_inc_nulls": 0.24214082956314087,
              "sum_tokens_in_value_count_group": 252,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 21
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.23638415336608887,
              "percentile_inc_nulls": 0.23739570379257202,
              "sum_tokens_in_value_count_group": 240,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 20
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.22998952865600586,
              "percentile_inc_nulls": 0.23100954294204712,
              "sum_tokens_in_value_count_group": 323,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 19
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2228623628616333,
              "percentile_inc_nulls": 0.22389179468154907,
              "sum_tokens_in_value_count_group": 360,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 18
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.2188236117362976,
              "percentile_inc_nulls": 0.21985840797424316,
              "sum_tokens_in_value_count_group": 204,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 17
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.20995426177978516,
              "percentile_inc_nulls": 0.21100085973739624,
              "sum_tokens_in_value_count_group": 448,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 16
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.20193618535995483,
              "percentile_inc_nulls": 0.20299339294433594,
              "sum_tokens_in_value_count_group": 405,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 15
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.194452702999115,
              "percentile_inc_nulls": 0.1955198049545288,
              "sum_tokens_in_value_count_group": 378,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 14
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.18158423900604248,
              "percentile_inc_nulls": 0.18266832828521729,
              "sum_tokens_in_value_count_group": 650,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 13
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.17279404401779175,
              "percentile_inc_nulls": 0.17388981580734253,
              "sum_tokens_in_value_count_group": 444,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 12
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.16538971662521362,
              "percentile_inc_nulls": 0.16649532318115234,
              "sum_tokens_in_value_count_group": 374,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 11
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.15588682889938354,
              "percentile_inc_nulls": 0.15700501203536987,
              "sum_tokens_in_value_count_group": 480,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 10
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.14394885301589966,
              "percentile_inc_nulls": 0.14508283138275146,
              "sum_tokens_in_value_count_group": 603,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 9
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.13143670558929443,
              "percentile_inc_nulls": 0.13258731365203857,
              "sum_tokens_in_value_count_group": 632,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 8
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.12145870923995972,
              "percentile_inc_nulls": 0.12262248992919922,
              "sum_tokens_in_value_count_group": 504,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 7
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.11076796054840088,
              "percentile_inc_nulls": 0.11194592714309692,
              "sum_tokens_in_value_count_group": 540,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 6
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.10017621517181396,
              "percentile_inc_nulls": 0.10136818885803223,
              "sum_tokens_in_value_count_group": 535,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 5
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.08655542135238647,
              "percentile_inc_nulls": 0.08776545524597168,
              "sum_tokens_in_value_count_group": 688,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 4
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.0702817440032959,
              "percentile_inc_nulls": 0.07151329517364502,
              "sum_tokens_in_value_count_group": 822,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 3
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0.047276854515075684,
              "percentile_inc_nulls": 0.04853886365890503,
              "sum_tokens_in_value_count_group": 1162,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 2
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 0,
              "percentile_inc_nulls": 0.0013247132301330566,
              "sum_tokens_in_value_count_group": 2388,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 1
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "percentile_ex_nulls": 1,
              "percentile_inc_nulls": 1,
              "sum_tokens_in_value_count_group": 2780,
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value_count": 2780
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "percentile_ex_nulls",
              "type": "quantitative"
             },
             {
              "field": "percentile_inc_nulls",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "percentile_ex_nulls",
             "sort": "descending",
             "title": "Percentile",
             "type": "quantitative"
            },
            "y": {
             "field": "value_count",
             "title": "Count of values",
             "type": "quantitative"
            }
           },
           "mark": {
            "interpolate": "step-after",
            "type": "line"
           },
           "title": {
            "subtitle": "In this col, 67 values (0.1%) are null and there are 4413 distinct values",
            "text": "Distribution of counts of values in column first_name"
           }
          },
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "william",
              "value_count": 2780
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "john",
              "value_count": 2736
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "thomas",
              "value_count": 1448
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "george",
              "value_count": 1415
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "henry",
              "value_count": 1306
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "james",
              "value_count": 1265
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "sir",
              "value_count": 1262
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "charles",
              "value_count": 1216
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "edward",
              "value_count": 911
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "robert",
              "value_count": 884
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value",
              "type": "nominal"
             },
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "value",
             "sort": "-y",
             "title": null,
             "type": "nominal"
            },
            "y": {
             "field": "value_count",
             "title": "Value count",
             "type": "quantitative"
            }
           },
           "mark": "bar",
           "title": "Top 10 values by value count"
          },
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "clifford,",
              "value_count": 1
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "feank",
              "value_count": 1
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "jerald",
              "value_count": 1
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "watts",
              "value_count": 1
             },
             {
              "distinct_value_count": 4413,
              "group_name": "first_name",
              "total_non_null_rows": 50511,
              "total_rows_inc_nulls": 50578,
              "value": "pzul",
              "value_count": 1
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value",
              "type": "nominal"
             },
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "value",
             "sort": "-y",
             "title": null,
             "type": "nominal"
            },
            "y": {
             "field": "value_count",
             "scale": {
              "domain": [
               0,
               2780
              ]
             },
             "title": "Value count",
             "type": "quantitative"
            }
           },
           "mark": "bar",
           "title": "Bottom 5 values by value count"
          }
         ]
        },
        {
         "hconcat": [
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9991315603256226,
              "percentile_inc_nulls": 0.9993277788162231,
              "sum_tokens_in_value_count_group": 34,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 34
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9955557584762573,
              "percentile_inc_nulls": 0.9965597987174988,
              "sum_tokens_in_value_count_group": 140,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 28
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9941765666007996,
              "percentile_inc_nulls": 0.9954921007156372,
              "sum_tokens_in_value_count_group": 54,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 27
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9935380220413208,
              "percentile_inc_nulls": 0.9949977993965149,
              "sum_tokens_in_value_count_group": 25,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 25
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9929249882698059,
              "percentile_inc_nulls": 0.994523286819458,
              "sum_tokens_in_value_count_group": 24,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 24
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9923630952835083,
              "percentile_inc_nulls": 0.994088351726532,
              "sum_tokens_in_value_count_group": 22,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 22
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9912903308868408,
              "percentile_inc_nulls": 0.9932579398155212,
              "sum_tokens_in_value_count_group": 42,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 21
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9902687072753906,
              "percentile_inc_nulls": 0.9924671053886414,
              "sum_tokens_in_value_count_group": 40,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 20
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9892981052398682,
              "percentile_inc_nulls": 0.9917157888412476,
              "sum_tokens_in_value_count_group": 38,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 19
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9856201410293579,
              "percentile_inc_nulls": 0.9888686537742615,
              "sum_tokens_in_value_count_group": 144,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 18
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9830149412155151,
              "percentile_inc_nulls": 0.9868519902229309,
              "sum_tokens_in_value_count_group": 102,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 17
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9772936105728149,
              "percentile_inc_nulls": 0.9824231863021851,
              "sum_tokens_in_value_count_group": 224,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 16
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9642674922943115,
              "percentile_inc_nulls": 0.9723397493362427,
              "sum_tokens_in_value_count_group": 510,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 15
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9413822889328003,
              "percentile_inc_nulls": 0.9546245336532593,
              "sum_tokens_in_value_count_group": 896,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 14
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.9061861634254456,
              "percentile_inc_nulls": 0.9273794889450073,
              "sum_tokens_in_value_count_group": 1378,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 13
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.8565335273742676,
              "percentile_inc_nulls": 0.8889437913894653,
              "sum_tokens_in_value_count_group": 1944,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 12
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.7893849611282349,
              "percentile_inc_nulls": 0.8369646668434143,
              "sum_tokens_in_value_count_group": 2629,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 11
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.7117388248443604,
              "percentile_inc_nulls": 0.7768595218658447,
              "sum_tokens_in_value_count_group": 3040,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 10
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.6372599005699158,
              "percentile_inc_nulls": 0.7192059755325317,
              "sum_tokens_in_value_count_group": 2916,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 9
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.5657438039779663,
              "percentile_inc_nulls": 0.6638458967208862,
              "sum_tokens_in_value_count_group": 2800,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 8
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.4999489188194275,
              "percentile_inc_nulls": 0.6129146814346313,
              "sum_tokens_in_value_count_group": 2576,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 7
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.4349713921546936,
              "percentile_inc_nulls": 0.5626161694526672,
              "sum_tokens_in_value_count_group": 2544,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 6
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.3731610178947449,
              "percentile_inc_nulls": 0.5147692561149597,
              "sum_tokens_in_value_count_group": 2420,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 5
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.3146199584007263,
              "percentile_inc_nulls": 0.4694530963897705,
              "sum_tokens_in_value_count_group": 2292,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 4
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.24496835470199585,
              "percentile_inc_nulls": 0.41553640365600586,
              "sum_tokens_in_value_count_group": 2727,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 3
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0.1625204086303711,
              "percentile_inc_nulls": 0.35171419382095337,
              "sum_tokens_in_value_count_group": 3228,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 2
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 0,
              "percentile_inc_nulls": 0.22590851783752441,
              "sum_tokens_in_value_count_group": 6363,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 1
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "percentile_ex_nulls": 1,
              "percentile_inc_nulls": 1,
              "sum_tokens_in_value_count_group": 34,
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value_count": 34
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "percentile_ex_nulls",
              "type": "quantitative"
             },
             {
              "field": "percentile_inc_nulls",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "percentile_ex_nulls",
             "sort": "descending",
             "title": "Percentile",
             "type": "quantitative"
            },
            "y": {
             "field": "value_count",
             "title": "Count of values",
             "type": "quantitative"
            }
           },
           "mark": {
            "interpolate": "step-after",
            "type": "line"
           },
           "title": {
            "subtitle": "In this col, 11,426 values (22.6%) are null and there are 12363 distinct values",
            "text": "Distribution of counts of values in column postcode_fake"
           }
          },
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "se1 7sg",
              "value_count": 34
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "sw1a 2jh",
              "value_count": 28
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "se1 8xz",
              "value_count": 28
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "l3 0ah",
              "value_count": 28
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "sw1p 3pl",
              "value_count": 28
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "sw1h 9aa",
              "value_count": 28
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "pl1 3dq",
              "value_count": 27
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "sw1a 2bj",
              "value_count": 27
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "se1 7eh",
              "value_count": 25
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "s8 8ly",
              "value_count": 24
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value",
              "type": "nominal"
             },
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "value",
             "sort": "-y",
             "title": null,
             "type": "nominal"
            },
            "y": {
             "field": "value_count",
             "title": "Value count",
             "type": "quantitative"
            }
           },
           "mark": "bar",
           "title": "Top 10 values by value count"
          },
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "tq1w 8df",
              "value_count": 1
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "cf83 4jg",
              "value_count": 1
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "wr11 7qw",
              "value_count": 1
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "sg18 8qx",
              "value_count": 1
             },
             {
              "distinct_value_count": 12363,
              "group_name": "postcode_fake",
              "total_non_null_rows": 39152,
              "total_rows_inc_nulls": 50578,
              "value": "ip14 5hh",
              "value_count": 1
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value",
              "type": "nominal"
             },
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "value",
             "sort": "-y",
             "title": null,
             "type": "nominal"
            },
            "y": {
             "field": "value_count",
             "scale": {
              "domain": [
               0,
               34
              ]
             },
             "title": "Value count",
             "type": "quantitative"
            }
           },
           "mark": "bar",
           "title": "Bottom 5 values by value count"
          }
         ]
        },
        {
         "hconcat": [
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.9430503845214844,
              "percentile_inc_nulls": 0.9558899402618408,
              "sum_tokens_in_value_count_group": 2231,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 2231
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.8983024954795837,
              "percentile_inc_nulls": 0.9212305545806885,
              "sum_tokens_in_value_count_group": 1753,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1753
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.8536056280136108,
              "percentile_inc_nulls": 0.886610746383667,
              "sum_tokens_in_value_count_group": 1751,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1751
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.8123548030853271,
              "percentile_inc_nulls": 0.854660153388977,
              "sum_tokens_in_value_count_group": 1616,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1616
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.7715379595756531,
              "percentile_inc_nulls": 0.8230456113815308,
              "sum_tokens_in_value_count_group": 1599,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1599
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.7308487892150879,
              "percentile_inc_nulls": 0.7915298938751221,
              "sum_tokens_in_value_count_group": 1594,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1594
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.6960816979408264,
              "percentile_inc_nulls": 0.7646012306213379,
              "sum_tokens_in_value_count_group": 1362,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1362
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.662922739982605,
              "percentile_inc_nulls": 0.7389180660247803,
              "sum_tokens_in_value_count_group": 1299,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1299
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.6321889162063599,
              "percentile_inc_nulls": 0.7151132822036743,
              "sum_tokens_in_value_count_group": 1204,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1204
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.6024760603904724,
              "percentile_inc_nulls": 0.6920993328094482,
              "sum_tokens_in_value_count_group": 1164,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1164
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.5732482671737671,
              "percentile_inc_nulls": 0.6694610118865967,
              "sum_tokens_in_value_count_group": 1145,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1145
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.5462412238121033,
              "percentile_inc_nulls": 0.6485428810119629,
              "sum_tokens_in_value_count_group": 1058,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1058
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.5202552676200867,
              "percentile_inc_nulls": 0.6284155249595642,
              "sum_tokens_in_value_count_group": 1018,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1018
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.500906229019165,
              "percentile_inc_nulls": 0.6134287714958191,
              "sum_tokens_in_value_count_group": 758,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 758
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.48467135429382324,
              "percentile_inc_nulls": 0.6008541584014893,
              "sum_tokens_in_value_count_group": 636,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 636
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.4693554639816284,
              "percentile_inc_nulls": 0.5889912843704224,
              "sum_tokens_in_value_count_group": 600,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 600
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.4556732773780823,
              "percentile_inc_nulls": 0.5783937573432922,
              "sum_tokens_in_value_count_group": 536,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 536
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.4441097378730774,
              "percentile_inc_nulls": 0.5694372653961182,
              "sum_tokens_in_value_count_group": 453,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 453
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.4328015446662903,
              "percentile_inc_nulls": 0.5606785416603088,
              "sum_tokens_in_value_count_group": 443,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 443
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.42269307374954224,
              "percentile_inc_nulls": 0.5528490543365479,
              "sum_tokens_in_value_count_group": 396,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 396
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.4129674434661865,
              "percentile_inc_nulls": 0.5453161597251892,
              "sum_tokens_in_value_count_group": 381,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 381
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.4039565920829773,
              "percentile_inc_nulls": 0.5383368134498596,
              "sum_tokens_in_value_count_group": 353,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 353
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.39514994621276855,
              "percentile_inc_nulls": 0.5315157175064087,
              "sum_tokens_in_value_count_group": 345,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 345
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.3872367739677429,
              "percentile_inc_nulls": 0.5253865718841553,
              "sum_tokens_in_value_count_group": 310,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 310
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.38008934259414673,
              "percentile_inc_nulls": 0.519850492477417,
              "sum_tokens_in_value_count_group": 280,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 280
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.3741927146911621,
              "percentile_inc_nulls": 0.5152833461761475,
              "sum_tokens_in_value_count_group": 231,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 231
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.36857688426971436,
              "percentile_inc_nulls": 0.5109336376190186,
              "sum_tokens_in_value_count_group": 220,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 220
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.36319082975387573,
              "percentile_inc_nulls": 0.5067617893218994,
              "sum_tokens_in_value_count_group": 211,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 211
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.35818761587142944,
              "percentile_inc_nulls": 0.5028866529464722,
              "sum_tokens_in_value_count_group": 196,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 196
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.3532865643501282,
              "percentile_inc_nulls": 0.4990904927253723,
              "sum_tokens_in_value_count_group": 192,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 192
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.34874284267425537,
              "percentile_inc_nulls": 0.49557119607925415,
              "sum_tokens_in_value_count_group": 178,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 178
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.34422463178634644,
              "percentile_inc_nulls": 0.49207162857055664,
              "sum_tokens_in_value_count_group": 177,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 177
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.33975750207901,
              "percentile_inc_nulls": 0.48861163854599,
              "sum_tokens_in_value_count_group": 175,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 175
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.33097636699676514,
              "percentile_inc_nulls": 0.4818102717399597,
              "sum_tokens_in_value_count_group": 344,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 172
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.32661134004592896,
              "percentile_inc_nulls": 0.47842937707901,
              "sum_tokens_in_value_count_group": 171,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 171
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.31823867559432983,
              "percentile_inc_nulls": 0.47194433212280273,
              "sum_tokens_in_value_count_group": 328,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 164
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.314103364944458,
              "percentile_inc_nulls": 0.46874135732650757,
              "sum_tokens_in_value_count_group": 162,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 162
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.31004464626312256,
              "percentile_inc_nulls": 0.4655976891517639,
              "sum_tokens_in_value_count_group": 159,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 159
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.3060370087623596,
              "percentile_inc_nulls": 0.4624935984611511,
              "sum_tokens_in_value_count_group": 157,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 157
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.3020803928375244,
              "percentile_inc_nulls": 0.4594290256500244,
              "sum_tokens_in_value_count_group": 155,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 155
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.29827696084976196,
              "percentile_inc_nulls": 0.45648306608200073,
              "sum_tokens_in_value_count_group": 149,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 149
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2944990396499634,
              "percentile_inc_nulls": 0.4535568952560425,
              "sum_tokens_in_value_count_group": 148,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 148
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.29074662923812866,
              "percentile_inc_nulls": 0.4506504535675049,
              "sum_tokens_in_value_count_group": 147,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 147
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.28704530000686646,
              "percentile_inc_nulls": 0.4477836489677429,
              "sum_tokens_in_value_count_group": 145,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 145
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2834460735321045,
              "percentile_inc_nulls": 0.44499582052230835,
              "sum_tokens_in_value_count_group": 141,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 141
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2799489498138428,
              "percentile_inc_nulls": 0.4422871470451355,
              "sum_tokens_in_value_count_group": 137,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 137
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2764773368835449,
              "percentile_inc_nulls": 0.4395982623100281,
              "sum_tokens_in_value_count_group": 136,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 136
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2731078267097473,
              "percentile_inc_nulls": 0.4369884133338928,
              "sum_tokens_in_value_count_group": 132,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 132
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2664709687232971,
              "percentile_inc_nulls": 0.43184781074523926,
              "sum_tokens_in_value_count_group": 260,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 130
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2631780505180359,
              "percentile_inc_nulls": 0.4292973279953003,
              "sum_tokens_in_value_count_group": 129,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 129
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.25993621349334717,
              "percentile_inc_nulls": 0.4267863631248474,
              "sum_tokens_in_value_count_group": 127,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 127
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.256924033164978,
              "percentile_inc_nulls": 0.4244533181190491,
              "sum_tokens_in_value_count_group": 118,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 118
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.25398850440979004,
              "percentile_inc_nulls": 0.42217957973480225,
              "sum_tokens_in_value_count_group": 115,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 115
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2512826919555664,
              "percentile_inc_nulls": 0.42008382081985474,
              "sum_tokens_in_value_count_group": 106,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 106
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.24862796068191528,
              "percentile_inc_nulls": 0.4180275797843933,
              "sum_tokens_in_value_count_group": 104,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 104
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.24599874019622803,
              "percentile_inc_nulls": 0.4159911274909973,
              "sum_tokens_in_value_count_group": 103,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 103
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.24339503049850464,
              "percentile_inc_nulls": 0.41397446393966675,
              "sum_tokens_in_value_count_group": 102,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 102
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.23308229446411133,
              "percentile_inc_nulls": 0.4059867858886719,
              "sum_tokens_in_value_count_group": 404,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 101
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.23058074712753296,
              "percentile_inc_nulls": 0.40404921770095825,
              "sum_tokens_in_value_count_group": 98,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 98
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.22810465097427368,
              "percentile_inc_nulls": 0.4021313786506653,
              "sum_tokens_in_value_count_group": 97,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 97
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.22567963600158691,
              "percentile_inc_nulls": 0.4002530574798584,
              "sum_tokens_in_value_count_group": 95,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 95
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2208806872367859,
              "percentile_inc_nulls": 0.39653605222702026,
              "sum_tokens_in_value_count_group": 188,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 94
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.21850669384002686,
              "percentile_inc_nulls": 0.39469730854034424,
              "sum_tokens_in_value_count_group": 93,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 93
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.21618378162384033,
              "percentile_inc_nulls": 0.3928980827331543,
              "sum_tokens_in_value_count_group": 91,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 91
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.21393746137619019,
              "percentile_inc_nulls": 0.39115822315216064,
              "sum_tokens_in_value_count_group": 88,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 88
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2117166519165039,
              "percentile_inc_nulls": 0.38943809270858765,
              "sum_tokens_in_value_count_group": 87,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 87
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2095213532447815,
              "percentile_inc_nulls": 0.3877377510070801,
              "sum_tokens_in_value_count_group": 86,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 86
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2052329182624817,
              "percentile_inc_nulls": 0.3844161629676819,
              "sum_tokens_in_value_count_group": 168,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 84
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.2032163143157959,
              "percentile_inc_nulls": 0.3828542232513428,
              "sum_tokens_in_value_count_group": 79,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 79
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.20122528076171875,
              "percentile_inc_nulls": 0.3813120126724243,
              "sum_tokens_in_value_count_group": 78,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 78
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.19933629035949707,
              "percentile_inc_nulls": 0.3798489570617676,
              "sum_tokens_in_value_count_group": 74,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 74
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.1974983811378479,
              "percentile_inc_nulls": 0.3784254193305969,
              "sum_tokens_in_value_count_group": 72,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 72
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.19568604230880737,
              "percentile_inc_nulls": 0.3770216107368469,
              "sum_tokens_in_value_count_group": 71,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 71
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.19211232662200928,
              "percentile_inc_nulls": 0.37425363063812256,
              "sum_tokens_in_value_count_group": 140,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 70
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.18858963251113892,
              "percentile_inc_nulls": 0.3715251684188843,
              "sum_tokens_in_value_count_group": 138,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 69
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.18685388565063477,
              "percentile_inc_nulls": 0.37018072605133057,
              "sum_tokens_in_value_count_group": 68,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 68
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.18343329429626465,
              "percentile_inc_nulls": 0.36753135919570923,
              "sum_tokens_in_value_count_group": 134,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 67
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.18174856901168823,
              "percentile_inc_nulls": 0.3662264347076416,
              "sum_tokens_in_value_count_group": 66,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 66
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.17541801929473877,
              "percentile_inc_nulls": 0.36132311820983887,
              "sum_tokens_in_value_count_group": 248,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 62
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.17386090755462646,
              "percentile_inc_nulls": 0.36011701822280884,
              "sum_tokens_in_value_count_group": 61,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 61
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.17232930660247803,
              "percentile_inc_nulls": 0.358930766582489,
              "sum_tokens_in_value_count_group": 60,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 60
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.16788768768310547,
              "percentile_inc_nulls": 0.355490505695343,
              "sum_tokens_in_value_count_group": 174,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 58
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.16497766971588135,
              "percentile_inc_nulls": 0.3532365560531616,
              "sum_tokens_in_value_count_group": 114,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 57
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.16216975450515747,
              "percentile_inc_nulls": 0.35106170177459717,
              "sum_tokens_in_value_count_group": 110,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 55
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.1607913374900818,
              "percentile_inc_nulls": 0.34999406337738037,
              "sum_tokens_in_value_count_group": 54,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 54
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.15808552503585815,
              "percentile_inc_nulls": 0.34789830446243286,
              "sum_tokens_in_value_count_group": 106,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 53
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.1567581295967102,
              "percentile_inc_nulls": 0.34687018394470215,
              "sum_tokens_in_value_count_group": 52,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 52
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.1554563045501709,
              "percentile_inc_nulls": 0.34586185216903687,
              "sum_tokens_in_value_count_group": 51,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 51
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.1516273021697998,
              "percentile_inc_nulls": 0.34289610385894775,
              "sum_tokens_in_value_count_group": 150,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 50
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.14912569522857666,
              "percentile_inc_nulls": 0.34095853567123413,
              "sum_tokens_in_value_count_group": 98,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 49
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.14667516946792603,
              "percentile_inc_nulls": 0.3390604853630066,
              "sum_tokens_in_value_count_group": 96,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 48
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.14427566528320312,
              "percentile_inc_nulls": 0.33720195293426514,
              "sum_tokens_in_value_count_group": 94,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 47
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.14310145378112793,
              "percentile_inc_nulls": 0.33629244565963745,
              "sum_tokens_in_value_count_group": 46,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 46
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.1419527530670166,
              "percentile_inc_nulls": 0.3354027271270752,
              "sum_tokens_in_value_count_group": 45,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 45
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.13858330249786377,
              "percentile_inc_nulls": 0.3327929377555847,
              "sum_tokens_in_value_count_group": 132,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 44
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.13638800382614136,
              "percentile_inc_nulls": 0.33109259605407715,
              "sum_tokens_in_value_count_group": 86,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 43
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.1331716775894165,
              "percentile_inc_nulls": 0.3286013603210449,
              "sum_tokens_in_value_count_group": 126,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 42
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.13107848167419434,
              "percentile_inc_nulls": 0.3269801139831543,
              "sum_tokens_in_value_count_group": 82,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 41
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.12903636693954468,
              "percentile_inc_nulls": 0.32539838552474976,
              "sum_tokens_in_value_count_group": 80,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 40
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.12704533338546753,
              "percentile_inc_nulls": 0.3238562345504761,
              "sum_tokens_in_value_count_group": 78,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 39
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.12510532140731812,
              "percentile_inc_nulls": 0.3223536014556885,
              "sum_tokens_in_value_count_group": 76,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 38
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.12038290500640869,
              "percentile_inc_nulls": 0.31869590282440186,
              "sum_tokens_in_value_count_group": 185,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 37
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.11946392059326172,
              "percentile_inc_nulls": 0.31798410415649414,
              "sum_tokens_in_value_count_group": 36,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 36
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.11499679088592529,
              "percentile_inc_nulls": 0.3145241141319275,
              "sum_tokens_in_value_count_group": 175,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 35
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.11152517795562744,
              "percentile_inc_nulls": 0.3118351697921753,
              "sum_tokens_in_value_count_group": 136,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 34
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.10899806022644043,
              "percentile_inc_nulls": 0.30987781286239624,
              "sum_tokens_in_value_count_group": 99,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 33
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.10409700870513916,
              "percentile_inc_nulls": 0.30608171224594116,
              "sum_tokens_in_value_count_group": 192,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 32
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.10093170404434204,
              "percentile_inc_nulls": 0.3036300539970398,
              "sum_tokens_in_value_count_group": 124,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 31
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.09786856174468994,
              "percentile_inc_nulls": 0.30125749111175537,
              "sum_tokens_in_value_count_group": 120,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 30
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.09268665313720703,
              "percentile_inc_nulls": 0.2972438335418701,
              "sum_tokens_in_value_count_group": 203,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 29
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.08982771635055542,
              "percentile_inc_nulls": 0.2950294613838196,
              "sum_tokens_in_value_count_group": 112,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 28
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.08638161420822144,
              "percentile_inc_nulls": 0.2923603057861328,
              "sum_tokens_in_value_count_group": 135,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 27
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.08306318521499634,
              "percentile_inc_nulls": 0.2897900342941284,
              "sum_tokens_in_value_count_group": 130,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 26
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.08051055669784546,
              "percentile_inc_nulls": 0.28781288862228394,
              "sum_tokens_in_value_count_group": 100,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 25
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.07622206211090088,
              "percentile_inc_nulls": 0.28449130058288574,
              "sum_tokens_in_value_count_group": 168,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 24
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.07328653335571289,
              "percentile_inc_nulls": 0.2822175621986389,
              "sum_tokens_in_value_count_group": 115,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 23
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.06935542821884155,
              "percentile_inc_nulls": 0.27917277812957764,
              "sum_tokens_in_value_count_group": 154,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 22
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.06506699323654175,
              "percentile_inc_nulls": 0.27585119009017944,
              "sum_tokens_in_value_count_group": 168,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 21
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.06047224998474121,
              "percentile_inc_nulls": 0.2722923159599304,
              "sum_tokens_in_value_count_group": 180,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 20
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.05465221405029297,
              "percentile_inc_nulls": 0.2677844166755676,
              "sum_tokens_in_value_count_group": 228,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 19
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.05097639560699463,
              "percentile_inc_nulls": 0.2649373412132263,
              "sum_tokens_in_value_count_group": 144,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 18
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.04533505439758301,
              "percentile_inc_nulls": 0.260567843914032,
              "sum_tokens_in_value_count_group": 221,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 17
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.04329293966293335,
              "percentile_inc_nulls": 0.25898611545562744,
              "sum_tokens_in_value_count_group": 80,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 16
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.03793233633041382,
              "percentile_inc_nulls": 0.2548341155052185,
              "sum_tokens_in_value_count_group": 210,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 15
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.033643901348114014,
              "percentile_inc_nulls": 0.2515125274658203,
              "sum_tokens_in_value_count_group": 168,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 14
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.029329955577850342,
              "percentile_inc_nulls": 0.2481711506843567,
              "sum_tokens_in_value_count_group": 169,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 13
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.02473515272140503,
              "percentile_inc_nulls": 0.24461227655410767,
              "sum_tokens_in_value_count_group": 180,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 12
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.020804107189178467,
              "percentile_inc_nulls": 0.2415674924850464,
              "sum_tokens_in_value_count_group": 154,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 11
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.01850670576095581,
              "percentile_inc_nulls": 0.23978805541992188,
              "sum_tokens_in_value_count_group": 90,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 10
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.015749812126159668,
              "percentile_inc_nulls": 0.2376527190208435,
              "sum_tokens_in_value_count_group": 108,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 9
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.01370769739151001,
              "percentile_inc_nulls": 0.23607099056243896,
              "sum_tokens_in_value_count_group": 80,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 8
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.010848760604858398,
              "percentile_inc_nulls": 0.23385661840438843,
              "sum_tokens_in_value_count_group": 112,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 7
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.007938742637634277,
              "percentile_inc_nulls": 0.23160266876220703,
              "sum_tokens_in_value_count_group": 114,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 6
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.0060242414474487305,
              "percentile_inc_nulls": 0.23011982440948486,
              "sum_tokens_in_value_count_group": 75,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 5
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.004084229469299316,
              "percentile_inc_nulls": 0.22861719131469727,
              "sum_tokens_in_value_count_group": 76,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 4
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.002552628517150879,
              "percentile_inc_nulls": 0.22743088006973267,
              "sum_tokens_in_value_count_group": 60,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 3
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0.0017358064651489258,
              "percentile_inc_nulls": 0.22679823637008667,
              "sum_tokens_in_value_count_group": 32,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 2
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 0,
              "percentile_inc_nulls": 0.22545373439788818,
              "sum_tokens_in_value_count_group": 68,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 1
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "percentile_ex_nulls": 1,
              "percentile_inc_nulls": 1,
              "sum_tokens_in_value_count_group": 2231,
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value_count": 2231
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "percentile_ex_nulls",
              "type": "quantitative"
             },
             {
              "field": "percentile_inc_nulls",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "percentile_ex_nulls",
             "sort": "descending",
             "title": "Percentile",
             "type": "quantitative"
            },
            "y": {
             "field": "value_count",
             "title": "Count of values",
             "type": "quantitative"
            }
           },
           "mark": {
            "interpolate": "step-after",
            "type": "line"
           },
           "title": {
            "subtitle": "In this col, 11,403 values (22.5%) are null and there are 537 distinct values",
            "text": "Distribution of counts of values in column substr(dob, 1,4)"
           }
          },
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1862",
              "value_count": 2231
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1860",
              "value_count": 1753
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1861",
              "value_count": 1751
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1859",
              "value_count": 1616
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1857",
              "value_count": 1599
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1858",
              "value_count": 1594
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1856",
              "value_count": 1362
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1855",
              "value_count": 1299
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1851",
              "value_count": 1204
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1850",
              "value_count": 1164
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value",
              "type": "nominal"
             },
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "value",
             "sort": "-y",
             "title": null,
             "type": "nominal"
            },
            "y": {
             "field": "value_count",
             "title": "Value count",
             "type": "quantitative"
            }
           },
           "mark": "bar",
           "title": "Top 10 values by value count"
          },
          {
           "data": {
            "values": [
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1092",
              "value_count": 1
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1085",
              "value_count": 1
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1338",
              "value_count": 1
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1262",
              "value_count": 1
             },
             {
              "distinct_value_count": 537,
              "group_name": "substr_dob_1_4_",
              "total_non_null_rows": 39175,
              "total_rows_inc_nulls": 50578,
              "value": "1295",
              "value_count": 1
             }
            ]
           },
           "encoding": {
            "tooltip": [
             {
              "field": "value",
              "type": "nominal"
             },
             {
              "field": "value_count",
              "type": "quantitative"
             },
             {
              "field": "total_non_null_rows",
              "type": "quantitative"
             },
             {
              "field": "total_rows_inc_nulls",
              "type": "quantitative"
             }
            ],
            "x": {
             "field": "value",
             "sort": "-y",
             "title": null,
             "type": "nominal"
            },
            "y": {
             "field": "value_count",
             "scale": {
              "domain": [
               0,
               2231
              ]
             },
             "title": "Value count",
             "type": "quantitative"
            }
           },
           "mark": "bar",
           "title": "Bottom 5 values by value count"
          }
         ]
        }
       ]
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 4 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initialise the linker, passing in the input dataset(s)\n",
    "settings = {\"link_type\": \"dedupe_only\"}\n",
    "linker = DuckDBLinker(df, settings, connection=\":temporary:\")\n",
    "\n",
    "import altair as alt\n",
    "alt.renderers.enable('mimetype')\n",
    "linker.profile_columns([\"first_name\", \"postcode_fake\", \"substr(dob, 1,4)\"], top_n=10, bottom_n=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16372982"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linker.count_num_comparisons_from_blocking_rule(\"l.first_name = r.first_name\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "243656"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linker.count_num_comparisons_from_blocking_rule(\"l.first_name = r.first_name and l.surname = r.surname\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1549081"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linker.count_num_comparisons_from_blocking_rule(\"l.dob = r.dob\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import splink.duckdb.duckdb_comparison_library as cl\n",
    "\n",
    "settings = {\n",
    "    \"probability_two_random_records_match\": 9/50_000,\n",
    "    \"link_type\": \"dedupe_only\",\n",
    "    \"blocking_rules_to_generate_predictions\": [\n",
    "        \"l.first_name = r.first_name and l.surname = r.surname\",\n",
    "        \"l.surname = r.surname and l.dob = r.dob\",\n",
    "        \"l.first_name = r.first_name and l.dob = r.dob\",\n",
    "        \"l.postcode_fake = r.postcode_fake and l.first_name = r.first_name\",\n",
    "    ],\n",
    "    \"comparisons\": [\n",
    "        cl.jaccard_at_thresholds(\"first_name\", [0.9, 0.5], term_frequency_adjustments=False),\n",
    "        cl.jaccard_at_thresholds(\"surname\", [0.9, 0.5], term_frequency_adjustments=False),\n",
    "        cl.levenshtein_at_thresholds(\"dob\", [1,2], term_frequency_adjustments=False),\n",
    "        cl.levenshtein_at_thresholds(\"postcode_fake\", 2),\n",
    "        cl.exact_match(\"birth_place\", term_frequency_adjustments=False),\n",
    "        cl.exact_match(\"occupation\",  term_frequency_adjustments=False),\n",
    "    ],\n",
    "    \"retain_matching_columns\": True,\n",
    "    \"retain_intermediate_calculation_columns\": True,\n",
    "    \"max_iterations\": 10,\n",
    "    \"em_convergence\": 0.01\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "linker.initialise_settings(settings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "----- Estimating u probabilities using random sampling -----\n",
      "\n",
      "Estimated u probabilities using random sampling\n",
      "\n",
      "Your model is not yet fully trained. Missing estimates for:\n",
      "    - first_name (no m values are trained).\n",
      "    - surname (no m values are trained).\n",
      "    - dob (no m values are trained).\n",
      "    - postcode_fake (no m values are trained).\n",
      "    - birth_place (no m values are trained).\n",
      "    - occupation (no m values are trained).\n"
     ]
    }
   ],
   "source": [
    "linker.estimate_u_using_random_sampling(target_rows=5e6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "----- Starting EM training session -----\n",
      "\n",
      "Estimating the m probabilities of the model by blocking on:\n",
      "l.first_name = r.first_name and l.surname = r.surname\n",
      "\n",
      "Parameter estimates will be made for the following comparison(s):\n",
      "    - dob\n",
      "    - postcode_fake\n",
      "    - birth_place\n",
      "    - occupation\n",
      "\n",
      "Parameter estimates cannot be made for the following comparison(s) since they are used in the blocking rules: \n",
      "    - first_name\n",
      "    - surname\n",
      "\n",
      "Iteration 1: Largest change in params was -0.529 in probability_two_random_records_match\n",
      "Iteration 2: Largest change in params was -0.0373 in probability_two_random_records_match\n",
      "Iteration 3: Largest change in params was 0.0162 in the m_probability of birth_place, level `Exact match`\n",
      "Iteration 4: Largest change in params was -0.00748 in the m_probability of dob, level `All other comparisons`\n",
      "\n",
      "EM converged after 4 iterations\n",
      "\n",
      "Your model is not yet fully trained. Missing estimates for:\n",
      "    - first_name (no m values are trained).\n",
      "    - surname (no m values are trained).\n"
     ]
    },
    {
     "data": {
      "application/vnd.vegalite.v4+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v5.2.json",
       "config": {
        "header": {
         "title": null
        },
        "mark": {
         "tooltip": null
        },
        "title": {
         "anchor": "middle"
        },
        "view": {
         "height": 60,
         "width": 400
        }
       },
       "data": {
        "values": [
         {
          "bayes_factor": 17.431670092481585,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.946 or one in  1.1 records.This is equivalent to a starting match weight of 4.124.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "",
          "log2_bayes_factor": 4.123638892619225,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 485.0929388661202,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 485.09 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.922117369231236,
          "m_probability": 0.95,
          "m_probability_description": "Amongst matching record comparisons, 95.00% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.001958387607580057,
          "u_probability_description": "Amongst non-matching record comparisons, 0.20% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.8345096561732536,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is  1.20 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": -0.2609993516713652,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01997180804724744,
          "u_probability_description": "Amongst non-matching record comparisons, 2.00% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 0.2189065270450917,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is  4.57 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": -2.1916131234877376,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.07613599690992119,
          "u_probability_description": "Amongst non-matching record comparisons, 7.61% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.01847881355513238,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  54.12 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -5.757984059269421,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9019338074352513,
          "u_probability_description": "Amongst non-matching record comparisons, 90.19% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 6743.111229314421,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 6,743.11 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.719198680141636,
          "m_probability": 0.95,
          "m_probability_description": "Amongst matching record comparisons, 95.00% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 43.84431950934584,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 43.84 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 5.454318033416854,
          "m_probability": 0.025000000000000022,
          "m_probability_description": "Amongst matching record comparisons, 2.50% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.025017789745374188,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  39.97 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -5.320901852878554,
          "m_probability": 0.025000000000000022,
          "m_probability_description": "Amongst matching record comparisons, 2.50% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 173.55730542773736,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 173.56 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.439268282584897,
          "m_probability": 0.95,
          "m_probability_description": "Amongst matching record comparisons, 95.00% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.05027519113332021,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  19.89 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.314009528478925,
          "m_probability": 0.050000000000000044,
          "m_probability_description": "Amongst matching record comparisons, 5.00% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 24.00662489341873,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 24.01 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.5853606833038105,
          "m_probability": 0.95,
          "m_probability_description": "Amongst matching record comparisons, 95.00% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.05206014541241721,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  19.21 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.26367684834618,
          "m_probability": 0.050000000000000044,
          "m_probability_description": "Amongst matching record comparisons, 5.00% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 0.7132680062605772,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.416 or one in  2.4 records.This is equivalent to a starting match weight of -0.487.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "",
          "log2_bayes_factor": -0.48748383216913177,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 302.04923795000485,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 302.05 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.238639936530836,
          "m_probability": 0.5915294844802893,
          "m_probability_description": "Amongst matching record comparisons, 59.15% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.001958387607580057,
          "u_probability_description": "Amongst non-matching record comparisons, 0.20% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 15.907246013825315,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 15.91 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": 3.991612181586529,
          "m_probability": 0.3176964639484612,
          "m_probability_description": "Amongst matching record comparisons, 31.77% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01997180804724744,
          "u_probability_description": "Amongst non-matching record comparisons, 2.00% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 0.5400401710794696,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is  1.85 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": -0.8888613682379283,
          "m_probability": 0.0411164967965398,
          "m_probability_description": "Amongst matching record comparisons, 4.11% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.07613599690992119,
          "u_probability_description": "Amongst non-matching record comparisons, 7.61% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.05505676177713806,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  18.16 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.1829364310117745,
          "m_probability": 0.04965755477470974,
          "m_probability_description": "Amongst matching record comparisons, 4.97% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9019338074352513,
          "u_probability_description": "Amongst non-matching record comparisons, 90.19% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4768.589767865864,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,768.59 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.219346960595827,
          "m_probability": 0.6718204884087544,
          "m_probability_description": "Amongst matching record comparisons, 67.18% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 242.4634315111324,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 242.46 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.921623365298087,
          "m_probability": 0.1382524773017912,
          "m_probability_description": "Amongst matching record comparisons, 13.83% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.19006218443264156,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.26 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.395456578989874,
          "m_probability": 0.18992703428945437,
          "m_probability_description": "Amongst matching record comparisons, 18.99% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 141.17060629859952,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 141.17 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.141295920298018,
          "m_probability": 0.7727250411796044,
          "m_probability_description": "Amongst matching record comparisons, 77.27% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.22852583989025707,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  4.38 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.129570791862923,
          "m_probability": 0.22727495882039547,
          "m_probability_description": "Amongst matching record comparisons, 22.73% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.66439779518055,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.66 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.502355923464063,
          "m_probability": 0.8968848391230607,
          "m_probability_description": "Amongst matching record comparisons, 89.69% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.10736380538956489,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  9.31 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.2194203826241186,
          "m_probability": 0.10311516087693921,
          "m_probability_description": "Amongst matching record comparisons, 10.31% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 0.6104459657599911,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.379 or one in  2.6 records.This is equivalent to a starting match weight of -0.712.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "",
          "log2_bayes_factor": -0.7120644956290275,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 304.47755405420094,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 304.48 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.25019206700886,
          "m_probability": 0.596285068646034,
          "m_probability_description": "Amongst matching record comparisons, 59.63% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.001958387607580057,
          "u_probability_description": "Amongst non-matching record comparisons, 0.20% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 16.635239952495354,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 16.64 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": 4.056170771194008,
          "m_probability": 0.3322358191511388,
          "m_probability_description": "Amongst matching record comparisons, 33.22% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01997180804724744,
          "u_probability_description": "Amongst non-matching record comparisons, 2.00% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 0.5086240212627599,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is  1.97 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": -0.9753284957411832,
          "m_probability": 0.03872459691117317,
          "m_probability_description": "Amongst matching record comparisons, 3.87% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.07613599690992119,
          "u_probability_description": "Amongst non-matching record comparisons, 7.61% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.03631587486979236,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  27.54 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.783255853965445,
          "m_probability": 0.032754515291653984,
          "m_probability_description": "Amongst matching record comparisons, 3.28% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9019338074352513,
          "u_probability_description": "Amongst non-matching record comparisons, 90.19% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4738.028806140032,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,738.03 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.210071254525381,
          "m_probability": 0.6675149219347322,
          "m_probability_description": "Amongst matching record comparisons, 66.75% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 241.64496927624043,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 241.64 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.91674514987962,
          "m_probability": 0.13778579071385275,
          "m_probability_description": "Amongst matching record comparisons, 13.78% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.1948378333812756,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.13 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.3596542494390476,
          "m_probability": 0.1946992873514151,
          "m_probability_description": "Amongst matching record comparisons, 19.47% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 147.6027209407126,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 147.60 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.205575506372836,
          "m_probability": 0.8079324840178524,
          "m_probability_description": "Amongst matching record comparisons, 80.79% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.19312462153008975,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.18 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.3723959891438655,
          "m_probability": 0.1920675159821474,
          "m_probability_description": "Amongst matching record comparisons, 19.21% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.91356095071162,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.91 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.5181297773063775,
          "m_probability": 0.906744825639508,
          "m_probability_description": "Amongst matching record comparisons, 90.67% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.09709755875335059,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  10.30 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.364421166207594,
          "m_probability": 0.093255174360492,
          "m_probability_description": "Amongst matching record comparisons, 9.33% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 0.5881025404062642,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.370 or one in  2.7 records.This is equivalent to a starting match weight of -0.766.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "",
          "log2_bayes_factor": -0.7658603724140379,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 309.4158033797555,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 309.42 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.27340307394161,
          "m_probability": 0.6059560749283407,
          "m_probability_description": "Amongst matching record comparisons, 60.60% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.001958387607580057,
          "u_probability_description": "Amongst non-matching record comparisons, 0.20% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 16.834530948170727,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 16.83 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": 4.073351619547119,
          "m_probability": 0.3362160206623122,
          "m_probability_description": "Amongst matching record comparisons, 33.62% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01997180804724744,
          "u_probability_description": "Amongst non-matching record comparisons, 2.00% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 0.4987744313951526,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is  2.00 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": -1.0035405845010226,
          "m_probability": 0.03797468856744904,
          "m_probability_description": "Amongst matching record comparisons, 3.80% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.07613599690992119,
          "u_probability_description": "Amongst non-matching record comparisons, 7.61% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.02201183244073406,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  45.43 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -5.505576938088312,
          "m_probability": 0.01985321584189805,
          "m_probability_description": "Amongst matching record comparisons, 1.99% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9019338074352513,
          "u_probability_description": "Amongst non-matching record comparisons, 90.19% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4796.178300638656,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,796.18 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.227669577414582,
          "m_probability": 0.6757072856515781,
          "m_probability_description": "Amongst matching record comparisons, 67.57% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 244.25263209501875,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 244.25 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.932230298703168,
          "m_probability": 0.13927267821031764,
          "m_probability_description": "Amongst matching record comparisons, 13.93% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.18515169451138488,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.40 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.4332203417484117,
          "m_probability": 0.18502003613810425,
          "m_probability_description": "Amongst matching record comparisons, 18.50% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 150.5580366850963,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 150.56 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.2341759099680045,
          "m_probability": 0.8241089852042774,
          "m_probability_description": "Amongst matching record comparisons, 82.41% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.17685908774977185,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.65 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.49932774231911,
          "m_probability": 0.17589101479572247,
          "m_probability_description": "Amongst matching record comparisons, 17.59% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.914030010839344,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.91 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.518159310201559,
          "m_probability": 0.9067633874791383,
          "m_probability_description": "Amongst matching record comparisons, 90.68% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.09707823211194505,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  10.30 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.364708353911897,
          "m_probability": 0.09323661252086166,
          "m_probability_description": "Amongst matching record comparisons, 9.32% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 0.5781206493677187,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.366 or one in  2.7 records.This is equivalent to a starting match weight of -0.791.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "",
          "log2_bayes_factor": -0.7905574912951311,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 312.16784723322775,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 312.17 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.286178139563434,
          "m_probability": 0.6113456435064976,
          "m_probability_description": "Amongst matching record comparisons, 61.13% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.001958387607580057,
          "u_probability_description": "Amongst non-matching record comparisons, 0.20% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 16.949819994845715,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 16.95 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": 4.083198047073621,
          "m_probability": 0.3385185513724552,
          "m_probability_description": "Amongst matching record comparisons, 33.85% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01997180804724744,
          "u_probability_description": "Amongst non-matching record comparisons, 2.00% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 0.4960511903664868,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is  2.02 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": -1.0114390866197067,
          "m_probability": 0.03776735189690557,
          "m_probability_description": "Amongst matching record comparisons, 3.78% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.07613599690992119,
          "u_probability_description": "Amongst non-matching record comparisons, 7.61% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.013713260465657417,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  72.92 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -6.188284562611205,
          "m_probability": 0.012368453224141701,
          "m_probability_description": "Amongst matching record comparisons, 1.24% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9019338074352513,
          "u_probability_description": "Amongst non-matching record comparisons, 90.19% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4829.005718588069,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,829.01 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.237510456636885,
          "m_probability": 0.6803321607265089,
          "m_probability_description": "Amongst matching record comparisons, 68.03% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 245.7183271533245,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 245.72 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.94086165637649,
          "m_probability": 0.14010841649677533,
          "m_probability_description": "Amongst matching record comparisons, 14.01% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.1796871954331449,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.57 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.4764404892853946,
          "m_probability": 0.17955942277671572,
          "m_probability_description": "Amongst matching record comparisons, 17.96% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 151.89709947284584,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 151.90 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.246950511121673,
          "m_probability": 0.8314386083810542,
          "m_probability_description": "Amongst matching record comparisons, 83.14% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.16948912362681862,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.90 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.5607353984762313,
          "m_probability": 0.16856139161894576,
          "m_probability_description": "Amongst matching record comparisons, 16.86% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.899086252274472,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.90 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.517218126152232,
          "m_probability": 0.9061720269401348,
          "m_probability_description": "Amongst matching record comparisons, 90.62% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.09769395842497895,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  10.24 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 4,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.3555868437512575,
          "m_probability": 0.0938279730598653,
          "m_probability_description": "Amongst matching record comparisons, 9.38% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.36633488675238196,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         }
        ]
       },
       "params": [
        {
         "bind": {
          "input": "range",
          "max": 4,
          "min": 0,
          "step": 1
         },
         "description": "Filter by the iteration number",
         "name": "iteration_number",
         "value": 4
        }
       ],
       "resolve": {
        "axis": {
         "y": "independent"
        },
        "scale": {
         "y": "independent"
        }
       },
       "selection": {
        "zoom_selector": {
         "bind": "scales",
         "encodings": [
          "x"
         ],
         "type": "interval"
        }
       },
       "title": {
        "subtitle": "Training session blocked on <splink.blocking.BlockingRule object at 0x7fe966bfc700>",
        "text": "Model parameters (components of final match weight)"
       },
       "transform": [
        {
         "filter": "(datum.iteration == iteration_number)"
        }
       ],
       "vconcat": [
        {
         "encoding": {
          "color": {
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             0,
             10
            ],
            "range": [
             "red",
             "orange",
             "green"
            ]
           },
           "title": "Match weight",
           "type": "quantitative"
          },
          "tooltip": [
           {
            "field": "comparison_name",
            "title": "Comparison name",
            "type": "nominal"
           },
           {
            "field": "probability_two_random_records_match",
            "format": ".4f",
            "title": "Probability two random records match",
            "type": "nominal"
           },
           {
            "field": "log2_bayes_factor",
            "format": ",.4f",
            "title": "Equivalent match weight",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor_description",
            "title": "Match weight description",
            "type": "nominal"
           }
          ],
          "x": {
           "axis": {
            "domain": false,
            "labels": false,
            "ticks": false,
            "title": ""
           },
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             10
            ]
           },
           "type": "quantitative"
          },
          "y": {
           "axis": {
            "title": "Prior (starting) match weight",
            "titleAlign": "right",
            "titleAngle": 0,
            "titleFontWeight": "normal"
           },
           "field": "label_for_charts",
           "sort": {
            "field": "comparison_vector_value",
            "order": "descending"
           },
           "type": "nominal"
          }
         },
         "height": 30,
         "mark": {
          "clip": true,
          "height": 20,
          "type": "bar"
         },
         "selection": {
          "zoom_selector": {
           "bind": "scales",
           "encodings": [
            "x"
           ],
           "type": "interval"
          }
         },
         "transform": [
          {
           "filter": "(datum.comparison_name == 'probability_two_random_records_match')"
          }
         ]
        },
        {
         "encoding": {
          "color": {
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             0,
             10
            ],
            "range": [
             "red",
             "orange",
             "green"
            ]
           },
           "title": "Match weight",
           "type": "quantitative"
          },
          "row": {
           "field": "comparison_name",
           "header": {
            "labelAlign": "left",
            "labelAnchor": "middle",
            "labelAngle": 0
           },
           "sort": {
            "field": "comparison_sort_order"
           },
           "type": "nominal"
          },
          "tooltip": [
           {
            "field": "comparison_name",
            "title": "Comparison name",
            "type": "nominal"
           },
           {
            "field": "label_for_charts",
            "title": "Label",
            "type": "ordinal"
           },
           {
            "field": "sql_condition",
            "title": "SQL condition",
            "type": "nominal"
           },
           {
            "field": "m_probability",
            "format": ".4f",
            "title": "M probability",
            "type": "quantitative"
           },
           {
            "field": "u_probability",
            "format": ".4f",
            "title": "U probability",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor",
            "format": ",.4f",
            "title": "Bayes factor = m/u",
            "type": "quantitative"
           },
           {
            "field": "log2_bayes_factor",
            "format": ",.4f",
            "title": "Match weight = log2(m/u)",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor_description",
            "title": "Match weight description",
            "type": "nominal"
           }
          ],
          "x": {
           "axis": {
            "title": "Comparison level match weight = log2(m/u)"
           },
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             10
            ]
           },
           "type": "quantitative"
          },
          "y": {
           "axis": {
            "title": null
           },
           "field": "label_for_charts",
           "sort": {
            "field": "comparison_vector_value",
            "order": "descending"
           },
           "type": "nominal"
          }
         },
         "mark": {
          "clip": true,
          "type": "bar"
         },
         "resolve": {
          "axis": {
           "y": "independent"
          },
          "scale": {
           "y": "independent"
          }
         },
         "selection": {
          "zoom_selector": {
           "bind": "scales",
           "encodings": [
            "x"
           ],
           "type": "interval"
          }
         },
         "transform": [
          {
           "filter": "(datum.comparison_name != 'probability_two_random_records_match')"
          }
         ]
        }
       ]
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 4 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blocking_rule = \"l.first_name = r.first_name and l.surname = r.surname\"\n",
    "training_session_names = linker.estimate_parameters_using_expectation_maximisation(blocking_rule)\n",
    "training_session_names.match_weights_interactive_history_chart()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "----- Starting EM training session -----\n",
      "\n",
      "Estimating the m probabilities of the model by blocking on:\n",
      "l.dob = r.dob\n",
      "\n",
      "Parameter estimates will be made for the following comparison(s):\n",
      "    - first_name\n",
      "    - surname\n",
      "    - postcode_fake\n",
      "    - birth_place\n",
      "    - occupation\n",
      "\n",
      "Parameter estimates cannot be made for the following comparison(s) since they are used in the blocking rules: \n",
      "    - dob\n",
      "\n",
      "Iteration 1: Largest change in params was -0.352 in the m_probability of first_name, level `Exact match`\n",
      "Iteration 2: Largest change in params was -0.0365 in the m_probability of first_name, level `Exact match`\n",
      "Iteration 3: Largest change in params was -0.00587 in the m_probability of surname, level `Exact match`\n",
      "\n",
      "EM converged after 3 iterations\n",
      "\n",
      "Your model is fully trained. All comparisons have at least one estimate for their m and u values\n"
     ]
    },
    {
     "data": {
      "application/vnd.vegalite.v4+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v5.2.json",
       "config": {
        "header": {
         "title": null
        },
        "mark": {
         "tooltip": null
        },
        "title": {
         "anchor": "middle"
        },
        "view": {
         "height": 60,
         "width": 400
        }
       },
       "data": {
        "values": [
         {
          "bayes_factor": 0.056200328561122004,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.053 or one in  18.8 records.This is equivalent to a starting match weight of -4.153.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "",
          "log2_bayes_factor": -4.15327762494924,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 74.08118264344488,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 74.08 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 6.211035225075873,
          "m_probability": 0.95,
          "m_probability_description": "Amongst matching record comparisons, 95.00% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01282376935816995,
          "u_probability_description": "Amongst non-matching record comparisons, 1.28% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 87.6946036801133,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 87.69 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 6.454416163515539,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00019005350349107303,
          "u_probability_description": "Amongst non-matching record comparisons, 0.02% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 0.4523384896189828,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is  2.21 times less likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": -1.1445253343372628,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.036845563773946086,
          "u_probability_description": "Amongst non-matching record comparisons, 3.68% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.017541263295388434,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  57.01 times less likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -5.833103537524471,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9501406133643929,
          "u_probability_description": "Amongst non-matching record comparisons, 95.01% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 1307.0145008460236,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,307.01 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.352059432056025,
          "m_probability": 0.95,
          "m_probability_description": "Amongst matching record comparisons, 95.00% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0007268473298383989,
          "u_probability_description": "Amongst non-matching record comparisons, 0.07% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 194.15009551098393,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 194.15 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 7.601028606746528,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 8.58442362482576e-05,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 0.44973174303970015,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is  2.22 times less likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": -1.1528633787484646,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.03705912896878935,
          "u_probability_description": "Amongst non-matching record comparisons, 3.71% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.01732270920069318,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  57.73 times less likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -5.851191610475199,
          "m_probability": 0.01666666666666668,
          "m_probability_description": "Amongst matching record comparisons, 1.67% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.962128179465124,
          "u_probability_description": "Amongst non-matching record comparisons, 96.21% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4829.005718588069,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,829.01 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.237510456636885,
          "m_probability": 0.6803321607265089,
          "m_probability_description": "Amongst matching record comparisons, 68.03% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 245.7183271533245,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 245.72 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.94086165637649,
          "m_probability": 0.14010841649677533,
          "m_probability_description": "Amongst matching record comparisons, 14.01% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.1796871954331449,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.57 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.4764404892853946,
          "m_probability": 0.17955942277671572,
          "m_probability_description": "Amongst matching record comparisons, 17.96% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 151.89709947284584,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 151.90 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.246950511121673,
          "m_probability": 0.8314386083810542,
          "m_probability_description": "Amongst matching record comparisons, 83.14% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.16948912362681862,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.90 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.5607353984762313,
          "m_probability": 0.16856139161894576,
          "m_probability_description": "Amongst matching record comparisons, 16.86% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.899086252274472,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.90 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.517218126152232,
          "m_probability": 0.9061720269401348,
          "m_probability_description": "Amongst matching record comparisons, 90.62% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.09769395842497895,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  10.24 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.3555868437512575,
          "m_probability": 0.0938279730598653,
          "m_probability_description": "Amongst matching record comparisons, 9.38% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 0.05517651572856884,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.052 or one in  19.1 records.This is equivalent to a starting match weight of -4.180.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "",
          "log2_bayes_factor": -4.179801833116615,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 46.65650519705579,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 46.66 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.544006341264239,
          "m_probability": 0.5983122617053012,
          "m_probability_description": "Amongst matching record comparisons, 59.83% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01282376935816995,
          "u_probability_description": "Amongst non-matching record comparisons, 1.28% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 43.12138756487968,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 43.12 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 5.430331696928683,
          "m_probability": 0.008195370782101774,
          "m_probability_description": "Amongst matching record comparisons, 0.82% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00019005350349107303,
          "u_probability_description": "Amongst non-matching record comparisons, 0.02% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 5.972145977799306,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 5.97 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 2.5782494298517644,
          "m_probability": 0.2200470854923199,
          "m_probability_description": "Amongst matching record comparisons, 22.00% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.036845563773946086,
          "u_probability_description": "Amongst non-matching record comparisons, 3.68% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.18254696155563488,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.48 times less likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.45366043918002,
          "m_probability": 0.17344528202027718,
          "m_probability_description": "Amongst matching record comparisons, 17.34% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9501406133643929,
          "u_probability_description": "Amongst non-matching record comparisons, 95.01% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 1132.5349710331207,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,132.53 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.14533988394649,
          "m_probability": 0.8231800196440322,
          "m_probability_description": "Amongst matching record comparisons, 82.32% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0007268473298383989,
          "u_probability_description": "Amongst non-matching record comparisons, 0.07% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 318.56580836721,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 318.57 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 8.315447620648742,
          "m_probability": 0.027347038514091933,
          "m_probability_description": "Amongst matching record comparisons, 2.73% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 8.58442362482576e-05,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 2.9245790946435246,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 2.92 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 1.5482290069985314,
          "m_probability": 0.10838235384781958,
          "m_probability_description": "Amongst matching record comparisons, 10.84% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.03705912896878935,
          "u_probability_description": "Amongst non-matching record comparisons, 3.71% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.042708018402391974,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  23.41 times less likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.5493492294524644,
          "m_probability": 0.041090587994056404,
          "m_probability_description": "Amongst matching record comparisons, 4.11% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.962128179465124,
          "u_probability_description": "Amongst non-matching record comparisons, 96.21% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 5077.019308821457,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 5,077.02 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.30976603156988,
          "m_probability": 0.7152734367501693,
          "m_probability_description": "Amongst matching record comparisons, 71.53% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 250.59599015831952,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 250.60 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.969219519737809,
          "m_probability": 0.1428896564952404,
          "m_probability_description": "Amongst matching record comparisons, 14.29% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.14193783645282318,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  7.05 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.81666887415251,
          "m_probability": 0.14183690675459024,
          "m_probability_description": "Amongst matching record comparisons, 14.18% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 157.92202067222217,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 157.92 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.303068544630657,
          "m_probability": 0.8644171979327953,
          "m_probability_description": "Amongst matching record comparisons, 86.44% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.13632902576639663,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  7.34 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.8748353363312464,
          "m_probability": 0.13558280206720466,
          "m_probability_description": "Amongst matching record comparisons, 13.56% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.827008555411652,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.83 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.512669903856632,
          "m_probability": 0.9033197387770264,
          "m_probability_description": "Amongst matching record comparisons, 90.33% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.1006637691555695,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  9.93 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 1,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.312383572068616,
          "m_probability": 0.09668026122297346,
          "m_probability_description": "Amongst matching record comparisons, 9.67% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 0.05922932703572757,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.056 or one in  17.9 records.This is equivalent to a starting match weight of -4.078.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "",
          "log2_bayes_factor": -4.077544495465789,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 43.81016501615015,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.81 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.453193743639373,
          "m_probability": 0.5618114517104754,
          "m_probability_description": "Amongst matching record comparisons, 56.18% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01282376935816995,
          "u_probability_description": "Amongst non-matching record comparisons, 1.28% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 40.244863150312085,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 40.24 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 5.330732744437247,
          "m_probability": 0.007648677239235594,
          "m_probability_description": "Amongst matching record comparisons, 0.76% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00019005350349107303,
          "u_probability_description": "Amongst non-matching record comparisons, 0.02% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 6.158133816708462,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 6.16 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 2.622493217594504,
          "m_probability": 0.22689991227202566,
          "m_probability_description": "Amongst matching record comparisons, 22.69% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.036845563773946086,
          "u_probability_description": "Amongst non-matching record comparisons, 3.68% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.2143261280634936,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  4.67 times less likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.2221203581258795,
          "m_probability": 0.20363995877826324,
          "m_probability_description": "Amongst matching record comparisons, 20.36% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9501406133643929,
          "u_probability_description": "Amongst non-matching record comparisons, 95.01% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 1094.209645167616,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,094.21 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.095673462525156,
          "m_probability": 0.7953233588735035,
          "m_probability_description": "Amongst matching record comparisons, 79.53% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0007268473298383989,
          "u_probability_description": "Amongst non-matching record comparisons, 0.07% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 317.6098547147231,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 317.61 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 8.311111866510924,
          "m_probability": 0.027264975402905463,
          "m_probability_description": "Amongst matching record comparisons, 2.73% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 8.58442362482576e-05,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 3.388719369390239,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 3.39 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 1.7607401676119296,
          "m_probability": 0.12558298814926738,
          "m_probability_description": "Amongst matching record comparisons, 12.56% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.03705912896878935,
          "u_probability_description": "Amongst non-matching record comparisons, 3.71% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.05386878659258979,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  18.56 times less likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.214406621285469,
          "m_probability": 0.051828677574323696,
          "m_probability_description": "Amongst matching record comparisons, 5.18% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.962128179465124,
          "u_probability_description": "Amongst non-matching record comparisons, 96.21% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4954.015871925087,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,954.02 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.274382775432663,
          "m_probability": 0.6979441563812567,
          "m_probability_description": "Amongst matching record comparisons, 69.79% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 253.94600998780402,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 253.95 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.988377996212089,
          "m_probability": 0.1447998354345956,
          "m_probability_description": "Amongst matching record comparisons, 14.48% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.1573679099579138,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.35 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.6677867145537273,
          "m_probability": 0.15725600818414764,
          "m_probability_description": "Amongst matching record comparisons, 15.73% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 155.74852270944174,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 155.75 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.283074668979893,
          "m_probability": 0.852520129932387,
          "m_probability_description": "Amongst matching record comparisons, 85.25% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.14829157311932936,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.74 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.7534914779739728,
          "m_probability": 0.147479870067613,
          "m_probability_description": "Amongst matching record comparisons, 14.75% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.68319142207502,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.68 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.503551730091991,
          "m_probability": 0.8976285482295683,
          "m_probability_description": "Amongst matching record comparisons, 89.76% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.10658945330497853,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  9.38 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 2,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.229863399917401,
          "m_probability": 0.10237145177043173,
          "m_probability_description": "Amongst matching record comparisons, 10.24% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 0.059989118532167554,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.057 or one in  17.7 records.This is equivalent to a starting match weight of -4.059.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "",
          "log2_bayes_factor": -4.059155356776684,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 43.42788509499787,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.43 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.440549791237722,
          "m_probability": 0.5569091821713592,
          "m_probability_description": "Amongst matching record comparisons, 55.69% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01282376935816995,
          "u_probability_description": "Amongst non-matching record comparisons, 1.28% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 39.99257176768589,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 39.99 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 5.321660153159371,
          "m_probability": 0.007600728378066878,
          "m_probability_description": "Amongst matching record comparisons, 0.76% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00019005350349107303,
          "u_probability_description": "Amongst non-matching record comparisons, 0.02% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 6.149087018356728,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 6.15 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 2.620372223156573,
          "m_probability": 0.22656657788640683,
          "m_probability_description": "Amongst matching record comparisons, 22.66% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.036845563773946086,
          "u_probability_description": "Amongst non-matching record comparisons, 3.68% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.21988693949665092,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  4.55 times less likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.185166179111879,
          "m_probability": 0.20892351156416705,
          "m_probability_description": "Amongst matching record comparisons, 20.89% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9501406133643929,
          "u_probability_description": "Amongst non-matching record comparisons, 95.01% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 1086.1401447009125,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,086.14 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.084994550806584,
          "m_probability": 0.7894580640061504,
          "m_probability_description": "Amongst matching record comparisons, 78.95% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0007268473298383989,
          "u_probability_description": "Amongst non-matching record comparisons, 0.07% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 315.98468298256677,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 315.98 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 8.303710816782589,
          "m_probability": 0.027125463776786245,
          "m_probability_description": "Amongst matching record comparisons, 2.71% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 8.58442362482576e-05,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 3.44947014172407,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 3.45 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 1.7863747727273322,
          "m_probability": 0.1278343588561404,
          "m_probability_description": "Amongst matching record comparisons, 12.78% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.03705912896878935,
          "u_probability_description": "Amongst non-matching record comparisons, 3.71% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.05776996719067391,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  17.31 times less likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.113536514446963,
          "m_probability": 0.05558211336092304,
          "m_probability_description": "Amongst matching record comparisons, 5.56% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.962128179465124,
          "u_probability_description": "Amongst non-matching record comparisons, 96.21% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4920.224045694047,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,920.22 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.264508295806355,
          "m_probability": 0.6931834111068901,
          "m_probability_description": "Amongst matching record comparisons, 69.32% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 253.21693876293475,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 253.22 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.98423010580442,
          "m_probability": 0.1443841195374006,
          "m_probability_description": "Amongst matching record comparisons, 14.44% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.1625480546465227,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.15 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.6210618047379532,
          "m_probability": 0.1624324693557093,
          "m_probability_description": "Amongst matching record comparisons, 16.24% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 154.99046610133217,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 154.99 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.276035663781599,
          "m_probability": 0.8483707581964681,
          "m_probability_description": "Amongst matching record comparisons, 84.84% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.1524637822614597,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.56 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.713461523408021,
          "m_probability": 0.15162924180353188,
          "m_probability_description": "Amongst matching record comparisons, 15.16% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.654213182791178,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.65 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.501707479492909,
          "m_probability": 0.8964818094671695,
          "m_probability_description": "Amongst matching record comparisons, 89.65% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.10778344103938929,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  9.28 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "iteration": 3,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.2137925435627768,
          "m_probability": 0.10351819053283055,
          "m_probability_description": "Amongst matching record comparisons, 10.35% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.05659408901785539,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         }
        ]
       },
       "params": [
        {
         "bind": {
          "input": "range",
          "max": 3,
          "min": 0,
          "step": 1
         },
         "description": "Filter by the iteration number",
         "name": "iteration_number",
         "value": 3
        }
       ],
       "resolve": {
        "axis": {
         "y": "independent"
        },
        "scale": {
         "y": "independent"
        }
       },
       "selection": {
        "zoom_selector": {
         "bind": "scales",
         "encodings": [
          "x"
         ],
         "type": "interval"
        }
       },
       "title": {
        "subtitle": "Training session blocked on <splink.blocking.BlockingRule object at 0x7fe966bfc4f0>",
        "text": "Model parameters (components of final match weight)"
       },
       "transform": [
        {
         "filter": "(datum.iteration == iteration_number)"
        }
       ],
       "vconcat": [
        {
         "encoding": {
          "color": {
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             0,
             10
            ],
            "range": [
             "red",
             "orange",
             "green"
            ]
           },
           "title": "Match weight",
           "type": "quantitative"
          },
          "tooltip": [
           {
            "field": "comparison_name",
            "title": "Comparison name",
            "type": "nominal"
           },
           {
            "field": "probability_two_random_records_match",
            "format": ".4f",
            "title": "Probability two random records match",
            "type": "nominal"
           },
           {
            "field": "log2_bayes_factor",
            "format": ",.4f",
            "title": "Equivalent match weight",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor_description",
            "title": "Match weight description",
            "type": "nominal"
           }
          ],
          "x": {
           "axis": {
            "domain": false,
            "labels": false,
            "ticks": false,
            "title": ""
           },
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             10
            ]
           },
           "type": "quantitative"
          },
          "y": {
           "axis": {
            "title": "Prior (starting) match weight",
            "titleAlign": "right",
            "titleAngle": 0,
            "titleFontWeight": "normal"
           },
           "field": "label_for_charts",
           "sort": {
            "field": "comparison_vector_value",
            "order": "descending"
           },
           "type": "nominal"
          }
         },
         "height": 30,
         "mark": {
          "clip": true,
          "height": 20,
          "type": "bar"
         },
         "selection": {
          "zoom_selector": {
           "bind": "scales",
           "encodings": [
            "x"
           ],
           "type": "interval"
          }
         },
         "transform": [
          {
           "filter": "(datum.comparison_name == 'probability_two_random_records_match')"
          }
         ]
        },
        {
         "encoding": {
          "color": {
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             0,
             10
            ],
            "range": [
             "red",
             "orange",
             "green"
            ]
           },
           "title": "Match weight",
           "type": "quantitative"
          },
          "row": {
           "field": "comparison_name",
           "header": {
            "labelAlign": "left",
            "labelAnchor": "middle",
            "labelAngle": 0
           },
           "sort": {
            "field": "comparison_sort_order"
           },
           "type": "nominal"
          },
          "tooltip": [
           {
            "field": "comparison_name",
            "title": "Comparison name",
            "type": "nominal"
           },
           {
            "field": "label_for_charts",
            "title": "Label",
            "type": "ordinal"
           },
           {
            "field": "sql_condition",
            "title": "SQL condition",
            "type": "nominal"
           },
           {
            "field": "m_probability",
            "format": ".4f",
            "title": "M probability",
            "type": "quantitative"
           },
           {
            "field": "u_probability",
            "format": ".4f",
            "title": "U probability",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor",
            "format": ",.4f",
            "title": "Bayes factor = m/u",
            "type": "quantitative"
           },
           {
            "field": "log2_bayes_factor",
            "format": ",.4f",
            "title": "Match weight = log2(m/u)",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor_description",
            "title": "Match weight description",
            "type": "nominal"
           }
          ],
          "x": {
           "axis": {
            "title": "Comparison level match weight = log2(m/u)"
           },
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             10
            ]
           },
           "type": "quantitative"
          },
          "y": {
           "axis": {
            "title": null
           },
           "field": "label_for_charts",
           "sort": {
            "field": "comparison_vector_value",
            "order": "descending"
           },
           "type": "nominal"
          }
         },
         "mark": {
          "clip": true,
          "type": "bar"
         },
         "resolve": {
          "axis": {
           "y": "independent"
          },
          "scale": {
           "y": "independent"
          }
         },
         "selection": {
          "zoom_selector": {
           "bind": "scales",
           "encodings": [
            "x"
           ],
           "type": "interval"
          }
         },
         "transform": [
          {
           "filter": "(datum.comparison_name != 'probability_two_random_records_match')"
          }
         ]
        }
       ]
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 4 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blocking_rule = \"l.dob = r.dob\"\n",
    "training_session_dob = linker.estimate_parameters_using_expectation_maximisation(blocking_rule)\n",
    "training_session_dob.match_weights_interactive_history_chart()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The final match weights can be viewed in the match weights chart:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v4+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v5.2.json",
       "config": {
        "header": {
         "title": null
        },
        "mark": {
         "tooltip": null
        },
        "title": {
         "anchor": "middle"
        },
        "view": {
         "height": 60,
         "width": 400
        }
       },
       "data": {
        "values": [
         {
          "bayes_factor": 0.00018003240583304995,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.000 or one in  5,555.6 records.This is equivalent to a starting match weight of -12.439.",
          "comparison_name": "probability_two_random_records_match",
          "comparison_sort_order": -1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "",
          "log2_bayes_factor": -12.439455764512674,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 43.42788509499787,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.43 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.440549791237722,
          "m_probability": 0.5569091821713592,
          "m_probability_description": "Amongst matching record comparisons, 55.69% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01282376935816995,
          "u_probability_description": "Amongst non-matching record comparisons, 1.28% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 39.99257176768589,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 39.99 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 5.321660153159371,
          "m_probability": 0.007600728378066878,
          "m_probability_description": "Amongst matching record comparisons, 0.76% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00019005350349107303,
          "u_probability_description": "Amongst non-matching record comparisons, 0.02% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 6.149087018356728,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 6.15 times more likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 2.620372223156573,
          "m_probability": 0.22656657788640683,
          "m_probability_description": "Amongst matching record comparisons, 22.66% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "jaccard(\"first_name_l\", \"first_name_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.036845563773946086,
          "u_probability_description": "Amongst non-matching record comparisons, 3.68% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.21988693949665092,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  4.55 times less likely to be a match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.185166179111879,
          "m_probability": 0.20892351156416705,
          "m_probability_description": "Amongst matching record comparisons, 20.89% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9501406133643929,
          "u_probability_description": "Amongst non-matching record comparisons, 95.01% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 1086.1401447009125,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,086.14 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.084994550806584,
          "m_probability": 0.7894580640061504,
          "m_probability_description": "Amongst matching record comparisons, 78.95% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0007268473298383989,
          "u_probability_description": "Amongst non-matching record comparisons, 0.07% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 315.98468298256677,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.9` then comparison is 315.98 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "jaccard >= 0.9",
          "log2_bayes_factor": 8.303710816782589,
          "m_probability": 0.027125463776786245,
          "m_probability_description": "Amongst matching record comparisons, 2.71% of records are in the jaccard >= 0.9 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.9",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 8.58442362482576e-05,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the jaccard >= 0.9 comparison level"
         },
         {
          "bayes_factor": 3.44947014172407,
          "bayes_factor_description": "If comparison level is `jaccard >= 0.5` then comparison is 3.45 times more likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "jaccard >= 0.5",
          "log2_bayes_factor": 1.7863747727273322,
          "m_probability": 0.1278343588561404,
          "m_probability_description": "Amongst matching record comparisons, 12.78% of records are in the jaccard >= 0.5 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "jaccard(\"surname_l\", \"surname_r\") >= 0.5",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.03705912896878935,
          "u_probability_description": "Amongst non-matching record comparisons, 3.71% of records are in the jaccard >= 0.5 comparison level"
         },
         {
          "bayes_factor": 0.05776996719067391,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  17.31 times less likely to be a match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -4.113536514446963,
          "m_probability": 0.05558211336092304,
          "m_probability_description": "Amongst matching record comparisons, 5.56% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.962128179465124,
          "u_probability_description": "Amongst non-matching record comparisons, 96.21% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 312.16784723322775,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 312.17 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.286178139563434,
          "m_probability": 0.6113456435064976,
          "m_probability_description": "Amongst matching record comparisons, 61.13% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.001958387607580057,
          "u_probability_description": "Amongst non-matching record comparisons, 0.20% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 16.949819994845715,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 16.95 times more likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": 4.083198047073621,
          "m_probability": 0.3385185513724552,
          "m_probability_description": "Amongst matching record comparisons, 33.85% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.01997180804724744,
          "u_probability_description": "Amongst non-matching record comparisons, 2.00% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 0.4960511903664868,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is  2.02 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": -1.0114390866197067,
          "m_probability": 0.03776735189690557,
          "m_probability_description": "Amongst matching record comparisons, 3.78% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.07613599690992119,
          "u_probability_description": "Amongst non-matching record comparisons, 7.61% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.013713260465657417,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  72.92 times less likely to be a match",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -6.188284562611205,
          "m_probability": 0.012368453224141701,
          "m_probability_description": "Amongst matching record comparisons, 1.24% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9019338074352513,
          "u_probability_description": "Amongst non-matching record comparisons, 90.19% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 4874.614882141059,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,874.61 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 3,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.251072528227413,
          "m_probability": 0.6867577859166996,
          "m_probability_description": "Amongst matching record comparisons, 68.68% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.00014088452165375115,
          "u_probability_description": "Amongst non-matching record comparisons, 0.01% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 249.4676329581296,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 249.47 times more likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.962708835798494,
          "m_probability": 0.14224626801708795,
          "m_probability_description": "Amongst matching record comparisons, 14.22% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0005701992933125814,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.1711176250398338,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.84 times less likely to be a match",
          "comparison_name": "postcode_fake",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.546939730454512,
          "m_probability": 0.1709959460662125,
          "m_probability_description": "Amongst matching record comparisons, 17.10% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9992889161850337,
          "u_probability_description": "Amongst non-matching record comparisons, 99.93% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 153.443782787089,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 153.44 times more likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 7.261566381855136,
          "m_probability": 0.8399046832887611,
          "m_probability_description": "Amongst matching record comparisons, 83.99% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "\"birth_place_l\" = \"birth_place_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.005473696412021928,
          "u_probability_description": "Amongst non-matching record comparisons, 0.55% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.16097645294413918,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.21 times less likely to be a match",
          "comparison_name": "birth_place",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.635078423349897,
          "m_probability": 0.16009531671123883,
          "m_probability_description": "Amongst matching record comparisons, 16.01% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9945263035879781,
          "u_probability_description": "Amongst non-matching record comparisons, 99.45% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 22.776649717532823,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.78 times more likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 5,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.509483647404612,
          "m_probability": 0.9013269182036521,
          "m_probability_description": "Amongst matching record comparisons, 90.13% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.039572409875093965,
          "u_probability_description": "Amongst non-matching record comparisons, 3.96% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.10273869973218412,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  9.73 times less likely to be a match",
          "comparison_name": "occupation",
          "comparison_sort_order": 5,
          "comparison_vector_value": 0,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -3.282948374818687,
          "m_probability": 0.09867308179634793,
          "m_probability_description": "Amongst matching record comparisons, 9.87% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.00018,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9604275901249061,
          "u_probability_description": "Amongst non-matching record comparisons, 96.04% of records are in the all other comparisons comparison level"
         }
        ]
       },
       "resolve": {
        "axis": {
         "y": "independent"
        },
        "scale": {
         "y": "independent"
        }
       },
       "selection": {
        "zoom_selector": {
         "bind": "scales",
         "encodings": [
          "x"
         ],
         "type": "interval"
        }
       },
       "title": {
        "subtitle": "Use mousewheel to zoom",
        "text": "Model parameters (components of final match weight)"
       },
       "vconcat": [
        {
         "encoding": {
          "color": {
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             0,
             10
            ],
            "range": [
             "red",
             "orange",
             "green"
            ]
           },
           "title": "Match weight",
           "type": "quantitative"
          },
          "tooltip": [
           {
            "field": "comparison_name",
            "title": "Comparison name",
            "type": "nominal"
           },
           {
            "field": "probability_two_random_records_match",
            "format": ".4f",
            "title": "Probability two random records match",
            "type": "nominal"
           },
           {
            "field": "log2_bayes_factor",
            "format": ",.4f",
            "title": "Equivalent match weight",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor_description",
            "title": "Match weight description",
            "type": "nominal"
           }
          ],
          "x": {
           "axis": {
            "domain": false,
            "labels": false,
            "ticks": false,
            "title": ""
           },
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             10
            ]
           },
           "type": "quantitative"
          },
          "y": {
           "axis": {
            "title": "Prior (starting) match weight",
            "titleAlign": "right",
            "titleAngle": 0,
            "titleFontWeight": "normal"
           },
           "field": "label_for_charts",
           "sort": {
            "field": "comparison_vector_value",
            "order": "descending"
           },
           "type": "nominal"
          }
         },
         "height": 30,
         "mark": {
          "clip": true,
          "height": 20,
          "type": "bar"
         },
         "selection": {
          "zoom_selector": {
           "bind": "scales",
           "encodings": [
            "x"
           ],
           "type": "interval"
          }
         },
         "transform": [
          {
           "filter": "(datum.comparison_name == 'probability_two_random_records_match')"
          }
         ]
        },
        {
         "encoding": {
          "color": {
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             0,
             10
            ],
            "range": [
             "red",
             "orange",
             "green"
            ]
           },
           "title": "Match weight",
           "type": "quantitative"
          },
          "row": {
           "field": "comparison_name",
           "header": {
            "labelAlign": "left",
            "labelAnchor": "middle",
            "labelAngle": 0
           },
           "sort": {
            "field": "comparison_sort_order"
           },
           "type": "nominal"
          },
          "tooltip": [
           {
            "field": "comparison_name",
            "title": "Comparison name",
            "type": "nominal"
           },
           {
            "field": "label_for_charts",
            "title": "Label",
            "type": "ordinal"
           },
           {
            "field": "sql_condition",
            "title": "SQL condition",
            "type": "nominal"
           },
           {
            "field": "m_probability",
            "format": ".4f",
            "title": "M probability",
            "type": "quantitative"
           },
           {
            "field": "u_probability",
            "format": ".4f",
            "title": "U probability",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor",
            "format": ",.4f",
            "title": "Bayes factor = m/u",
            "type": "quantitative"
           },
           {
            "field": "log2_bayes_factor",
            "format": ",.4f",
            "title": "Match weight = log2(m/u)",
            "type": "quantitative"
           },
           {
            "field": "bayes_factor_description",
            "title": "Match weight description",
            "type": "nominal"
           }
          ],
          "x": {
           "axis": {
            "title": "Comparison level match weight = log2(m/u)"
           },
           "field": "log2_bayes_factor",
           "scale": {
            "domain": [
             -10,
             10
            ]
           },
           "type": "quantitative"
          },
          "y": {
           "axis": {
            "title": null
           },
           "field": "label_for_charts",
           "sort": {
            "field": "comparison_vector_value",
            "order": "descending"
           },
           "type": "nominal"
          }
         },
         "mark": {
          "clip": true,
          "type": "bar"
         },
         "resolve": {
          "axis": {
           "y": "independent"
          },
          "scale": {
           "y": "independent"
          }
         },
         "selection": {
          "zoom_selector": {
           "bind": "scales",
           "encodings": [
            "x"
           ],
           "type": "interval"
          }
         },
         "transform": [
          {
           "filter": "(datum.comparison_name != 'probability_two_random_records_match')"
          }
         ]
        }
       ]
      },
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqgAAAJVCAYAAAAStAebAAAAAXNSR0IArs4c6QAAIABJREFUeF7snQm8ldP6x58ikiGZGtxCVIYyNBv+JTIP0YDKlAjdDCUhU8ZShIpIKQkhQsaMkaKoZB6SSKUSma50u/0/33Wtc/fZ7X3O3mftfc679/mtz+d8Ou3zrvWu9XvW3u93P+tZ66mwfv369aYiBaSAFJACUkAKSAEpIAUiokAFAWpELKFuSAEpIAWkgBSQAlJACjgFBKiaCFJACkgBKSAFpIAUkAKRUkCAGilzqDNSQApIASkgBaSAFJACAlTNASkgBaSAFJACUkAKSIFIKSBAjZQ51Jl0FPD7+ypUqFCoWrLX02k71WtL816p9ilXr1u4cKFtuummVqtWrVwdQs71+88//7Tff//dtt1224R9/+WXX9zrVapUcddhn8qVKwePk/dN/Ps2uNEcbAB90aJq1aoZ7/3XX3/tbKX3U8alVYOlpIAAtZSE1m0yq8AFF1xgI0aMcI0OGjTILrvssoLfr7jiCvd7v3797JZbbkn7xtOnT7f/+7//sxNPPNGefPLJIuuffPLJ9thjj9lbb71lBx10UNr3imqFsWPHOoA488wzS6WLs2fPtubNm9vdd99t559/fqncM+o3mT9/vs2aNctatGhhjRo1ynh3u3XrZuPGjbMddtjBfvjhh0Lt//vf/7YDDjjAsAvvg+7du9uxxx5rvXv3tqFDhwb15d5777XzzjvPbrrpJuvfv39QW0VVXrFihT399NNWp04dO/zww4u9z1ZbbWW//vqrA8bSKh7S16xZY5tsskmRt+W98c9//tPuuOMOu+iiiza4Nn68Z599tvtsWrBggW2//falNSTdRwpkTAEBasakVEOlqUAsoB5xxBH24osvutsfeeSR9tJLL7nfL730Uhs8eHDa3QI2W7VqZe3atbOnnnoqJUB98803HdTmS/EPztJ4WK9du9aaNm1qABmAsMUWW+SLjEHjuP32261Pnz42fPhw69WrV1Bb8ZWXL19u1atXty233NJuvfVW69GjR6FL3njjDWvTpo3tueee7gvgTjvtZPfcc497rVOnTkF9oR2+hNx444125ZVXBrVVVOX33nvPmjVrltIXTdopzTnv+33hhRcaXmzgc+ONNy5SC+YB12OvSy65ZINr48f7wQcf2L777munnXaajR8/Pms6q2EpkC0FBKjZUlbtZlUBD6g8YIEaPBAUliD9ax5Q8QbxsMdb9Mknn1jjxo3t3HPPtXPOOcc9lP71r3/Ztddea0888YRtvfXWduihh9qQIUMKABUPBF7ZadOmWY0aNeyEE06wq6++2j1QvAc1EaDiheSnffv2NmPGDJs5c6Z7wOPVrVmzpuszD3+8tKtWrXKeMtoDAD766CPr2bOn8/wAE++8847zpj3yyCN211132Zdffmn77LOPHXLIIda3b1+rWLGi6zfeMF5/6KGH3O9AwHPPPefGxjLisGHDrGXLlvaf//zH8GSNHj3aFi9ebE2aNHHXok3Xrl3t4Ycfdnoed9xxri36kEwDHpZ42oCOkSNHOm07d+7sgAbA50F58MEHO73btm27wbzgGrx0QBgPYQqeL2zAOKmL5rRJ+eyzz9yY0XyzzTaz1q1bu4c2njIgF+8SY/zxxx/t1Vdftf32289uuOEGu/766w3vOGMFCFj6ZB7RvzPOOMMmTJhgS5YssVNOOcV59phLxc0d5gFtYrPHH3/c6cQXm2uuucYtrzI3BwwYUPBFhy8x9HW77bZzcyhZ3ddff90uvvhi++KLL6x+/fpuNQAv5gsvvOD6+corr7hxoQveyEQFu2PTjz/+2HnQ0BgN8LQBLXwR471y1FFH2aOPPlrQxGuvveY8pWj5j3/8w92XL36M6ZhjjnH6FjfuZcuWufmCjXif4I3FvmhfFKCma79k90Ff5vHbb7/txoi3+M4773TvK2z77rvvuvfJYYcd5uYAtvKAyrxjHtNv3oPx3ko/R/gi9fzzz7v3JfNx5513diCIroA38xa90ZP5x2dPvXr1nBfUz2V0/eOPP9w13A9tuPdvv/3m9OO9x/uW94gHVL5MEA7DvAXAeZ33f6LxsirBe5Pl/l122SWrn8lqXApkWgEBaqYVVXulooAHVA8GABwFMPGveUC97rrrHCRQ8LZ6D6tfTo71xuIx4kFCATR4QOy+++4O4njgfPvtt+7DHhijflGAysMJ8KXEtksfeXACryzD1a1b1w488EB78MEH3bVfffWVW3LlNV8ABaCEvvDApW9Tp051QPTAAw+4h5P3wPB3wI2/+QKs8n/aYQw8BHlQci2hCYAPBSAEYj0onnXWWTZw4EAHFsk0iPVa0wb9qVatmh1//PHOqwwIEI5BH4D9eA/p5Zdf7qAdTQAZoNKDLJrjzaM8++yzDuLRC/Dbe++93YMce9A2D+05c+YUeLIZK32O14D/A13AOZ5ygCLeRkAMS9DFzR0AEZimAPfcnzJp0iTr0KGDnXTSSQ5cgcxtttnGfdHgd+YYf09Wl2uBHtoDQvCkAuJANeOnLl9WGB+eM+wTWwAevrBQ+BIDpKAZ7w2WiPHE8aUF3ZjD2NwXvkxhE3Rh3vLFArBiiR8YZu4UN24AGJBmDjOWKVOmFIQSFAWoPryGvqRiv2T3QTe+KAHpjJH3K/N91113dTowL+fNm+d+JzyIL4qxMbH0m/lIoU+x70VWFfiiyvuJ+cwXTD5rKMSU8gWE9z7j5L2OF5MC3APs3HPixIlO99glfr7AMt/j5xLvUdr1gOr/znuVtoBsPkvix8tnng8LYK7wxUtFCuSSAgLUXLKW+lqggIdKHqx+2YsPe7x5/jUeGsCVB7f333/fQQSgg2eQByBw42O//IMe7wYPECAQ8AMyeKABXnjlAAb/MPJxXok8qB5Q/RLbd99957x8FLyAeMc+/fRTBx546niAAByAKIDpH4rEkeH9whMGXO+1114OVPBi8QAGgr1Hl7bxkG200UYODCg8yLh+8803d/8H6tCB+/Nw5QFKX/Ew81CjvdjlTgCrKA2AHvrFv4AmnhpswgOVLwRoROHBzpjiN23sv//+DtwY+4477mgdO3Z08DRq1CgHR8AeninssXr1amdjH36xbt06F7sKkABELEVjK8aHZ4vQD8AKqMN7hucMLzbgABB4QB0zZowB49yLcQA1eMqKmjvY04Oaj4MGFGgHm/Ma46FgZzyygDdzDghnngKoieoy3vglfjRBG8YCQOMVxQsP8PIlILbw5QAo9PGKK1euLIhD/Oabb5zXDvgEiPDWxhe+mJ166qkOVHkPMc8SAWqyvuPlZ54B0nj66DeF+/I+SrbE7wE1VfsVdR/eL7FL/FyLF7dLly7uiyfvQTzTeD6xhZ/zzAvmh19JwBbYM7b4zx/Aj/cQ7xEKqyy8f/hShc7AKl/OgGAAkvcJY+d9wFhjARWb3nbbbXb//fc7j69/38UDKh5vPLfMKWzoY4gThTTwxfPoo49270c8yCpSIJcUEKDmkrXU1w0AFaDgIbPHHnu4Za5FixY5bwYPFP7lg7l27dquHsvaPBB++umnAnjzS6j+IcB1fskZCMKj5zdgxcsP+LGUB0AWBagefqjvl9yAAryMgB99iC0vv/yy2zXNQwzYom0K/eah6b2dvg5Lr/x4mIo/WcD/328CoR3unah4mI4FVOJ4i9KAhz4PXsbkPUBAiQ9P8PehbbxKjC22+HsBNEA0sMVSr/9CEXstD3faACiABAp9o494O/EWopkHWMAV76KPU6Zd2gdcAFgPqMBKgwYN3E517+EFJIHBZHMHOMR+QKaHGg8EvM4yOfCdqACyzLNkdVlyjwdU+kZ/vZeWdpkjzC/6HlvwFNJ/riUUgIIO/B87EX+aCUBNNG76zpcFvkjEevHpA2MA+IoD1FTtV9R98FLHAir9YYNXsk1Gfh4SE817yc8rPO143GOLtzNzkPHypQRYZymfL4t493kPxK8u+DY8VMYCKnaiz4AnKyWxccKxHlS+MPDFwX+O+c+uRIDKFz/mYOznSMIJqRelQAQVEKBG0CjqUvEKeA8GgAoc3Xfffa4SUID3xQMqDwx/LI6Pw2IJkwc7H+wsd3tYw8vEcTt4Goj/4yFJ/CjxiYAPng3K999/77yBwCZepuIAlWtYvsdLWqlSJdcGDyGW+AAmPI30lyVAlhVjATV2oxZLvnhY2FlPbCEPTjye6QIqD7vddtvNPQDxJjds2NAtFeKdBFz4WyygAhRFaYA3GhvEAiUaE1dLYXkZqOR+3jsUa2Huyd88JAIDXIeXCG8RgI4HDzibO3euAwC/RE87HgLwuPJFpSSA6iHEP+SZG3iii5o76Og9qHhmmQ+xgHrzzTc7LWkLbfDUM8fYFANQ4lkGUBPVTQSozBm8ttQnDAQvIMCZaBMMXmK8eMw75h918MpTfAhLJgA1Ud/xVPpd48Aw98HGmQZU5kdR94kHVLQ4/fTTC5b08djz3mMlhc+T+E1SRQEqnmC+TPl4d+Y/qxB+ox9flnhf+pMS/JwFfgnfYV7xhSEWUOkDc5jPE7z4eP+Z//EeVL9JKhVA9V5YPNl8VqpIgVxSQICaS9ZSXwsUiAVUvDIAFIUPYTY9eUDFswbQsazIgxKgALh4OPnYM2L1iNnD6wEE+eOrgEO8LX5zAVDIpgz+DgQDXoQDFAeo9AvvDaBFHZZkWX4knhLAIE4M7yGeYAoPO7x4QHQsoPpYTeIAiRvFCwVYXnXVVQ5UU/WgUocNGIyDe/AwpA94cgEmlof9g5NriHHzHrpEGrCEGA+o3j7YAYDlCwTgzeYrv0HEG5P6gN0zzzzjrvUhGOjEA566fLkgJhIgwbtEAVKxtd/QxZI8MYElAVQ8WswTwhyAZR9rWdzcKQpQ6RdzClCiHcbjj1XiSw4bcIoCVLRHR7y/AA9wiq1YfiaekrmEJ5lNM2x4iy2MA+2AG+7DcjJffvAo4lX7/PPPM+JBTQSozCU2KXFvPJyTJ092ulKY54BiJjyozJmi7sN85n2KF5yNT4TScD2F+GLmIxDvj7tKB1Bpg5AHvjhR+DJGeIs/1g7vPFrHhoxge/rMe4WYYr5sxgIqfcHW6MYXYh+fnCqg8uUtdrx8ufZhRgAyv6tIgVxSQICaS9ZSXzcAVOLyiMn00MKDH9gEjPw5qHgaAFhi8nzh72z2wAvC5hoeNt6zxPKsj/skvoyHEEt4frkS0MUzwb9FnYPqHw5AoPes4K0BxFjOY4kXjw7AyOucdckDDNjkQcXSXCyg4jXjdWCNQiwdEET7xL6lCqhAAh5gYuI83NGe98by0IzdOIbXD6hJpoH3YMZ6UIm/A5D8Bifax4uDZ5D42NiCtxTY5OGOzfD28f/YvuGN9OfbYnPaRjcKD3AgiC8aQBjw7nXzD22/xP/hhx86aIyPQQXCARYKYQrYF5gpbu54QPVA4r1ezCHGCggy14BJ31cAjf4VV5cvMYAK8w6gQR9s4DfT0R5zkM1yPtbV60psLlDLCoIvwAvjAth8/CLtEw8bX4itZH6hOdoni0FNNm5CX6jn9eTLA8DIvQiNSXYOarr2K+o+zAMgkfeeX3L3YRN+vMxJvrzyORAPqP4LYWyITqxO/jxXbICtfGgQX3aWLl3qQo4o6BB7nBbx5Mx5TgzxYTeEi3A9nxl8Kabg+eYzyocD+E1SgC3z4eeff3arP/7vf/311wbj9Z9PicJl9DiRAlFXQIAadQupfxlTgA1OACwblXg4xBbiUwE/oCT+b1zH8jzeOZZJiWlNJQuOB1TiDXnY47VlQ4Z/cNEuDxXCDIgZTKVNwJJ+sBnIhwuECIQmPEyBlvjYUJawGTcPQfpWEg0YM0vlbIzyy7zx/QUCGQ9fMgAeXwBjwAZt/AYv/zffF2AXD3c89KaqiY9BBZqwPUu3iTLvFDV3irsXNkMH2mYsxZ13Gdse4wREsI23D3r5ucpcjJ1P8X3hnswXQldKO6MQ/cTDne37FnUftOfvaA4MUlhxQT+OevMe1eJsGPp37skXYe6ZLGsXKyx8FrBpkS/dhDDwxZnQH1YVUimx4+V6vvhyAgOAqiIFck0BAWquWUz9zRkFYgG1tDIy5Yw4cR31ni08bHj1SqvEAirxoipSoKwU4PQOPPmU2CPLSppdDU8rceusKBGTrCIFck0BAWquWUz9zRkFOKqKZWpOEkh0QH3ODKQUOoonmTg5PEeEPZRWYQmb5VmWa/FuqUiBslSAmFW+pBGKwsoA3tOSnl/KF2Ri2QkHUJECuaiAADUXraY+SwEpIAWkgBSQAlIgjxUQoOaxcTU0KSAFpIAUkAJSQArkogIC1Fy0mvosBaRAZBVgBz2lpBu3IjswdUwKSAEpUIoKCFBLUWzdSgrkmwIcxcMxPuw4ji3s+ufYHY5TKm+FVJecFsHh/5koHFPEsVCxx0vRLlnTOPWAs1FVpIAUkAL5poAANd8sqvFIgVJUAECdN2+ey3oTD6icTcpZn+WtZBpQOUieM3rJHBVbONeXjGccI6UiBaSAFMg3BQSo+WZRjUcKlKICqQAqB8xzMDlncpIcgEPL8bDi/eMYHA6z5yB9jpqKPyOSa8heRVpIPJIcvA6UkZWHs0s5IcEfgk4CBHblk3GJo6rIxMQ5kKTe5CxMMm5ReJ1zKckkRr50PJQkSOAepGQlaxanCpCqkoPtORf30ksvdbupuZZD1TkEH68mCSBojzSnZLziHEvOJeXcSc5uJVEB7eL9ZGzJ2qVfZCm76KKLXJ199tnHHcxPQodEgMqZmpwMwXmePo0l/SFrFPfmTFn0JIlDbPnoo4/cYf+xhYxmHAKfrD4hC+R/56QD+t+1a1fXJ84EJl0rZ+iy85yzXtEfzy624mxbDsHn7FcVKSAFpEC6CghQ01VM10sBKVCgQHGAytFNpE4lbSPZpUjLyXmPwEuNGjVcxqqOHTsaKWlJDMBh5bHFnw0JhJEVp1evXi6zks+2A6wCpNTliCr+D0iSQYkD7kmPCjRx6L9fcqfPZKECXMnWRQpMsvvQJq9zKDrXAKdAImANiJJXnnuRcQuvJUDKeAYNGuRgl4xaeIxJfMARP0A0gNm7d28H5gBysnbpA6AIDJNNC+CjXcCafsZ7UOnTnXfe6foMXDN2znElGxn3JZ0mfyf7GJmGfKE9so5RCENAT473AvKT1cd2pOoFUmkfQOVcXzQmWxPZn9AWTzqwTqpZvnj4LxZkQ1KRAlJACqSrgAA1XcV0vRSQAikD6tq1a420nwAXh+IDWttvv71LA8pB4ngc8abiRcXrB9wBrr54QMXjSSYlUqByLbnlyf7FRiSAixADAHLmzJmu6meffWZ77LGH8+Zx0HkyQMW7R7/wjNI+bTdt2tS1i9eTdJMUzmb18Ej2JvoJSHKGKhBOStxNN93U3ZezbxkvfaQApYAt7SVrFwDmvnhGyTZGRiB0oO+0WdwSv0+xCqiT+pVC5iS8np07d95gxmIXUsNSOMgdnY866igH+vH18cTi4fZpU4FRPMiMCUDFkwu0481FT59sAXgGbrGLihSQAlIgXQUEqOkqpuulgBQoUGDIkCEOQjyM8QfSc5KGFe8lHkqgkqVewKdbt27OE4fHEW9bfOGA8oYNGxYCVDx2P/zwg3sNjytpK4EmD2FsxgLEyEc/dOhQ9zopNgHO9957z3lCYwGVg8tZpsczCUyRhhavLOEDxI/i9YzPbU+bgDZgTQgA4/7nP//pABIPMV5EPLxAG20AxuPGjXN9AX7pB+NN1i5eZNqPL3gfSbNaHKAy/mHDhjlg9gV4pF28yvEFuMYmeD0B7KLqE+oAXPsNb2gGlALR3INlfjyqpKVlfNgKry19J242dm7orSMFpIAUSFUBAWqqSuk6KSAFNlCA5fDjjjvO8Mj5HPPEMu65554OVABFPIv84KVjSZmlXzynLEMTb0mhPvGRQOImm2xSCFCBWsDPAyrw60EULyGAOmPGDAdbZOKhALp77723/fnnnw4Y6cvYsWPd31iG9zAL+BHvyb0BVtrCi1qtWjUjptV7Gbk/3k+WwVm+5v8srRNGAOAdfPDBLsYUYI3fJOUBFcgjs0+idvE+4qUkDtV7MOkboRCjRo0qFlAJjSBvO0v49BN4rFq1qvMsH3DAAYXshreXkAm+QACYlKLqs/wP6ALkFMCTLyW0Tf3zzjvPeZo9oAL7eMkFqPrAkAJSIEQBAWqIeqorBcq5AoAmEIWXDa8cgERcJpBHnCZxkIDcI4884oAJkONYJLyQABnL80ApHkkAkbhID7pIC9SmAqgsv9M2x10BTT6FKTAINBJOQFv0j01LbCYCcgEp+kZKSf7evn17B4mAaZ06dZxnEfCir8RqMja/nO49qiz/A5+AH7GsyQCVv+NRTtQu96Uv6MDGr5dfftkBJ+AMNCfyoD733HNuwxOaAaQAMvXxWFMfj+6qVasK6elDKYiLBbQp6A2QJ6uPXfCEkrqXTWd4T+kvnlkBajn/ANDwpUAWFRCgZlFcNS0FyoMCQCAeNDYYUdjVDSgCoXguia1khzeeQbyaXL/NNtu4jTfeE8qSMKDI7vvYkgqg4s0DTlmqB6K4DzvMOYaJ3fVANLvZ6QOARY5zNvsAXsRJAmvcH28vm7XwCHK2K9BKHQrgyfI+3tvffvvN3YP6gC4nFACFLMUzrkSAOmfOHOfdLapd2o/Nu47nFdBOdg4q4Ew4AnAOeNIfPNT0DVuMGTNmgx37tOVPPfA6Yy++TCSrD7AD6PSdwhcKPOdolghQV6xYYdttt53zoGLT+I1v5eE9oTFKASkQroAANVxDtSAFyr0CePCAHI5WYsk7thDvyRI2XrrYDVBcA8wAsSydV65cOVhHoA04Y7NOrCeWDVVsbKpVq5bblBVb8DICooQlxNYhlpZd8CzLA7aZKkW1y2YwNkoRgoBexRXaAqx9WAAhB4wF6CQGN92SrD5HTdEvwi/YJBavYbr30fVSQApIgeIUEKAWp5D+LgWkgBSQAlJACkgBKVCqCghQS1Vu3UwKSAEpIAWkgBSQAlKgOAUEqMUppL9LASkgBaSAFJACUkAKlKoCAtRSlVs3kwJSQApIASkgBaSAFChOAQFqcQrp71JACkgBKSAFpIAUkAKlqoAAtVTl1s2kgBSQAlJACkgBKSAFilNAgFqcQvp7XijAuY0cI6QiBaSAFJACUiBdBTi6TeV/CnC8HYUzp7NVBKjZUlbtRkqBW265xeV/j2rhnNAofwCqf2EzR/pJvzAFwmpr/kVHP9IykziEQgY6kpxwHjSJLyjXX3+9y1qXrJBQg+QcXbt2TXoNWebIREc66cMPPzxs8ElqkyKaVNIkCUlWOGOaDHWMh3HFltWrV9uIESNckhWvR3w7AtSsmE6NRk0BAWqYRfSAk35hCoTV1vyTfmEKhNXO5PyLBVSy1pFxzQMlvbzuuutc6uhkhUQZe+yxh02dOjXpNWTra9euncvod9RRR4UNPknt4cOHG4lCyMSXrJBdj6xyl19+uQ0cOLDQZT5NNq/z90RFgJoV06nRqCkgQA2zSCY/oMN6kri2+hemqvSTfmEKhNUuT/PPAyoeU5+mGA/juHHjXBY4D6iAKx7GlStXWuvWrV1a5tGjR7tUymSO69Onjw0YMMABLtfhzTz66KNdauTXX3/dAeppp51mb7/9tjPOoEGDrFOnTgWG+v77710KZ35IqXzkkUcaGfcAX+AT7+7kyZNdprq7777bHnroIZe+mn40btzYtf3XX3+5FNCffvqpnXvuuc6b2qNHDwfG3bp1s86dOztABZIZx6JFi1z6ZfpI+uSXX37ZZekbNmyY60d8EaCGva9UO0cUEKCGGao8PUDClBJAS79sKBDWpt6/0dHPAypL+xMmTHDL++3bt7edd97ZQSGACtzVqVPHQdxxxx1nvXr1cj8nnHCCAznSSQON1CFNNJB3/PHHu9eGDBli9evXd4DK6927d7ehQ4e6FNBr164tlM4ZbyygzB4N0kBTvvvuOwefpHnm9zZt2tgnn3ziPKX33Xef6y9po1u1amW//fabg1LA+IUXXrCTTjrJ5s2bZ1988YULqbv00ksdoFL69+9vTz31lGtr1qxZDpxpk/HccMMNttdeewlQk01T8oXzDaRq1aphMznDtXGR06fYHOEZvkXS5nDfb7TRRmWWd5s3U6VKlTIy3HUVts9IO2pECuSjAosWvKsY6ADDCgADxDOz8qSfB1S8lBdccIFNmjTJOnbs6Dyh559/vgNUwA5P5pw5cxwoPvzwwy5Oc/r06Ra7xH/77bc7TypAWK9ePQeJm2++uf38888OUEeNGmXnnHOOA148tL/88ovzvvpy4YUXOm/pI488YhdddJHzlgKhp5xyimv3kksusR133NFatmxphx12mL3xxhv21ltvuftceeWVDlBnz57tGAUQxsMLfLZo0aIQoHbp0sV5YMePH29nnHGGq8+m5Z122skt/efcEj/uZ+g/tkD1GG+XXXYp9PqTTz7pDPHiiy+W+F2Cofi2cswxxyRsg8lSo0YN23vvvVO+x5gxY9w3HkCPurjPK1SokHJ9LrzzzjutYsWKbiJnsjCBcc3jlk9WcOczoZhsmS68GfgWmGxi/vDDD06z9evXb3DrkthCgJppC6q9fFJAgBpmzfIEWGFKJa5dnvTzgHr//fc7bqG88847DkZZOodxWG7nueuX0olJhT3iARWIBFKXLFninqcstePUwUsZG4Pas2dPB8A892OdcIAi3k8AlE26AOry5cudd/OVV16xypUr20EHHeTgGE+qLyeffLIDTQAVYK1evbpdfPHFri8ffPCB7bvvvoUA1cegwhzALyEAeExzHlBxMVerVs3YDYb7eP/993fu6tjy008/GUCz++67l+i9g1H5BvPRRx8lBcgzzzzTGZGg5lQLRnv11Vddv+bOnWvNmjVLtWrBdXzj4ZsRfdt++8x5AflWxuSn/bIAVGJR92zuAAAgAElEQVRReCP5ZYX4PhQFqCWxhQA17amnCuVIAQFqmLHLE2CFKSVAjQXUzz77zAYPHuy8iUAbS/YAKs9FPJ8s+QOJxI7GelC32mortyzPvGvbtq37O2ECxHwCisBkKoD6+++/2xZbbOGMQhwrgIr3lsKKMn9nFz4hAwD1xIkT7d1337Wnn37anRLgl/i5H95Vluzfe+89B62xS/xFASrP81tvvdXdJ75ENgbVe1CBTwKJKbiicVMTAIxRiM0gyBaDEsxLoC0GJ0iXOAiAFhc1O94g/kaNGtnYsWOdwD4ugnb79u3rvHX8y1I/BsYryz0QfOHChc59zXlfuKm5FsHffPNNO+CAA9wE49tN7D3wxNJfvpXg2eVICO7LtSzbY0AmF8BL/zlyAS8pY8OTyzcdjMaExe0PpPJtKbYQV8lOOL6xEBfCWNm9h8udvuB9ZQLhHWby822Fycdkbt68ubuOSYx2HAFx2223ueUDdGBJAA8q42OcFIKj0dwXIJIjLPhWRwgC37Roh7ETyA2Y86YhAJwlBGJd+B2vMvdiQjI5mezEp7A0QRvYG88uOqM/mtN3ljkA/VhbYKNUigA1FZV0TXlVQIAaZnkBqvRLVYFYQOVZxzOS+FKeuzzveYbCBTiQ2DQFHLIKy7McFmFjFHGmxG4CjNTnFAAKHIJXFEjk2c7veGP/+c9/uvjUeA8qdXz8KHwC/xBbCjvh7aRMmTLF3RMPL4VnNyzSpEkTB7Eff/yxffjhh3bjjTe6+3FsFHXYeAWTwFr8zsaoxx9/3LUNjxAywFFVeHuBY/qYc4AKlED4ACtG4VsF8LLNNtu4gQI2CxYssHvuucfFbAA/BA2zhAygMXjE8GLigj7kkEMKxXRC/0wOwIqgZSAW0KMebbz22msuaBioA/aANIzIPYBiQA+gjb0H3lagEvgDUnG/s8Tv4Ze2mXxMDoD1gQcecG51xkgfcOljeL5Zce4Z7eONjS3cnwnKpODnqquucksGeJmBcbykTzzxhFs+IK6ESUQgNffjdyYSfeA6JjMgyk67Qw891O22Ywx8k6PuY4895iYsMBxbgFj6x7e8//u//zOWEvg/oMubBv3REE3R6fTTT3f24jU0YaIzgWmffwF53pjAsAdUwJ2Jjl15E8TaItUYVQFqqh+fuq48KiBADbO6AFX6hSmwYW3C2+AeYkDjQwMBQ3iiSpUqruKyZcscXLKxKt0wwlT7jTOMZ7S/p69HP3DGEdvK8x0nIpwGxxR1Viv1qcsqLm1usskmuQeofEsAQoBUvingXWQ3GoDq4y6ANwCVYGPiK/y3BI5A2HTTTd2OM+ATimcpP7YwCajDtwc8rT7Al280eAeBNcSLXVbm/nhuATvc7EwW6gGosffwS/wszfsYVACVvuNhpHANYM23B7yreCxxs3NPD6j8Ha8h4Q7xgMoYiS3h/ngT8TizsWnXXXd10AtkTps2zYEigMiSAf1nQvslfr6xMblvuukm1zy78gDTpk2bOlBmkvEaHtH4mFDc+MApOvMhjQeY/jIO7oH98DwDwxQAk6Bo+gKgcg/G7c9042/z588vANRvvvnGeU+J5wV4sbWW+FP9SNF1UiA1BQSoqemU7CoBqvQLUyC3a7MyyrPdF5xyONzgnpCSU0v8fqBAGJDnj0zwgIoLG6+lf92DHqCEpxKvHF6+2OIhltAAvIEUlpHxcNIusAf8xUKRz8zAUjmxsXgpPaDG3iMZoPpldA+ohABwzhkeUh+nSowJnk88qAAmYQvxGRvwoPq2AD6Cj/FOUjyg4v1liZwlAyYNHuZ4QGVswKffsAT4E/dLoLPfJAWMo208oAKkTE4AFe8v550RkoCXlm9VuPAB4dj4W2zAbj8AlSUOIJ8fyr333utiWbwH1d+Pb2V4tQWoIW931ZUCiRUQoIbNDAGq9AtTIPdr40yCnXDI7bfffi5kL7TkFaDihWNnGB7BDh06uKBelqdnzpzpvJuJABUBgTw8sHhZ+RaAq7lfv37OAwhEAVssMQNsZ599tvPmsrEKqAMSaT8ZoLL0jos+1oOaCFAJMcCoHKZLejICkL0HleVxdtQBmrElFUBlCR13Oz/AHbp4QAU4iW0hBhS45L7EwqIhsS54QYsDVDyvfvMWXxwIv8DbCXCiJ95RjsBAX3QkvII2+UFbYk4BViY3QI8XGH2KAlTiY7EF40+1aIk/VaV0XXlUQIAaZnUBqvQLU0C1EymQ04Dqz+mMPWaKZX48m2zI4bgEHyQMoBL3CBDFF6AK7ykwh5cSSGKJnfaJdyRAGND1GRLwrhIzCahyLRuA8PCxRB57D5bGCRbGEwr0+RhU4kQAXg/HwBxQzHI53kMCnXGP+7PN8GziMQaa4wHVt0VQNBujYj2oQCDgh0cT6GOMQC8xtAQsE9IAPOPBBSbff/99txzPsj4hB1xPSACxt8k8qPSHWGAKmhATy05CArr5BoU3lrhUIB57+KMogEvCCBgzcEwsL95sjhZDK16LPWYq1oMaa4tU07gJUPUBKAWSKyBADZsdAlTpF6aAaucUoIaYCzgiuJgdcSw9F1eIP/U7/wkwBkzxkLIpB2DzBe8iUMqyP95VluEBWY5iYBd7/L1YniYeluXy4gqQyLI8fSZsADgD2OgPwMZSPTBZkoLHFE8nAc6EPxD6QN8B5j/++KPgmAlidfGGZiMpAKBKu3xxiC2cBACU+9hgPMl4UpOdj5rIFqloIkBNRSVdU14VEKCGWV6AKv3CFFDtcgOoJTE1cZikHkv12KKS3KOoOngeOQKKpX2W8tkUxYkDvI4nFa9mPha+DADgLNnjYcWLSzytjwcuyZg59srnH/b1+15+a0maUh0pUC4UEKCGmVmAKv3CFFBtAWoRcwDvKD8smZdVYaMWQcZkymJZnQxSLK2zFB5/tENZ9TEb98XjDVRyGgJHZOFBzXSRBzXTiqq9fFJAgBpmTQGq9EtVgU2vTy+bZHy7a64pnF0RJw+rrbGMwOooK6WZ2KiU6riycV1kY1CzMVi1WX4VEKCWX9tr5MUrIEAtXqOirhCgSr9UFah+Yxig/nDVfwF13bp1LsMkm5IJO2TFlcKmajYZExIIvLIim8kslKmOMxPXCVAzoaLaiLwCAtTIm0gdLEMFBKhh4gtQpV+qCjS4OQxQP+//X0Bl5ZHQPzZIswkcQAVIOTfenwVPYiFOy+F89lwsAtRctJr6nLYCAtS0JVOFcqSAADXM2AJU6ZeqAk0H/g9QQc1YXE3l/+9fUXiJn8yNhAICqBwPyT4af5oPZ5GT9h0vay4WAWouWk19TlsBAWrakqlCOVJAgBpmbAGq9EtVgYMGVzDzJFqCf6dflhxQycLIMZs+qQ/HRJLoh8Q4uVgEqLloNfU5bQUEqGlLpgrlSAEBapixBajSL1UFDhsStsT/8qXJAdVnz2STFEdU+rjU3r17p9q9SF0nQI2UOdSZbCkgQM2Wsmo3HxQQoIZZUYAq/VJV4LjbwgB1yiXJAZU+kF6ctO+NGjWyI444wq677jrz6dlT7WNUrhOgRsUS6kdWFRCgZlVeNZ7jCghQwwwoQJV+qSrQ8faYGNT1ZhVieHV9Cv9/os+GgErM6dChQ10XnnnmGXemO4UMkQ8//LDzpuZiEaDmotXU57QVEKCmLZkqlCMFBKhhxhagSr9UFeh6RxgsPnRxYUBNdF8yRHKuOzv4c7kIUHPZeup7ygrccsstdtlll6V8fWlfqAdcmOLST/qFKRBWW/NP+qWqwFnDwgD1/guLB9RU+xL16wSoUbeQ+pcRBQSoYTLqASz9whQIq635J/3CFAirncn5d/6IMEAd2UuAGmZN1ZYCEVNAgBpmkEx+QIf1JHFt9S9MVekn/cIUCKtdnubfxXclj0EtOH7qbznjY1L5+x0C1LDJptpSIGoKCFDDLFKeHiBhSgmgpV82FAhrU+/f6Oh32cjYXVHpn9R/S095UMOsqdpSIGIKCFDDDKIHnPQLUyCstuaf9AtTIKx2JufftfeELfFfd54ANcyaqi0FIqbAlClTbK+99opYr9QdKSAFpIAUyAUF6tatm5Fu3nxvGKD2P1eAmhFDqBEpIAWkgBSQAlJACkiB/ypw26iYGNT0V/itbw8BquaSFJACUkAKSAEpIAWkQAYVGDE6zIPa62wBagbNoaakgBSQAlJACkgBKSAF7hsTBqjndBegahZJASkgBaSAFJACUkAKZFCB8feHAerpZwlQM2gONSUFpIAUkAJSQApIASkwcWwYoJ7STYCqWSQFpIAUkAJSQApIASmQQQUmjwsD1BPPFKBm0BxqSgpIASkgBaSAFJACUuD5B8IA9egzBKiaRVJACkgBKSAFpIAUkAIZVODV8WGAeujpAtQMmkNNSQEpIAWkgBSQAlJACkx/sIKBmGDq+vVmFSok+LeIv/+fAFWTSApIASkgBaSAFJACUiCTCsx+MMyD2uw0eVAzaQ+1JQWkgBSQAlJACkiBcq/A/AlhgLr3qQLUcj+JJIAUkAJSQApIASkgBTKpwOeBgNpAgJpJc6gtKSAFpIAUkAJSQApIgW8eIuj0v0GmPgb1f0GpPvg0+d93jgPU33//3TbbbDOrWLFi3olbYf16JFKRAlJACkgBKSAFpIAUyKYCSx+K2SRVmEeTcWqh12t1/S+yrVy50rp06WIbb7yxLVq0yC699FI788wzs9n1Um9bgFrqkuuGUkAKSAEpIAWkQHlUYBUe1ICyzd+AOmjQIPv111/tpptusmXLllnNmjUNb2qVKlUCWo9WVQFqtOyh3kgBKSAFpIAUkAJ5qsAfD4cBapUu//Wgnn322da2bVs75ZRTjIVwlvgXLFhgdevWzRvlBKh5Y0oNRApIASkgBaSAFIiyAutiAbXgQNS4tX4/gAR/3+hvD+pJJ51k/HTs2NFdXb16dXv33Xdt5513jvLw0+qbADUtuXSxFJACUkAKSAEpIAVKqEDgLn77e5PU9ddfb1tttZVdfPHFtm7dOqtWrZr9/PPPebVZSoBawjmmalJACkgBKSAFpIAUSEuBB8KW+O2M/y7xP/PMMzZixAibOnWqPf744zZ06FCbOXNmWl2J+sUC1KhbSP2TAlJACkgBKSAF8kOB+wMB9az/Auq//vUvO/roo+2TTz5xv7/88svWokWL/NDo71EIUPPKnBqMFJACUkAKSAEpEFkFRgUCao/CJ4N+9913VqNGDatUqVJkh1zSjglQS6qc6uWUAlOmTLG99torp/qszkqB0lFgvS3/aVbp3Ep3yZgCO1TLL29ZxoTJUkMZ2x1/Twyg+k1Qvs+p/P/88nN0vQA1S5NZzUZLgVtuucUuu+yyaHUqpjdff/11pI8HUf/Cpk609VtvN9+bf1lowiwW7drVt9vPuneYk3Inoz3/zMpV/0YEelB7CVBTnvi6UArkggIC1DArlasHSJhUCWtHWz8BahZMntUmBahZlXeDxjP6/h0WCKgXClBL1/q6W6kqMH/+fJcmLbYQw7LnnnuWaj9CbrZ48WKXRWOPPfZIqRkBakoyJb0oox/QYV3JQQCMuodIgJqFKZnVJgWoWZU3u4A6NBBQ+whQS9f6ulupKsDBvv/5z3+sUaNGBffdd9997cQTTwzqx4QJE2zt2rXWrVu3oHZ85YMOOsimTZtmG2200QbtjRs3zu1eHDx4cEr3EqCmJJMANUymHNVPgJols2etWQFq1qTN/hfgIYExqP0EqKVrfd2tVBUAUEmP5jNQ+JsDmJyt9uijj7ojKzhjbfLkyTZ27Fi7/fbbbeONN7bevXvbmWeeaUuXLnUgOnfuXDv22GOtX79+1qpVK3fcxZNPPulSsFG+/fZbO++886xq1ao2a9Ysu/DCC13bs2fPdjmESdf29ttv2xVXXGELFy60Tp06ufPc+vbta7fddpu1a9fOnfHWp08fe+CBB9w9hg8f7sD1wQcftD/++MPwpt5666128sknJ9VRgBo2xeRBzWf9BKhh1i392gLU0tU8o59/AwM9qFcIUEvX+rpbqSoAmP7yyy/WoEGDgvuee+65tttuu7md7v3797ebb77Zhg0bZgcccIDttNNODijJ9dukSRNbsWKFDRo0yMEh4Ni9e3fr0aOHzZs3z9asWWMDBgwoOPLiq6++snr16jnIrVy5snXu3NlGjRpl//jHP1wdjsho3ry5A1ygtnXr1nbfffe5fmyxxRb2448/unZpc9KkSTZy5Ej7/fffXTgCgExqt88//9z15+OPPxagZmkmZfQDOgt9VP9CRBWghqhXFnUFqKWrekY/X24MBNSrBKila33drVQVAFC33XZba9asWcF9Dz/8cKtTp4699tprduihh7ocv3hS169fb0AmXs4PP/zQeTd/+uknd83DDz/sIHf58uVuaX/ixIn2559/2pVXXlnQLnUbN25sq1atcrBJrCvX8lOlShXXPvGwr7/+ugNNluwfeeQRO+aYY6xChQruuquuusrq169vZ511loPiRYsWOTB96qmn3M/q1att6623dm1Rpk+f7vobX7SLv+TTLKMf0CXvRtKa6l+IqALUEPXKoq4AtXRVz+jny4BAQB0gQC1d6+tupapAsiV+OjFjxgw78MAD3TI+Xk/gcfvtt3dHNDVt2tQtwQObeFU/++wzq1WrloPGf//7387zmQhQO3ToYB988IHzvDZs2NB++OEHFwoAoBILS7t4b4866ii7//773b1iAbVLly7uvvxwH/INP/vsswUxqHhU8bZ6QE0kppb4w6ZYRj+gw7qSsLb6FyKqADVEvbKoK0AtXdUz+vlyTRExqPHDij8Xlb9fL0AtXevrbqWqAIAK7AGOvuCtXLdune23337Wq1cvB4lPP/202ylP/CkeTuJTiQkFUIklxYsKyHbt2tXat29v33zzjf3222927bXXFvKgFgWoy5Ytc15VIJd2WfoHPj2gArIPPfSQA2dCAwBYPLl4Zf0mKQFq9qdPRj+gs9Bd9S9EVAFqiHplUVeAWrqqZ/Tz5YpAD+pAAWrpWl93K1UFWL5n41FsYQc/saBsYHrllVfc8v0111xjM2fOtP3339+FBGyyySbO8wmksjGKzU9sluKop+eff97eeOMNB6psrAJeKSzxFwWoeD0B5k8//dRdT7wqoPrmm28au/jxjLJ5q2fPnq4veF2fe+45t4wvQC29aZPRD+gsdFv9CxFVgBqiXlnUFaCWruoZ/XzpFwiogwWopWt93S3SCrAMj6eT5XyW2FnGBxwpLLcT/+kLAAvIJjoaqqhBAro77LCDq0eMa7Vq1dy9/vrrLwelie6Vjmha4k9HrQ2vzegHdFhXEtZW/0JEFaCGqFcWdQWopat6Rj9fegcC6u0C1NK1vu4mBbKsgAA1TOCMfkCHdUWAmnH9BKgZlzTLDQpQsyxwXPMZ/fy7MBBQhwlQS9f6upsUyLICAtQwgTP6AR3WFQFqxvUToGZc0iw3KEDNssDZBNTzAwF1pAC1dK2vu0mBLCsgQA0TWICaz/qtt3fee8Rq1qoZNsgs1l66ZKn6F6fvTrXapKy43r8pS5X9L8A9AgF1lAA1zJqqLQUipoAANcwgesDls37r7d33J7oY86iWJUuWqH9/G6d2zdZpm0nv37QlK1Qho/p1CwTUsQLUMGuqthSImAIC1DCDZPQDOqwr2fdwlLv+rbdbRlXMwqjVZKYVaHvAndak4YVpN6v3b9qSZQ9QTw88B3W8ADXMmqotBSKmgAA1zCB6wOWzfgLUMOuWXm0BaulpHXunjH7+nRLoQZ0oQC2bWaC7BivAgfccZN+iRYvgttJpgKOheBM3adIknWpFXsvxVhxbtfnmmye9jkQCHHlFooGiigA1zCwZ/YAO64o8qBnXT4CacUmz1KAANUvCFtNsRj//OgUC6uOpASrHQXJsY6VKlQpGl8oztWwUTnzXCuuLyg8ZpZ6qLykpQBpRUo6Sv740yzvvvGODBg2yp556KiO3Jc3qHXfcYTvuuKM7D5VsUqRc9eXzzz+366+/3ipWrGiLFi2yHj162Kmnnpr03gLUMLNk9AM6rCsC1IzrJ0DNuKRZalCAmiVhSxNQ2wUC6tNFAyqZFXFS9evXz3r37m0k4aEU90wtG2WLvqsANYpWCejTd999Z/3797cHH3zQRo8e7dKUbrzxxm6ikpZ07dq11qdPH3vggQesVatWNnz4cJdeNP61OnXquNeefPJJ22WXXWzIkCHOK3v00Udby5Ytbf78+TZixAg766yzDFjkNd4YsYBK27Vr13ZZp+hDqgUg5VsfSQCqVq3q0qrWrFnTrrjiioImANL69etb37597csvv3S/r1mzxiUJSFQEqKmqn/g6AWo+6ydADbNu6dUWoJae1rF3yujn3zGBgPpc0YD6wQcfuOc7qcpvvfVWB6ipPFPLRlkBahR1z1qffGpR0o7utNNONnv2bOdlZOl9xYoV9vbbb9uAAQNs0qRJNnLkSAeVRx555AavtWnTxgYPHmyPPfaYvfTSSzZx4kR79tlnHXDus88+NnDgQPeNjKV9QPbiiy+2LbfcshCgvvrqq3bdddfZwoULCwB5m222salTp9rcuXMLaUD/2rZt617jen5fsGCB+z+gO2/ePBszZkxBHbynQCzQiveWdKzff/990p2+AtSwKZfRD+iwriSsrf6FiCpADVGvNOsKUEtT7f/dK6OfL4cFAurLqS3xk0K8a9euDlBTeaaWjbIC1CjqnrU+eUAF6PgdIMXdP3ToUAeTLMPjbcTzSbwqy+N824p/rV69evb+++/bnDlz7PXXX3ewyP8B1BdeeMEaNmxoDRo0sCeeeML9DsDyk2iJn/uPGjXK3ee0006zQw45xD7++ONCGjRr1syOOuoo9xre2U6dOjnPLAVv8LRp05xH2JdPPvnEeW3bt29vM2fOtC+++MJ8POr06dPduOPLZZddljXdQxvO6AdgaGcS1Ff/wkSNtn4C1DDrll5tAWrpaR17p4y+f1sHAuq0/wIqz8lvvvmmkCA777yz7b333u61WEBN5ZlaNsoKUKOoe9b65AEV7yUxm0BZ06ZNHfABqCyN8zs/uP1ZRu/Zs+cGr+GV7N69u/OMbrXVVnb//fcXACpe2R122MG23nprB4Y1atSw5557zsW+JgJUwpzfeustu/HGG523880339xgyR/Pqj+HkY1RVapUMQK62fxEmAKFMIXYsnr1aufV3XXXXV3/CW9IVuRBDZtyGf2ADutKwtrqX4ioAtQQ9UqzrgC1NNX+370y+vnSMhBQ3/kvoPJMZnUztuDkIZQvHlBTfaaWjbrJ76oY1KhZJLA/HlBvuOEGB3Z4P5955hlr166drVq1ysWUzpgxw3k0meB4N1myj38NMNxuu+3s6quvtksuucQIGfAeVAAVKMUbevDBBztvbK9evdwSeyyg4tF85JFH7M4773Q78S+66CLr0KGD3XbbbTZ+/PhCI+VNFRtjSp/uvvtua9SokR1xxBEuVODwww+3d999171GeAEfGrTF0j8e1vg2Y28gQA2bWBn9gA7rigA14/oJUDMuaZYaFKBmSdhims3o51/jIs5BhT1j+TXR/+emv8TP8JI9U8tG0dTuWiqAiqeLZVeWcVeuXGl77rlnar3TVWkr4AEVL2Xjxo1t2223dRuH+AYFpOIt5YdlcbyUeD7xhMa/tnz5che7Qhwrm6RY1iduFa/qe++9Z9WrV3fe0DPOOMOFCrDRio1MALAvZ599tuE9Pf/8850XN50CVPtd+cccc4w9/PDDzpuKNxdIpe9s2OL/ixcvdq8VlQlHgJqO+htem9EP6LCuCFAzrp8ANeOSZqlBAWqWhC1NQK0X6EH9MnVA5Rl6wgknuNEle6aWjaKp3bVUAJUl2HPPPdcOO+wwF3tY1HFAvtsTJkxwO867deuW2kh0lVOA2Ew8m3g7WSJftmyZAzeW8zkXjTNDKSztA6axJf41wJM6LL/zOzvx43fJ0y4wmwgO+Vs6u/fjTcg9+XID+CYrLOsDxzoHNbtvAAFqmL7R1k+AGmbd0qstQC09rWPvlNH37w6BgLo8NUBNpFQqz9SyUTjxXbMOqNdee607r5KNMZzNyeYVloQ5OggvGMu1LN1eeumlxvEIACmetwMPPNB5/fDI+d3dsUNgKZk4S7x5eNDYcd65c2fXPkvF7FojLpHNQVw7ZcoUt+mGZWfuPWzYMAdfgDCeXY4qIj4TDyQ734iXDIGrsjIyXwZOP/10Q3eV/ykgD2rYbMjoB3RYVxLWVv9CRBWghqhXmnUFqKWp9v/uldHPl00DAXVNyQG1bNQr+V2zDqgcY4QHDDhkSZglZeIa2W12+eWXO1gkNpHl5IMOOsjFKHIoO7GLnGvJkUixmRD8UMeNG+fqsvkGSAVE8c42b97cHVAL1LZu3dpt3MGrCPjOmjXLASlwyu/0ZcmSJS4ekx1vHFVEvXPOOcfFVAJ6uVZ++eUXt+ytUlgBAWrYjMjoB3RYVwSoGddPgJpxSbPUoAA1S8IW02zUP//KRpXs3zXrgMoQACYyHL344osFgIrnlF3lpOLiYHlgkVjF4447zmUPYvMLHs4rr7wyoQoAKnGHnOWJV5R7EO9IjCsbgwBivKqALud/En+BN5YfNgex+xsPLPfhqCS8ubTFEjav0x7/V8kPBQSoYXaM+ge0+hdi3/X27vsTC4fSeCdPcc6aZM6g+HqBTqOlS5ZazVrJQ31CRp+JuqXZvzo1D067y3p/pC1ZoQpR1y9sdNGtXWaAyrK+P+cSTykbcIBIDoYnrvDRRx8tFlDxjAKhxFWwS5yYSzbj7Lbbbu5MTUCUY5YAVH8ty/38ALgeUPGyEseIF9aX3XffveBczuiaTz1LVQEBaqpKJb4u6h/Q6l+IfdfbZ3MnWs1atUIayWrdpUuWpN2/qtVbZ7VPsY1r/oVJLf3C9MvX2pEA1EMPPdQdxsDvg6wAACAASURBVM5GG5baTz75ZPv222/tt99+SxpLCWDGAyobgjj+CGDlSCWgE09pcYD6/PPPuyMYaHPfffe1a665xvUFj65KfiggQA2zox4g+azfent7fMWwAUas9q4tR1qN+ueVWq/0/giTWvqF6ZevtSMBqPfee6/bJAUcsqzPEj3HIJElaPLkyQbAxpdEgMoSP4D76aefusvJhgSosikrmQeVs0KBWPLWkrKTUrlyZXeYPEc0qeSHAgLUMDvqAZLP+glQw6xr7kzmunXrhjaTtfrqX5i0UdcvbHTRrV0qgJrK8H/88Ue3KSr2uCJ28RMGwM76+MLZmJtttlnCppcuXeoyHRHfSpxrtWrVUumCO4oJzy0buMhfn4sF7zGH77do0aJUu4/OvImbNGmSsfsSssEcIHwjtAhQwxSM+ge0+hdiXwFqiHrU1fwLU1D6hemXr7UjA6jJBCZXPMAZXziEPv5Mznw1UjrjYjMaMbUc6VWahRMaBg0alDDVaUn6QaaoO+64w22Y44sDJzuQutUXvszw/yOPPNK91KBBAyN7VrIiQC2JFf5XRw+QfNZPgBpmXQGq9AtVQPUTKRB5QJXZ0lOADWacikBM7+jRo126U85zJY896URJfkAowwMPPGCtWrWy4cOHu1jd+Nfq1KnjXuPUAzJJDRkyxHllyd7UsmVLmz9/vo0YMcKFT+Dl5jWOFItNdUrbtWvXtmOPPTatM2UBUo4WI3FA1apV3XFiHFUWmwr1s88+c7HCZJhK5bxaAWp68yj+agFqPusnQA2zrgBV+oUqoPoC1HIwB3yq0zfeeMOlKZ09e7YLV2Dpnc1iJDLgbFlOTeAYLaASL2T8a23atHEnJHCqwksvvWQTJ050sboAJxvKBg4caHg5WdoHZC+++GLbcsstCwEqpyRcd911LmmCB2SyUk2dOtXmzp1byBr0zydk4Hp+X7BggbsG0J03b56NGTOmoA596dKliztijEQLACh9lgc1O5NcgBqma7T1E6CGWVeAKv1CFVB9AWo5mAMeUAE6fgdIiUnlCC1gkmX4+vXrO88n8aqEUOBNjX+NDWakS50zZ47btAYs8n8A9YUXXrCGDRu6ZXXOkOV3AJafWA+ql5v7kwyB+5CGlaxiJFWILUAmR4NR8M6SBcwfQ4Y3eNq0ac4j7AvwS99IqMCRZAAzXlVSnk6fPt2NO75w5FhUS7QBRg/g0HkTbfsKUPPbvnr/5rt9Q8cX1fpa4o+qZUrYLw+oABwxmkAZZ8MCfAAq6Vz5nR+W0llG79mz5wavEVPavXt35xklaQFnynpAxSvLJrStt97avvjiC3e0F4kWiH1NBKicrkDGL9LH0i4nJMQvy+NZ9Rvk2BhF4gQ2SQGchClQ8ML68tdff7lNcPysW7fOtUd4A+EKiYqW+Es4of6uFm3A0gM4zLoC1DD9NP+kX6gCqp9IAQFqns0LD6hsGALs8H6SAKFdu3buyC1iSmfMmOE8mkAn3k2W7ONfAwy32247l5b2kksuMUIGYgEVKMUbevDBBztvLJ7M77//vhCgsvxOJq8777zT7cQnpS2pbMneNX78+ELKEx8bG2NKn+6++24j49gRRxzhQgVI7kD2MF4DOAlZ4Br6Tl98SIAANfOTWoAapmm09ROghllXgCr9QhVQfQFqOZgDHlDxUnLSAWe5ctoBXkkgFW8pP5wzi5cSzyee0PjXli9fbieeeKKLY2WTFMv6xK3iVX3vvfesevXqzht6xhlnuFABPJdsZAKAfTn77LNd+lkSHuDFTacA1aeeeqqrwpFibIYCmvHmAqn0mThVPKf8ALBsxkpW5EFNR/0Nr402YAkQwqwrQA3TT/NP+oUqoPoC1HIwB0hIgDcRbydL5GTXYumc5XySIGyxxRZOBZb2gbzYEv8a4Ekdlt/5HRCMP9qLdoHZ2PNrfZv8LZUd9snMwj1Xr15dOEd43MU+e1hxphWgFqdQ0X8XoOazfgLUMOsKUKVfqAKqL0AtB3Ng1113tdNPPz1pithyIEHCIQpQwywvQM1n/QSoYdYVoEq/UAVUX4BaDubAL7/84pbBVQorIEANmxEC1HzWT4AaZl0BqvQLVUD1BaiaA+VWAQFqmOkFqPms33r7bO7EIkNpwkYfXptsgsS4p1Oq1jg4ncuDrtX7I0g+pYoNky9va2sXf96aVgOLVUCAGjYf9ABOU7/F0wpVWLJ0qdVKE7DSvGPQ5V//Vdvq1q0b1EY2K2v+hakr/fJbv7DRRbe2ADW6tlHPMqiAADVMTD3g0tTv6WPNFj6XZqWyuryCfX38VwLUAPn1/ggQzxQiEaZe/tYWoOavbTWyGAUEqGHTQQ/gNPUToKYpWNGXa/6FySn98lu/sNFFt7YANQu24exPDsjnsHnOIaUQQ8UGJs4VJQ1py5Yt07oz7R100EFWqVKltOpF7WKOjiKVabrjDx2HADVMQT3g0tRPgJqmYALUjAoW15jev2HqRl2/sNFFt7YANQu2IQ88MHnttdfagAED3B0eeughd8A8WZnIiOTzzBd1+wkTJtjatWutW7du7mB8cs1Xq1YtCz0uvSY5M5WUqFdeeWXp3dTMZZ4i7WtUS9Q/ANW/NGeOADVNwQSoGRVMgJpROaP++ZfRwUaoMQFqFoxx7rnnOk/pyJEjbdGiRVaxYsViAXXs2LF2zTXXuN4Ata1bt7YDDzzQZYAiO1PXrl2tffv27vdmzZq5FKJbbrmlywTVr18/dxA/aUnJvnT99dc7sCX7EiDsD8vHiwvszp0712Vduvfee+3HH3902aHIPAVUk/6UPPfnnXeeVa1a1WbNmmUXXnihvfzyyzZ79my76aabjAxRffr0sXXr1tmzzz7rdtc++OCDLuPU6NGjXYpV7tm7d28jhWlsf1555RXr37+/6//EiRNt4MCB9ttvv9mgQYOsU6dO7n7nnHOOLVy40GWpoi3u8eqrr7oMVosXL7bBgwdb586dE9ZPZk4BathEj/oHdOT6J0ANm3ACLOmXUQXCGovc50vYcHKmtgA1w6b6/fff3WaDb775xgEfeej5tygPKoDGAfuPPvqoUZ/c9qQsBc7WrFnjgJVUoieccIL17dvXHcSPNxDI3HHHHR2okfHp5JNPtjlz5jiAA+gAZLy1vpAOlCV24BIoBULJOPXdd985aMTju8cee9hJJ51k9erVM6C5cuXKDgZHjRrl+tCjRw93vU8rOmbMGBsxYoTr5xVXXOHAHJAFyps0aWIrVqywCy64oKA/aEPKVZ8u9cMPP7QffvjBgfPHH3/stAKyDzjgAPca+qEFkPzWW2+5ekOHDk1aX4Ca4Qn9d3NR/4COXP8EqBmdiJGzrwBa9s2oAmoskQIC1AzPC0AUMGQJe/z48S6dKP8vClDxggKZeCEpACjexC+++MKBJ22xxP/CCy9Y48aN7fLLL3ewuN9++7lr/XL5XXfdZZdeeqm99tpr7joANLYAjHhVGzRo4NKT4mU9/vjjC17D2wqcPvHEE67+qlWrnIe1Ro0a7lp+qlSpYsTYAqh4UoHm77//3lq1auVAkh9CHABPQPKnn35yHl7fH/4OoH700UfWokUL1zapWY888kjXPoAMsFLw5pK6tWHDhi48Ah1//fVXl4iANK7x9fEoU6ZPn+76EF+0xF/yyS5ASFM7AWqaghV9ueZfmJzSL7/1CxtddGsLUDNsm0MPPdQ222wztzwORAGfABdL88liUO+55x4HowAdheX8jh07usOLYwHVx6CylL/DDju4H8ICYkGU++PRPOKII6xDhw4FowMq6RNt1KpVy3lSAcLatWvbggULbPvtt3d/o87TTz/t/v3ggw+cBxRAZAyEG8QCKvG0bdq0ccvxbAijv7QDCDZt2tRBtgdU3x8PqHhLgWS8xo8//rhrn3/x1vI3CuEC/E7bgCqeYfq9+eabO0iOr//pp586z22ioiX+sImuB1ya+glQ0xRMgJpRweIa0/s3TN2o6xc2uujWFqBm0DYs6++9997O8+jjPlmyZkkduEsGqF9++aXzRPJ3PITNmzd3O/3HjRvn4jNZeo/dJOUBtUuXLrbvvvsaYLbRRhs5qLzhhhtcbGk8oDJMPJUALHGhPqYVGG3btq3zxBIPClCyjJ8KoBILS9/w3AK3hx12mAtL4MSBZ555xnlK0QJgjQdUvLz05/nnn7cKFSq4zV+MA08r3k88xHhz6cvKlSs3ANRly5ZtUN/DtwA1g5P676ai/gEduf4JUDM6CSNnXwGg7JtRBdRYIgUEqBmcFwAem3gANl/wirLkDPCxAQhYTbSLnw1FLL8DsgDksGHDjA1FbIyaPHmyAaOxHlSAtVevXm4DErviV69ebccdd5yx859NWiyZUze2EJ+Kt5XNUiylA4e8xqYkQJLNUYDjpptumhKgApSA+M8//2zTpk1zm6UATI7W2mSTTdxYgFQ08f2J9aACqFOnTnVxq4QrANaERRCyQF+IV2Vz1qRJkxJ6UBPVT2ZOeVDDJroAIU39BKhpCiYPakYFE0BnVM6of/5ldLARakyAGiFj4L0kLMCfnUrXgDxgDw9pssL5qnhaWbpPpQCUxMbGFk4bYMOV9/wW1w4xqMTCsrmLUAPfP/qPd5O+cLIAIQpbbLFF0ua4L6EHsf0BTlnyx4uKd7Wokqh+ousFqMVZVIAQplBcbQFqRuWMOiCof2Hmln5h+uVrbQFqvlo2y+PygEoIQy4UAWqYlfQASVM/AWqagukLUkYFi2tM798wdaOuX9jooltbgBpd20S6Z8TKshksVY9rWQ9GgBpmgah/QEeufwLUsAknwJJ+GVUgrLHIfb6EDSdnagtQc8ZU6miIAgLUEPXMndBATHBUS+T6t/iNQlItXbLEaqYYglMWGn/9Vx3ZN0D4yM0/AX6ANTesGnX7ZnSwEWpMgBohY6gr2VNgypQpttdee2XvBmpZCkgBKSAF8laBKH9Bz1fRBaj5almNSwpIASkgBaSAFJACOaqAADVHDaduSwEpIAWkgBSQAlIgXxUQoOarZTUuKSAFpIAUkAJSQArkqAIC1Bw1nLotBaSAFJACUkAKSIF8VUCAmq+W1bikgBSQAlJACkgBKZCjCghQc9Rw6rYUkAJSQApIASkgBfJVAQFqvlpW45ICUkAKSAEpIAWkQI4qIEDNUcOp21JACkgBKSAFpIAUyFcFBKj5almNSwpIASkgBaSAFJACOaqAADVHDaduSwEpIAWkgBSQAlIgXxUQoOarZTUuKSAFpIAUkAJSQArkqAIC1Bw1nLotBaSAFJACUkAKSIF8VUCAmq+W1bikgBSQAlJACkgBKZCjCghQc9Rw6rYUkAJSQApIASkgBfJVAQFqvlpW45ICUkAKSAEpIAWkQI4qIEDNUcOp21JACkgBKSAFpIAUyFcFBKj5almNSwpIASkgBaSAFJACOaqAADVHDaduSwEpIAWkgBSQAlIgXxUQoOarZTUuKSAFpIAUkAJSQArkqAIC1Bw1nLotBaSAFJACUkAKSIF8VUCAmq+W1bikgBSQAlJACkgBKZCjCghQc9Rw6nZ6CkyZMsX22muv9CrpailQDhRYb+tt9g+zy8FINUQpUDIFmldvbnXr1i1ZZdUqsQIC1BJLp4q5pMAtt9xil112WWS7/PXXX0f6A1D9C5s6UdYPQK04YOOwAaq2FMhjBRac/mWkP5/zVXoBar5aVuMqpIAANWxCRBmwGJn6V3L7ClBLrp1qlg8FBKhlY2cBatnoXqZ3nT9/vq1cubJQH2rUqGF77rlnmfYrnZsvXrzYfv31V9tjjz1SqiZATUmmpBcJAPNXPwFqmG1VO/8VEKCWjY0FqGWje5netWPHjvaf//zHGjVqVNCPfffd10488cSgfk2YMMHWrl1r3bp1C2rHVz7ooINs2rRpttFGG23Q3rhx4+yTTz6xwYMHp3QvAWpKMglQw2TKSf0EqFkyuprNGwUEqGVjSgFq2ehepncFUE855RTj39gCYD7zzDP26KOP2ssvv2wjRoywyZMn29ixY+3222+3jTfe2Hr37m1nnnmmLV261IHo3Llz7dhjj7V+/fpZq1at7F//+pc9+eST1rZtW9f0t99+a+edd55VrVrVZs2aZRdeeKFre/bs2XbTTTfZ2WefbW+//bZdccUVtnDhQuvUqZMNHTrU+vbta7fddpu1a9fOHn/8cevTp4898MAD7h7Dhw934Prggw/aH3/8YXhTb731Vjv55JOT6ipADZty8qDmr34C1DDbqnb+KyBALRsbC1DLRvcyvStg+ssvv1iDBg0K+nHuuefabrvt5na69+/f326++WYbNmyYHXDAAbbTTjs5oKxYsaI1adLEVqxYYYMGDXJwCDh2797devToYfPmzbM1a9bYgAEDrFKlSq7tr776yurVq+cgt3Llyta5c2cbNWqU/eMf/3B1vvvuO2vevLkDXKC2devWdt9997l+bLHFFvbjjz+6dmlz0qRJNnLkSPv9999dOAKA/O6779rnn3/u+vPxxx8LULM0swSoYcJGWT8BaphtVTv/FRCglo2NBahlo3uZ3hVA3Xbbba1Zs2YF/Tj88MOtTp069tprr9mhhx5qJ510kvOkrl+/3kEmXs4PP/zQeTd/+uknd83DDz/sIHf58uVuaX/ixIn2559/2pVXXlnQLnUbN25sq1atcrBJrCvX8lOlShXXPvGwr7/+ugNNluwfeeQRO+aYY6xChQruuquuusrq169vZ511loPiRYsWOTB96qmn3M/q1att6623dm1Rpk+f7vobX7SLv+TTLsqAxajUv5LbVoBacu1Us3woIEAtGzsLUMtG9zK9a7Ilfjo1Y8YMO/DAA90yPl5P4HH77bd3RzQ1bdrULcEDm3hVP/vsM6tVq5aDxn//+9/O85kIUDt06GAffPCB87w2bNjQfvjhBxcKAKASC0u7eG+POuoou//++929YgG1S5cu7r78cJ+ff/7Znn322YIYVDyqeFs9oCYSV0v8YVNOAJi/+glQw2yr2vmvgAC1bGwsQC0b3cv0rgAqsAc4+oK3ct26dbbffvtZr169HCQ+/fTTbqc88ad4OIlPJSYUQCWWFC8qINu1a1dr3769ffPNN/bbb7/ZtddeW8iDWhSgLlu2zHlVgVzaZekf+PSACsg+9NBDDpwJDQBg8eTilfWbpASo2Z9OAtQwjaOsnwA1zLaqnf8KCFDLxsYC1LLRvUzvyvI9G49iCzv4iQVlA9Mrr7zilu+vueYamzlzpu2///4uJGCTTTZxnk8glY1RbH5isxRHPT3//PP2xhtvOFBlYxXwSmGJvyhAxesJMH/66afueuJVAdU333zT2MWPZ5TNWz179nR9wev63HPPuWV8AWrpTaMoAxYqqH8lnwsC1JJrp5rlQwEBatnYWYBaNrrn1F1ZhsfTyXI+S+ws4wOOFJbbif/0BYAFZBMdDVXUoAHdHXbYwdUjxrVatWruXn/99ZeD0kT3SkdELfGno9aG1woA81c/AWqYbVU7/xUQoJaNjQWoZaO77lrKCghQwwQXoOavfgLUMNuqdv4rIEAtGxsLUMtGd921lBUQoIYJLkDNX/0EqGG2Ve38V0CAWjY2FqCWje66aykrIEANE1yAmr/6AagTZ0y0mrVqhQ0yi7WXLlmi/gXoK/0CxDOzOv+pbXXr1g1rRLXTVkCAmrZkqpCLCghQw6wmQM1f/QDUl2Y9arVq1gwbZBZrL1m6VP0L0DdX9dtl+0a2ZeVtAkaemapR//zLzCij14oANXo2UY+yoIAANUzUqH9Aq38lty+AetiQiiVvQDWlQJYUmHzBjwLULGmbC80KUHPBSupjsAIC1DAJBYD5q58ANcy2qp09BQSo2dM2F1oWoOaCldLoIwfec5B9ixYt0qgVfilHQwExTZo0CW/s7xY43opjqzbffPOkbZJIgCOvSDRQVBGghplFgJq/+glQw2yr2tlTQICaPW1zoWUBai5YKY0+kkaUlKPkry/N8s4779igQYPsqaeeyshtSbN6xx132I477ujOQyWbFClXffn888/t+uuvt4oVK9qiRYusR48eduqppya9twA1zCwC1PzVT4AaZlvVzp4CAtTsaZsLLQtQc8FKafTxu+++s/79+9uDDz5oo0ePdmlKN954Y+vdu7dLS7p27Vrr06ePPfDAA9aqVSsbPny4Sy8a/1qdOnXca08++aTtsssuNmTIEOeVPfroo61ly5Y2f/58GzFihJ111lkGLPIaKUdjAZW2a9eu7bJO0YdUC0BaqVIllwSgatWqLq1qzZo17YorrihoAiCtX7++9e3b17788kv3+5o1a1ySgERFgJqq+omvE6Dmr34C1DDbqnb2FBCgZk/bXGhZgJoLVkqjjz61KGlHd9ppJ5s9e7bzMrL0vmLFCnv77bdtwIABNmnSJBs5cqSDyiOPPHKD19q0aWODBw+2xx57zF566SWbOHGiPfvssw4499lnHxs4cKDh5WRpH5C9+OKLbcsttywEqK+++qpdd911tnDhwgJA3mabbWzq1Kk2d+7cQqOif23btnWvcT2/L1iwwP0f0J03b56NGTOmoA7eUyAWaMV7SzrW77//3mW7EqCmMWFSvFSAmqJQSS6Lsn4C1DDbqnb2FBCgZk/bXGhZgJoLVkqjjx5QATp+B0iJSR06dKiDSZbh8Tbi+SReleVxvKnxr9WrV8/ef/99mzNnjr3++usOFvk/gPrCCy9Yw4YNrUGDBvbEE0+43wFYfhIt8XP/UaNGufucdtppdsghh9jHH39caFTNmjWzo446yr2Gd7ZTp07OM0vBGzxt2jTnEfblk08+cV7b9u3b28yZM+2LL74wH486ffp0N+74ctlll6WhZOleGmWAQQn1L2w+RFk/AWqYbVU7ewoIULOnbS60LEDNBSul0UcPqHgvidkEypo2beqAD0BlaZzf+WEpnWX0nj17bvAaXsnu3bs7z+hWW21l999/fwGg4pXdYYcdbOutt3ZgWKNGDXvuuedc7GsiQF2/fr299dZbduONNzpv55tvvrnBkj+eVe/9ZGNUlSpVjE1SbH4iTIFCmEJsWb16tfPq7rrrrq7/hDckK1riT2MSJbg0yoAlgA6zrQA1TD/Vzp4CAtTsaZsLLQtQc8FKafTRA+oNN9zgwA7v5zPPPGPt2rWzVatWuZjSGTNmOI8m0Il3kyX7+NcAw+22286uvvpqu+SSS4yQAe9BBVCBUryhBx98sPPG9urVyy2xxwIqHs1HHnnE7rzzTrcT/6KLLrIOHTrYbbfdZuPHjy80KuJjY2NM6dPdd99tjRo1siOOOMKFChx++OH27rvvutcILwCaaIulfzys8W3G3kCAmsYkEqCGiZVj+glQM25uNZghBQSoGRIyR5sRoOao4ZJ12wMqXsrGjRvbtttu6zYO4ZUEUvGW8sOyOF5KPJ94QuNfW758uZ144okujpVNUizrE7eKV/W9996z6tWrO2/oGWec4UIF2GjFRiYA2Jezzz7b8J6ef/75zoubTgGq/a78Y445xh5++GHnTcWbC6TSdzZs8f/Fixe715LFn3JfAWo66m94rTyo+aufADXMtqqdPQUEqNnTNhdaFqDmgpXS6COxmXg28XayRL5s2TIHbizn//nnn+7MUApL+4BpbIl/DfCkDsvv/M5O/Phd8rQLzCaCQ/6Wzu79+GFyT5bxAd9khWV94FjnoKYxSUpwqQC1BKLFVImyfgLUMNuqdvYUEKBmT9tcaLnMAZW4SJZ+2SnOTuxx48bZK6+84jbvsLTMsnSXLl3sxRdfdOdeckwS3jR2mLPsjDcPDxr/79y5s9scw1IxO8GJS2RzEMvOU6ZMcZtuWHbm2KJhw4Y5+JowYYKxQYejiojPxAPZtWtXFy8ZAldlZXziMU8//XS79tpry6oLkbyvPKhhZokyYDEy9a/k9hWgllw71cyuAgLU7Oob9dbLHFA56og4yJtuuslt6GndurXz1gGMQCcHz7M8jKcMgGTzD3WWLFniQJPNN0AqIMrO8ObNm1u/fv3cMUW0xcYdvIrdunWzWbNmOSAFTvmd5W3aIR6zY8eODpCpd84557iYSkAv18ovv/zilr1VCisgQA2bEQLA/NVPgBpmW9XOngIC1Oxpmwstlzmg4hk96aSTHFQef/zxtvfeezuoTAaoxFWed955ztNK3CGwilcUKCPeceXKlW5jEN5SAJdNOpz/SUwj8ZH8sDmI3d/ALptsOCqJmEbaYgmb12mP/6vkhwIC1DA7ClDzVz8A9aV3J1rNWslDacJGH1576ZKl6l+AjLmqX93t97YtK28TMPLMVI36519mRhm9VsocUJEEIGQZnnMygUeW/T2gApp4Nb0HlR3d7AQHUPGMAqF4XNklTswl3tbddtvNnalJW3hlAVR/Lffhh/oeUAFi4hjxwvqy++67F5zLGT2zqUfpKiBATVexwtdH/QNa/Qux73r7at5Eq1VErHdI6yWpW6X6wYWqyb4lUfF/daRffusXNrro1i5zQCVWkixCeE/vuusuB5Kk4CT3+uTJk503leX/VACVDUEcfwSwErsKdOIpLQ5Qn3/+eXfUEtC677772jXXXOM2/bD7XCU/FBCghtlRD7h81m+9fTahYtgAM1h785qHW+1DXxKgZlBTvX/DxIy6fmGji27tMgdUYkHxkLK0j+eUrEHAIUcUcd4m4Pjpp58WACppOckelMiDyhI/saRcTyEbEqDKOZ3JPKicFQrEPv300y5lJ6Vy5cruMHmOaFLJDwUEqGF2jPoHtPoXYl8Baoh61NX8C1NQ+oXpl6+1yxxQERaPJ5uV2IEee1wQRwxVrVo1be2XLl3qMh1ttNFGDnqrVauWUhsci/Ttt9/azjvv7PLX52JBSzadtWjRolS7j858yOANz1QhZIPzWwnfCC0C1DAF9QDJZ/0EqGHWFaBKv1AFVD+Rz9j9tAAAIABJREFUApEAVJkmcwpw6gFhEVdddVXmGk2hJQ7tHzRoUMJUpylU3+ASMkXdcccdtuOOO7ozXAn5IHWrL2vWrHH/x6NOadCggZE9K1kRoJbECv+rI0DNZ/0EqGHWFaBKv1AFVF+AWg7mALG6/fv3d6ESo0ePdulOOc+VPPakE+UcWUIZ2JBGrO/w4cNdrG78a3Xq1HGvceoBmaSGDBnivLJkb2rZsqXNnz/fRowY4cIn2MjGa7///nshQKXt2rVr27HHHpvWmbIAaaVKlVwyATzoHCfGYf2xqVA/++wzFyvMmbipnFcrQA2b/ALUfNZPgBpmXQGq9AtVQPUFqOVgDvhUpyQxIE0pcbyEK7D0zmYxEhkMGDDApS3lGC2gEi9k/Gtt2rRxJyQ89thjLonCxIkTXawuwElc8MCBAw0vJ0v7gOzFF19sW265ZSFA5ZSE6667ziVN8IBMVqqpU6fa3LlzC1mD/hGLTOF6fl+wYIH7P6A7b948GzNmTEEd+kICB44YI9ECAEqf5UHNziQXoIbpGm39BKhh1hWgSr9QBVRfgFoO5oAHVICO3wFSYlI5QguYZBm+fv36zvNJvCoZu/Cmxr/GBjPSpc6ZM8edKwss8n8A9YUXXrCGDRu6ZXXOkOV3AJYfjvCKL9yfZAjchzSshxxyiEuqEFuATI4Go+CdJQsYnlkK3uBp06Y5j7AvwC99I6HCo48+6oAZryoxzNOnT3fjji8cORbVEm2A0QM4dN5E274C1Py2r96/+W7f0PFFtb5iUKNqmRL2ywMqAEeMJlDG2bAAH4BKNi5+54eldJbRe/bsucFrxJR2797deUZJWsCZsh5Q8cqyCW3rrbe2L774wh3tRVYuYl8TASqnK5Dxi/SxtMsJCfHL8nhWOb2BwsYoEiewSQrgJEyBghfWl7/++sttguNn3bp1rj3CGwhXSFS0xF/CCfV3tWgDlh7AYdYVoIbpp/kn/UIVUP1ECghQ82xeeEBlwxBgh/eTLFrt2rVzR24RUzpjxgzn0QQ68W6yZB//GmC43Xbb2dVXX22XXHKJETIQC6hAKd7Qgw8+2Hlj8WR+//33hQCV5Xcyed15551uJ/5FF13kkiyQvWv8+PGFlCc+NjbGlD7dfffd1qhRIyM5A6EChx9+uMsexmsAJyELXEPf6YsPCRCgZn5SC1DDNI22fgLUMOsKUKVfqAKqL0AtB3PAAypeStLCcpYr6VvxSgKpeEv5mTlzpvNS4vnEExr/2vLly+3EE090caxskmJZn7hVvKrvvfeeVa9e3XlDzzjjDBcqgOeSjUwAsC+cZYv3lIQHeHHTKUD1qaee6qocc8wxbjMU0Iw3F0ilz8Sp4jnlB4BlM1ayIg9qOupveG20AUuAEGZdAWqYfpp/0i9UAdUXoJaDOUBCAryJeDtZIie7FkvnLOf/+eeftsUWWzgVWNoH8mJL/GuAJ3VYfud3QBDYjS20C8z65fn4v6Wywz6ZWbgnZ+ECvsmKzx5WnGkFqMUpVPTfBaj5rJ8ANcy6AlTpF6qA6gtQy8EcINnB6aefbqSQVfmfAgLUsNkgQM1n/QSoYdYVoEq/UAVUX4BaDubAL7/84pbBVQorIEANmxEC1HzWT4AaZl0BqvQLVUD1BaiaA+VWAQFqmOkFqPms33r7au4jVqtW8lCasNGnX7tK9cJnGmv+pa9hbA3pl9/6hY0uurW1iz+6tlHPMqiAADVMzEg+4Ba+UTCopUuWWs0IAVa82lHv39fr61jdunXDJkkWa0dy/sWMV/0LM770C9MvX2sLUPPVshpXIQUEqGETIpIPkK+mmo09Imxgqm1WoYJ9fdZXAtSAuRDJ94cAOsCihatG3b4ZG2jEGhKgRswg6k52FBCghukayQ9oAWqYUX1tAWqwjpF8fwhQg+3qG4i6fTM20Ig1JEDNgkE4+5MD8jlsnnNIKUuXLjU2MHGuKGlIW7Zsmdadae+ggw6ySpUqpVUvahdzdBSpTNMdf+g4BKhhCkbyA1qAGmZUAWpm9DNtkgoVMpKfLzkE+KH6R7W+ADULliEPPDDJUU8DBgxwd3jooYfcAfNkZSIjks8zX9TtJ0yYYGvXrrVu3bq5g/HJNV+tWrUs9Lj0muTMVFKiXnnllaV3UzOXeYq0r1Et+oAugWUEqCUQLUEVeVCDddT7N0xC6RemX77WFqBmwbLnnnuu85SOHDnSFi1aZBUrViwWUMeOHWvXXHON6w1Q27p1azvwwANdBiiyM3Xt2tXat2/vfm/WrJlLIbrlllu6TFD9+vVzB/GTlpTsS9dff70DW7IvAcL+sHy8uMDu3LlzXdale++913788UeXHYrMU0A16U/Jc3/eeedZ1apVbdasWXbhhRfayy+/bLNnz7abbrrJyBDVp08fW7dunT377LPuIP0HH3zQZZwaPXq0S7HKPXv37m2kMI3tzyuvvGL9+/d3/Z84caINHDjQfvvtNxs0aJB16tTJ3e+cc86xhQsXuixVtMU9Xn31VZfBavHixTZ48GDr3LlzwvrJzClADZvokXyACFDDjCoPamb0kwc1WMdIfr7Igxps19AGBKihCsbV//33391mg2+++cYBH3no+bcoDyqAxgH7jz76qFGf3PakLAXO1qxZ44CVVKInnHCC9e3b1x3EjzcQyNxxxx0dqJHx6eSTT7Y5c+Y4gAPoAGS8tb6QDpQlduASKAVCyTj13XffOWjE47vHHnvYSSedZPXq1TOguXLlyg4GR40a5frQo0cPd71PKzpmzBgbMWKE6+cVV1zhwByQBcqbNGliK1assAsuuKCgP2hDylWfLvXDDz+0H374wYHzxx9/7LQCsg844AD3GvqhBZD81ltvuXpDhw5NWl+AmuEJ/XdzkXyACFAzY2x5UIN1jOT7I4cAS/oFT8G8bECAmmGzAqKAIUvY48ePd+lE+X9RgIoXFMjEC0kBQPEmfvHFFw48aYsl/hdeeMEaN25sl19+uYPF/fbbz13rl8vvuusuu/TSS+21115z1wGgsQVgxKvaoEEDl54UL+vxxx9f8BreVuD0iSeecPVXrVrlPKw1atRw1/JTpUoVI8YWQMWTCjR///331qpVKweS/BDiAHgCkj/99JPz8Pr+8HcA9aOPPrIWLVq4tknNeuSRR7r2AWSAlYI3l9StDRs2dOER6Pjrr7+6RASkcY2vj0eZMn36dNeH+KIl/pJP9kg+QASoJTdobE0BarCOkXx/CFCD7eobiLp9MzbQiDUkQM2wQQ499FDbbLPN3PI4EAV8AlwszSeLQb3nnnscjAJ0FJbzO3bsaLwpYgHVx6CylL/DDju4H8ICYkGU++PRPOKII6xDhw4FowMq6RNt1KpVy3lSAcLatWvbggULbPvtt3d/o87TTz/t/v3ggw+cBxRAZAyEG8QCKvG0bdq0ccvxbAijv7QDCDZt2tRBtgdU3x8PqHhLgWS8xo8//rhrn3/x1vI3CuEC/E7bgCqeYfq9+eabO0iOr//pp586z22ioiX+sIkeyQ9oAWqYUX1tAWqwjpF8fwhQg+0qQM2YhCVqSIBaItkSV2JZf++993aeRx/3yZI1S+rAXTJA/fLLL50nkr/jIWzevLnb6T9u3DgXn8nSe+wmKQ+oXbp0sX333dcAs4022shB5Q033OBiS+MBlR7jqQRgiQv1Ma3AaNu2bZ0nlnhQgJJl/FQAlVhY+obnFrg97LDDXFgCJw4888wzzlOKFgBrPKDi5aU/zz//vFWoUMFt/mIceFrxfuIhxptLX1auXLkBoC5btmyD+h6+BagZnNR/NxXJB7AANTOGFqAG6xjJ94cANdiuAtSMSViihgSoJZItcSUAj008AJsveEVZcgb42AAErCbaxc+GIpbfAVkActiwYcaGIjZGTZ482YDRWA8qwNqrVy+3AYld8atXr7bjjjvO2PnPJi2WzKkbW4hPxdvKZimW0oFDXmNTEiDJ5ijAcdNNN00JUAFKQPznn3+2adOmuc1SACZHa22yySZuLEAqmvj+xHpQAdSpU6e6uFXCFQBrwiIIWaAvxKuyOWvSpEkJPaiJ6iczpzyoYRM9kg9gAWqYUeVBzYx+2iQVrGMkP19yCPCDDRDRBgSoETIM3kvCAvzZqXQNyAP28JAmK5yviqeVpftUCkBJbGxs4bQBNlx5z29x7RCDSiwsm7sINfD9o/94N+kLJwsQorDFFlskbY77EnoQ2x/glCV/vKh4V4sqieonul6AWpxFi/57JB8gAtQwowpQM6OfADVYx0h+vghQg+0a2oAANVTBclrfAyohDLlQBKhhVorkA0SAGmZUAWpm9BOgBusYyc8XAWqwXUMbEKCGKlhO6xMry2awVD2uZS2TADXMApF8gAhQw4wqQM2MfgLUYB0j+fkiQA22a2gDAtRQBVU/JxQQoIaZKZIPkK9fLxgUcdXEQEe1RL1/X9tOLuY7qiWS8y+HAEb6hc3sqOsXNrro1hagRtc26lkGFZgyZYrttddeGWxRTUkBKSAFpEB5USDKX+Dy1QYC1Hy1rMYlBaSAFJACUkAKSIEcVUCAmqOGU7elgBSQAlJACkgBKZCvCghQ89WyGpcUkAJSQApIASkgBXJUAQFqjhpO3ZYCUkAKSAEpIAWkQL4qIEDNV8tqXFJACkgBKSAFpIAUyFEFBKg5ajh1WwpIASkgBaSAFJAC+aqAADVfLatxSQEpIAWkgBSQAlIgRxUQoOao4dRtKSAFpIAUkAJSQArkqwIC1Hy1rMYlBaSAFJACUkAKSIEcVUCAmqOGU7elgBSQAlJACkgBKZCvCghQ89WyGpcUkAJSQApIASkgBXJUAQFqjhpO3ZYCUkAKSAEpIAWkQL4qIEDNV8tqXFJACkgBKSAFpIAUyFEFBKg5ajh1WwpIASkgBaSAFJAC+aqAADVfLatxSQEpIAWkgBSQAlIgRxUQoOao4dRtKSAFpIAUkAJSQArkqwIC1Hy1rMYlBaSAFJACUkAKSIEcVUCAmqOGU7elgBSQAlJACkgBKZCvCghQ89WyGpcUkAJSQApIASkgBXJUAQFqjhpO3ZYCUkAKSAEpIAWkQL4qIEDNV8tqXFJACkgBKSAFpIAUyFEFBKg5ajh1Oz0Fnn32Wdtzzz3Tq6SrpUA5UGC9rbcvV8wqByPVEHNNgfrbt4hMl+vWrRuZvpSXjghQy4uly/k4b7nlFrvssssiq8LXX39tUf4AVP/Cpk6U9QNQDx9SMWyAqi0FMqxA052PsIGdXsxwqyVrLsrv35KNKDdqCVBzw07qZaACAtQwAaP+Aa3+ldy+AtSSa6ea2VNAgJo9bXOlZQFqrlgqg/2cP3++rVy5slCLNWrUyKkl8MWLF/8/e+8BJVWVtX/vJoNIGCRKMAEioGQV+OB1QMKIouSgSBJQgRmQIBlElGEQ/wIOCEqQLEgGXyUJggooUZKARCVnkAzfeva81VPdVDdVtau6b916zlq9GrrvPnefZ5+q+vU++9wjFy5ckCJFivilDAHVL5kSvIgA6F79CKi22NI6PAoQUMOjayT1SkCNpGiFyNd69erJrVu3pHjx4rE9lihRQl566SXTHSZPnizXr1+XFi1amPrxGFesWFFWrlwpKVOmvKO/CRMmyPbt22XIkCF+3YuA6pdMBFSbTBGpHwE1TEFntyYFCKgm+VxhTEB1RRgDGwQAtVGjRoLv3g2AOX/+fJkxY4YsWbJERo4cKXPmzJHx48fLhx9+KKlSpZJOnTpJ8+bN5ciRIwqiGzdulFq1akm3bt2kUqVKcvnyZZk9e7ZUrVpVuz548KC0a9dOMmfOLOvWrZOOHTtq3+vXr5dBgwZJ69atZc2aNdKjRw/Zt2+f1K9fX4YNGyZdunSRDz74QGrXri0zZ86Uzp07y8SJE/UeI0aMUHCdNGmS/Pnnn4Js6tChQ6Vhw4YJCkFADWyOxL+aGVT36kdAtcWW1uFRgIAaHl0jqVcCaiRFK0S+AkzPnz8vhQsXju2xbdu28sgjj0jRokWlZ8+e8t5778nw4cOlfPnyUqBAAQXKFClSSOnSpeXEiRMyePBghUOAY6tWraRNmzayadMmuXr1qvTv319Sp06tfe/Zs0cKFiyokJsuXTpp3LixjBkzRvLmzas2hw4dknLlyingAmorV64sY8eOVT8yZswop06d0n7R56xZs2TUqFFy6dIlLUcAIK9du1Z27dql/mzbto2AGqI5QkANrZBOBnwCamhjzd5CowABNTQ6RnIvBNRIjl6QvgNQs2XLJmXLlo3toVq1apI/f35Zvny5VKlSRRo0aKCZ1Nu3bytkIsu5detWzW6eOXNGr5k6dapC7vHjx3Vpf/r06XLlyhXp1atXbL+wLVWqlJw+fVphE7WuuBZfGTJk0P5RD7tixQoFTSzZT5s2TZ577jmJiYnR63r37i2FChWSli1bKhQfOHBAwXTu3Ln6de7cOcmSJYv2hbZ69Wr1N37jLv4gJ4yIOBmwMCr6F3xsCajBa0fL8ClAQA2ftpHSMwE1UiIVQj8TWuLHLb7//nupUKGCLuMj6wl4zJ49uz6iqUyZMroED9hEVnXnzp2SJ08ehcYbN25o5tMXoNatW1c2b96smddixYrJsWPHtBQAgIpaWPSL7G3NmjVl3Lhxei9vQG3SpIneF1+4z9mzZwXPNfXUoCKjimyrB1B9ScUlftsEIgC6Vz8Cqi22tA6PAgTU8OgaSb0SUCMpWiHyFYAK2AM4ehqylTdv3pSSJUtK+/btFRLnzZunO+VRf4oMJ+pTURMKQEUtKbKoANmmTZtKnTp1ZP/+/XLx4kXp169fnAxqYoB69OhRzaoCctEvlv4Bnx5ABchOmTJFwRmlAQBYZHKRlSWghmhC+NENAdUPkRK5xMn6EVBtsaV1eBQgoIZH10jqlYAaSdEKka9YvsfGI++GHfyoBcUGpqVLl+ryfd++feWHH36Qp59+WksC0qRJo5lPQCo2RmHzEzZL4VFPixcvlm+//VZBFRurAK9oWOJPDFCR9QQw79ixQ69HvSpAddWqVYJd/MiMYvPWG2+8ob4g67po0SJdxieghmhC+NGNkwEL7tM/P4KYwCUE1OC1o2X4FCjzYHV5vx4f1B8+hZ3fMwHV+TFKdg+xDI9MJ5bzscSOZXyAIxqW21H/6WkAWICsr0dDJTYQgG6OHDnUDjWuWbNm1Xtdu3ZNodTXvQIRhkv8gah157UEQPfqR0C1xZbW4VGAGdTw6BpJvRJQIyla9DVoBQioQUunhgRU9+pHQLXFltbhUYAZ1PDoGkm9ElAjKVr0NWgFCKhBS0dAtUnneP0IqCEIMLsIuQIE1JBLGnEdElAjLmR0OBgFCKjBqPZfG2ZQ3asfAPV/106TPLlz2wYZRus/jhyhfwZ9I1W/J/I/Yxh16Eyd/v4XupE6qycCqrPiQW/CpAAB1Sas09+g6Z8lvrdl04bpkjuPcwH1yB9HwuPffx6dbG6ooc+dO7fkzP0/5r7C0QFfHzZVna6fbXTOtSagOjc29CyEChBQbWI6/Q2a/lnie1smj0th6YC2IlLgwYby/z0z3ZFa8PVhC4vT9bONzrnWBFTnxoaehVABAqpNTKe/QdM/S3wJqBb1PLYE1OBV5Os3eO3cbElAjZDoYgnp/PnzerRoUrXDhw/rg/rxnNNA2h9//KF2SeFr/EdRJeQnATWQCN55LT9A3KwfAdUW3f9YE1CDV5HvL8Fr52ZLAmqERBenKeH8+eHDhyeZxxMmTIh9GL73TTt37iyvvPKKnjrlq+HUp99//11PqwpXw6lXv/zyi3z22Wf67FScdpVYI6DaIsEPEDfrR0C1RZeAatWP7y9WBd1pT0CNkLh6AypOUurWrZs+yL5Pnz7y8ssvS40aNeT9999XaNy4caMMGjRIT4vCz0aNGqVHiH788cd6RChOeXrqqafk008/ldKlS8v06dP1mNO///3v8vXXX+vJUYDTuXPnyqRJk/QYUmRThw4dqmo1atRI+9u8ebOe7hTflwULFsju3bv1hCgciZo+fXpZvny5+vnBBx8kqDge+o9TowoVKiTVqlVLNDI4UhUnXf300086BgJqeCcyP0Bs+jpbPwKqLboEVKt+zn59OP850Fb9nWpPQHVqZOL55QHUYcOGyf33369QiROdGjZsKBs2bJB///vfeqIToK1fv36aVXzxxRelcePG8sUXX+gRpoDDAwcO6G7Tpk2bSocOHRQie/furSdF4Yx7gG337t2lcuXKcv36dWnRooVmbnft2iWDBw/Wez3//PMKs4BIgGp8X3AM6ZYtW/R4Uhxdivs/8MADepTqsWPH9MQo74a+R48erdlQ+IzsLO7j3e677z5p1arVHdECdOM4VQJqeCcyP0Bs+jpbPwKqLboEVKt+zn59EFCt8Q3WnoAarHJJbOcBVABps2bNpFevXuoBAK1r166SP39+ef311zWridrPOXPmyLx582TNmjVSp04dvRaAd/DgQSlTpoz8/PPPCpc9e/bUI0wfeeQRadCggWZDX3jhBXn88cdjs6jIpJ47d04B+Pbt21KrVi2F2BQpUvj0BeDsAdTq1avL3r179f4A1E8++SS2NGDHjh3qO3xu3769NG/eXHLmzCnbtm2TL7/88g5ABfDGb74AFYCMccdv8NmpjW/QtshQP4t+BFSLeh5b1qAGryJfv8Fr52ZLAmqERNcDqBUqVNAsabt27WI9x5J90aJFFTRxHbKbgDwAHTYsIRvqaYDARx99VHbu3Knn3SPbiuwksqnLli3TZf2JEyfKuHHjBMvo27dvlyFDhsilS5ckY8aMcQAVffvyBdlPD6DWrVtXARStYsWKMmLEiFhARakCYLdTp07qKyAZDTB86NChOJFJly5d7O+9f8EMatJMYH6A2HR2tn4EVFt0/2NNQA1eRWe/PphBDT6yNksCqk2/JLP2ACqAskSJEoLsI5bxAYADBw7UOswuXbpo5rRt27aaCV24cKF8/vnnMm3aNN20VLNmTa1PzZcv3x2AevLkSe0D2VNAH8C0bNmyCQIqNko98cQTPn1BqYA/gArxTpw4oZlabP4qXry4vPnmm5qZ/cc//hFHW2SF58+ff4feBNSkmYL8ALHp7Gz9CKi26BJQrfo5+/VBQLXGN1j7qAHU9evX69J3pkyZfGr1zTff3HVjTrAih8IOgLpu3Tr56KOP5J133pGxY8dqphH1oNhYFBMTIz/++KNucMKS+kMPPaSbm5AxxXI+3gBGjhypAIhldE8GtX///ppBxfJ71apVdWn/zJkzujkKkOkrgzpgwACteQUko8/4viAD6y+gerTBhq/Fixdrf6hBxZc/DYCK8aI2N7HGXfz+qJnwNfwAcbN+BFRbdAmoVv34/mJV0J32UQOof/vb33QDUEKPRgLgob4yUhqeiYoleCzr361hA1TatGl1ST+xBqDFsv3DDz+swJtYw/09sB+IL3fzFb8HrKZKlcqfS/2+hoDqt1Q+L+QHiJv1I6DaoktAterH9xergu60dzWgYkMO6ieROcXudWQasYyMjTnYMf7ggw/qph08cglA1rJlS30003PPPae/z5AhgzujHoWjIqDags4PEDfrR0C1RZeAatWP7y9WBd1p71pARW0jMoGzZ88W1FficUvYvIPl67fffltrM7FJB8vK2BwEQMUubyyJt27dWiG2du3a7ox6FI6KgGoLOj9A3KwfAdUWXQKqVT++v1gVdKe9awEVjynC8zdnzJihkQOszpo1S5/ZiQfBe56p6XmsUfbs2XV5G88IxQ521G2ivpHNHQoQUG1x5AeIm/W7LRs3TJc8uXPbBhlG6z+OHIkI/3Lm/p8wqhB813z9Bq8dLJ2un210zrV2LaACLrERCI81QsMOdZychIfNI5uKB8KjoY4SD3oHqF67dk1Sp06tj2rCczSxEYjNHQoQUG1xdPobdFT5d2yNyK3rtoDGs/7tcn7dWOnUFlXxDUMQqJ9NVKfrZxudc61dC6g4nQiPTMLOd2ziwYPsscS/adMmhU/UnmLXO56/ied0Yol/6dKl+sxQ7CDHCUt4hBObOxQgoNri6PQ36KjyD4C6oKItoHGsY+S3v+4hoBoUjar5Z9ApIVPqFwZRXdClawEVsQF8YmMUzoLHKUhY3sfpSVjex/NAcb48fobTmQCoeMQSdrzjGpwdnzlzZheEmEOAAgRU2zzgB4iD9COg2oIRBmu+PmyiUj+bfm61djWgImjHjx/XZXycROTdUG8KaPXeqY/HTB05ckTrUO/2mKWknhDwC5lgPJEgqRoA/sKFC1KkSJGAbgltYZcUvmLTG+KLQwsSawTUgEJ4x8X8AHGQfgRUWzDCYM3Xh01U6mfTz63WrgdUtwTOc5IUTlxKqoYTnjwP6ve+J06RQhlEQs+U/f777/Xkqvr164fN1YMHD2rmG5vb8MzUUqVKSe/evRO8HwHVFgp+gDhIPwKqLRhhsObrwyYq9bPp51ZrAmqERNYbUPF4LBxligfa9+nTR15++WWpUaOGvP/++wqNKF/AoQR4pit+NmrUKC1bwMYxgFyVKlX02a/YNIbNYyhzuHnzpvz973+Xr7/+Wk+jApzOnTtXT5TCA/yRTR06dKiq1ahRI+0Ptbs//PDDHb4sWLBAdu/erXW8HTt21BILlEzAzw8++CBBxc+ePaslGXjKQrVq1RKNzLvvvivXr18XnGp15coVvQegOKGDCwiotonODxAH6UdAtQUjDNZ8fdhEpX42/dxqTUCNkMh6ABVHet5///0KlQAzZBGx+QtPHEDJQt++faVfv3665I0nFeCJBXjc1pIlSxQOcWABShiaNm0qHTp0UIhE5hG1t1u3blWwxfNgsVkMANiiRQtZu3atYNMZnoCAe+F4VcAsIBKgGt+X1atXxx51WrBgQb3/Aw88oMepHjt2THLkyBFHdfQ9evRoPRwBPiM7i/t4NxzH6nlqqjsLAAAgAElEQVQ0GH5++fJlLcNA6ca8efMEWV08jSGh0gwCqm2i8wPEQfoRUG3BCIM1Xx82UamfTT+3WhNQIySyHkAFkDZr1kx69eqlniMrikMF8JSC119/XbOaqP2cM2eOghueWFCnTh29FoCHpfEyZcroc14Blz179tSs4yOPPCINGjTQbCiefoANY54sKjKp586dUwBGnW6tWrUUYlOkSOHTF4Dzli1bdJNa9erVZe/evXp/ACqenuApDdixY4f6Dp/bt2+vhyR4nkuL59jGB1T0593wWDBkiAHe8PGvf/2r/hqAjHHHb/DZqY1v0LbIRJV+BFTbZAmDdVTNP+oXBgXYpS8FCKgRMi88gFqhQgXNkrZr1y7WcyzZFy1aVEET1yG7uW3bNgVEbFhCNtTTAIGPPvqoPiM2a9asmm1FdhLZVJyoBdCbOHGiHlZw8eLF2BrUS5cuScaMGeMAKvr25Quynx5AxaO6AKBoFStW1OfSegAVpQqA3U6dOqmvgGQ0wPChQ4fiRAaZUs/v8QtAMIA6TZo02ieywok1ZlBtE50fwA7Sj4BqC0YYrPn6sIlK/Wz6udWagBohkfUAKoCyRIkSguwjlvEBgAMHDtRa0i5dumjmtG3btpoJXbhwoR7pOm3aNK3PrFmzptan5suX7w5AxXGw6APZU2RlsTmqbNmyCQIqltSfeOIJn76gVMAfQIX0OJIWmVps/ipevLi8+eabmpn9xz/+EScyyArPnz8/9mdjxozR8Xn/jIAavsnMDxCbtiHVj4BqC0YYrEMaX/oXBgVsXTo9vrbROdeagOrc2MTxDICKQwc++ugjeeedd2Ts2LGaaUQ9KDYWofYSBw9ggxOW1HEqDDY3IWOK5Xy8wEaOHKkAiGV0Twa1f//+mkHF8nvVqlV1aR+PbsLmKECmZxe/dwYVG5NQ8wpIRp/xfUEG1l9A9QwSG74WL16s/aEGFV+JNdTGAmy926+//iqoefXVmEG1TXSnv0FHlX8EVNtkDoN1VM0/6hcGBdilLwUIqBE6L/BMVCzBJ7Rr3XtY2ACVNm1aXdJPrAFosWz/8MMP3/U5sLg/nj+KFogv/sgNWMWjo0LZCKg2NfkB7CD9CKi2YITBmq8Pm6jUz6afW60JqG6NLMcVRwECqm1C8APEQfoRUG3BCIM1Xx82UamfTT+3WhNQ3RpZjouAGsI5wA8Qm5gh1Y+AagtGGKxDGl/6FwYFbF06Pb620TnXmoDq3NjQsxAqwAyqTUynv0FHlX8A1FvXbQGNZ/3b5fxat+7UFlXxDUMQqJ9NVKfrZxudc60JqM6NDT0LoQIE1ADF3LUyjsEffxyRPHkSf5RXgHcI6eX0zybnb6nzEVANEjodYOifIbgiusnYyX/A2UbnXGsCqnNjQ89CqAABNUAx36sosvvOww4C7IWXR4ICMTHyW+89jv4Adjog0D/bRKd+Nv3cak1AdWtkOa44ChBQA5wQBNQABYvgywmo5uARsGwSUj+bfm61JqCGIbI4DnTFihX6IPts2bLpHY4cOaKPYypQoIBs2rRJnnrqqYDujP5wElPq1KkDsnPaxXiUFZ6RGuj4reMgoAaoIAE1QMEi+HICqjl4BCybhNTPpp9brQmoYYgszoEHTOLUJzwIH81zEtRbb70l1apVk127dt31zngA//Xr1wUPpfd+uP5dDR18wfHjx/XB/r169UpSLwmoAcpNQA1QsAi+nIBqDh4ByyYh9bPp51ZrAmoYIoujRpEpHTVqlBw4cECP7rwboI4fP17PtUcD1FauXFkqVKggly9fltmzZ0vTpk2lTp06+m8cQYrjS++9917BefY41hQPt+/Tp4+8/PLLetIUwHbq1KkKwp6H3iOLC9jFcae1atWSTz75RE6dOiWtWrWSVatWKVSPGzdOrl27Ju3atZPMmTPr6VUdO3aUJUuWyPr162XQoEHSunVrwVGnN2/e1ONGc+fOrSdPPfjgg/Lpp5/Khx9+qPfs1KmTnmTl7c/SpUulZ8+e6v/06dPl/fff1wMHBg8eLPXr19f7vfbaa7Jv3z559dVXtS/cY9myZfLTTz/J4cOHZciQIdK4cWOf9gmFk4Aa4EQnoAYoWARfTkA1B4+AZZOQ+tn0c6s1ATXEkcWRoNjtt3//fgU+HE2K74kBKgANpzfNmDFDYN+yZUvZs2ePwtnVq1cVWPPmzSsvvviidOnSRZo1aybdu3dXyLz//vsV1K5cuSINGzaUDRs2KMAB6ADIyNZ6Go4oxRI74BJQCgjFMaiHDh1SaETGt0iRItKgQQM9MhTQnC5dOoXBMWPGqA9t2rTR63FvtM8++0yPJ4WfPXr0UDAHyALKS5cuLSdOnJAOHTrE+gNtateurbCJrPDWrVvl2LFjCs7btm1TrQDZ5cuX159BP2gBSP7uu+/UbtiwYQnaE1BDNKEJqCESMgK6IaCag0TAsklI/Wz6udWagBriyAJEAYZYwv78888lS5Ys+v/EABVZUEAmspBoAFBkE3G2PMATfQHmvvrqKylVqpS8/fbbCoslS5bUaz3L5R9//LF07dpVli9frtcBQL0bgBFZ1cKFCwuW2pFlfeGFF2J/hmwr4PTLL79U+9OnT2uGNVeuXHotvjJkyCCosQWgIpMKaP7999+lUqVKCpL4QokDwBMgeebMGc3wevzB7wGov/zyizz55JPa9yuvvCI1atTQ/gHIAFY0ZHO3b98uxYoVk7Vr16qOFy5c0CNWb926dYc9Mspoq1evVh/iN0C9U5vj3qAJqE6dKqH3i4Bq1tRxr994I6J/thA7XT/b6JxrTUANcWyqVKki6dOn1+VxQBTgE8CFpXlAlq8a1NGjRyuMAujQsJxfr149ffaaN6Du3LlTsmbNqkv5OXLk0C+UBXiDKO6PjGb16tWlbt26saMDVMIn9JEnTx7NpAII8+XLJ3v37pXs2bPr72Azb948/b5582bNgAIQMQaUG3gDKsbyzDPP6HI8NoTBX/QDECxTpoxCtgdQPf54ABXZUkAyssYzZ87U/vEd2Vr8Dg3lAvg3+gaoIjMMv++55x6F5Pj2O3bs0Mytr8Yl/gAnOgE1QMEi+HICqjl4TgcY+mcLsdP1s43OudYE1BDGBsv6jz/+uGYePXWfWLLGkjrgLiFA3b17t2Yi8XtkCMuVK6c7/SdMmKD1mVh6994k5QHUJk2aSIkSJQRgljJlSoXKgQMHam1pfEDFMJGpBMCiLtRT0woYrVq1qmZiUQ8KoMQyvj+AilpY+IbMLeD22Wef1bIEPHFg/vz5mimFFgDW+ICKLC/8Wbx4scTExCh4YxzItCL7iQwxsrnw5eTJk3cA6tGjR++w98A3ATUEk5qAGgIRI6QLAqo5UE4HGPpnC7HT9bONzrnWBNQQxgaAh008ADZPQ1YUS84APmwAAqz62sWPDUVYfgfIAiCHDx8u2FCEjVFz5swRwKh3BhXA2r59e92AhF3x586dk+eff16w8x+btLBkDlvvhvpUZFuxWQpL6YBD/AybkgCS2BwFcEybNq1fgAqgBIifPXtWVq5cqZulAJh4tFaaNGl0LIBUaOLxxzuDCkD95ptvtG4V5QoAa5RFoGQBvqBeFZuzZs2a5TOD6ss+oXAygxrgRCegBihYBF9OQDUHz+kAQ/9sIXa6frbROdeagOqg2CB7ibIAz7NT4RogD7CHDGlCDc9XRaYVS/f+NAAlamO9G542gA1Xnszv3fpBDSpqYbG5C6UGHv/gP7Kb8AVPFkCJQsaMGRPsDvdF6YG3P4BTLPkji4rsamLNl72v6wmod4tovN8TUAMULIIvJ6Cag+d0gKF/thA7XT/b6JxrTUB1bmwc7ZkHUFHCEAmNgBpglAioAQoWwZcTUM3BczrA0D9biJ2un210zrUmoDo3No72DLWy2Azmb8Y1uQdDQA0wAgTUAAWL4MsJqObgOR1g6J8txE7XzzY651oTUJ0bG3oWQgUIqAGKufPbOAaoW0aNsVMb/bNF5rc0+bXm26nN6YBA/2wzh/rZ9HOrNQHVrZHluOIosGDBAilatChVoQJUgApQASoQsAJO/gMu4MFEiAEBNUICRTepABWgAlSAClABKhAtChBQoyXSHCcVoAJUgApQASpABSJEAQJqhASKblIBKkAFqAAVoAJUIFoUIKBGS6Q5TipABagAFaACVIAKRIgCBNQICRTdpAJUgApQASpABahAtChAQI2WSHOcVIAKUAEqQAWoABWIEAUIqBESKLpJBagAFaACVIAKUIFoUYCAGi2R5jipABWgAlSAClABKhAhChBQIyRQdJMKUAEqQAWoABWgAtGiAAE1WiLNcVIBKkAFqAAVoAJUIEIUIKBGSKDoJhWgAlSAClABKkAFokUBAmq0RJrjpAJUgApQASpABahAhChAQI2QQNFNKkAFqAAVoAJUgApEiwIE1GiJNMdJBagAFaACVIAKUIEIUYCAGiGBoptUgApQASpABagAFYgWBQio0RJpjpMKUAEqQAWoABWgAhGiAAE1QgJFN6kAFaACVIAKUAEqEC0KEFCjJdIcJxWgAlSAClABKkAFIkQBAmqEBIpuUgEqQAWoABWgAlQgWhQgoEZLpDlOKkAFqAAVoAJUgApEiAIE1AgJFN2kAlSAClABKkAFqEC0KEBAjZZIR/k4Fy5cKI899liUq8DhUwEqQAWoQDAKPPTQQ8GY0cagAAHVIB5NI0eBf/7zn9K9e3fHOvzbb7+Jk98A6Z9t6jhdv5jOD9gGSGsq4GIF9rZf7uj3Z7dKT0B1a2Q5rjgKEFBtE8LpgEX/bPEloNr0o7W7FSCgJk98CajJozvvmsQKEFBtghMA3a0fAdUWX1q7WwECavLEl4CaPLrzrkmsAAHVJjgB1d36EVBt8aW1uxUgoCZPfAmoyaN7st51y5YtcvLkyTg+5MqVK6I2ER0+fFguXLggRYoU8UtLAqpfMiV4EQHV3foRUG3xpbW7FSCgJk98CajJo3uy3rVevXpy69YtKV68eKwfJUqUkJdeesnk1+TJk+X69evSokULUz8e44oVK8rKlSslZcqUd/Q3YcIE2b59uwwZMsSvexFQ/ZKJgGqTKWL1I6CGKfDs1hUKEFCTJ4wE1OTRPVnvCkBt1KiR4Lt3A2DOnz9fZsyYIUuWLJGRI0fKnDlzZPz48fLhhx9KqlSppFOnTtK8eXM5cuSIgujGjRulVq1a0q1bN6lUqZJcvnxZZs+eLVWrVtWuDx48KO3atZPMmTPLunXrpGPHjtr3+vXrZdCgQdK6dWtZs2aN9OjRQ/bt2yf169eXYcOGSZcuXeSDDz6Q2rVry8yZM6Vz584yceJEvceIESMUXCdNmiR//vmnIJs6dOhQadiwYYK6ElBtU44ZVHfrR0C1xZfW7laAgJo88SWgJo/uyXpXgOn58+elcOHCsX60bdtWHnnkESlatKj07NlT3nvvPRk+fLiUL19eChQooECZIkUKKV26tJw4cUIGDx6scAhwbNWqlbRp00Y2bdokV69elf79+0vq1Km17z179kjBggUVctOlSyeNGzeWMWPGSN68edXm0KFDUq5cOQVcQG3lypVl7Nix6kfGjBnl1KlT2i/6nDVrlowaNUouXbqk5QgA5LVr18quXbvUn23bthFQwzSzCKg2YZ2uHwHVFl9au1sBAmryxJeAmjy6J+tdAajZsmWTsmXLxvpRrVo1yZ8/vyxfvlyqVKkiDRo00Ezq7du3FTKR5dy6datmN8+cOaPXTJ06VSH3+PHjurQ/ffp0uXLlivTq1Su2X9iWKlVKTp8+rbCJWldci68MGTJo/6iHXbFihYImluynTZsmzz33nMTExOh1vXv3lkKFCknLli0Vig8cOKBgOnfuXP06d+6cZMmSRftKqDGDaptyTgcs+meLLwHVph+t3a0AATV54ktATR7dk/WuCS3xw6nvv/9eKlSooMv4yHoCHrNnz64PuS9TpowuwQM2kVXduXOn5MmTR6Hxxo0bmvn0Bah169aVzZs3a+a1WLFicuzYMS0FAKCiFhb9Intbs2ZNGTdunN7LG1CbNGmi98UX7nP27FnByVCeGlRkVJFt9QDq6tWrFajjNz6oP/hpRwAMXjtYOl0/AqotvrR2twIE1OSJLwE1eXRP1rsCUAF7AEdPQ7by5s2bUrJkSWnfvr1C4rx583SnPOpPkeFEfSpqQgGoqCVFFhUg27RpU6lTp47s379fLl68KP369YuTQU0MUI8ePapZVUAu+sXSP+DTA6gA2SlTpig4ozQAAItMLrKyCQGqL3GZQbVNOacDFv2zxZeAatOP1u5WgICaPPEloCaP7sl6VyzfY+ORd8MOftSCYgPT0qVLdfm+b9++8sMPP8jTTz+tJQFp0qTRzCcgFRujsPkJm6XwqKfFixfLt99+q6CKjVWAVzQs8ScGqMh6Aph37Nih16NeFaC6atUqwS5+ZEaxeeuNN95QX5B1XbRokSBLSkBNumlEALRp7XT9CKi2+NLa3QoQUJMnvgTU5NE9ou6KZXhkOrGcjyV2LOMDHNGw3I76T08DwAJkfT0aKrFBA3Rz5MihdqhxzZo1q97r2rVrCqW+7hWIiMygBqLWndc6HbDony2+BFSbfrR2twIE1OSJLwE1eXTnXZNYAQKqTXACoLv1I6Da4ktrdytAQE2e+BJQk0d33jWJFSCg2gQnoLpbPwKqLb60drcCBNTkiS8BNXl0512TWAECqk1wAqq79SOg2uJLa3crQEBNnvgSUJNHd941iRUgoNoEJ6C6Wz8Cqi2+tHa3AgTU5IkvATV5dOddk1gBAqpNcAKqu/UjoNriS2t3K0BATZ74ElCTR3feNYkVIKDaBCegUj+bAjZrzj/qZ1PAZu30+WcbnXOtCajOjQ09C6ECBFSbmE5/g6Z/jK9NAZs15x/1sylAa18KEFA5L6JCAQKqLcz8AHavfrfltrzwQQrbAGlNBYJU4L0Gy+We6wXkoYceCrKH8Js5/f0v/Aokzx0IqMmjO++axAoQUG2CO/0Nmv4FH18CavDa0dKuAAHVrqFbeyCgujWyHFccBQiotglBAHSvfgRUW2xpbVOAgGrTz83WBFQ3R5dji1WAgGqbDARU9+pHQLXFltY2BQioNv3cbO1oQN27d6+ezf7AAw8EFIMDBw7I77//LuXLl/dp991330nFihUlJiYmoH6T82KcVX/+/HkpXLhwkrlx+PBhuXDhghQpUiSge/7xxx9qlxS+njlzRjJlyqTzJLFGQA0ohHdcTEB1r34EVFtsaW1TgIBq08/N1mED1G3btsno0aNlxIgRQev3r3/9S9KlSycdOnTwuw9A1WOPPSZffPGF1KhRw6cdgObkyZOSJk0av/tN7gunTJkia9euleHDhyeZKxMmTJDt27fLkCFD4tyzc+fO8sorr0jJkiV9+vL999/rHwj169cPm68HDx6Uhg0bSvbs2SVVqlRSqlQp6d27d4L3I6DaQkFAda9+BFRbbGltU4CAatPPzdYJAiqAaPXq1bJu3To5e/asjBo1SqpVq6b/f+2112Tfvn3y6quvyocffqjZsr///e/y9ddfy9NPPy0Am8qVK8uWLVtk8ODB0qxZM2nRooVs3LhRatWqJZ988omcOnVKWrVqJatWrdJs5rhx4yRXrlwCuHn99dcVOgAfzz33nLRv317ef/999SFv3rzy8ccfK5D4auhrzZo10q9fP+nTp48ApmbPni0PPvigAHiffPJJzbgBUHfu3ClvvPGGYKzXrl2TNm3ayJ49e6Rp06by7rvvqg9Oad6AumjRIunWrZvcuHFDx/jyyy8rjEMjQCN0HjRokMycOdOnblWqVJGnnnpKPv30UyldurRMnz5dbt68eUcM586dK5MmTZI///xTAP5Dhw5VORo1aqRx2Lx5s/zwww93+LJgwQLZvXu31KtXTzp27Cjp06eX5cuXq58ffPBBgpJink2ePFkKFSqkcy2xhvhcv35dBgwYIFeuXNF7AIrz5Mnj04yAapvJBFT36kdAtcWW1jYFCKg2/dxsnSCgDhs2TPChDmDcsWOHAt/PP/+sMAnQwPI5oPOjjz6SrVu36hegqHv37gqnOXPmlP79+8s333yjYAjIASwCStu1a6d9HTp0SHr27Kl9YxkZ/y5atKgCapkyZaR27dqaFUN/jRs31qzokiVLFHKwjJ8ixZ2PRoG///jHP2TFihUKv8j+wQ7wDBBbuHChAipA+9lnn5Xx48dL1apVFaYA1/g3ABxQDLB2SvMAKuJy//3361gAZsgibtiwQf79739LlixZpG/fvqonlrxffPFFn7rlzp1bIRyZaYwbGh89evSOGAIAEWNkbnft2qV/bOBezz//vMIsIBKgGt8X/GGDP04A/wULFlT9UaZRrlw5OXbsmOTIkSOOrOgb2fbPPvtMfUZ2Fvfxbvfdd5/OHU+7fPmylmggwz5v3jydW/jjIqGyDQKqbSYTUN2rHwHVFlta2xQgoNr0c7N1ooCKrBmACO3hhx8WZMaeeeYZhQw0ZD2xBAyoa9CggWbSXnjhBXn88ccVAHv06CHLli3TLN3UqVO1JvH48eOa+cJ1np8BUGD/v//7v3oNMmmAT4AqlusvXryoWdE6derofQEqWOLNly/fHbH55ZdfFIABScgwAoQBOwBW1LTi/wDUzJkzy9/+9jfN5gJ2MmTIoBlaLPvDZ1yD/zuleQAVQApw7tWrl7qGbHLXrl0lf/78qheymtBwzpw5Cm6+dAP8QwfAJf4oQNbxkUceuSOGyIQji4qvc+fOKQDfvn1bs+D4QwQx8uULwNkDqNWrV1fd0QCo0NtTGoA/fOA7fMYfBM2bN9c/bFAe8uWXX94BqABe74asN7LG+IMFPv71r39NMFwEVNtMJqC6Vz8Cqi22tLYpQEC16edm60QBFWCIjBwaYBDZUiwpAyDQsESMf2OZH1AHSJg4caKCK4AJgLp06VKFQSynA4SQSQWgoj+AC5bx8bu6devKtGnTtG4RwIr29ttva7YQIIONN8ikehpgJmvWrIkCKrKlgFlkVAGc8MsDqPjZ//t//08AtMg2AtY8MI5OH330UalZs6ZjYu8B1AoVKmhMAOGehiV7ZJ6hL65DdhNxAdD50g1jg+bQD9lWZCeRTY0fQ8TfU4N66dIlyZgxYxxARd++fMEfBB5ARVwBoGjIvqMm2QOoKFUA7Hbq1El9BSSjAYaRXfduyJR6fo+fA4LxRw3+oECfyAp7Gv44AZjHb4BqpzYCoC0y1C94/QiowWtHS7sCBFS7hm7tIVFABXB+++23CnEAR4AiQAAf/gA6AALqNtevX69ZUmRFkdED1KA+FcuugAUs2QKiAJVYWkYmFNk9ZF6RgUMWDLuxsYSM/leuXCkFChTQOsnWrVtr/ejnn3+uAIs6Q4Aj6ix9bXLyzqAC1ABfgOq33npLx+IBVNSg4ueopcUS9BNPPKG1syVKlFDoAuwhI+mU5gFUACV8RCwA1gDAgQMHqv5dunTRzGnbtm01mw1A96Ub/jiID6jQI34My5YtmyCgIrbQzJcvKPfwB1Ch7YkTJ1R3bP4qXry4vPnmm5qZxR8Q3g1Z4fnz58f+aMyYMTo+758lFitmUG0zmQDoXv0IqLbY0tqmAAHVpp+brRMFVGRGsfSNRxyNHTtW6x0BPFhexvIqjiZDTahnmR9L+wBNbKwBYAIssYSMWkVk/NAPak0XL16sy+6o9Tx9+rT2hU00yAKiJhRgiU0v6AMA27JlS4VbwCU+KEeOHKkg46t5AyrqUV966aVYX7766iuZNWuW1l1ikxYyhLgHYBmPcAJ0oSFbh/rVbNmyOSb2AFSUTSCL/c4772g8kGlEPSg2FqH28scff9Q6WmSmERtkq33phmV0D6CiThgQj+V3/MHgHUNApq8MKjYmoeYVkIxYxPcFWXR/AdUjMMoxMC/QH/6gwVdiDbWxAFvv9uuvv2rNq69GQLVNZQKqe/UjoNpiS2ubAgRUm35utk4UULGMiixi2rRpFVQ9DUCJOlRkUT2bUgBDWPJFrarnZ4COW7duxWY6UVuKOkbvhs1OWMb33jGPpwLANv4SPjbywBf8HAAEIInfsOsfcOtp8Avj+Mtf/qLAhvsk9Hgp3BO1rdjQ42sDlpMmAoAagJ3QrnVvX711S2wMvmKY0PW4P8om0ALxxR8NEYdQP0GBgOqP8glfQ0B1r34EVFtsaW1TgIBq08/N1ncFVGRAndgAtsjIxm94/FQkPd/Uidq60ScCqi2qBFT36kdAtcWW1jYFCKg2/dxsnSCgYmc7mnc20s1CcGzuVoCAaosvAdW9+hFQbbGltU0BAqpNPzdbh+0kKTeLxrFFngIEVFvMCKju1Y+AaostrW0KEFBt+rnZmoDq5uhybLEKEFBtk4GA6l79CKi22NLapgAB1aafm60JqG6OLsdGQA3RHCCg2oR0sn4EVFtsaW1TgIBq08/N1gRUN0eXYyOghmgOOBmwMET6Zws09aN+NgVs1px/Nv3cak1AdWtkOa44CnCJ3zYh+AFC/WwK2Kw5/6ifTQGbtdPnn210zrUmoDo3NvQshAoQUG1iOv0NOln8G5FF5Oo5m7COsI6R3+rs0cM9nNqSJb4BiEH/AhDLx6XUz6afW60JqG6NLMfFDGoI5wA/QHyISUAN4QxLvCvOP5vU1M/d+tlG51xrAqpzY0PPQqgAM6g2MfkBR0C1zSCbNecf9bMpYLN2+vyzjc651gRU58aGnoVQAQKqTUynv0Eni3/MoNomVQDWyRJf+heAArZLGV+bfm61JqCGIbK3b9+WFStWyBNPPCHZsmXTO+BYVpxZX6BAAdm0aZM89dRTAd0Z/VWsWFFSp04dkJ3TLv7zzz9ly5YtAY/fOg4Cqk1BfoAwg2qbQTZrzj/qZ1PAZu30+WcbnXOtCahhiM2aNWsUJvv16yf9+/fXO0yZMoY7M5QAACAASURBVEXWrl0rb731llSrVk127dp11ztPnjxZrl+/Li1atJCcOXPKzp07JWvWrHe1c/IFx48fl7Fjx0qvXr2S1E0Cqk1up79BJ4t/zKDaJlUA1skSX/oXgAK2Sxlfm35utSaghiGybdu21UzpqFGj5MCBA5IiRYq7Aur48eOlb9++6g2gtnLlylKhQgW5fPmyzJ49W5o2bSp16tTRf5ctW1amTZsm9957ryxatEi6desmN27ckD59+sjLL78s77zzjoLt1KlTFYRTpUql/SKLC9jduHGj1KpVSz755BM5deqUtGrVSlatWqVQPW7cOLl27Zq0a9dOMmfOLOvWrZOOHTvKkiVLZP369TJo0CBp3bq1dO7cWW7evCkLFy6U3Llzy6RJk+TBBx+UTz/9VD788EO9Z6dOnaR58+Zx/Fm6dKn07NlT/Z8+fbq8//77cvHiRRk8eLDUr19f7/faa6/Jvn375NVXX9W+cI9ly5bJTz/9JIcPH5YhQ4ZI48aNfdonFE4Cqm2i8wOEGVTbDLJZc/5RP5sCNmunzz/b6JxrTUANcWwuXbqkj2vZv3+/At9HH32k3xPLoALQHn74YZkxY4bAvmXLlrJnzx6Fs6tXryqw5s2bV1588UXp0qWLNGvWTLp3766Qef/99yuoXblyRRo2bCgbNmxQgAPQAZCRrfW0AQMGCJbYAZeAUkDozz//LIcOHVJoRMa3SJEi0qBBAylYsKAAmtOlS6cwOGbMGPWhTZs2ej3ujfbZZ5/JyJEj1c8ePXoomANkAeWlS5eWEydOSIcOHWL9gTa1a9dW2ERWeOvWrXLs2DEF523btqlWgOzy5cvrz6AftAAkf/fdd2o3bNiwBO0JqCGe0P/XndPfoJPFP2ZQwzPZfPSaLPENYHT0LwCxGF+bWFFkTUANcbABogBDLGF//vnnkiVLFv1/YoCKLCggE1lINAAosom//vqrgif6Asx99dVXUqpUKXn77bcVFkuWLKnXepbLP/74Y+natassX75crwOAejcAI7KqhQsXFiy1I8v6wgsvxP4M2VbA6Zdffqn2p0+f1gxrrly59Fp8ZciQQVBjC0BFJhXQ/Pvvv0ulSpUUJPGFEgeAJ0DyzJkzmuH1+IPfA1B/+eUXefLJJ7XvV155RWrUqKH9A5ABrGjI5m7fvl2KFSum5RHQ8cKFC5IpUya5devWHfbIKBNQQzyhCagJC0pADc9kI8CEXFcCtE1Sp+tnG51zrQmoIY5NlSpVJH369Lo8DogCfAK4sDSfUA3q6NGjFUYBdGhYzq9Xr54e3+gNqJ4aVCzl58iRQ79QFuANorg/MprVq1eXunXrxo4OUAmf0EeePHk0kwogzJcvn+zdu1eyZ8+uv4PNvHnz9PvmzZs1AwpAxBhQbuANqKinfeaZZ3Q5HhvC4C/6QXa3TJkyCtkeQPX44wFUZEsBycgaz5w5U/vHd2Rr8Ts0lAvg3+gboIrMMPy+5557FJLj2+/YsUMzt6tXr1ZIjt/gl1Ob098A6Z+PmUNATbKXE+efTWrq5279bKNzrjUBNYSxwbL+448/rplHT90nlqyxpA64SwhQd+/erZlI/B4ZwnLlyulO/wkTJmh9JpbevTdJeQC1SZMmUqJECQGYpUyZUqFy4MCBWlsaH1AxTGQqAbCoC/XUtAJGq1atqplY1IMCKLGM7w+gohYWviFzC7h99tlntSwBTxyYP3++ZkqhBcAwPqAiywt/Fi9eLDExMbr5C+NAphVwiQwxsrnw5eTJk3cA6tGjR++w98C3r5CyBtU20fkBR0C1zSCbNecf9bMpYLN2+vyzjc651gTUEMYGgIdNPAA2T0NWFBk9AB82AAFWfe3ix4YiLL8DZAGQw4cPF2wowsaoOXPmCGDUO4MKYG3fvr1uQMKu+HPnzsnzzz8v2PmPTVpYMoetd0N9KrKt2CyFpXTAIX6GTUkASWyOAjimTZvWL0AFUALEz549KytXrtTNUgBMPForTZo0OhZAKjTx+OOdQQWgfvPNN1q3inIFgDXKIlCyAF9Qr4rNWbNmzfKZQfVln1A4Cai2ie70N+hk8Y8ZVNukCsA6WeJL/wJQwHYp42vTz63WBFQHRRbZS5QFeJ6dCtcAeYA9ZEgTani+KjKtWLr3pwEoURvr3fC0AWy48mR+79YPalBRC4vNXSg18PgH/5HdhC94sgBKFDJmzJhgd7gvSg+8/QGcYskfWVRkVxNrvux9XU9AvVtEE/89P0B86ENAtU2qAKw5/wIQy8el1M/d+tlG51xrAqpzY+NozzyAihKGSGgEVFuU+AFHQLXNIJs15x/1sylgs3b6/LONzrnWBFTnxsbRnqFWFpvB/M24JvdgCKi2CDj9DTpZ/GMG1TapArBOlvjSvwAUsF3K+Nr0c6s1AdWtkeW44ihAQLVNCH6AMINqm0E2a84/6mdTwGbt9PlnG51zrQmozo0NPQuhAgRUm5hOf4NOFv+YQbVNqgCskyW+9C8ABWyXMr42/dxqTUB1a2Q5LmZQQzgH+AHCDGoIp1PAXXH+BSxZHAPq5279bKNzrjUB1bmxoWchVIAZVJuY/IDzpd9tv0V1vn779LFuTm3O1+836meYPIyvQTwXmxJQXRxcDu2/CixYsECKFi1KSagAFaACVIAKBKyAk/+AC3gwEWJAQI2QQNFNKkAFqAAVoAJUgApEiwIE1GiJNMdJBagAFaACVIAKUIEIUYCAGiGBoptUgApQASpABagAFYgWBQio0RJpjpMKUAEqQAWoABWgAhGiAAE1QgJFN6kAFaACVIAKUAEqEC0KEFCjJdIcJxWgAlSAClABKkAFIkQBAmqEBIpuUgEqQAWoABWgAlQgWhQgoEZLpDlOKkAFqAAVoAJUgApEiAIE1AgJFN2kAlSAClABKkAFqEC0KEBAjZZIc5xUgApQASpABagAFYgQBQioERIoukkFqAAVoAJUgApQgWhRgIAaLZHmOKkAFaACVIAKUAEqECEKEFAjJFB0kwpQASpABagAFaAC0aIAATVaIs1xUgEqQAWoABWgAlQgQhQgoEZIoOgmFaACVIAKUAEqQAWiRQECarREmuOkAlSAClABKkAFqECEKEBAjZBA0U0qQAWoABWgAlSACkSLAgTUaIk0x0kFqAAVoAJUgApQgQhRgIAaIYGim1SAClABKkAFqAAViBYFCKjREmmOkwpQASpABagAFaACEaIAATVCAkU3qQAVoAJUgApQASoQLQoQUKMl0lE+zgULFkjRokWjXAUOnwpQASpABYJR4KGHHgrGjDYGBQioBvFoGjkK/POf/5Tu3bs71uHffvtNnPwGSP9sU8fJ+t2W2zL+fztJ1qxZbYMMo/WZM2eS3b/btxMeoBP8S0z+pPCvWP6qUihPhaBmgZNfHxiQ0/0LSvQIMCKgRkCQ6KJdAQKqTUOnv0HTv+DjC0Bt9lGK4DugJRUQkT71VxNQORNCqgABNaRysjOnKkBAtUWGAOhe/QiottjS+j8KEFA5E0KtAAE11IqyP0cqQEC1hYWA6l79CKi22NKagMo5EB4FCKjh0ZW9OkwBAqotIARU9+pHQLXFltYEVM6B8ChAQA2PruzVYQoQUG0BIaC6Vz8Cqi22tCagcg6ERwECanh0Za8OU4CAagsIAdW9+hFQbbGlNQGVcyA8ChBQw6Mre3WYAgRUW0AIqO7Vj4Bqiy2tCaicA+FRgIAaHl3Zq8MUIKDaAkJAda9+BFRbbGlNQOUcCI8CBNTw6OroXrds2SInT56M42OuXLnksccec7Tf3s4dPnxYLly4IEWKFPHLZwKqXzIleBEB1b36EVBtsaU1AZVzIDwKEFDDo6uje61Xr57cunVLihcvHutniRIl5KWXXjL5PXnyZLl+/bq0aNHC1I/HuGLFirJy5UpJmTLlHf1NmDBBtm/fLkOGDPHrXgRUv2QioNpkikj9CKhhCnqUdcvnoEZZwJNguATUJBDZabcAoDZq1Ejw3bsBMOfPny8zZsyQJUuWyMiRI2XOnDkyfvx4+fDDDyVVqlTSqVMnad68uRw5ckRBdOPGjVKrVi3p1q2bVKpUSS5fviyzZ8+WqlWratcHDx6Udu3aSebMmWXdunXSsWNH7Xv9+vUyaNAgad26taxZs0Z69Ogh+/btk/r168uwYcOkS5cu8sEHH0jt2rVl5syZ0rlzZ5k4caLeY8SIEQqukyZNkj///FOQTR06dKg0bNgwQakJqLZZyAyqe/UjoNpiS2tmUDkHwqMAATU8ujq6V4Dp+fPnpXDhwrF+tm3bVh555BEpWrSo9OzZU9577z0ZPny4lC9fXgoUKKBAmSJFCildurScOHFCBg8erHAIcGzVqpW0adNGNm3aJFevXpX+/ftL6tSpte89e/ZIwYIFFXLTpUsnjRs3ljFjxkjevHnV5tChQ1KuXDkFXEBt5cqVZezYsepHxowZ5dSpU9ov+pw1a5aMGjVKLl26pOUIAOS1a9fKrl271J9t27YRUMM08wioNmGdrB8B1RZbWhNQOQfCowABNTy6OrpXAGq2bNmkbNmysX5Wq1ZN8ufPL8uXL5cqVapIgwYNNJN6+/ZthUxkObdu3arZzTNnzug1U6dOVcg9fvy4Lu1Pnz5drly5Ir169YrtF7alSpWS06dPK2yi1hXX4itDhgzaP+phV6xYoaCJJftp06bJc889JzExMXpd7969pVChQtKyZUuF4gMHDiiYzp07V7/OnTsnWbJk0b4Sasyg2qakkwELI6N/wceXgBq8drT8rwJc4udsCLUCBNRQKxoB/SW0xA/Xv//+e6lQoYIu4yPrCXjMnj27dO/eXcqUKaNL8IBNZFV37twpefLkUWi8ceOGZj59AWrdunVl8+bNmnktVqyYHDt2TEsBAKiohUW/yN7WrFlTxo0bp/fyBtQmTZroffGF+5w9e1YWLlwYW4OKjCqyrR5AXb16tQJ1/IZ+ndoIWLbIUL/g9SOgBq8dLQmonAPhU4CAGj5tHdszABWwB3D0NGQrb968KSVLlpT27dsrJM6bN093yqP+FBlO1KeiJhSAilpSZFEBsk2bNpU6derI/v375eLFi9KvX784GdTEAPXo0aOaVQXkol8s/QM+PYAKkJ0yZYqCM0oDALDI5CIr69kkFR9QfQnPDKptOhIA3asfAdUWW1r/RwFmUDkTQq0AATXUikZAf1i+x8Yj74Yd/KgFxQampUuX6vJ937595YcffpCnn35aSwLSpEmjmU9AKjZGYfMTNkvhUU+LFy+Wb7/9VkEVG6sAr2hY4k8MUJH1BDDv2LFDr0e9KkB11apVgl38yIxi89Ybb7yhviDrumjRIkGWlICadJONgGrT2sn6EVBtsaU1AZVzIDwKEFDDo6uresUyPDKdWM7HEjuW8QGOaFhuR/2npwFgAbK+Hg2VmCgA3Rw5cqgdalyzZs2q97p27ZpCqa97BSIyM6iBqHXntU4GLHhL/4KPLwE1eO1o+V8FmEHlbAi1AgTUUCvK/hypAAHVFhYCoHv1I6DaYktrZlA5B8KjAAE1PLqyV4cpQEC1BYSA6l79CKi22NKagMo5EB4FCKjh0ZW9OkwBAqotIARU9+pHQLXFltYEVM6B8ChAQA2PruzVYQoQUG0BIaC6Vz8Cqi22tCagcg6ERwECanh0Za8OU4CAagsIAdW9+hFQbbGlNQGVcyA8ChBQw6Mre3WYAgRUW0AIqO7Vj4Bqiy2tCaicA+FRIGyAig80tIceeiiO5zgvHScT4ZFCvtrhw4f14fB4tmawDc/UxHM7E7pHsP3SLnIVIKDaYkdAda9+BFRbbGlNQOUcCI8CYQNUnNmO52d26dIljucTJkyQJ554Qk8s8m54KPvKlStl0qRJsQ9gD3bIbdq0kerVq8c5KSnYvmjnDgUIqLY4ElDdqx8B1RZbWhNQOQfCo0BYARUnEu3atUs9Hz58uB5fOWDAAKlcubKeFHT9+nU9sQgnE+E4TXx/8cUXFVJx9CWyqUOHDpWGDRv6HD2OwMSJQuvWrdMHxo8aNUqqVasmHkBFX507d5bZs2fLgw8+KP/617/kySeflB9//FFatWqlD5zv0KGD/OMf/5Ddu3erHU4+wtGd7777rqRKlSo8qrPXJFeAgGqTnIDqXv0IqLbY0pqAyjkQHgXCCqiAgq+++ko2b96soInlfZzdDnDE0ZjLli1TqKxQoYKeTHTq1Ck9771Fixaydu1ahdvBgwerna+GLC3ugXPasayPM+B//vnnWEDFCURDhgyRL774Qr7++muZPn26nvOObO1bb70ljz/+uPzP//yP2uAoTRzpWbVqVXnttdf0PPpmzZqFR3X2muQKEFBtkhNQ3asfAHXcV5309LaYmLjjvH077v8D+b23bSB23nf09HH27H9Ol4vfPL+P37/nuoR+H6hdfB3i++E5/c42S8JnnRT+FctfVQrlqRDUIPj+EpRsrjcKK6Du27dPRowYoSLmzJlTNm7cKD179owF1FKlSul57mgxMTGaUcW563PnztWvc+fO6TGaOK89IUBFlhWgivbwww/Ld999J/3799clfmRkAZ8bNmyQFStWyN69e/Wc+UcffVSOHTumNvgZ4DhXrlwKyzimE+CcKVMm/T+bOxQgoNriyA8Q6mdTwGbN+Uf9bArYrJ0+/2yjc651WAEVS+gAUrR8+fLp0nqvXr1iAdW7TtQbULdv366Zz0uXLik8JgaoFy9elL59+8beY9GiRTJy5EgF1LRp0+pSPpbwAZzjxo0T1MDWq1cvtvQAf1miPACbuTygi84AsTVr1nRu5OhZQAoQUAOS646Lnf4GTf8YX5sCNmvOP+pnU4DWvhQIK6AiC/rtt9/K1q1bpUGDBgqF3kv88QH18uXLugwfCKB67vHLL79I/fr1ZefOndK2bVsFVNS53nfffdKnTx9d0ocv69evl9y5c2vtav78+eWxxx7T6/72t78pvJYoUUKBN0+ePPL6669z1rhEAQKqLZD8APbS7/sBAYsZ9BKr78WjgO9/N4Pfcr9yxxNX7maTlL/n/LOpTf3crZ9tdM61DhugYtPT6NGj5caNG4IXx8SJE7Wm0xtQa9SoIXXq1FF1UBeKbGmjRo0CAlTcB7WmR44ckbFjx+qGKmx2Qt9Ytn/ppZekQIECukkK9bCzZs2S06dPayb32rVr6s/AgQNl3rx5uqEKLV26dAqteFQVmzsUIKDa4sgPOC/9Zj8vsnehTVBHWcfIb3X3EFANMeHrwyCeiDJC/EdS2noMrbXT/QvtaJ3TW9gA1TNE1JGmTJlS4TOxBpAFMAI247ctW7bIr7/+esfPUT968+ZNzXRiOd+XLZ4GgFKDv/zlL/pkAOzMR50p7oV7etvg/wcPHpQHHnhAUqRI4Zwo0ROzAgRUm4ROf4NOUv8IqLbJFIR1ksaX/gWhgM2E8bXp51brsANqKIQ7cOCAZkjjN2yIwsYqT51rKO7FPtypAAHVFld+gHjpR0C1TaYgrDn/ghDNy4T6uVs/2+icax0RgJqQfKhZRUufPr1zFaZnjlCAgGoLAz/gCKi2GWSz5vyjfjYFbNZOn3+20TnXOqIB1bmy0jOnKUBAtUXE6W/QSeofM6i2yRSEdZLGl/4FoYDNhPG16edWawKqWyPLccVRgIBqmxD8AGEG1TaDbNacf9TPpoDN2unzzzY651oTUJ0bG3oWQgUIqDYxnf4GnaT+MYNqm0xBWCdpfOlfEArYTBhfm35utSagujWyHBczqCGcA/wAYQY1hNMp4K44/wKWLI4B9XO3frbROdeagBqG2ODkKxyt+sQTT8Q+SxVPITh//rw+k3XTpk3y1FNPBXRn9IdnxaZOnTogO6ddjEd94bFhgY7fOg5mUG0K8gOOgGqbQTZrzj/qZ1PAZu30+WcbnXOtCahhiM2aNWsUJvv16yf9+/fXO0yZMkXWrl2rJ1pVq1Yt9qjVxG4/efJkfYxWixYtJGfOnHpKVtasWcPgcdJ1efz4cT1QAQclJGUjoNrUdvobdJL6xyV+22QKwjpJ40v/glDAZsL42vRzqzUBNQyRxVGryJSOGjVK8AxXPPT/boA6fvx4PWIVDVBbuXJlqVChguBRWrNnz5amTZvqqVv4d9myZWXatGly7733yqJFi6Rbt2566ACOdH355ZflnXfeUbCdOnWqgjAOJ0BDFhewu3HjRqlVq5Z88skncurUKWnVqpWenAWoHjdunB5i0K5dO8mcObOsW7dOOnbsKEuWLNFjYgcNGiStW7fWU7dwSMLChQv16NhJkybpaV2ffvqp4HQv3LNTp056Upe3P0uXLtXn1sJ/HGv7/vvvy8WLF2Xw4MF6VC3u99prr8m+ffvk1Vdf1b5wj2XLlslPP/0khw8fliFDhkjjxo192icUTgKqbaLzA4QZVNsMsllz/lE/mwI2a6fPP9vonGtNQA1xbC5duqRHtu3fv1+B76OPPtLviQEqAO3hhx+WGTNmCOxbtmwpe/bsUTi7evWqAmvevHnlxRdflC5duuiRsd27d1fIvP/++xXUcFoWjnndsGGDAhyADoCMbK2nDRgwQE/TAlwCSgGhP//8sxw6dEihERnfIkWKSIMGDaRgwYICaMaxr4DBMWPGqA84RhbX495on332mYwcOVL97NGjh4I5QBZQXrp0aTlx4oR06NAh1h9oU7t2bYVNZIW3bt0qx44dU3Detm2bagXILl++vP4M+kELQDIOZoDdsGHDErQnoIZ4Qv9fd05/g05S/5hBDc8kS6TXJI1vEKOjf0GI5mVC/Wz6udWagBriyAJEAYZYwv78888lS5Ys+v/EABVZUEAmspBoAFBkE3G8K8ATfQHmvvrqKylVqpS8/fbbCoslS5bUaz3L5R9//LF07dpVli9frtcBQL0bgBFZ1cKFCwuW2pFlfeGFF2J/hmwr4PTLL79U+9OnT2uGNVeuXHotvnA0LGpsAajIpAKaf//9d6lUqZKCJL5Q4gDwBEieOXNGM7wef/B7AOovv/wiTz75pPb9yiuvSI0aNbR/ADKAFQ3Z3O3bt0uxYsW0PAI6XrhwQTJlyiS3bt26wx4ZZQJqiCc0AfVOQQmo4ZlkBNSw6UoAtEnrdP1so3OuNQE1xLGpUqWKnmyF5XFAFOATwIWl+YRqUEePHq0wCqBDw3J+vXr1BC8Kb0D11KBiKT9Hjhz6hbIAbxDF/ZHRrF69utStWzd2dIBK+IQ+8uTJo5lUAGG+fPlk7969kj17dv0dbObNm6ffN2/erBlQACLGgHIDb0BFPe0zzzyjy/HYEAZ/0Q+yu2XKlFHI9gCqxx8PoCJbCkhG1njmzJnaP74jW4vfoaFcAP9G3wBVZIbh9z333KOQHN9+x44dmrldvXq1QnL8Br+c2pz+Bkj/vGYOATXJX0acfzbJqZ+79bONzrnWBNQQxgbL+o8//rhmHj11n1iyxpI64C4hQN29e7dmIvF7ZAjLlSunO/0nTJig9ZlYevfeJOUB1CZNmkiJEiUEYJYyZUqFyoEDB2ptaXxAxTCRqQTAoi7UU9MKGK1atapmYlEPCqDEMr4/gIpaWPiGzC3g9tlnn9WyBDxxYP78+ZophRYAw/iAiiwv/Fm8eLHExMTo5i+MA5lWwCUyxMjmwpeTJ0/eAahHjx69w94D375CyhpU20TnBxwB1TaDbNacf9TPpoDN2unzzzY651oTUEMYGwAeNvEA2DwNWVFk9AB82AAEWPW1ix8birD8DpAFQA4fPlywoQgbo+bMmSOAUe8MKoC1ffv2ugEJu+LPnTsnzz//vGDnPzZpYckctt4N9anItmKzFJbSAYf4GTYlASSxOQrgmDZtWr8AFUAJED979qysXLlSN0sBMLNlyyZp0qTRsQBSoYnHH+8MKgD1m2++0bpVlCsArFEWgZIF+IJ6VWzOmjVrls8Mqi/7hMJJQLVNdKe/QSepf8yg2iZTENZJGl/6F4QCNhPG16afW60JqA6KLLKXKAsA4HkaIA+whwxpQg3PV0WmFUv3/jQAJWpjvRueNoANV57M7936QQ0qamGxuQulBh7/4D+ym/AFTxZAiULGjBkT7A73RemBtz+AUyz5I4uK7GpizZe9r+sJqHeLaOK/5weIlz4EVNtkCsKa8y8I0bxMqJ+79bONzrnWBFTnxsbRnnkAFSUMkdAIqLYo8QOOgGqbQTZrzj/qZ1PAZu30+WcbnXOtCajOjY2jPUOtLDaD+ZtxTe7BEFBtEXD6G3SS+scMqm0yBWGdpPGlf0EoYDNhfG36udWagOrWyHJccRQgoNomBD9AmEG1zSCbNecf9bMpYLN2+vyzjc651gRU58aGnoVQAQKqTUynv0EnqX/MoNomUxDWSRpf+heEAjYTxtemn1utCahujSzHxQxqCOcAP0CYQQ3hdAq4K86/gCWLY0D93K2fbXTOtSagOjc29CyECjCDahOTH3AEVNsMsllz/lE/mwI2a6fPP9vonGtNQHVubOhZCBUgoNrEdPobdJL6xyV+22QKwjpJ40v/glDAZsL42vRzqzUB1a2R5bjiKEBAtU0IfoAwg2qbQTZrzj/qZ1PAZu30+WcbnXOtCajOjQ09C6ECBFSbmE5/g05S/5hBtU2mIKyTNL70LwgFbCaMr00/t1oTUN0aWY6LGdQQzgF+gHiJuWaACA44u+2HwP933ZmzZyRrlqwJG3j6i//dj1uE4pLfcr+iRws7tXH+2SJD/dytn210zrUmoDo3NvQshAowg2oTkx9w1M+mgM2a84/62RSwWTt9/tlG51xrAqpzY0PPQqjAggULpGjRoiHskV1RASpABahAtCjg5BUGt8aAgOrWyHJcVIAKUAEqQAWoABWIUAUIqBEaOLpNBagAFaACVIAKUAG3KkBAdWtkOS4qQAWoABWgAlSACkSoAgTUCA0c3aYCVIAKUAEqQAWogFsVIKC6NbIcFxWgAlSAHN8YJwAAIABJREFUClABKkAFIlQBAmqEBo5uUwEqQAWoABWgAlTArQoQUN0aWY6LClABKkAFqAAVoAIRqgABNUIDR7epABWgAlSAClABKuBWBQiobo0sx0UFqAAVoAJUgApQgQhVgIAaoYGj21SAClABKkAFqAAVcKsCBFS3RpbjogJUgApQASpABahAhCpAQI3QwNFtKkAFqAAVoAJUgAq4VQECqlsjy3FRASpABagAFaACVCBCFSCgRmjg6DYVoAJUgApQASpABdyqAAHVrZHluKgAFaACVIAKUAEqEKEKEFAjNHB0mwpQASpABagAFaACblWAgOrWyHJcVIAKUAEqQAWoABWIUAUIqBEaOLpNBagAFaACVIAKUAG3KkBAdWtkOS4qQAWoABWgAlSACkSoAgTUCA0c3aYCVIAKUAEqQAWogFsVIKC6NbIcFxWgAlSAClABKkAFIlQBAmqEBo5uB6bAggULpGjRooEZ8WoqQAWoABWgAiLy0EMPUYckVoCAmsSC83bJo8A///lP6d69e/Lc3I+7/vbbb45+A6R/fgQxkUucrd9tWba8k2TNmtU2yDBanzlzxqd/t2+H8aYBdJ2Qf/G7eKBAbcmWrUQAPYfmUmfPPxH6F5o4u60XAqrbIsrx+FSAgGqbGPwAcbN+t2XsZylsA6S1XwrUeXEjAdWHUnx/8Wv6RN1FBNSoC3l0DpiAaos7P0DcrB8B1RZd/60JqL614vuL/3Momq4koEZTtKN4rARUW/D5AeJm/Qiotuj6b01AJaD6P1t4JQGVcyAqFCCg2sJMQHWzfgRUW3T9tyagElD9ny28koDKORAVChBQbWEmoLpZPwKqLbr+WxNQCaj+zxZeSUDlHIgKBQiotjATUN2sHwHVFl3/rQmoBFT/ZwuvJKByDkSFAgRUW5gJqG7Wj4Bqi67/1gRUAqr/s4VXElA5B6JCAQKqLcwEVDfrR0C1Rdd/awIqAdX/2cIrCahROAe2bNkiJ0+ejDPyXLlyyWOPPRYxahw+fFguXLggRYoU8ctnAqpfMiV4EQHVzfoRUG3R9d+agEpA9X+28EoCahTOgXr16smtW7ekePHisaMvUaKEvPTSSyY1Jk+eLNevX5cWLVqY+vEYV6xYUVauXCkpU6a8o78JEybI9u3bZciQIX7di4Dql0wEVJtMEaofATVMYb+jWwIqATWp5pob7kNAdUMUAxwDALVRo0aC794NgDl//nyZMWOGLFmyREaOHClz5syR8ePHy4cffiipUqWSTp06SfPmzeXIkSMKohs3bpRatWpJt27dpFKlSnL58mWZPXu2VK1aVbs+ePCgtGvXTjJnzizr1q2Tjh07at/r16+XQYMGSevWrWXNmjXSo0cP2bdvn9SvX1+GDRsmXbp0kQ8++EBq164tM2fOlM6dO8vEiRP1HiNGjFBwnTRpkvz555+CbOrQoUOlYcOGCSpBQA1wksS7nBlUN+tHQLVF139rAioB1f/ZwisJqFE4BwCm58+fl8KFC8eOvm3btvLII49I0aJFpWfPnvLee+/J8OHDpXz58lKgQAEFyhQpUkjp0qXlxIkTMnjwYIVDgGOrVq2kTZs2smnTJrl69ar0799fUqdOrX3v2bNHChYsqJCbLl06ady4sYwZM0by5s2rNocOHZJy5cop4AJqK1euLGPHjlU/MmbMKKdOndJ+0eesWbNk1KhRcunSJS1HACCvXbtWdu3apf5s27aNgBqm+UxAtQnrbP0IqLbo+m9NQCWg+j9beCUBNQrnAAA1W7ZsUrZs2djRV6tWTfLnzy/Lly+XKlWqSIMGDTSTevv2bYVMZDm3bt2q2c0zZ87oNVOnTlXIPX78uC7tT58+Xa5cuSK9evWK7Re2pUqVktOnTytsotYV1+IrQ4YM2j/qYVesWKGgiSX7adOmyXPPPScxMTF6Xe/evaVQoULSsmVLheIDBw4omM6dO1e/zp07J1myZNG+EmrMoNomurMBS4T+WeJLQLWoF4gtAZWAGsh8ifZrCahROAMSWuKHFN9//71UqFBBl/GR9QQ8Zs+eXbp37y5lypTRJXjAJrKqO3fulDx58ig03rhxQzOfvgC1bt26snnzZs28FitWTI4dO6alAABU1MKiX2Rva9asKePGjdN7eQNqkyZN9L74wn3Onj0rCxcujK1BRUYV2VYPoK5evVqBOn5Dv05tBCxbZKifRT8CqkW9QGwJqATUQOZLtF9LQI3CGQBABewBHD0N2cqbN29KyZIlpX379gqJ8+bN053yqD9FhhP1qagJBaCilhRZVIBs06ZNpU6dOrJ//365ePGi9OvXL04GNTFAPXr0qGZVAbnoF0v/gE8PoAJkp0yZouCM0gAALDK5yMp6NknFB1RfIWUG1TbRCYBu1o+Aaouu/9YEVAKq/7OFVxJQo3AOYPkeG4+8G3bwoxYUG5iWLl2qy/d9+/aVH374QZ5++mktCUiTJo1mPgGp2BiFzU/YLIVHPS1evFi+/fZbBVVsrAK8omGJPzFARdYTwLxjxw69HvWqANVVq1YJdvEjM4rNW2+88Yb6gqzrokWLBFlSAmrSTV4Cqk1rZ+tHQLVF139rAioB1f/ZwisJqJwDd1UAy/DIdGI5H0vsWMYHOKJhuR31n54GgAXI+no0VGI3AujmyJFD7VDjmjVrVr3XtWvXFEp93euujntdwAxqIGrdea2zAYs1qLboElBt+vlvTUAloPo/W3glAZVzICoUIKDawkxAdbN+BFRbdP23JqASUP2fLbySgMo5EBUKEFBtYSagulk/Aqotuv5bE1AJqP7PFl5JQOUciAoFCKi2MBNQ3awfAdUWXf+tCagEVP9nC68koHIORIUCBFRbmAmobtaPgGqLrv/WBFQCqv+zhVcSUDkHokIBAqotzARUN+tHQLVF139rAioB1f/ZwisJqJwDUaEAAdUWZgKqm/UjoNqi6781AZWA6v9s4ZVRA6g4px0nIuFRRmzRpwAB1RZzAqqb9SOg2qLrvzUBlYDq/2zhla4GVEDp6NGjZcSIETJhwgR54okn9KQktuhTgIBqizkB1c36EVBt0fXfmoBKQPV/tvDKsAIqzlxv1apV7KlAOKYSx1riyMo+ffroee7vvvuuVKtW7Y6f4Wz2t99+W7744guNEk4oGjlypAwdOlSP5MRxmLlz55ZJkybJgw8+KJ9++qkeyZkqVSrp1KmTHsEJIN2yZYsMHjxYHy5fuXJlPZ2oa9eu8tlnn6ndJ598Ik899ZSefITv6Kd06dIyffp0yZQpE2eISxQgoNoCSUB1s34EVFt0/bcmoBJQ/Z8tvDKsgDpgwAA5dOiQ9OzZU89nx5GYOMP94YcfVmjdt2+fnvm+Zs2aO34GQGzUqJFs3rxZo1S4cGH5+uuv9Zx4NAAmgPXq1avSo0cPhd3169dLihQpFDBPnDgh3333nfTv31+++eYbPSrzxRdfVOgE+H7++ed6ZCaO6Fy2bJnkzJlTz5Tv0KGDHr3Zu3dvwfGfbO5QgIBqiyMB1c363ZZlyztJlixZJSZG5Pbt/47V8//4P/dcEezv49vd7b6e0+Xudj9blIK39vh3tx4eKFBbsmUrcbfLQv57vn5tkjpdP9vonGsdVkAFKOJMd8Dlrl27BGfAv/feezJjxgwFRLSNGzfK77//rplS75/huMwmTZr4BNTWrVsrbMKuUqVKet47vgC6W7dulWHDhulxmb/++qvCKwAUGVXY4Dz3QoUKaWYXDWCKUoCiRYvKzz//LHnz5lWgxrGeHhh2bvjomb8KEFD9VYoZDptS1I/6hUMBW59OByz6Z4uvW63DBqi3b9+WzJkzy969e3Vz0s6dO6Vu3boKfdu3b9e6UDSc8T579mzZsWNHnJ+dPn1aGjduHAuoyHxiuR72b731ljzzzDOagcUyPiY37oFsbJkyZaR+/foJAuq0adO0X8AqGvoF3BYvXlx9xBnwyPbed999mk1lc4cCBFRbHPkBQv1sCtisOf+on00Bm7XT559tdM61DhugYsjNmjWTqlWr6vf3339foRGZyxdeeEGzlVj+x3L6l19+KbVr147zMyznP/bYY5olPXDggAIkgBSAWrZsWYXIjz/+WKHy2Wef1frTFStWyPz587UvAO7u3bulc+fOsnr16tgMKnxAphW1pz/++KMu/aOMAJlUAqpzJ6rVMwKqTUGnv0EH7d+qATZh/LT2dwnYz+5CftlveV+Rhx56KOT9hqrDoOMbKgfu0g/9swlN/Wz6udU6rIC6bt06ee211xQWr127JsuXL9eldNSATpkyRTVFJhXZTF8/Q70qygGw7H748GFZu3atAiqyrdgMdfbsWVm5cqVulipVqpRky5ZNUBpw+fJlhdQ333xTN0JhyR7wiftUqFBBIRmlBegTta4NGzaMA6ioW0UGlUv87pn2BFRbLF37ATK7kcj2GTZxIt46Rn5rsoeAaoija18fBk0CMaV+gagVPdeGFVA9MiIDev/99ytUetqFCxckffr0d/3Z+fPn4+ymr1WrlsIsNlrhmaYpU6bULm/duqXlAqgdvXHjhu7az5gxo/4bvwO4erc//vhDsmTJIhkyZIieaEfxSAmotuC79gOEgCoiBFTbq0O0zIwZ6OBVpH7Ba+dmyyQB1FAK6AFUPC6KjQr4qwAB1V+lfF/n2g8QAioB1fbSUGvXvj5CoI0/XVA/f1SKvmsiDlB9ZV6jL2wccaAKEFADVSzu9a79ACGgElBtLw0CKvULgQLswpcCEQeoDCMVCEYBAmowqv3XhoBq08/Z1lzit8bHta8PqzB+2lM/P4WKsssIqFEW8GgdLgHVFnnXfoAwg8oMqu2lwQwq9QuBAuyCGVTOgahVgIBqCz0B1aafs62ZQbXGx7WvD6swftpTPz+FirLLmEGNsoBH63AJqLbIu/YDhBlUZlBtLw1mUKlfCBRgF8ygJtEcwClaODQAp1zh2axoR44cETwyq0CBArJp0yZ56qmnAvIG/eHJBalTpw7IzmkX//nnn3oiWKDjt46DgGpTkIBq08/Z1sygWuPj2teHVRg/7amfn0JF2WXMoIYh4DipCjCJ067w0H80HEyAgwZwTGu1atVk165dd73z5MmT5fr169KiRYs4Bwnc1dDBFxw/flzGjh0rvXr1SlIvCag2uV37AcIMKjOotpcGM6jULwQKsAtmUJNoDrRt21YzpaNGjdJjWlOkSHFXQB0/frz07dtXPQTUVq5cWU+9wqlYs2fPlqZNm0qdOnX03zjqddq0aXLvvffKokWLpFu3bnogQZ8+feTll1+Wd955R8F26tSpCsKeAxKQxQXs4hQtPE8Wx72eOnVKT9ZatWqVQvW4ceP01K927dpJ5syZBaeB4USvJUuWyPr162XQoEHSunVrPUL25s2bsnDhQj3Ja9KkSXpq16effqrHzuKenTp10iNmvf1ZunSpnuwF/3GKF47AvXjxogwePFjq16+v98PpYzjW9tVXX9W+cI9ly5bJTz/9pKd/DRkyRBo3buzTPqEQE1Btk5+AatPP2dbMoFrj49rXh1UYP+2pn59CRdllzKCGOOCXLl3SE0X279+vwPfRRx/p98QyqAA0nIyFY11h37JlS9mzZ4/C2dWrVxVYcdwrjmrt0qWLNGvWTLp3766QiRO6AHo4OQtHtm7YsEEBDkAHQEa21tMGDBggWGIHXAJKAaE///yzHDp0SKERGd8iRYpIgwYNpGDBggJoTpcuncLgmDFj1Ic2bdro9bg32meffSYjR45UP3v06KFgDpAFlJcuXVpOnDghHTp0iPUH2uAYWsBmzpw5ZevWrXLs2DEF523btqlWgOzy5cvrz6AftAAkf/fdd2o3bNiwBO0JqCGe0P/XnWs/QJhBZQY1BC8Z174+QqCNP11QP39Uir5rCKghjjlAFGCIJezPP/9cj1PF/xMDVGRBAZnIQqIBQJFN/PXXXxU80Rdg7quvvpJSpUrpUa+AxZIlS+q1nuXyjz/+WLp27SrLly/X6wCg3g3AiKxq4cKFBUvtyLK+8MILsT9DthVw+uWXX6r96dOnNcOaK1cuvRZfOBoWNbYAVGRSAc2///67VKpUSUESXyhxAHgCJM+cOaMZXo8/+D0A9ZdffpEnn3xS+37llVekRo0a2j8AGcCKhmzu9u3bpVixYloeAR1xUEOmTJn0+Nr49sgoE1BDPKEJqOER1FG9MoNqDQcBy6Yg9bPp51ZrAmqII1ulShVJnz69Lo8DogCfAC4szSdUgzp69GiFUQAdGpbz69Wrp8fneQPqzp07JWvWrLqUnyNHDv1CWYA3iOL+yGhWr15d6tatGzs6QCV8Qh958uTRTCqAMF++fLJ3717Jnj27/g428+bN0++bN2/WDCgAEWNAuYE3oKKe9plnntHleGwIg7/oB9ndMmXKKGR7ANXjjwdQkS0FJCNrPHPmTO0f35Gtxe/QUC6Af6NvgCoyw/D7nnvuUUiOb79jxw7N3K5evVohOX6DX05tfIO2RSZo/ZhBZQbVNvXUOuj5F4J7+9MF/fNHpYSvcbp+ttE515qAGsLYYFn/8ccf18yjp+4TS9ZYUgfcJQSou3fv1kwkfo8MYbly5XSn/4QJE7Q+E0vvyKDGB9QmTZpIiRIlBGCWMmVKhcqBAwdqbWl8QMUwkakEwKIu1FPTChitWrWqZmJRDwqgxDK+P4CKWlj4hswtfHv22We1LAFPHJg/f75mSqEFwDA+oCLLC38WL14sMTExCt4YBzKtgEtkiJHNhS8nT568A1CPHj16h70Hvn2FlDWotonu9DfooP0joBJQbS8NAir1C4EC7MKXAgTUEM4LAB428QDYPA1ZUWT0AHzYAARY9bWLHxuKsPwOkAVADh8+XLChCBuj5syZI4BRb0AFsLZv3143IGFX/Llz5+T5558X7PzHJi0smcPWu6E+FdlWbJbCUjrgED/DpiSAJDZHARzTpk3rF6ACKAHiZ8+elZUrV+pmKQAmHq2VJk0aHQsgFZp4/PHOoAJQv/nmG61bRbkCwBplEShZgC+oV8XmrFmzZvnMoPqyTyicBFTbRA8aAG239ds6aP8IqARUv2dZ5GbYgn59hEAbf7qgf/6oFH3XEFAdFHNkL1EW4Hl2KlwD5AH2kCFNqOH5qsi0YunenwagRG2sd8PTBrDhypP5vVs/qEFFLSw2d6HUwOMf/Ed2E77gyQIoUciYMWOC3eG+KD3w9gdwiiV/ZFGRXU2s+bL3dT0B9W4RTfz3rv0AIaASUG0vDbV27esjBNr40wX180el6LuGgBp9MQ/JiD2AihKGSGgEVFuUXPsBQkAloNpeGgRU6hcCBdiFLwUIqJwXQSmAWllsBvM34xrUTUJoREC1iUlAtennbGvu4rfGx7WvD6swftpTPz+FirLLCKhRFvBoHS4B1RZ5136AMIPKDKrtpcEMKvULgQLsghlUzoGoVYCAags9AdWmn7OtmUG1xse1rw+rMH7aUz8/hYqyy5hBjbKAR+twCai2yLv2A4QZVGZQbS8NZlCpXwgUYBfMoHIORK0CBFRb6AmoNv2cbc0MqjU+rn19WIXx0576+SlUlF3GDGqUBTxah0tAtUXetR8gzKAyg2p7aTCDSv1CoAC7YAaVcyBqFSCg2kJPQLXp52xrZlCt8XHt68MqjJ/21M9PoaLsMmZQoyzg0TpcAqot8q79AGEGlRlU20uDGVTqFwIF2AUzqJwDUasAAdUWetcC6qoBNmH8tMYpcVmzZvXz6qS/7Le8r+jRwk5trp1/SSQ49bMJ7XT9bKNzrjUzqM6NDT0LoQIEVJuYTn+Dpn+Mr00BmzXnH/WzKUBrZlA5B6JWgQULFkjRokWjdvwcOBWgAlSACgSvgJNXGIIflbMtmUF1dnzoXYgUmDp1qhw6dChEvbEbKkAFqAAViBYF8ufPL40bN46W4TpmnARUx4SCjoRTAacv8dM/W/SpH/WzKWCz5vyjfjYFaO1LAQIq50VUKMAPEFuYqR/1sylgs+b8o342BWzWTp9/ttE515qA6tzY0LMQKuD0Nxj6Zws29aN+NgVs1px/1M+mAK2ZQeUciFoF+AFiCz31o342BWzWnH/Uz6aAzdrp8882OudaM4Pq3NjQsxAqsHr1aqlYsWIIewxtV/TPpif1o342BWzWnH/Uz6YArZlB5RygAlSAClABKkAFqAAVcLwCzKA6PkR0kApQASpABagAFaAC0aUAATW64s3RJqLA+fPnJVOmTHGuuHTpkqRPn15SpEiRZNqdO3dOMmfOnGT3C/RGp06dUp1Sp04dqGmSXf/HH39Injx5kux+gdwI+mXLli0Qk7Bfe+HCBbn33nvDfp9gb+BEzeKPxelzzsmv2eR+z7t48aLcc889EhMTExvW5HjvD/b14VY7AqpbI8tx+a3Anj17ZOXKlTJs2DDZtm2b2p08eVKaNGkiqVKlkgMHDkjXrl2lefPmfvcZzIXr1q2T99577/9v70xgqyq6OH4SIqBo4gLaKgqKLKJxAaxFVFpFFpVVFCJuyCJQC7gvdWkVAgIGFBXZVxVQrAtqXUDFgCKyaIkRSBVRFDQq7iv1y+/km+by6Ht9be/rGx5nkgZt35t75j9z7/3Nf869I4ceeqh8++23UlBQIGeeeWZVqkrId9ChX79+cuyxx2p8ubm50qNHj4QcqzqVLl26VLp27SqlpaV73HCqU2cY312+fLkMHTpUWrVqJdz8Bg4cKN26dQuj6irXsXbtWrnuuuukUaNGOs5nzpwpbdq0qXJ9YX/RR83Ka6OvY873czbZ1zyuYx9//LH06tVLuA8ceeSRSbn2h33epEp9Bqip0pPWjiorMHbsWCkuLpYNGzaUASq/w1UaPXq07NixQ9LT0xUqDjrooCofp6Iv9u/fXzIyMhRieGr0q6++ksmTJ1f0tRr7O4DepUsX6dOnj+BocWHPzs6usePHc6CtW7dKXl6esHOYb4CKVsTWoUMHWbFihQwaNEg2bdoUT7MS9pmOHTvKLbfcIvy7ZMkSmTp1qrz++usJO15lK/ZRs8g2+DzmfD9nk33NY8yvXLlSJk6cKDt37lRATca1v7Lnxf7yeQPU/aWnrZ0xFQC22MrOOai4W4BE37595b///tMl/pKSEknkfsy4RTiS3bt3lxdeeEFBITMz05uea9u2rbqnRUVFkpWVJePGjZMWLVp4E99ff/0lF110kcydO1fj9A1QmfAwwalVq5bcdNNN8scff8iUKVOSqh86rVq1SvVav369dO7cWW/UvhQfNQtq4/uY8/2c9eWax9K+A9RkXPt9Od98i8MA1bcesXgSosCff/5ZrjPUqVMnqVOnjrqBQUC9/PLLhZ/evXtrPEcddZSsXr1aGjduXO34li1bpm5ssLCsOm3aNHnmmWfUoXzyySdl8ODBcvPNN1f7eJWtIFp8Z5xxhrRv317Gjx+vzu6uXbtkxowZla2+2p+PFh+u8+mnny64MtxwkgWo0eIjJ5abYE5OjmzZskVefPFFXVpPZiEvEReXFYLPPvtM+/fLL79MZkh7Hds3zYIBjhgxwosxF63DuG75cM5Giy8/P9+La14QUBN57ffqxNoHgjFA3Qc6yUKsvgLA1PXXX79XRUAhDyRFAur999+vDwKNHDlSdu/eLYcddpgCWRgPS+Gebd++fY9Y+B35iIWFhXL22WcLzgI5ns7Rrb4C8dcQLT5yFSdMmKDL/LjJxJkMt628+EiLYDnY5eyuWbNG/5v3U9auXTv+xofwyWj6AYHEePXVV8vtt98udevWDeFo1avivPPO0+XN1q1by4cffiiMe8DZl7Jt2zbvNHPaMMk8+OCDvRhz0frr5JNP9uKcjQXQPlzzgoCayGu/L+fVvhKHAeq+0lMWZ0IViARUbtKPPvqouq64mjxA9d577yU0BpangRfSCnBQWeZfvHhxQo9ZmcoHDBigTjIPci1YsEBee+01mT9/fmWqSNhnScPgIQdXmjVrJp9++qnwb/DJ3IQFEEfFOPRAMwDrS8Ghr1+/vtx2222aiwpwPfDAA76Ep6savmnmxNkXxpzP5yw6+nLNCwJqMq793pxwngVigOpZh1g4yVEgElDJD+Ti+cknn2iu4BtvvCFnnXVWQoN7++235aqrrtI8Rd4eMH36dHUpfSksAQ8bNkxd3YYNG8rjjz8uLPv7WJK5xB/LLeKpYVd4ICMZDnQwPueE8zveHsEk7PDDD/emS5kQ+aZZNHF8HHO+n7O+XPPoO8ZZgwYN9Hpf09d+b044zwIxQPWsQywcvxQgHy8tLa3G3vmJK8PT+zy04mvhFVy4blZSQ4F///1XeIcnY84Xtzk1lPWnFT6fs75e82r62u/PaPEnEgNUf/rCIjEFTAFTwBQwBUwBU8AUEBEDVBsGpoApYAqYAqaAKWAKmAJeKWCA6lV3WDCmgClgCpgCpoApYAqYAgaoNgZMAVPAFDAFTAFTwBQwBbxSwADVq+6wYEwBU8AUMAVMAVPAFDAFDFBtDJgCpkBoCvz8889lGxuEVmlIFfH6GMqBBx4YUo1WTVgK/PPPP8IPr1hL1cK5wRjk1VnxFnbA27Bhg5x00km6oUhk+eGHH/RXPr0aLN622edMgYoUMECtSCH7uylgClSoAC/JZ6epd999Vz97wgkn6Htczz///Aq/W1MfOO2004QbPlt7JqoAIYAEL+N/6KGHQjtMRkaGsO/7Rx99FFqdVPTyyy9rnXfddVfMejk+O3KxM1ciCjtrjRs3Tnbs2BET4GLF8dhjj+m7XPv161elEMPQmBhmzZqlbbjiiiv0h93neJUS71G+8MILZfTo0XHHt3DhQt2sgNdEHXHEEXt9b/bs2XLHHXfI559/ntJwH7dg9sGUUsAANaW60xpjCtS8ArxHk60y2ezgyiuvlDp16sjMmTM1EPacP/HEE2s+qHKOOHnyZCHWG2+8MWHx/PTTTwqgeghAAAALMklEQVRJHIPdx8Iq6Atch731Ldv/st1vaWlpzHegAvf06wcffBBWk/aoB1AGfvPy8nQ3q2glVhy8xxWnkd3fqlKqq/Fbb72lEzI2scDR5Hx45513hO1k2Y2OPd7Z3ax58+Zxh9exY0fdZnnRokXlfueXX37RLZnHjBmjoGrFFEglBQxQU6k3rS2mQBIUePXVV3Xnla5du5bt4842rWyfyU2zf//+MnXqVHXI2NkmMzNTt5EFCB588EF5/vnnpX379rq9a9u2baVXr15l223yvXPOOUe6deumN32WgdkzfuDAgbqrVb169WTKlClaHy4T9UyaNEndPmLq1KmTgg+fLyoqkr///ltv9mvWrJHx48fr79itCzA699xzdQkWN/Gpp57S/+7cubMAtjhi7PIFyFFWrFghwAN/Cy5LRwIqn8NVY8ecrKws/W+OP3fuXI0Z1w5dnn32Wf0hPnboQgtAhra0atVKtSoPUOPVjzhwKdEfyLvnnnuE5eHBgwfrDjoXX3yxLF26VNvlPtehQwfVgv3c+c6vv/6qO4cBgLQdhxx4coU9zKnjpZde0h2p2BL3zjvvFOq54IIL5JJLLpHc3FyZMWOG/tStW1duvfVWndQ8/PDD2ma2z8WBHjVqlNafnp6u/U8fAoD0VXlx0A/33XefHHLIIepe5+fnV/pMCGrMZhmML/qNlBBWB2jfAQccoOOV8YJbfu2112rMEydOlGXLlmkM33zzje4Ex65EQ4YM0fFJGxhPa9eujXvMf/HFF9K4cWPh/GLMswUnwEs8jO1GjRpp3TjGaM7Ys40WKt3t9gWPFTBA9bhzLDRTYF9QAKeQPd2feOIJwZGLLLhGOFvsqQ6kTJgwQcGDpfahQ4fq9wAkbvikCAAZwCCgxndw7XDHgAbgtlatWrJy5UpheRPwOe644xREAeQbbrhBfwCjY445RkMh3QAgYq954IZ4WrRooX8DOgApB39z5syRQYMG6dIsUAp4ACqAJZCGK8ZxKEDB008/LX379i1rchBQqZu2tGvXTkEXKCRO9roHTIE/lnubNGmiriFL7dnZ2bq9Lg4sgPbdd98p8ODClQeo8ei3evXqMhcbIATMAaglS5YoQKHlvHnzFBTREp1HjhypLjhg9corr5S1nb3dmQgwASGFARh0ZcGCBdpvQCrgD4BTP/oA50A/qSD33nuvAiTHZQth/qU+t8S/detW7We2ggXE6BPKrl27VAf6IDIO+qZnz546rhg3jAtXmChRf2QhXsaXK0FABfqIF1hnvPD9xYsX60QJsKYwTliCx8VEIyZHpGEA7W5pnrGek5OjUAm80u54+owxz7gE3Ldv36460h+MXyZljM2WLVsqtDvd0ZaxZMUUSBUFDFBTpSetHaZAkhQAsu6++269gV922WV7RQGQASUbN25UN47lSOBs3bp1urwMoOLmcYMF3HAFgQqAAVhiv3oAlaVMII58PhzSLl26KGTh6FEX3wcqAEJiAVABXxwvgIz6uMEDxriTmzdvVtcUmMEJw5ECEHG5gCFyB3EK33//fYUyYIY42QJx27Zt+nnAHAhxJQiouMEs6/LD8QBq2gOIt2nTRmPCNQU0HnnkEbn00ks1ZuCMXEXcO4AdBw3HLhagxtKPvE4czTfffFOBF3Ck4OiNGDGibImfdtNGcihxvXHtgCM0RDvqAZbJdyRtg0kAEOUKfwcQGQsAG+kU/FxzzTUKehybvkFbJjKA6Pz587We3bt3lwEqgAnccxwcRCY1pAA4QI0WR7QlftqDYx5ZgPDgcrsDVICeiQX9htuNTrjk3bt318kPfUPcAD2QzsSH9jIeKbi5w4cPLwNI2kn+KasBONYOUGP1GTo3bdpUJ0qcP6wARANU+pWYIidLSboc2GFNgdAUMEANTUqryBTYPxVwbhGuFk4lhaVIwJUbNaDFjdwBB4DEUi+f4aYKoAKOJSUl6tRxgwcEgFVu1AChA1RyMFlmJ88TkGMpmJs/y+C4d4DwqaeeWgaoQYgKAiqACTAAiIAqQAL4AJUOQmkH6QaFhYUaH8vLLg+TuI4++uiYgMoSLyAOZODiugKsAWH8SzvRg/poP44lEAcou9KnTx+FvFiAGks/JgbNmjXT6nDzAHpcy0hAZQm5R48emi/Zu3dv1f3HH3+UU045RZf2XduBVvIsiX/s2LF7DHo0pjBhAOLoE9pDwe1jYoEzCqi5AhQTj3NQmaDgXDMx4bOuDxygRosjGqCy9I62kQXwpQ9dcYDKeOW4bjwD2aSX8HeAHqf2ueee039x1ulDABWXHFcVkKTP3UNbuKGMUf4fN94Baqw+Y6KFW+xyVl2uMBqQAsFkjXjQFLcaODZA3T+vv6ncagPUVO5da5spUAMK/Pbbbwpg5DJyg+a/AQwKOY8AEjDJTRvwIS+VGy2gA6jGC6i4j9zkASfACLADVjgmMMTSKw5u0EHFzcOBojhAxSFjKRt3tKCgQF1AgICldZw6lsH5XVpamuYeskTNzT/4gE48gAqwsKRPuwEMnFZcU+CGNhADxeXuAsY8qQ1M4voB/sTK8jIwVVVAZRmdFAvgiDiAYJalf//9d13Kx8XGJcbJBXwAfJaimVzw/0wKgm2PBag4ny4XFMfULYeTVoDTyASAY+EKozdt5DuAlgNUoIyUAOLBVXb9Fw+gEi8pBscff3zZyAeSWQaPLO4BpkhApb2kFpD3ykSCOHH9mdCw9E86BhMa3HMmZIxLAJXzgPFH/6Ed+aBAM+MMHQBZJl/xACrjBh3cWxPcqgMTMCYN1ING6EYaAP3IuRZsdw2c+nYIUyChChigJlReq9wU2D8UIC+Q1+FwM6dwA2d5FGeHJXngCBB1xaUD8Hse9CjPTQJmuPk7B5UHiCiAMBCFE8gyPO4pnwMMcLuoa9WqVboMHQmoDvK4qQOh1EWsxAH0AlU4Yzh6FJxCYAoXMQhpbjk7cok/8jVTuIHk6LrjkG7gXr0FuACPuLiAGIW8VuAQB5Li8jyjvQIpXv3cMjltRTvaxzFId6DfKPSTS9dwfUjbAf5g27/++mtNRSjPQQWomYy4h65In1i+fLkUFxerhowP0jeYCFBYRmdyQX4ugIprilvJRASgBziBesYX2gLXzkGNjIN6AWHncLqxRp8zLiILeaHkM7sS1Jj0BtpAv1FYQsdZxr0EgNEJIASicVzRiYlEcIzzPcY/8OoccYCyoj5jvDMJcKkW1EP6Cv1APByTz+A84w4DvEx60M6KKZBKChigplJvWltMgSQrADTgzuGi8sRzsJCfyQ2W3E2WeitTcKLIXwVsuBGT6+ieWAasuKEDTZV5ipnvsdQPhAZBhbhwSHngBZcTCK5OIb+S49AGYCuewufr168f6rst6RuWriNjwEnl7QjuRfD0E6BO/me88cbTpuBnqB/Ny3vBPEvivE0AB5o0CdxMxg1QVlEBRkkBCeOF//QbT9K7h5I4NhritjMWeSgKp5QlffJMGdfRiksFYJwG0woqak/w74xX9z5UNyZpL8fFpXerFpWp0z5rCvisgAGqz71jsZkCpoAq4ACVfDsrqa0A7jRuOBMdVyLfGJAsBYBfckOdw04cwderRYuLyQ7OJ44xy/VhFeCYNz7gsDL5sGIKpJICBqip1JvWFlMgRRXgaXZcMfcQToo205r1fwVwdUlB+P777xXseKLdl4KzSp4qm1DgMvMAWTwuO+BNHq17xVkY7eHBQ44dy70N4zhWhymQDAUMUJOhuh3TFDAFTAFTwBQwBUwBUyCqAgaoNjhMAVPAFDAFTAFTwBQwBbxSwADVq+6wYEwBU8AUMAVMAVPAFDAF/gfWnI81WDsBTQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<VegaLite 4 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linker.match_weights_chart()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>match_weight</th>\n",
       "      <th>match_probability</th>\n",
       "      <th>unique_id_l</th>\n",
       "      <th>unique_id_r</th>\n",
       "      <th>first_name_l</th>\n",
       "      <th>first_name_r</th>\n",
       "      <th>gamma_first_name</th>\n",
       "      <th>bf_first_name</th>\n",
       "      <th>surname_l</th>\n",
       "      <th>surname_r</th>\n",
       "      <th>...</th>\n",
       "      <th>bf_postcode_fake</th>\n",
       "      <th>birth_place_l</th>\n",
       "      <th>birth_place_r</th>\n",
       "      <th>gamma_birth_place</th>\n",
       "      <th>bf_birth_place</th>\n",
       "      <th>occupation_l</th>\n",
       "      <th>occupation_r</th>\n",
       "      <th>gamma_occupation</th>\n",
       "      <th>bf_occupation</th>\n",
       "      <th>match_key</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19.641479</td>\n",
       "      <td>0.999999</td>\n",
       "      <td>Q2296770-1</td>\n",
       "      <td>Q2296770-14</td>\n",
       "      <td>thomas</td>\n",
       "      <td>thomas</td>\n",
       "      <td>3</td>\n",
       "      <td>43.427885</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>...</td>\n",
       "      <td>249.467633</td>\n",
       "      <td>devon</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>politician</td>\n",
       "      <td>politician</td>\n",
       "      <td>1</td>\n",
       "      <td>22.77665</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.048632</td>\n",
       "      <td>0.970672</td>\n",
       "      <td>Q2296770-10</td>\n",
       "      <td>Q2296770-14</td>\n",
       "      <td>thomas</td>\n",
       "      <td>thomas</td>\n",
       "      <td>3</td>\n",
       "      <td>43.427885</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>chudleigh</td>\n",
       "      <td>...</td>\n",
       "      <td>0.171118</td>\n",
       "      <td>devon</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>politician</td>\n",
       "      <td>politician</td>\n",
       "      <td>1</td>\n",
       "      <td>22.77665</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25.497744</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Q1443188-1</td>\n",
       "      <td>Q1443188-3</td>\n",
       "      <td>frank</td>\n",
       "      <td>frank</td>\n",
       "      <td>3</td>\n",
       "      <td>43.427885</td>\n",
       "      <td>brightman</td>\n",
       "      <td>brightman</td>\n",
       "      <td>...</td>\n",
       "      <td>4874.614882</td>\n",
       "      <td>bristol</td>\n",
       "      <td>bristol, city of</td>\n",
       "      <td>0</td>\n",
       "      <td>0.160976</td>\n",
       "      <td>liturgist</td>\n",
       "      <td>liturgist</td>\n",
       "      <td>1</td>\n",
       "      <td>22.77665</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>25.497744</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Q1443188-2</td>\n",
       "      <td>Q1443188-3</td>\n",
       "      <td>frank</td>\n",
       "      <td>frank</td>\n",
       "      <td>3</td>\n",
       "      <td>43.427885</td>\n",
       "      <td>brightman</td>\n",
       "      <td>brightman</td>\n",
       "      <td>...</td>\n",
       "      <td>4874.614882</td>\n",
       "      <td>bristol</td>\n",
       "      <td>bristol, city of</td>\n",
       "      <td>0</td>\n",
       "      <td>0.160976</td>\n",
       "      <td>liturgist</td>\n",
       "      <td>liturgist</td>\n",
       "      <td>1</td>\n",
       "      <td>22.77665</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9.131831</td>\n",
       "      <td>0.998221</td>\n",
       "      <td>Q1443188-4</td>\n",
       "      <td>Q1443188-5</td>\n",
       "      <td>francis</td>\n",
       "      <td>francis</td>\n",
       "      <td>3</td>\n",
       "      <td>43.427885</td>\n",
       "      <td>brightman</td>\n",
       "      <td>brightman</td>\n",
       "      <td>...</td>\n",
       "      <td>0.171118</td>\n",
       "      <td>NaN</td>\n",
       "      <td>bristol, city of</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>liturgist</td>\n",
       "      <td>liturgist</td>\n",
       "      <td>1</td>\n",
       "      <td>22.77665</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   match_weight  match_probability  unique_id_l  unique_id_r first_name_l  \\\n",
       "0     19.641479           0.999999   Q2296770-1  Q2296770-14       thomas   \n",
       "1      5.048632           0.970672  Q2296770-10  Q2296770-14       thomas   \n",
       "2     25.497744           1.000000   Q1443188-1   Q1443188-3        frank   \n",
       "3     25.497744           1.000000   Q1443188-2   Q1443188-3        frank   \n",
       "4      9.131831           0.998221   Q1443188-4   Q1443188-5      francis   \n",
       "\n",
       "  first_name_r  gamma_first_name  bf_first_name  surname_l  surname_r  ...  \\\n",
       "0       thomas                 3      43.427885  chudleigh  chudleigh  ...   \n",
       "1       thomas                 3      43.427885  chudleigh  chudleigh  ...   \n",
       "2        frank                 3      43.427885  brightman  brightman  ...   \n",
       "3        frank                 3      43.427885  brightman  brightman  ...   \n",
       "4      francis                 3      43.427885  brightman  brightman  ...   \n",
       "\n",
       "   bf_postcode_fake  birth_place_l     birth_place_r gamma_birth_place  \\\n",
       "0        249.467633          devon               NaN                -1   \n",
       "1          0.171118          devon               NaN                -1   \n",
       "2       4874.614882        bristol  bristol, city of                 0   \n",
       "3       4874.614882        bristol  bristol, city of                 0   \n",
       "4          0.171118            NaN  bristol, city of                -1   \n",
       "\n",
       "   bf_birth_place  occupation_l occupation_r gamma_occupation  bf_occupation  \\\n",
       "0        1.000000    politician   politician                1       22.77665   \n",
       "1        1.000000    politician   politician                1       22.77665   \n",
       "2        0.160976     liturgist    liturgist                1       22.77665   \n",
       "3        0.160976     liturgist    liturgist                1       22.77665   \n",
       "4        1.000000     liturgist    liturgist                1       22.77665   \n",
       "\n",
       "   match_key  \n",
       "0          0  \n",
       "1          0  \n",
       "2          0  \n",
       "3          0  \n",
       "4          0  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_predict = linker.predict()\n",
    "df_e = df_predict.as_pandas_dataframe(limit=5)\n",
    "df_e"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can also view rows in this dataset as a waterfall chart as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v4+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v5.2.0.json",
       "config": {
        "view": {
         "continuousHeight": 300,
         "continuousWidth": 400
        }
       },
       "data": {
        "values": [
         {
          "bar_sort_order": 0,
          "bayes_factor": 0.00018003240583304995,
          "bayes_factor_description": null,
          "column_name": "Prior",
          "comparison_vector_value": null,
          "label_for_charts": "Starting match weight (prior)",
          "log2_bayes_factor": -12.439455764512674,
          "m_probability": null,
          "record_number": 0,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 1,
          "bayes_factor": 43.42788509499787,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.43 times more likely to be a match",
          "column_name": "first_name",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.440549791237722,
          "m_probability": 0.5569091821713592,
          "record_number": 0,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.01282376935816995,
          "value_l": "thomas",
          "value_r": "thomas"
         },
         {
          "bar_sort_order": 2,
          "bayes_factor": 1086.1401447009125,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,086.14 times more likely to be a match",
          "column_name": "surname",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.084994550806584,
          "m_probability": 0.7894580640061504,
          "record_number": 0,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.0007268473298383989,
          "value_l": "chudleigh",
          "value_r": "chudleigh"
         },
         {
          "bar_sort_order": 3,
          "bayes_factor": 16.949819994845715,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 16.95 times more likely to be a match",
          "column_name": "dob",
          "comparison_vector_value": 2,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": 4.083198047073621,
          "m_probability": 0.3385185513724552,
          "record_number": 0,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "term_frequency_adjustment": false,
          "u_probability": 0.01997180804724744,
          "value_l": "1630-08-01",
          "value_r": "1638-08-01"
         },
         {
          "bar_sort_order": 4,
          "bayes_factor": 249.4676329581296,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 249.47 times more likely to be a match",
          "column_name": "postcode_fake",
          "comparison_vector_value": 1,
          "label_for_charts": "levenshtein <= 2",
          "log2_bayes_factor": 7.962708835798494,
          "m_probability": 0.14224626801708795,
          "record_number": 0,
          "sql_condition": "levenshtein(\"postcode_fake_l\", \"postcode_fake_r\") <= 2",
          "term_frequency_adjustment": false,
          "u_probability": 0.0005701992933125814,
          "value_l": "tq13 8df",
          "value_r": "tq1w 8df"
         },
         {
          "bar_sort_order": 5,
          "bayes_factor": 1,
          "bayes_factor_description": "If comparison level is `null` then comparison is 1.00 times more likely to be a match",
          "column_name": "birth_place",
          "comparison_vector_value": -1,
          "label_for_charts": "Null",
          "log2_bayes_factor": 0,
          "m_probability": null,
          "record_number": 0,
          "sql_condition": "\"birth_place_l\" IS NULL OR \"birth_place_r\" IS NULL",
          "term_frequency_adjustment": false,
          "u_probability": null,
          "value_l": "devon",
          "value_r": "nan"
         },
         {
          "bar_sort_order": 6,
          "bayes_factor": 22.776649717532823,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.78 times more likely to be a match",
          "column_name": "occupation",
          "comparison_vector_value": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.509483647404612,
          "m_probability": 0.9013269182036521,
          "record_number": 0,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.039572409875093965,
          "value_l": "politician",
          "value_r": "politician"
         },
         {
          "bar_sort_order": 7,
          "bayes_factor": 817851.3420256926,
          "bayes_factor_description": null,
          "column_name": "Final score",
          "comparison_vector_value": null,
          "label_for_charts": "Final score",
          "log2_bayes_factor": 19.641479107808358,
          "m_probability": null,
          "record_number": 0,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 0,
          "bayes_factor": 0.00018003240583304995,
          "bayes_factor_description": null,
          "column_name": "Prior",
          "comparison_vector_value": null,
          "label_for_charts": "Starting match weight (prior)",
          "log2_bayes_factor": -12.439455764512674,
          "m_probability": null,
          "record_number": 1,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 1,
          "bayes_factor": 43.42788509499787,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.43 times more likely to be a match",
          "column_name": "first_name",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.440549791237722,
          "m_probability": 0.5569091821713592,
          "record_number": 1,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.01282376935816995,
          "value_l": "thomas",
          "value_r": "thomas"
         },
         {
          "bar_sort_order": 2,
          "bayes_factor": 1086.1401447009125,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,086.14 times more likely to be a match",
          "column_name": "surname",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.084994550806584,
          "m_probability": 0.7894580640061504,
          "record_number": 1,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.0007268473298383989,
          "value_l": "chudleigh",
          "value_r": "chudleigh"
         },
         {
          "bar_sort_order": 3,
          "bayes_factor": 1,
          "bayes_factor_description": "If comparison level is `null` then comparison is 1.00 times more likely to be a match",
          "column_name": "dob",
          "comparison_vector_value": -1,
          "label_for_charts": "Null",
          "log2_bayes_factor": 0,
          "m_probability": null,
          "record_number": 1,
          "sql_condition": "\"dob_l\" IS NULL OR \"dob_r\" IS NULL",
          "term_frequency_adjustment": false,
          "u_probability": null,
          "value_l": "nan",
          "value_r": "1638-08-01"
         },
         {
          "bar_sort_order": 4,
          "bayes_factor": 0.1711176250398338,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.84 times less likely to be a match",
          "column_name": "postcode_fake",
          "comparison_vector_value": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.546939730454512,
          "m_probability": 0.1709959460662125,
          "record_number": 1,
          "sql_condition": "ELSE",
          "term_frequency_adjustment": false,
          "u_probability": 0.9992889161850337,
          "value_l": "tq13 8jr",
          "value_r": "tq1w 8df"
         },
         {
          "bar_sort_order": 5,
          "bayes_factor": 1,
          "bayes_factor_description": "If comparison level is `null` then comparison is 1.00 times more likely to be a match",
          "column_name": "birth_place",
          "comparison_vector_value": -1,
          "label_for_charts": "Null",
          "log2_bayes_factor": 0,
          "m_probability": null,
          "record_number": 1,
          "sql_condition": "\"birth_place_l\" IS NULL OR \"birth_place_r\" IS NULL",
          "term_frequency_adjustment": false,
          "u_probability": null,
          "value_l": "devon",
          "value_r": "nan"
         },
         {
          "bar_sort_order": 6,
          "bayes_factor": 22.776649717532823,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.78 times more likely to be a match",
          "column_name": "occupation",
          "comparison_vector_value": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.509483647404612,
          "m_probability": 0.9013269182036521,
          "record_number": 1,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.039572409875093965,
          "value_l": "politician",
          "value_r": "politician"
         },
         {
          "bar_sort_order": 7,
          "bayes_factor": 33.09709053362309,
          "bayes_factor_description": null,
          "column_name": "Final score",
          "comparison_vector_value": null,
          "label_for_charts": "Final score",
          "log2_bayes_factor": 5.048632494481733,
          "m_probability": null,
          "record_number": 1,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 0,
          "bayes_factor": 0.00018003240583304995,
          "bayes_factor_description": null,
          "column_name": "Prior",
          "comparison_vector_value": null,
          "label_for_charts": "Starting match weight (prior)",
          "log2_bayes_factor": -12.439455764512674,
          "m_probability": null,
          "record_number": 2,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 1,
          "bayes_factor": 43.42788509499787,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.43 times more likely to be a match",
          "column_name": "first_name",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.440549791237722,
          "m_probability": 0.5569091821713592,
          "record_number": 2,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.01282376935816995,
          "value_l": "frank",
          "value_r": "frank"
         },
         {
          "bar_sort_order": 2,
          "bayes_factor": 1086.1401447009125,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,086.14 times more likely to be a match",
          "column_name": "surname",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.084994550806584,
          "m_probability": 0.7894580640061504,
          "record_number": 2,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.0007268473298383989,
          "value_l": "brightman",
          "value_r": "brightman"
         },
         {
          "bar_sort_order": 3,
          "bayes_factor": 312.16784723322775,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 312.17 times more likely to be a match",
          "column_name": "dob",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.286178139563434,
          "m_probability": 0.6113456435064976,
          "record_number": 2,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.001958387607580057,
          "value_l": "1856-06-18",
          "value_r": "1856-06-18"
         },
         {
          "bar_sort_order": 4,
          "bayes_factor": 4874.614882141059,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,874.61 times more likely to be a match",
          "column_name": "postcode_fake",
          "comparison_vector_value": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.251072528227413,
          "m_probability": 0.6867577859166996,
          "record_number": 2,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.00014088452165375115,
          "value_l": "bs2 0el",
          "value_r": "bs2 0el"
         },
         {
          "bar_sort_order": 5,
          "bayes_factor": 0.16097645294413918,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.21 times less likely to be a match",
          "column_name": "birth_place",
          "comparison_vector_value": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.635078423349897,
          "m_probability": 0.16009531671123883,
          "record_number": 2,
          "sql_condition": "ELSE",
          "term_frequency_adjustment": false,
          "u_probability": 0.9945263035879781,
          "value_l": "bristol",
          "value_r": "bristol, city of"
         },
         {
          "bar_sort_order": 6,
          "bayes_factor": 22.776649717532823,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.78 times more likely to be a match",
          "column_name": "occupation",
          "comparison_vector_value": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.509483647404612,
          "m_probability": 0.9013269182036521,
          "record_number": 2,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.039572409875093965,
          "value_l": "liturgist",
          "value_r": "liturgist"
         },
         {
          "bar_sort_order": 7,
          "bayes_factor": 47379001.850928925,
          "bayes_factor_description": null,
          "column_name": "Final score",
          "comparison_vector_value": null,
          "label_for_charts": "Final score",
          "log2_bayes_factor": 25.497744469377196,
          "m_probability": null,
          "record_number": 2,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 0,
          "bayes_factor": 0.00018003240583304995,
          "bayes_factor_description": null,
          "column_name": "Prior",
          "comparison_vector_value": null,
          "label_for_charts": "Starting match weight (prior)",
          "log2_bayes_factor": -12.439455764512674,
          "m_probability": null,
          "record_number": 3,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 1,
          "bayes_factor": 43.42788509499787,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.43 times more likely to be a match",
          "column_name": "first_name",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.440549791237722,
          "m_probability": 0.5569091821713592,
          "record_number": 3,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.01282376935816995,
          "value_l": "frank",
          "value_r": "frank"
         },
         {
          "bar_sort_order": 2,
          "bayes_factor": 1086.1401447009125,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,086.14 times more likely to be a match",
          "column_name": "surname",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.084994550806584,
          "m_probability": 0.7894580640061504,
          "record_number": 3,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.0007268473298383989,
          "value_l": "brightman",
          "value_r": "brightman"
         },
         {
          "bar_sort_order": 3,
          "bayes_factor": 312.16784723322775,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 312.17 times more likely to be a match",
          "column_name": "dob",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.286178139563434,
          "m_probability": 0.6113456435064976,
          "record_number": 3,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.001958387607580057,
          "value_l": "1856-06-18",
          "value_r": "1856-06-18"
         },
         {
          "bar_sort_order": 4,
          "bayes_factor": 4874.614882141059,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 4,874.61 times more likely to be a match",
          "column_name": "postcode_fake",
          "comparison_vector_value": 2,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 12.251072528227413,
          "m_probability": 0.6867577859166996,
          "record_number": 3,
          "sql_condition": "\"postcode_fake_l\" = \"postcode_fake_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.00014088452165375115,
          "value_l": "bs2 0el",
          "value_r": "bs2 0el"
         },
         {
          "bar_sort_order": 5,
          "bayes_factor": 0.16097645294413918,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.21 times less likely to be a match",
          "column_name": "birth_place",
          "comparison_vector_value": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.635078423349897,
          "m_probability": 0.16009531671123883,
          "record_number": 3,
          "sql_condition": "ELSE",
          "term_frequency_adjustment": false,
          "u_probability": 0.9945263035879781,
          "value_l": "bristol",
          "value_r": "bristol, city of"
         },
         {
          "bar_sort_order": 6,
          "bayes_factor": 22.776649717532823,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.78 times more likely to be a match",
          "column_name": "occupation",
          "comparison_vector_value": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.509483647404612,
          "m_probability": 0.9013269182036521,
          "record_number": 3,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.039572409875093965,
          "value_l": "liturgist",
          "value_r": "liturgist"
         },
         {
          "bar_sort_order": 7,
          "bayes_factor": 47379001.850928925,
          "bayes_factor_description": null,
          "column_name": "Final score",
          "comparison_vector_value": null,
          "label_for_charts": "Final score",
          "log2_bayes_factor": 25.497744469377196,
          "m_probability": null,
          "record_number": 3,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 0,
          "bayes_factor": 0.00018003240583304995,
          "bayes_factor_description": null,
          "column_name": "Prior",
          "comparison_vector_value": null,
          "label_for_charts": "Starting match weight (prior)",
          "log2_bayes_factor": -12.439455764512674,
          "m_probability": null,
          "record_number": 4,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         },
         {
          "bar_sort_order": 1,
          "bayes_factor": 43.42788509499787,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 43.43 times more likely to be a match",
          "column_name": "first_name",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 5.440549791237722,
          "m_probability": 0.5569091821713592,
          "record_number": 4,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.01282376935816995,
          "value_l": "francis",
          "value_r": "francis"
         },
         {
          "bar_sort_order": 2,
          "bayes_factor": 1086.1401447009125,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 1,086.14 times more likely to be a match",
          "column_name": "surname",
          "comparison_vector_value": 3,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 10.084994550806584,
          "m_probability": 0.7894580640061504,
          "record_number": 4,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.0007268473298383989,
          "value_l": "brightman",
          "value_r": "brightman"
         },
         {
          "bar_sort_order": 3,
          "bayes_factor": 16.949819994845715,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 16.95 times more likely to be a match",
          "column_name": "dob",
          "comparison_vector_value": 2,
          "label_for_charts": "levenshtein <= 1",
          "log2_bayes_factor": 4.083198047073621,
          "m_probability": 0.3385185513724552,
          "record_number": 4,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "term_frequency_adjustment": false,
          "u_probability": 0.01997180804724744,
          "value_l": "1856-06-18",
          "value_r": "1856-86-18"
         },
         {
          "bar_sort_order": 4,
          "bayes_factor": 0.1711176250398338,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  5.84 times less likely to be a match",
          "column_name": "postcode_fake",
          "comparison_vector_value": 0,
          "label_for_charts": "All other comparisons",
          "log2_bayes_factor": -2.546939730454512,
          "m_probability": 0.1709959460662125,
          "record_number": 4,
          "sql_condition": "ELSE",
          "term_frequency_adjustment": false,
          "u_probability": 0.9992889161850337,
          "value_l": "bs2 0el",
          "value_r": "cf83 4jg"
         },
         {
          "bar_sort_order": 5,
          "bayes_factor": 1,
          "bayes_factor_description": "If comparison level is `null` then comparison is 1.00 times more likely to be a match",
          "column_name": "birth_place",
          "comparison_vector_value": -1,
          "label_for_charts": "Null",
          "log2_bayes_factor": 0,
          "m_probability": null,
          "record_number": 4,
          "sql_condition": "\"birth_place_l\" IS NULL OR \"birth_place_r\" IS NULL",
          "term_frequency_adjustment": false,
          "u_probability": null,
          "value_l": "nan",
          "value_r": "bristol, city of"
         },
         {
          "bar_sort_order": 6,
          "bayes_factor": 22.776649717532823,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 22.78 times more likely to be a match",
          "column_name": "occupation",
          "comparison_vector_value": 1,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 4.509483647404612,
          "m_probability": 0.9013269182036521,
          "record_number": 4,
          "sql_condition": "\"occupation_l\" = \"occupation_r\"",
          "term_frequency_adjustment": false,
          "u_probability": 0.039572409875093965,
          "value_l": "liturgist",
          "value_r": "liturgist"
         },
         {
          "bar_sort_order": 7,
          "bayes_factor": 560.9897268980236,
          "bayes_factor_description": null,
          "column_name": "Final score",
          "comparison_vector_value": null,
          "label_for_charts": "Final score",
          "log2_bayes_factor": 9.131830541555354,
          "m_probability": null,
          "record_number": 4,
          "sql_condition": null,
          "term_frequency_adjustment": null,
          "u_probability": null,
          "value_l": "",
          "value_r": ""
         }
        ]
       },
       "height": 450,
       "layer": [
        {
         "layer": [
          {
           "encoding": {
            "color": {
             "value": "black"
            },
            "size": {
             "value": 0.5
            },
            "y": {
             "field": "zero",
             "type": "quantitative"
            }
           },
           "mark": "rule"
          },
          {
           "encoding": {
            "color": {
             "condition": {
              "test": "(datum.log2_bayes_factor < 0)",
              "value": "red"
             },
             "value": "green"
            },
            "opacity": {
             "condition": {
              "test": "datum.column_name == 'Prior match weight' || datum.column_name == 'Final score'",
              "value": 1
             },
             "value": 0.5
            },
            "tooltip": [
             {
              "field": "column_name",
              "title": "Comparison column",
              "type": "nominal"
             },
             {
              "field": "value_l",
              "title": "Value (L)",
              "type": "nominal"
             },
             {
              "field": "value_r",
              "title": "Value (R)",
              "type": "nominal"
             },
             {
              "field": "label_for_charts",
              "title": "Label",
              "type": "ordinal"
             },
             {
              "field": "sql_condition",
              "title": "SQL condition",
              "type": "nominal"
             },
             {
              "field": "comparison_vector_value",
              "title": "Comparison vector value",
              "type": "nominal"
             },
             {
              "field": "bayes_factor",
              "format": ",.4f",
              "title": "Bayes factor = m/u",
              "type": "quantitative"
             },
             {
              "field": "log2_bayes_factor",
              "format": ",.4f",
              "title": "Match weight = log2(m/u)",
              "type": "quantitative"
             },
             {
              "field": "prob",
              "format": ".4f",
              "title": "Adjusted match score",
              "type": "quantitative"
             },
             {
              "field": "bayes_factor_description",
              "title": "Match weight description",
              "type": "nominal"
             }
            ],
            "x": {
             "axis": {
              "grid": true,
              "labelAlign": "center",
              "labelAngle": -20,
              "labelExpr": "datum.value == 'Prior' || datum.value == 'Final score' ? '' : datum.value",
              "labelPadding": 10,
              "tickBand": "extent",
              "title": "Column"
             },
             "field": "column_name",
             "sort": {
              "field": "bar_sort_order",
              "order": "ascending"
             },
             "type": "nominal"
            },
            "y": {
             "axis": {
              "grid": false,
              "orient": "left",
              "title": "log2(Bayes factor)"
             },
             "field": "previous_sum",
             "type": "quantitative"
            },
            "y2": {
             "field": "sum"
            }
           },
           "mark": {
            "type": "bar",
            "width": 60
           }
          },
          {
           "encoding": {
            "color": {
             "value": "white"
            },
            "text": {
             "condition": {
              "field": "log2_bayes_factor",
              "format": ".2f",
              "test": "abs(datum.log2_bayes_factor) > 1",
              "type": "nominal"
             },
             "value": ""
            },
            "x": {
             "axis": {
              "labelAngle": 0,
              "title": "Column"
             },
             "field": "column_name",
             "sort": {
              "field": "bar_sort_order",
              "order": "ascending"
             },
             "type": "nominal"
            },
            "y": {
             "axis": {
              "orient": "left"
             },
             "field": "center",
             "type": "quantitative"
            }
           },
           "mark": {
            "fontWeight": "bold",
            "type": "text"
           }
          },
          {
           "encoding": {
            "color": {
             "value": "black"
            },
            "text": {
             "field": "column_name",
             "type": "nominal"
            },
            "x": {
             "axis": {
              "labelAngle": 0,
              "title": "Column"
             },
             "field": "column_name",
             "sort": {
              "field": "bar_sort_order",
              "order": "ascending"
             },
             "type": "nominal"
            },
            "y": {
             "field": "sum_top",
             "type": "quantitative"
            }
           },
           "mark": {
            "baseline": "bottom",
            "dy": -25,
            "fontWeight": "bold",
            "type": "text"
           }
          },
          {
           "encoding": {
            "color": {
             "value": "grey"
            },
            "text": {
             "field": "value_l",
             "type": "nominal"
            },
            "x": {
             "axis": {
              "labelAngle": 0,
              "title": "Column"
             },
             "field": "column_name",
             "sort": {
              "field": "bar_sort_order",
              "order": "ascending"
             },
             "type": "nominal"
            },
            "y": {
             "field": "sum_top",
             "type": "quantitative"
            }
           },
           "mark": {
            "baseline": "bottom",
            "dy": -13,
            "fontSize": 8,
            "type": "text"
           }
          },
          {
           "encoding": {
            "color": {
             "value": "grey"
            },
            "text": {
             "field": "value_r",
             "type": "nominal"
            },
            "x": {
             "axis": {
              "labelAngle": 0,
              "title": "Column"
             },
             "field": "column_name",
             "sort": {
              "field": "bar_sort_order",
              "order": "ascending"
             },
             "type": "nominal"
            },
            "y": {
             "field": "sum_top",
             "type": "quantitative"
            }
           },
           "mark": {
            "baseline": "bottom",
            "dy": -5,
            "fontSize": 8,
            "type": "text"
           }
          }
         ]
        },
        {
         "encoding": {
          "x": {
           "axis": {
            "labelAngle": 0,
            "title": "Column"
           },
           "field": "column_name",
           "sort": {
            "field": "bar_sort_order",
            "order": "ascending"
           },
           "type": "nominal"
          },
          "x2": {
           "field": "lead"
          },
          "y": {
           "axis": {
            "labelExpr": "format(1 / (1 + pow(2, -1*datum.value)), '.2r')",
            "orient": "right",
            "title": "Probability"
           },
           "field": "sum",
           "scale": {
            "zero": false
           },
           "type": "quantitative"
          }
         },
         "mark": {
          "color": "black",
          "strokeWidth": 2,
          "type": "rule",
          "x2Offset": 30,
          "xOffset": -30
         }
        }
       ],
       "params": [
        {
         "bind": {
          "input": "range",
          "max": 4,
          "min": 0,
          "step": 1
         },
         "description": "Filter by the interation number",
         "name": "record_number",
         "value": 0
        }
       ],
       "resolve": {
        "axis": {
         "y": "independent"
        }
       },
       "title": {
        "subtitle": "How each comparison contributes to the final match score",
        "text": "Match weights waterfall chart"
       },
       "transform": [
        {
         "filter": "(datum.record_number == record_number)"
        },
        {
         "frame": [
          null,
          0
         ],
         "window": [
          {
           "as": "sum",
           "field": "log2_bayes_factor",
           "op": "sum"
          },
          {
           "as": "lead",
           "field": "column_name",
           "op": "lead"
          }
         ]
        },
        {
         "as": "sum",
         "calculate": "datum.column_name === \"Final score\" ? datum.sum - datum.log2_bayes_factor : datum.sum"
        },
        {
         "as": "lead",
         "calculate": "datum.lead === null ? datum.column_name : datum.lead"
        },
        {
         "as": "previous_sum",
         "calculate": "datum.column_name === \"Final score\" || datum.column_name === \"Prior match weight\" ? 0 : datum.sum - datum.log2_bayes_factor"
        },
        {
         "as": "top_label",
         "calculate": "datum.sum > datum.previous_sum ? datum.column_name : \"\""
        },
        {
         "as": "bottom_label",
         "calculate": "datum.sum < datum.previous_sum ? datum.column_name : \"\""
        },
        {
         "as": "sum_top",
         "calculate": "datum.sum > datum.previous_sum ? datum.sum : datum.previous_sum"
        },
        {
         "as": "sum_bottom",
         "calculate": "datum.sum < datum.previous_sum ? datum.sum : datum.previous_sum"
        },
        {
         "as": "center",
         "calculate": "(datum.sum + datum.previous_sum) / 2"
        },
        {
         "as": "text_log2_bayes_factor",
         "calculate": "(datum.log2_bayes_factor > 0 ? \"+\" : \"\") + datum.log2_bayes_factor"
        },
        {
         "as": "dy",
         "calculate": "datum.sum < datum.previous_sum ? 4 : -4"
        },
        {
         "as": "baseline",
         "calculate": "datum.sum < datum.previous_sum ? \"top\" : \"bottom\""
        },
        {
         "as": "prob",
         "calculate": "1. / (1 + pow(2, -1.*datum.sum))"
        },
        {
         "as": "zero",
         "calculate": "0*datum.sum"
        }
       ],
       "width": {
        "step": 75
       }
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 4 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from splink.charts import waterfall_chart\n",
    "records_to_plot = df_e.to_dict(orient=\"records\")\n",
    "linker.waterfall_chart(records_to_plot, filter_nulls=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Completed iteration 1, root rows count 945\n",
      "Completed iteration 2, root rows count 312\n",
      "Completed iteration 3, root rows count 289\n",
      "Completed iteration 4, root rows count 149\n",
      "Completed iteration 5, root rows count 77\n",
      "Completed iteration 6, root rows count 47\n",
      "Completed iteration 7, root rows count 44\n",
      "Completed iteration 8, root rows count 19\n",
      "Completed iteration 9, root rows count 0\n"
     ]
    }
   ],
   "source": [
    "clusters = linker.cluster_pairwise_predictions_at_threshold(df_predict, threshold_match_probability=0.95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "linker.cluster_studio_dashboard(df_predict, clusters, \"50k_cluster.html\", sampling_method='by_cluster_size', overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"1200\"\n",
       "            src=\"./50k_cluster.html\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x7fe94b63b040>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import IFrame\n",
    "\n",
    "IFrame(\n",
    "    src=\"./50k_cluster.html\", width=\"100%\", height=1200\n",
    ")  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  },
  "vscode": {
   "interpreter": {
    "hash": "cc173ace240fa2ad02472fa75d75ddd885c067b637b7c50cecea7593995ac3de"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}