{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estimating m from a sample of pairwise labels\n",
    "\n",
    "In this example, we estimate the m probabilities of the model from a table containing pairwise record comparisons which we know are 'true' matches.  For example, these may be the result of work by a clerical team who have manually labelled a sample of matches.\n",
    "\n",
    "The table must be in the following format:\n",
    "\n",
    "|source_dataset_l|unique_id_l|source_dataset_r|unique_id_r|\n",
    "|----------------|-----------|----------------|-----------|\n",
    "|df_1            |1          |df_2            |2          |\n",
    "|df_1            |1          |df_2            |3          |\n",
    "\n",
    "It is assumed that every record in the table represents a certain match.\n",
    "\n",
    "Note that the column names above are the defaults.  They should correspond to the values you've set for [`unique_id_column_name`](https://moj-analytical-services.github.io/splink/settings_dict_guide.html#unique_id_column_name) and  [`source_dataset_column_name`](https://moj-analytical-services.github.io/splink/settings_dict_guide.html#source_dataset_column_name), if you've chosen custom values.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RendererRegistry.enable('mimetype')"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd \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_l</th>\n",
       "      <th>source_dataset_l</th>\n",
       "      <th>unique_id_r</th>\n",
       "      <th>source_dataset_r</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>3</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>3</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>3</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>5</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>10</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026</th>\n",
       "      <td>978</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>979</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2027</th>\n",
       "      <td>985</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>986</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2028</th>\n",
       "      <td>624</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>626</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2029</th>\n",
       "      <td>625</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>626</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2030</th>\n",
       "      <td>624</td>\n",
       "      <td>fake_1000</td>\n",
       "      <td>625</td>\n",
       "      <td>fake_1000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2031 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      unique_id_l source_dataset_l  unique_id_r source_dataset_r\n",
       "0               0        fake_1000            3        fake_1000\n",
       "1               1        fake_1000            3        fake_1000\n",
       "2               2        fake_1000            3        fake_1000\n",
       "3               4        fake_1000            5        fake_1000\n",
       "4               7        fake_1000           10        fake_1000\n",
       "...           ...              ...          ...              ...\n",
       "2026          978        fake_1000          979        fake_1000\n",
       "2027          985        fake_1000          986        fake_1000\n",
       "2028          624        fake_1000          626        fake_1000\n",
       "2029          625        fake_1000          626        fake_1000\n",
       "2030          624        fake_1000          625        fake_1000\n",
       "\n",
       "[2031 rows x 4 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pairwise_labels = pd.read_csv(\"./data/pairwise_labels_to_estimate_m.csv\")\n",
    "pairwise_labels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now proceed to estimate the Fellegi Sunter model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>first_name</th>\n",
       "      <th>surname</th>\n",
       "      <th>dob</th>\n",
       "      <th>city</th>\n",
       "      <th>email</th>\n",
       "      <th>cluster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Robert</td>\n",
       "      <td>Alan</td>\n",
       "      <td>1971-06-24</td>\n",
       "      <td>NaN</td>\n",
       "      <td>robert255@smith.net</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Robert</td>\n",
       "      <td>Allen</td>\n",
       "      <td>1971-05-24</td>\n",
       "      <td>NaN</td>\n",
       "      <td>roberta25@smith.net</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   unique_id first_name surname         dob city                email  cluster\n",
       "0          0     Robert    Alan  1971-06-24  NaN  robert255@smith.net        0\n",
       "1          1     Robert   Allen  1971-05-24  NaN  roberta25@smith.net        0"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"./data/fake_1000.csv\")\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from splink.duckdb.duckdb_linker import DuckDBLinker\n",
    "from splink.duckdb import duckdb_comparison_library as cl\n",
    "\n",
    "settings = {\n",
    "    \"link_type\": \"dedupe_only\",\n",
    "    \"blocking_rules_to_generate_predictions\": [\n",
    "        \"l.first_name = r.first_name\",\n",
    "        \"l.surname = r.surname\",\n",
    "    ],\n",
    "    \"comparisons\": [\n",
    "        cl.levenshtein_at_thresholds(\"first_name\", 2),\n",
    "        cl.levenshtein_at_thresholds(\"surname\", 2),\n",
    "        cl.levenshtein_at_thresholds(\"dob\"),\n",
    "        cl.exact_match(\"city\", term_frequency_adjustments=True),\n",
    "        cl.levenshtein_at_thresholds(\"email\"),\n",
    "    ],\n",
    "    \"retain_matching_columns\": True,\n",
    "    \"retain_intermediate_calculation_columns\": True,\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "linker = DuckDBLinker(df, settings, set_up_basic_logging=False)\n",
    "deterministic_rules = [\n",
    "    \"l.first_name = r.first_name and levenshtein(r.dob, l.dob) <= 1\",\n",
    "    \"l.surname = r.surname and levenshtein(r.dob, l.dob) <= 1\",\n",
    "    \"l.first_name = r.first_name and levenshtein(r.surname, l.surname) <= 2\",\n",
    "    \"l.email = r.email\"\n",
    "]\n",
    "\n",
    "linker.estimate_probability_two_random_records_match(deterministic_rules, recall=0.7)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "linker.estimate_u_using_random_sampling(max_pairs=1e6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Register the pairwise labels table with the database, and then use it to estimate the m values\n",
    "labels_df = linker.register_labels_table(pairwise_labels, overwrite=True)\n",
    "linker.estimate_m_from_pairwise_labels(labels_df)\n",
    "\n",
    "\n",
    "# Not if the labels table already existing in the dataset you could run\n",
    "# linker.estimate_m_from_pairwise_labels(\"labels_tablename_here\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<EMTrainingSession, blocking on l.first_name = r.first_name, deactivating comparisons first_name>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_blocking_rule = \"l.first_name = r.first_name\"\n",
    "linker.estimate_parameters_using_expectation_maximisation(training_blocking_rule)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v4+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v5.2.0.json",
       "config": {
        "title": {
         "anchor": "middle"
        },
        "view": {
         "continuousHeight": 300,
         "continuousWidth": 400
        }
       },
       "data": {
        "values": [
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.0057935713975033705,
          "estimated_probability_as_log_odds": -7.422948662194144,
          "m_or_u": "u",
          "sql_condition": "\"first_name_l\" = \"first_name_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 2,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.4953998584571833,
          "estimated_probability_as_log_odds": -0.026547154611559317,
          "m_or_u": "m",
          "sql_condition": "\"first_name_l\" = \"first_name_r\""
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.010119901990634016,
          "estimated_probability_as_log_odds": -6.611986562330359,
          "m_or_u": "u",
          "sql_condition": "levenshtein(\"first_name_l\", \"first_name_r\") <= 2"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 1,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.27034677990092004,
          "estimated_probability_as_log_odds": -1.4323997893323595,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"first_name_l\", \"first_name_r\") <= 2"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.9840865266118626,
          "estimated_probability_as_log_odds": 5.950464503226166,
          "m_or_u": "u",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "first_name",
          "comparison_sort_order": 0,
          "comparison_vector_value": 0,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.23425336164189667,
          "estimated_probability_as_log_odds": -1.7087973770195977,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.004889975550122249,
          "estimated_probability_as_log_odds": -7.668884984266247,
          "m_or_u": "u",
          "sql_condition": "\"surname_l\" = \"surname_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.4485294117647059,
          "estimated_probability_as_log_odds": -0.2980813529329946,
          "m_or_u": "m",
          "sql_condition": "\"surname_l\" = \"surname_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 2,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.6250962916703164,
          "estimated_probability_as_log_odds": 0.7375583478807848,
          "m_or_u": "m",
          "sql_condition": "\"surname_l\" = \"surname_r\""
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.007373772654946249,
          "estimated_probability_as_log_odds": -7.072703827504528,
          "m_or_u": "u",
          "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 2"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.2897058823529412,
          "estimated_probability_as_log_odds": -1.2938275593793966,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 2"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 1,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.31577702531443674,
          "estimated_probability_as_log_odds": -1.115560337169851,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 2"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.9877362517949315,
          "estimated_probability_as_log_odds": 6.331653973987579,
          "m_or_u": "u",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.26176470588235295,
          "estimated_probability_as_log_odds": -1.495810122984374,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "surname",
          "comparison_sort_order": 1,
          "comparison_vector_value": 0,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.059126683015246775,
          "estimated_probability_as_log_odds": -3.9921192358597515,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 3,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.0017477477477477479,
          "estimated_probability_as_log_odds": -9.157763635801777,
          "m_or_u": "u",
          "sql_condition": "\"dob_l\" = \"dob_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 3,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.42146725750861647,
          "estimated_probability_as_log_odds": -0.4569780550512149,
          "m_or_u": "m",
          "sql_condition": "\"dob_l\" = \"dob_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 3,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.3984390015055559,
          "estimated_probability_as_log_odds": -0.5943521609750629,
          "m_or_u": "m",
          "sql_condition": "\"dob_l\" = \"dob_r\""
         },
         {
          "comparison_level_label": "Levenshtein <= 1",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.0016016016016016017,
          "estimated_probability_as_log_odds": -9.283956487665113,
          "m_or_u": "u",
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1"
         },
         {
          "comparison_level_label": "Levenshtein <= 1",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.08862629246676514,
          "estimated_probability_as_log_odds": -3.3622360835408993,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1"
         },
         {
          "comparison_level_label": "Levenshtein <= 1",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 2,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.13058655173623407,
          "estimated_probability_as_log_odds": -2.735036080978985,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.015517517517517518,
          "estimated_probability_as_log_odds": -5.987395855901018,
          "m_or_u": "u",
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.19940915805022155,
          "estimated_probability_as_log_odds": -2.005333444303041,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 1,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.18478518670418353,
          "estimated_probability_as_log_odds": -2.1413311591928785,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.9811331331331331,
          "estimated_probability_as_log_odds": 5.700522147427907,
          "m_or_u": "u",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.29049729197439683,
          "estimated_probability_as_log_odds": -1.288283475925544,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "dob",
          "comparison_sort_order": 2,
          "comparison_vector_value": 0,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.2861892600540266,
          "estimated_probability_as_log_odds": -1.318572075787331,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "city",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.0551475711801453,
          "estimated_probability_as_log_odds": -4.098719767635731,
          "m_or_u": "u",
          "sql_condition": "\"city_l\" = \"city_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "city",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.5661492978566149,
          "estimated_probability_as_log_odds": 0.38398388881490503,
          "m_or_u": "m",
          "sql_condition": "\"city_l\" = \"city_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "city",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.5617206210359919,
          "estimated_probability_as_log_odds": 0.3580019642355839,
          "m_or_u": "m",
          "sql_condition": "\"city_l\" = \"city_r\""
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "city",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.9448524288198547,
          "estimated_probability_as_log_odds": 4.09871976763573,
          "m_or_u": "u",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "city",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.43385070214338506,
          "estimated_probability_as_log_odds": -0.3839838888149049,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "city",
          "comparison_sort_order": 3,
          "comparison_vector_value": 0,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.43827937896400826,
          "estimated_probability_as_log_odds": -0.35800196423558334,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 3,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.0021938713143283602,
          "estimated_probability_as_log_odds": -8.829136816196753,
          "m_or_u": "u",
          "sql_condition": "\"email_l\" = \"email_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 3,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.5340642129992169,
          "estimated_probability_as_log_odds": 0.1968820708288665,
          "m_or_u": "m",
          "sql_condition": "\"email_l\" = \"email_r\""
         },
         {
          "comparison_level_label": "Exact match",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 3,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.5901618249897193,
          "estimated_probability_as_log_odds": 0.5260562294318846,
          "m_or_u": "m",
          "sql_condition": "\"email_l\" = \"email_r\""
         },
         {
          "comparison_level_label": "Levenshtein <= 1",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 2,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.0007334349848487773,
          "estimated_probability_as_log_odds": -10.411984784067583,
          "m_or_u": "u",
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 1"
         },
         {
          "comparison_level_label": "Levenshtein <= 1",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 2,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.17854346123727485,
          "estimated_probability_as_log_odds": -2.201908948412526,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 1"
         },
         {
          "comparison_level_label": "Levenshtein <= 1",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 2,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.1765409703056437,
          "estimated_probability_as_log_odds": -2.2216938398112753,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 1"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.000620846281034272,
          "estimated_probability_as_log_odds": -10.652580302192339,
          "m_or_u": "u",
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 2"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.1417384494909945,
          "estimated_probability_as_log_odds": -2.598186195877321,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 2"
         },
         {
          "comparison_level_label": "Levenshtein <= 2",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.1271138693200017,
          "estimated_probability_as_log_odds": -2.779672013828551,
          "m_or_u": "m",
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 2"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "estimate_description": "estimate u by random sampling",
          "estimated_probability": 0.9964518474197885,
          "estimated_probability_as_log_odds": 8.133588228885133,
          "m_or_u": "u",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "estimate_description": "estimate m from pairwise labels",
          "estimated_probability": 0.1456538762725137,
          "estimated_probability_as_log_odds": -2.552276575215576,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         },
         {
          "comparison_level_label": "All other comparisons",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 0,
          "estimate_description": "EM, blocked on: l.first_name = r.first_name",
          "estimated_probability": 0.10618333538463538,
          "estimated_probability_as_log_odds": -3.073421578476505,
          "m_or_u": "m",
          "sql_condition": "ELSE"
         }
        ]
       },
       "encoding": {
        "color": {
         "field": "estimate_description",
         "type": "nominal"
        },
        "column": {
         "field": "m_or_u",
         "title": null,
         "type": "nominal"
        },
        "row": {
         "field": "comparison_name",
         "header": {
          "labelAlign": "left",
          "labelAnchor": "middle",
          "labelAngle": 0
         },
         "sort": {
          "field": "comparison_sort_order"
         },
         "title": null,
         "type": "nominal"
        },
        "shape": {
         "field": "estimate_description",
         "scale": {
          "range": [
           "circle",
           "square",
           "triangle",
           "diamond"
          ]
         },
         "type": "nominal"
        },
        "tooltip": [
         {
          "field": "comparison_name",
          "type": "nominal"
         },
         {
          "field": "estimate_description",
          "type": "nominal"
         },
         {
          "field": "estimated_probability",
          "type": "quantitative"
         }
        ],
        "x": {
         "field": "estimated_probability_as_log_odds",
         "type": "quantitative"
        },
        "y": {
         "axis": {
          "grid": true,
          "title": null
         },
         "field": "comparison_level_label",
         "sort": {
          "field": "comparison_vector_value",
          "order": "descending"
         },
         "type": "nominal"
        }
       },
       "mark": {
        "filled": false,
        "opacity": 0.7,
        "size": 100,
        "type": "point"
       },
       "resolve": {
        "scale": {
         "y": "independent"
        }
       },
       "selection": {
        "selection_zoom": {
         "bind": "scales",
         "encodings": [
          "x"
         ],
         "type": "interval"
        }
       },
       "title": {
        "subtitle": "Use mousewheeel to zoom",
        "text": "Comparison of parameter estimates across training sessions"
       }
      },
      "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": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linker.parameter_estimate_comparisons_chart()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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": 0.0033430420247643373,
          "bayes_factor_description": "The probability that two random records drawn at random match is 0.003 or one in  300.1 records.This is equivalent to a starting match weight of -8.225.",
          "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": -8.224622793739668,
          "m_probability": null,
          "m_probability_description": null,
          "max_comparison_vector_value": 0,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": null,
          "tf_adjustment_column": null,
          "tf_adjustment_weight": null,
          "u_probability": null,
          "u_probability_description": null
         },
         {
          "bayes_factor": 85.50854463805632,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 85.51 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": "Exact match",
          "log2_bayes_factor": 6.417996686710603,
          "m_probability": 0.4953998584571833,
          "m_probability_description": "Amongst matching record comparisons, 49.54% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "\"first_name_l\" = \"first_name_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0057935713975033705,
          "u_probability_description": "Amongst non-matching record comparisons, 0.58% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 26.71436740702888,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 26.71 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": "Levenshtein <= 2",
          "log2_bayes_factor": 4.739543949609427,
          "m_probability": 0.27034677990092004,
          "m_probability_description": "Amongst matching record comparisons, 27.03% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "levenshtein(\"first_name_l\", \"first_name_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.010119901990634016,
          "u_probability_description": "Amongst non-matching record comparisons, 1.01% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.2380414275647221,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  4.20 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": -2.0707154199210716,
          "m_probability": 0.23425336164189667,
          "m_probability_description": "Amongst matching record comparisons, 23.43% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9840865266118626,
          "u_probability_description": "Amongst non-matching record comparisons, 98.41% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 109.77822817623102,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 109.78 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": "Exact match",
          "log2_bayes_factor": 6.778448149248172,
          "m_probability": 0.5368128517175111,
          "m_probability_description": "Amongst matching record comparisons, 53.68% of records are in the exact match comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "\"surname_l\" = \"surname_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.004889975550122249,
          "u_probability_description": "Amongst non-matching record comparisons, 0.49% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 41.05652126806665,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 41.06 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": "Levenshtein <= 2",
          "log2_bayes_factor": 5.359539487508672,
          "m_probability": 0.302741453833689,
          "m_probability_description": "Amongst matching record comparisons, 30.27% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.007373772654946249,
          "u_probability_description": "Amongst non-matching record comparisons, 0.74% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.16243779061185123,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  6.16 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": -2.622040785111616,
          "m_probability": 0.16044569444879986,
          "m_probability_description": "Amongst matching record comparisons, 16.04% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 2,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9877362517949315,
          "u_probability_description": "Amongst non-matching record comparisons, 98.77% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 234.56081121281733,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 234.56 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": 7.873818187831652,
          "m_probability": 0.4099531295070862,
          "m_probability_description": "Amongst matching record comparisons, 41.00% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "\"dob_l\" = \"dob_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0017477477477477479,
          "u_probability_description": "Amongst non-matching record comparisons, 0.17% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 68.4355097996238,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 68.44 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": 6.096673199508259,
          "m_probability": 0.1096064221014996,
          "m_probability_description": "Amongst matching record comparisons, 10.96% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0016016016016016017,
          "u_probability_description": "Amongst non-matching record comparisons, 0.16% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 12.379375255117102,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 12.38 times more 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": 3.6298666033743396,
          "m_probability": 0.19209717237720253,
          "m_probability_description": "Amongst matching record comparisons, 19.21% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "levenshtein(\"dob_l\", \"dob_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.015517517517517518,
          "u_probability_description": "Amongst non-matching record comparisons, 1.55% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.2938880222028803,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  3.40 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": -1.766661533654638,
          "m_probability": 0.2883432760142117,
          "m_probability_description": "Amongst matching record comparisons, 28.83% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9811331331331331,
          "u_probability_description": "Amongst non-matching record comparisons, 98.11% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 10.225925591612203,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 10.23 times more likely to be a match",
          "comparison_name": "city",
          "comparison_sort_order": 3,
          "comparison_vector_value": 1,
          "has_tf_adjustments": true,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 3.3541595283715577,
          "m_probability": 0.5639349594463035,
          "m_probability_description": "Amongst matching record comparisons, 56.39% of records are in the exact match comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "\"city_l\" = \"city_r\"",
          "tf_adjustment_column": "city",
          "tf_adjustment_weight": 1,
          "u_probability": 0.0551475711801453,
          "u_probability_description": "Amongst non-matching record comparisons, 5.51% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 0.46151655777437467,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  2.17 times less likely to be a match",
          "comparison_name": "city",
          "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": -1.1155456866902822,
          "m_probability": 0.43606504055369666,
          "m_probability_description": "Amongst matching record comparisons, 43.61% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 1,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9448524288198547,
          "u_probability_description": "Amongst non-matching record comparisons, 94.49% of records are in the all other comparisons comparison level"
         },
         {
          "bayes_factor": 256.2196858691119,
          "bayes_factor_description": "If comparison level is `exact match` then comparison is 256.22 times more likely to be a match",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 3,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Exact match",
          "log2_bayes_factor": 8.001237514848016,
          "m_probability": 0.5621130189944681,
          "m_probability_description": "Amongst matching record comparisons, 56.21% of records are in the exact match comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "\"email_l\" = \"email_r\"",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0021938713143283602,
          "u_probability_description": "Amongst non-matching record comparisons, 0.22% of records are in the exact match comparison level"
         },
         {
          "bayes_factor": 242.06946687723885,
          "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 242.07 times more likely to be a match",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 2,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Levenshtein <= 1",
          "log2_bayes_factor": 7.919277308092294,
          "m_probability": 0.1775422157714593,
          "m_probability_description": "Amongst matching record comparisons, 17.75% of records are in the levenshtein <= 1 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 1",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.0007334349848487773,
          "u_probability_description": "Amongst non-matching record comparisons, 0.07% of records are in the levenshtein <= 1 comparison level"
         },
         {
          "bayes_factor": 216.52084181217398,
          "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 216.52 times more likely to be a match",
          "comparison_name": "email",
          "comparison_sort_order": 4,
          "comparison_vector_value": 1,
          "has_tf_adjustments": false,
          "is_null_level": false,
          "label_for_charts": "Levenshtein <= 2",
          "log2_bayes_factor": 7.758362092010155,
          "m_probability": 0.13442615940549812,
          "m_probability_description": "Amongst matching record comparisons, 13.44% of records are in the levenshtein <= 2 comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "levenshtein(\"email_l\", \"email_r\") <= 2",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.000620846281034272,
          "u_probability_description": "Amongst non-matching record comparisons, 0.06% of records are in the levenshtein <= 2 comparison level"
         },
         {
          "bayes_factor": 0.12636697513746203,
          "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is  7.91 times less likely to be a match",
          "comparison_name": "email",
          "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": -2.9843086173742144,
          "m_probability": 0.12591860582857453,
          "m_probability_description": "Amongst matching record comparisons, 12.59% of records are in the all other comparisons comparison level",
          "max_comparison_vector_value": 3,
          "probability_two_random_records_match": 0.0033319033319033323,
          "sql_condition": "ELSE",
          "tf_adjustment_column": null,
          "tf_adjustment_weight": 1,
          "u_probability": 0.9964518474197885,
          "u_probability_description": "Amongst non-matching record comparisons, 99.65% 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": 20,
         "mark": {
          "clip": true,
          "height": 15,
          "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"
          }
         },
         "height": {
          "step": 12
         },
         "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": [
    "linker.match_weights_chart()"
   ]
  }
 ],
 "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": "3.10.8"
  },
  "vscode": {
   "interpreter": {
    "hash": "3b53fa520a31e303a9636a08ff10a3bbc14893ee50cb37445791fa59628fc75b"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}