{ "cells": [ { "cell_type": "markdown", "id": "d729e302", "metadata": {}, "source": [ "# Specifying and estimating a linkage model\n", "\n", "\n", "In the last tutorial we looked at how we can use blocking rules to generate pairwise record comparisons.\n", "\n", "Now it's time to estimate a probabilistic linkage model to score each of these comparisons. The resultant match score is a prediction of whether the two records represent the same entity (e.g. are the same person). \n", "\n", "The purpose of estimating the model is to learn the relative importance of different parts of your data for the purpose of data linking. \n", "\n", "For example, a match on date of birth is a much stronger indicator that two records refer to the same entity than a match on gender. A mismatch on gender may be a stronger indicate against two records referring than a mismatch on name, since names are more likely to be entered differently.\n", "\n", "The relative importance of different information is captured in the (partial) 'match weights', which can be learned from your data. These match weights are then added up to compute the overall match score.\n", "\n", "The match weights are are derived from the `m` and `u` parameters of the underlying Fellegi Sunter model. Splink uses various statistical routines to estimate these parameters. Further details of the underlying theory can be found [here](https://www.robinlinacre.com/intro_to_probabilistic_linkage/), which will help you understand this part of the tutorial." ] }, { "cell_type": "code", "execution_count": 1, "id": "aa6a9e30", "metadata": {}, "outputs": [], "source": [ "# Begin by reading in the tutorial data again\n", "from splink.duckdb.duckdb_linker import DuckDBLinker\n", "import pandas as pd \n", "import altair as alt\n", "alt.renderers.enable(\"mimetype\")\n", "df = pd.read_csv(\"./data/fake_1000.csv\")" ] }, { "cell_type": "markdown", "id": "0f104340", "metadata": {}, "source": [ "## Specifying a linkage model\n", "\n", "To build a linkage model, the user defines the partial match weights that `splink` needs to estimate. This is done by defining how the information in the input records should be compared.\n", "\n", "To be concrete, here is an example comparison:\n", "\n", "\n", "first_name_l|first_name_r|surname_l|surname_r|dob_l |dob_r |city_l|city_r|email_l |email_r |\n", "------------|------------|---------|---------|----------|----------|------|------|-------------------|-------------------|\n", "Robert |Rob |Allen |Allen |1971-05-24|1971-06-24|nan |London|roberta25@smith.net|roberta25@smith.net|\n", "\n", "What functions should we use to assess the similarity of `Rob` vs. `Robert` in the the `first_name` field? \n", "\n", "Should similarity in the `dob` field be computed in the same way, or a different way?\n", "\n", "Your job as the developer of a linkage model is to decide what comparisons are most appropriate for the types of data you have. \n", "\n", "Splink can then estimate how much weight to place on a fuzzy match of `Rob` vs. `Robert`, relative to an exact match on `Robert`, or a non-match.\n", "\n", "Defining these scenarios is done using `Comparison`s." ] }, { "cell_type": "markdown", "id": "8a520392", "metadata": {}, "source": [ "### Comparisons\n", "\n", "The concept of a `Comparison` has a specific definition within Splink: it defines how data from one or more input columns is compared, using SQL expressions to assess similarity.\n", "\n", "For example, one `Comparison` may represent how similarity is assessed for a person's date of birth. \n", "\n", "Another `Comparison` may represent the comparison of a person's name or location.\n", "\n", "A model is composed of many `Comparison`s, which between them assess the similarity of all of the columns being used for data linking. \n", "\n", "Each `Comparison` contains two or more `ComparisonLevels` which define _n_ discrete gradations of similarity between the input columns within the Comparison.\n", "\n", "As such `ComparisonLevels`are nested within `Comparisons` as follows:\n", "\n", "```\n", "Data Linking Model\n", "├─-- Comparison: Date of birth\n", "│ ├─-- ComparisonLevel: Exact match\n", "│ ├─-- ComparisonLevel: One character difference\n", "│ ├─-- ComparisonLevel: All other\n", "├─-- Comparison: Surname\n", "│ ├─-- ComparisonLevel: Exact match on surname\n", "│ ├─-- ComparisonLevel: All other\n", "│ etc.\n", "```\n", "\n", "Our example data would therefore result in the following comparisons, for `dob` and `surname`:\n", "\n", "|dob_l |dob_r |comparison_level |interpretation |\n", "|----------|----------|------------------------|---------------|\n", "|1971-05-24|1971-05-24|Exact match |great match |\n", "|1971-05-24|1971-06-24|One character difference|ok match |\n", "|1971-05-24|2000-01-02|All other |bad match |\n", "\n", "\n", "\n", "surname_l|surname_r|comparison_level |interpretation |\n", "---------|---------|-----------------|-----------------------------------------------------|\n", "Rob |Rob |Exact match |great match |\n", "Rob |Jane |All other |bad match |\n", "Rob |Robert |All other |bad match, this comparison has no notion of nicknames|\n", "\n", "More information about comparisons can be found [here](https://moj-analytical-services.github.io/splink/comparison.html).\n", "\n", "\n", "We will now use these concepts to build a data linking model." ] }, { "cell_type": "markdown", "id": "02000a24", "metadata": {}, "source": [ "### Specifying the model using comparisons\n", "\n", "Splink includes libraries of comparison functions to make it simple to get started:\n", "\n", "Let's start by looking at a `Comparison` for `first_name`:" ] }, { "cell_type": "code", "execution_count": 2, "id": "bd6143e7", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Comparison 'Exact match vs. levenshtein at threshold 2 vs. anything else' of \"first_name\".\n", "Similarity is assessed using the following ComparisonLevels:\n", " - 'Null' with SQL rule: \"first_name_l\" IS NULL OR \"first_name_r\" IS NULL\n", " - 'Exact match' with SQL rule: \"first_name_l\" = \"first_name_r\"\n", " - 'levenshtein <= 2' with SQL rule: levenshtein(\"first_name_l\", \"first_name_r\") <= 2\n", " - 'All other comparisons' with SQL rule: ELSE\n", "\n" ] } ], "source": [ "import splink.duckdb.duckdb_comparison_library as cl\n", "\n", "first_name_comparison = cl.levenshtein_at_thresholds(\"first_name\", 2)\n", "print(first_name_comparison.human_readable_description)\n" ] }, { "cell_type": "markdown", "id": "47b7677a", "metadata": {}, "source": [ "## Specifying the full settings dictionary\n", "\n", "`Comparisons` are specified as part of the Splink `settings`, a Python dictionary which controls all of the configuration of a Splink model:" ] }, { "cell_type": "code", "execution_count": 3, "id": "0fa0611a", "metadata": {}, "outputs": [], "source": [ "settings = {\n", " \"link_type\": \"dedupe_only\",\n", " \"comparisons\": [\n", " cl.exact_match(\"first_name\"),\n", " cl.levenshtein_at_thresholds(\"surname\"),\n", " cl.levenshtein_at_thresholds(\"dob\", 1),\n", " cl.exact_match(\"city\", term_frequency_adjustments=True),\n", " cl.levenshtein_at_thresholds(\"email\"),\n", " ],\n", " \"blocking_rules_to_generate_predictions\": [\n", " \"l.first_name = r.first_name\",\n", " \"l.surname = r.surname\",\n", " ],\n", " \"retain_matching_columns\": True,\n", " \"retain_intermediate_calculation_columns\": True,\n", "}\n", "\n", "linker = DuckDBLinker(df, settings)" ] }, { "cell_type": "markdown", "id": "657a1fb8", "metadata": {}, "source": [ "In words, this setting dictionary says:\n", "\n", "\n", "* We are performing a `dedupe_only` (the other options are `link_only`, or `link_and_dedupe`, which may be used if there are multiple input datasets).\n", "* When comparing records, we will use information from the `first_name`, `surname`, `dob`, `city` and `email` columns to compute a match score.\n", "* The `blocking_rules_to_generate_predictions` states that we will only check for duplicates amongst records where either the `first_name` or `surname` is identical.\n", "* We have enabled term frequency adjustments for the 'city' column, because some values (e.g. `London`) appear much more frequently than others.\n", "* We have set `retain_intermediate_calculation_columns` and `additional_columns_to_retain` to `True` so that Splink outputs additional information that helps the user understand the calculations. If they were `False`, the computations would run faster." ] }, { "cell_type": "markdown", "id": "afa31386", "metadata": {}, "source": [ "## Estimate the parameters of the model\n", "\n", "Now that we have specified our linkage model, we need to estimate the [`probability_two_random_records_match`](https://moj-analytical-services.github.io/splink/settings_dict_guide.html#probability_two_random_records_match), `u`, and `m` parameters.\n", "\n", "- The `probability_two_random_records_match` parameter is the probability that two records taken at random from your input data represent a match (typically a very small number).\n", "\n", "- The `u` values are the proportion of records falling into each `ComparisonLevel` amongst truly *non-matching* records.\n", "\n", "- The `m` values are the proportion of records falling into each `ComparisonLevel` amongst truly *matching* records\n", "\n", "You can read more about [the theory of what these mean](https://www.robinlinacre.com/maths_of_fellegi_sunter/).\n", "\n", "We can estimate these parameters using unlabeled data. If we have labels, then we can estimate them even more accurately." ] }, { "cell_type": "markdown", "id": "c2871ac6", "metadata": {}, "source": [ "### Estimation of `probability_two_random_records_match`\n", "\n", "In some cases, the `probability_two_random_records_match` will be known. For example, if you are linking two tables of 10,000 records and expect a one-to-one match, then you should set this value to `1/10_000` in your settings instead of estimating it.\n", "\n", "More generally, this parameter is unknown and needs to be estimated. \n", "\n", "It can be estimated accurately enough for most purposes by combining a series of deterministic matching rules and a guess of the recall corresponding to those rules. For further details of the rationale behind this appraoch see [here](https://github.com/moj-analytical-services/splink/issues/462#issuecomment-1227027995).\n", "\n", "In this example, I guess that the following deterministic matching rules have a recall of about 70%:" ] }, { "cell_type": "code", "execution_count": 4, "id": "cbf92120", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Probability two random records match is estimated to be 0.00333.\n", "This means that amongst all possible pairwise record comparisons, one in 300.13 are expected to match. With 499,500 total possible comparisons, we expect a total of around 1,664.29 matching pairs\n" ] } ], "source": [ "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)" ] }, { "cell_type": "markdown", "id": "7712860b", "metadata": {}, "source": [ "### Estimation of `u` probabilities\n", "\n", "Once we have the `probability_two_random_records_match` parameter, we can estimate the `u` probabilities.\n", "\n", "We estimate `u` using the `estimate_u_using_random_sampling` method, which doesn't require any labels.\n", "\n", "It works by sampling random pairs of records, since most of these pairs are going to be non-matches. Over these non-matches we compute the distribution of `ComparisonLevel`s for each `Comparison`.\n", "\n", "For instance, for `gender`, we would find that the the gender matches 50% of the time, and mismatches 50% of the time. \n", "\n", "For `dob` on the other hand, we would find that the `dob` matches 1% of the time, has a \"one character difference\" 3% of the time, and everything else happens 96% of the time.\n", "\n", "The larger the random sample, the more accurate the predictions. You control this using the `max_pairs` parameter. For large datasets, we recommend using at least 10 million - but the higher the better and 1 billion is often appropriate for larger datasets." ] }, { "cell_type": "code", "execution_count": 5, "id": "b8d49e7a", "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", " - city (no m values are trained).\n", " - email (no m values are trained).\n" ] } ], "source": [ "linker.estimate_u_using_random_sampling(max_pairs=1e6)" ] }, { "cell_type": "markdown", "id": "a73921b7", "metadata": {}, "source": [ "### Estimation of `m` probabilities\n", "\n", "`m` is the trickiest of the parameters to estimate, because we have to have some idea of what the true matches are.\n", "\n", "If we have labels, we can directly estimate it.\n", "\n", "If we do not have labelled data, the `m` parameters can be estimated using an iterative maximum likelihood approach called Expectation Maximisation. \n", "\n", "#### Estimating directly\n", "\n", "If we have labels, we can estimate `m` directly using the `estimate_m_from_label_column` method of the linker.\n", "\n", "For example, if the entity being matched is persons, and your input dataset(s) contain social security number, this could be used to estimate the m values for the model.\n", "\n", "Note that this column does not need to be fully populated. A common case is where a unique identifier such as social security number is only partially populated.\n", "\n", "For example (in this tutorial we don't have labels, so we're not actually going to use this):\n", "\n", "```python\n", "linker.estimate_m_from_label_column(\"social_security_number\")\n", "```\n", "\n", "#### Estimating with Expectation Maximisation\n", "\n", "This algorithm estimates the `m` values by generating pairwise record comparisons, and using them to maximise a likelihood function. \n", "\n", "Each estimation pass requires the user to configure an estimation blocking rule to reduce the number of record comparisons generated to a manageable level.\n", "\n", "In our first estimation pass, we block on `first_name` and `surname`, meaning we will generate all record comparisons that have `first_name` and `surname` exactly equal. \n", "\n", "Recall we are trying to estimate the `m` values of the model, i.e. proportion of records falling into each `ComparisonLevel` amongst truly matching records.\n", "\n", "This means that, in this training session, we cannot estimate parameter estimates for the `first_name` or `surname` columns, since they will be equal for all the comparisons we do.\n", "\n", "We can, however, estimate parameter estimates for all of the other columns. The output messages produced by Splink confirm this." ] }, { "cell_type": "code", "execution_count": 6, "id": "098f0a40", "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", " - city\n", " - email\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 -0.514 in the m_probability of dob, level `Exact match`\n", "Iteration 2: Largest change in params was 0.0474 in probability_two_random_records_match\n", "Iteration 3: Largest change in params was 0.0212 in probability_two_random_records_match\n", "Iteration 4: Largest change in params was 0.0113 in probability_two_random_records_match\n", "Iteration 5: Largest change in params was 0.00694 in probability_two_random_records_match\n", "Iteration 6: Largest change in params was 0.00463 in probability_two_random_records_match\n", "Iteration 7: Largest change in params was 0.00328 in probability_two_random_records_match\n", "Iteration 8: Largest change in params was 0.00243 in probability_two_random_records_match\n", "Iteration 9: Largest change in params was 0.00186 in probability_two_random_records_match\n", "Iteration 10: Largest change in params was 0.00146 in probability_two_random_records_match\n", "Iteration 11: Largest change in params was 0.00117 in probability_two_random_records_match\n", "Iteration 12: Largest change in params was 0.000954 in probability_two_random_records_match\n", "Iteration 13: Largest change in params was 0.000787 in probability_two_random_records_match\n", "Iteration 14: Largest change in params was 0.000658 in probability_two_random_records_match\n", "Iteration 15: Largest change in params was 0.000555 in probability_two_random_records_match\n", "Iteration 16: Largest change in params was 0.000471 in probability_two_random_records_match\n", "Iteration 17: Largest change in params was 0.000404 in probability_two_random_records_match\n", "Iteration 18: Largest change in params was 0.000347 in probability_two_random_records_match\n", "Iteration 19: Largest change in params was 0.000301 in probability_two_random_records_match\n", "Iteration 20: Largest change in params was 0.000261 in probability_two_random_records_match\n", "Iteration 21: Largest change in params was 0.000228 in probability_two_random_records_match\n", "Iteration 22: Largest change in params was 0.0002 in probability_two_random_records_match\n", "Iteration 23: Largest change in params was 0.000175 in probability_two_random_records_match\n", "Iteration 24: Largest change in params was 0.000154 in probability_two_random_records_match\n", "Iteration 25: Largest change in params was 0.000136 in probability_two_random_records_match\n", "\n", "EM converged after 25 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" ] } ], "source": [ "training_blocking_rule = \"l.first_name = r.first_name and l.surname = r.surname\"\n", "training_session_fname_sname = linker.estimate_parameters_using_expectation_maximisation(training_blocking_rule)" ] }, { "cell_type": "markdown", "id": "92bd4a31", "metadata": {}, "source": [ "In a second estimation pass, we block on dob. This allows us to estimate parameters for the `first_name` and `surname` comparisons.\n", "\n", "Between the two estimation passes, we now have parameter estimates for all comparisons." ] }, { "cell_type": "code", "execution_count": 7, "id": "ac8d3264", "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", " - city\n", " - email\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 -0.392 in the m_probability of surname, level `Exact match`\n", "Iteration 2: Largest change in params was 0.137 in probability_two_random_records_match\n", "Iteration 3: Largest change in params was 0.0416 in probability_two_random_records_match\n", "Iteration 4: Largest change in params was 0.0171 in probability_two_random_records_match\n", "Iteration 5: Largest change in params was 0.00853 in probability_two_random_records_match\n", "Iteration 6: Largest change in params was 0.0047 in probability_two_random_records_match\n", "Iteration 7: Largest change in params was 0.00274 in probability_two_random_records_match\n", "Iteration 8: Largest change in params was 0.00165 in probability_two_random_records_match\n", "Iteration 9: Largest change in params was 0.00101 in probability_two_random_records_match\n", "Iteration 10: Largest change in params was 0.000629 in probability_two_random_records_match\n", "Iteration 11: Largest change in params was 0.000394 in probability_two_random_records_match\n", "Iteration 12: Largest change in params was 0.000247 in probability_two_random_records_match\n", "Iteration 13: Largest change in params was 0.000156 in probability_two_random_records_match\n", "Iteration 14: Largest change in params was 9.86e-05 in probability_two_random_records_match\n", "\n", "EM converged after 14 iterations\n", "\n", "Your model is fully trained. All comparisons have at least one estimate for their m and u values\n" ] } ], "source": [ "from numpy import fix\n", "\n", "\n", "training_blocking_rule = \"l.dob = r.dob\"\n", "training_session_dob = linker.estimate_parameters_using_expectation_maximisation(training_blocking_rule)" ] }, { "cell_type": "markdown", "id": "efdb0c5f", "metadata": {}, "source": [ "Note that Splink includes other algorithms for estimating m and u values, which are documented [here](https://moj-analytical-services.github.io/splink/linkerest.html)." ] }, { "cell_type": "markdown", "id": "38355535", "metadata": {}, "source": [ "## Visualising model parameters\n", "\n", "Splink can generate a number of charts to help you understand your model. For an introduction to these charts and how to interpret them, please see [this](https://www.youtube.com/watch?v=msz3T741KQI&t=507s) video.\n", "\n", "The final estimated match weights can be viewed in the match weights chart:" ] }, { "cell_type": "code", "execution_count": 8, "id": "3a1e15cc", "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": 85.80338234594069, "bayes_factor_description": "If comparison level is `exact match` then comparison is 85.80 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": "Exact match", "log2_bayes_factor": 6.422962614378335, "m_probability": 0.4971080217684876, "m_probability_description": "Amongst matching record comparisons, 49.71% of records are in the exact match comparison level", "max_comparison_vector_value": 1, "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": 0.5058224969822424, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 1.98 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": -0.9832968910951042, "m_probability": 0.5028919782315123, "m_probability_description": "Amongst matching record comparisons, 50.29% 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": 0.9942064286024966, "u_probability_description": "Amongst non-matching record comparisons, 99.42% of records are in the all other comparisons comparison level" }, { "bayes_factor": 89.48089949369762, "bayes_factor_description": "If comparison level is `exact match` then comparison is 89.48 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": 6.483507853838869, "m_probability": 0.43755941072712773, "m_probability_description": "Amongst matching record comparisons, 43.76% of records are in the exact match comparison level", "max_comparison_vector_value": 3, "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": 78.52365983637108, "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 78.52 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": "levenshtein <= 1", "log2_bayes_factor": 6.295055510492454, "m_probability": 0.18566006785783215, "m_probability_description": "Amongst matching record comparisons, 18.57% of records are in the levenshtein <= 1 comparison level", "max_comparison_vector_value": 3, "probability_two_random_records_match": 0.0033319033319033323, "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 1", "tf_adjustment_column": null, "tf_adjustment_weight": 1, "u_probability": 0.002364383782476692, "u_probability_description": "Amongst non-matching record comparisons, 0.24% of records are in the levenshtein <= 1 comparison level" }, { "bayes_factor": 22.372672328973053, "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 22.37 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": 4.483665685906399, "m_probability": 0.11207341581216516, "m_probability_description": "Amongst matching record comparisons, 11.21% of records are in the levenshtein <= 2 comparison level", "max_comparison_vector_value": 3, "probability_two_random_records_match": 0.0033319033319033323, "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 2", "tf_adjustment_column": null, "tf_adjustment_weight": 1, "u_probability": 0.005009388872469557, "u_probability_description": "Amongst non-matching record comparisons, 0.50% of records are in the levenshtein <= 2 comparison level" }, { "bayes_factor": 0.2679937130198923, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 3.73 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, "label_for_charts": "All other comparisons", "log2_bayes_factor": -1.8997289386125742, "m_probability": 0.2647071056028749, "m_probability_description": "Amongst matching record comparisons, 26.47% 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": 0.9877362517949315, "u_probability_description": "Amongst non-matching record comparisons, 98.77% of records are in the all other comparisons comparison level" }, { "bayes_factor": 222.50382383655824, "bayes_factor_description": "If comparison level is `exact match` then comparison is 222.50 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": "Exact match", "log2_bayes_factor": 7.797686319483587, "m_probability": 0.3888805569756063, "m_probability_description": "Amongst matching record comparisons, 38.89% of records are in the exact match comparison level", "max_comparison_vector_value": 2, "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": 92.7048728526088, "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 92.70 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 <= 1", "log2_bayes_factor": 6.534573268209908, "m_probability": 0.1484762728370111, "m_probability_description": "Amongst matching record comparisons, 14.85% of records are in the levenshtein <= 1 comparison level", "max_comparison_vector_value": 2, "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": 0.46419793122630476, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 2.15 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": -1.107188001796095, "m_probability": 0.4626431701873826, "m_probability_description": "Amongst matching record comparisons, 46.26% 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": 0.9966506506506506, "u_probability_description": "Amongst non-matching record comparisons, 99.67% of records are in the all other comparisons comparison level" }, { "bayes_factor": 10.264353562890939, "bayes_factor_description": "If comparison level is `exact match` then comparison is 10.26 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": 3.3595708659307335, "m_probability": 0.5660541687277061, "m_probability_description": "Amongst matching record comparisons, 56.61% 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": 0.4592736580190661, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 2.18 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": -1.1225740557985633, "m_probability": 0.4339458312722939, "m_probability_description": "Amongst matching record comparisons, 43.39% 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": 255.301619971695, "bayes_factor_description": "If comparison level is `exact match` then comparison is 255.30 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": 7.99605888191739, "m_probability": 0.560098900557462, "m_probability_description": "Amongst matching record comparisons, 56.01% 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": 235.63777521191432, "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 235.64 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": 7.880427026661069, "m_probability": 0.17282498809234997, "m_probability_description": "Amongst matching record comparisons, 17.28% 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": 206.57183856156232, "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 206.57 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": 7.690499778262131, "m_probability": 0.128249357737358, "m_probability_description": "Amongst matching record comparisons, 12.82% 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": 0.13932108608389637, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 7.18 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": -2.8435144702111006, "m_probability": 0.13882675361282992, "m_probability_description": "Amongst matching record comparisons, 13.88% 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": [ "\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": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "linker.match_weights_chart()" ] }, { "cell_type": "code", "execution_count": 9, "id": "8576c042", "metadata": {}, "outputs": [ { "data": { "application/vnd.vegalite.v4+json": { "$schema": "https://vega.github.io/schema/vega-lite/v4.json", "config": { "header": { "title": null }, "title": { "anchor": "middle", "offset": 10 }, "view": { "height": 300, "width": 400 } }, "data": { "values": [ { "bayes_factor": 85.80338234594069, "bayes_factor_description": "If comparison level is `exact match` then comparison is 85.80 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": "Exact match", "log2_bayes_factor": 6.422962614378335, "m_probability": 0.4971080217684876, "m_probability_description": "Amongst matching record comparisons, 49.71% of records are in the exact match comparison level", "max_comparison_vector_value": 1, "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": 0.5058224969822424, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 1.98 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": -0.9832968910951042, "m_probability": 0.5028919782315123, "m_probability_description": "Amongst matching record comparisons, 50.29% 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": 0.9942064286024966, "u_probability_description": "Amongst non-matching record comparisons, 99.42% of records are in the all other comparisons comparison level" }, { "bayes_factor": 89.48089949369762, "bayes_factor_description": "If comparison level is `exact match` then comparison is 89.48 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": 6.483507853838869, "m_probability": 0.43755941072712773, "m_probability_description": "Amongst matching record comparisons, 43.76% of records are in the exact match comparison level", "max_comparison_vector_value": 3, "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": 78.52365983637108, "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 78.52 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": "levenshtein <= 1", "log2_bayes_factor": 6.295055510492454, "m_probability": 0.18566006785783215, "m_probability_description": "Amongst matching record comparisons, 18.57% of records are in the levenshtein <= 1 comparison level", "max_comparison_vector_value": 3, "probability_two_random_records_match": 0.0033319033319033323, "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 1", "tf_adjustment_column": null, "tf_adjustment_weight": 1, "u_probability": 0.002364383782476692, "u_probability_description": "Amongst non-matching record comparisons, 0.24% of records are in the levenshtein <= 1 comparison level" }, { "bayes_factor": 22.372672328973053, "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 22.37 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": 4.483665685906399, "m_probability": 0.11207341581216516, "m_probability_description": "Amongst matching record comparisons, 11.21% of records are in the levenshtein <= 2 comparison level", "max_comparison_vector_value": 3, "probability_two_random_records_match": 0.0033319033319033323, "sql_condition": "levenshtein(\"surname_l\", \"surname_r\") <= 2", "tf_adjustment_column": null, "tf_adjustment_weight": 1, "u_probability": 0.005009388872469557, "u_probability_description": "Amongst non-matching record comparisons, 0.50% of records are in the levenshtein <= 2 comparison level" }, { "bayes_factor": 0.2679937130198923, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 3.73 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, "label_for_charts": "All other comparisons", "log2_bayes_factor": -1.8997289386125742, "m_probability": 0.2647071056028749, "m_probability_description": "Amongst matching record comparisons, 26.47% 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": 0.9877362517949315, "u_probability_description": "Amongst non-matching record comparisons, 98.77% of records are in the all other comparisons comparison level" }, { "bayes_factor": 222.50382383655824, "bayes_factor_description": "If comparison level is `exact match` then comparison is 222.50 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": "Exact match", "log2_bayes_factor": 7.797686319483587, "m_probability": 0.3888805569756063, "m_probability_description": "Amongst matching record comparisons, 38.89% of records are in the exact match comparison level", "max_comparison_vector_value": 2, "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": 92.7048728526088, "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 92.70 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 <= 1", "log2_bayes_factor": 6.534573268209908, "m_probability": 0.1484762728370111, "m_probability_description": "Amongst matching record comparisons, 14.85% of records are in the levenshtein <= 1 comparison level", "max_comparison_vector_value": 2, "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": 0.46419793122630476, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 2.15 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": -1.107188001796095, "m_probability": 0.4626431701873826, "m_probability_description": "Amongst matching record comparisons, 46.26% 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": 0.9966506506506506, "u_probability_description": "Amongst non-matching record comparisons, 99.67% of records are in the all other comparisons comparison level" }, { "bayes_factor": 10.264353562890939, "bayes_factor_description": "If comparison level is `exact match` then comparison is 10.26 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": 3.3595708659307335, "m_probability": 0.5660541687277061, "m_probability_description": "Amongst matching record comparisons, 56.61% 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": 0.4592736580190661, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 2.18 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": -1.1225740557985633, "m_probability": 0.4339458312722939, "m_probability_description": "Amongst matching record comparisons, 43.39% 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": 255.301619971695, "bayes_factor_description": "If comparison level is `exact match` then comparison is 255.30 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": 7.99605888191739, "m_probability": 0.560098900557462, "m_probability_description": "Amongst matching record comparisons, 56.01% 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": 235.63777521191432, "bayes_factor_description": "If comparison level is `levenshtein <= 1` then comparison is 235.64 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": 7.880427026661069, "m_probability": 0.17282498809234997, "m_probability_description": "Amongst matching record comparisons, 17.28% 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": 206.57183856156232, "bayes_factor_description": "If comparison level is `levenshtein <= 2` then comparison is 206.57 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": 7.690499778262131, "m_probability": 0.128249357737358, "m_probability_description": "Amongst matching record comparisons, 12.82% 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": 0.13932108608389637, "bayes_factor_description": "If comparison level is `all other comparisons` then comparison is 7.18 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": -2.8435144702111006, "m_probability": 0.13882675361282992, "m_probability_description": "Amongst matching record comparisons, 13.88% 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" } ] }, "hconcat": [ { "encoding": { "color": { "value": "green" }, "row": { "field": "comparison_name", "header": { "labelAlign": "left", "labelAnchor": "middle", "labelAngle": 0 }, "sort": { "field": "comparison_sort_order" }, "type": "nominal" }, "tooltip": [ { "field": "m_probability_description", "title": "m probability description", "type": "nominal" }, { "field": "comparison_name", "title": "Comparison column name", "type": "nominal" }, { "field": "label_for_charts", "title": "Label", "type": "ordinal" }, { "field": "sql_condition", "title": "SQL condition", "type": "nominal" }, { "field": "m_probability", "format": ".4p", "title": "m probability", "type": "quantitative" }, { "field": "u_probability", "format": ".4p", "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" } ], "x": { "axis": { "title": "Proportion of record comparisons" }, "field": "m_probability", "type": "quantitative" }, "y": { "axis": { "title": null }, "field": "label_for_charts", "sort": { "field": "comparison_vector_value", "order": "descending" }, "type": "nominal" } }, "height": { "step": 12 }, "mark": "bar", "resolve": { "scale": { "y": "independent" } }, "title": { "fontSize": 12, "fontWeight": "bold", "text": "Amongst matching record comparisons:" }, "transform": [ { "filter": "(datum.bayes_factor != 'no-op filter due to vega lite issue 4680')" } ], "width": 150 }, { "encoding": { "color": { "value": "red" }, "row": { "field": "comparison_name", "header": { "labels": false }, "sort": { "field": "comparison_sort_order" }, "type": "nominal" }, "tooltip": [ { "field": "u_probability_description", "title": "u probability description", "type": "nominal" }, { "field": "comparison_name", "title": "Comparison column name", "type": "nominal" }, { "field": "label_for_charts", "title": "Label", "type": "ordinal" }, { "field": "sql_condition", "title": "SQL condition", "type": "nominal" }, { "field": "m_probability", "format": ".4p", "title": "m probability", "type": "quantitative" }, { "field": "u_probability", "format": ".4p", "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" } ], "x": { "axis": { "title": "Proportion of record comparisons" }, "field": "u_probability", "type": "quantitative" }, "y": { "axis": { "title": null }, "field": "label_for_charts", "sort": { "field": "comparison_vector_value", "order": "descending" }, "type": "nominal" } }, "height": { "step": 12 }, "mark": "bar", "resolve": { "scale": { "y": "independent" } }, "title": { "fontSize": 12, "fontWeight": "bold", "text": "Amongst non-matching record comparisons:" }, "transform": [ { "filter": "(datum.bayes_factor != 'no-op filter2 due to vega lite issue 4680')" } ], "width": 150 } ], "title": { "subtitle": "(m and u probabilities)", "text": "Proportion of record comparisons in each comparison level by match status" } }, "image/png": "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", "text/plain": [ "\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.m_u_parameters_chart()" ] }, { "cell_type": "markdown", "id": "c44fcc26", "metadata": {}, "source": [ "### Saving the model\n", "\n", "We can save the model, including our estimated parameters, to a `.json` file, so we can use it in the next tutorial." ] }, { "cell_type": "code", "execution_count": 10, "id": "992703a7", "metadata": {}, "outputs": [], "source": [ "settings = linker.save_settings_to_json(\"./demo_settings/saved_model_from_demo.json\", overwrite=True)" ] }, { "cell_type": "markdown", "id": "d07e6901-110f-449e-9534-8e24bbf1d5fb", "metadata": {}, "source": [ "## Detecting unlinkable records\n", "\n", "An interesting application of our trained model that is useful to explore before making any predictions is to detect 'unlinkable' records.\n", "\n", "Unlinkable records are those which do not contain enough information to be linked. A simple example would be a record containing only 'John Smith', and null in all other fields. This record may link to other records, but we'll never know because there's not enough information to disambiguate any potential links. Unlinkable records can be found by linking records to themselves - if, even when matched to themselves, they don't meet the match threshold score, we can be sure they will never link to anything.\n" ] }, { "cell_type": "code", "execution_count": 11, "id": "b0f17f7c-fa83-41b5-b2da-25ae18e11d81", "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 } }, "data": { "values": [ { "cum_prop": 0.0020000000949949026, "match_probability": 0.42655, "match_weight": -0.43, "prop": 0.0020000000949949026 }, { "cum_prop": 0.003000000142492354, "match_probability": 0.96071, "match_weight": 4.61, "prop": 0.0010000000474974513 }, { "cum_prop": 0.006000000168569386, "match_probability": 0.98457, "match_weight": 6, "prop": 0.003000000026077032 }, { "cum_prop": 0.013000000384636223, "match_probability": 0.9852, "match_weight": 6.06, "prop": 0.007000000216066837 }, { "cum_prop": 0.014000000432133675, "match_probability": 0.9856, "match_weight": 6.1, "prop": 0.0010000000474974513 }, { "cum_prop": 0.019000000320374966, "match_probability": 0.99414, "match_weight": 7.41, "prop": 0.004999999888241291 }, { "cum_prop": 0.02100000041536987, "match_probability": 0.99438, "match_weight": 7.47, "prop": 0.0020000000949949026 }, { "cum_prop": 0.027000000467523932, "match_probability": 0.99476, "match_weight": 7.57, "prop": 0.006000000052154064 }, { "cum_prop": 0.030000000493600965, "match_probability": 0.99709, "match_weight": 8.42, "prop": 0.003000000026077032 }, { "cum_prop": 0.038000000873580575, "match_probability": 0.99802, "match_weight": 8.98, "prop": 0.00800000037997961 }, { "cum_prop": 0.04100000089965761, "match_probability": 0.9987, "match_weight": 9.58, "prop": 0.003000000026077032 }, { "cum_prop": 0.04200000094715506, "match_probability": 0.99888, "match_weight": 9.8, "prop": 0.0010000000474974513 }, { "cum_prop": 0.04300000099465251, "match_probability": 0.99931, "match_weight": 10.51, "prop": 0.0010000000474974513 }, { "cum_prop": 0.04400000104214996, "match_probability": 0.99939, "match_weight": 10.67, "prop": 0.0010000000474974513 }, { "cum_prop": 0.04500000108964741, "match_probability": 0.99942, "match_weight": 10.75, "prop": 0.0010000000474974513 }, { "cum_prop": 0.047000001184642315, "match_probability": 0.99945, "match_weight": 10.82, "prop": 0.0020000000949949026 }, { "cum_prop": 0.04900000127963722, "match_probability": 0.99946, "match_weight": 10.84, "prop": 0.0020000000949949026 }, { "cum_prop": 0.05000000132713467, "match_probability": 0.99948, "match_weight": 10.9, "prop": 0.0010000000474974513 }, { "cum_prop": 0.05100000137463212, "match_probability": 0.99952, "match_weight": 11.03, "prop": 0.0010000000474974513 }, { "cum_prop": 0.05200000142212957, "match_probability": 0.99954, "match_weight": 11.09, "prop": 0.0010000000474974513 }, { "cum_prop": 0.05300000146962702, "match_probability": 0.99958, "match_weight": 11.2, "prop": 0.0010000000474974513 }, { "cum_prop": 0.054000001517124474, "match_probability": 0.99959, "match_weight": 11.26, "prop": 0.0010000000474974513 }, { "cum_prop": 0.055000001564621925, "match_probability": 0.99969, "match_weight": 11.67, "prop": 0.0010000000474974513 }, { "cum_prop": 0.05800000159069896, "match_probability": 0.99971, "match_weight": 11.73, "prop": 0.003000000026077032 }, { "cum_prop": 0.05900000163819641, "match_probability": 0.99974, "match_weight": 11.9, "prop": 0.0010000000474974513 }, { "cum_prop": 0.06300000182818621, "match_probability": 0.99976, "match_weight": 12.03, "prop": 0.004000000189989805 }, { "cum_prop": 0.06400000187568367, "match_probability": 0.99977, "match_weight": 12.1, "prop": 0.0010000000474974513 }, { "cum_prop": 0.06500000192318112, "match_probability": 0.99979, "match_weight": 12.25, "prop": 0.0010000000474974513 }, { "cum_prop": 0.06600000197067857, "match_probability": 0.9998, "match_weight": 12.32, "prop": 0.0010000000474974513 }, { "cum_prop": 0.06800000206567347, "match_probability": 0.99981, "match_weight": 12.33, "prop": 0.0020000000949949026 }, { "cum_prop": 0.09200000227428973, "match_probability": 0.99982, "match_weight": 12.48, "prop": 0.024000000208616257 }, { "cum_prop": 0.09500000230036676, "match_probability": 0.99984, "match_weight": 12.61, "prop": 0.003000000026077032 }, { "cum_prop": 0.09600000234786421, "match_probability": 0.99985, "match_weight": 12.71, "prop": 0.0010000000474974513 }, { "cum_prop": 0.09700000239536166, "match_probability": 0.99987, "match_weight": 12.9, "prop": 0.0010000000474974513 }, { "cum_prop": 0.09800000244285911, "match_probability": 0.99991, "match_weight": 13.42, "prop": 0.0010000000474974513 }, { "cum_prop": 0.1210000024875626, "match_probability": 0.99993, "match_weight": 13.9, "prop": 0.023000000044703484 }, { "cum_prop": 0.15500000433530658, "match_probability": 0.99994, "match_weight": 14.05, "prop": 0.03400000184774399 }, { "cum_prop": 0.15600000438280404, "match_probability": 0.99995, "match_weight": 14.42, "prop": 0.0010000000474974513 }, { "cum_prop": 0.16800000448711216, "match_probability": 0.99997, "match_weight": 14.9, "prop": 0.012000000104308128 }, { "cum_prop": 0.21600000490434468, "match_probability": 0.99998, "match_weight": 16, "prop": 0.04800000041723251 }, { "cum_prop": 0.24600000423379242, "match_probability": 0.99999, "match_weight": 17.58, "prop": 0.029999999329447746 } ] }, "height": 400, "layer": [ { "encoding": { "x": { "axis": { "format": "+", "title": "Threshold match weight" }, "field": "match_weight", "type": "quantitative" }, "y": { "axis": { "format": "%", "title": "Percentage of unlinkable records" }, "field": "cum_prop", "type": "quantitative" } }, "mark": "line" }, { "encoding": { "opacity": { "value": 0 }, "tooltip": [ { "field": "match_weight", "format": "+.5", "title": "Match weight", "type": "quantitative" }, { "field": "match_probability", "format": ".5", "title": "Match probability", "type": "quantitative" }, { "field": "cum_prop", "format": ".3%", "title": "Proportion of unlinkable records", "type": "quantitative" } ], "x": { "field": "match_weight", "type": "quantitative" }, "y": { "field": "cum_prop", "type": "quantitative" } }, "mark": "point", "selection": { "selector112": { "empty": "none", "fields": [ "match_weight", "cum_prop" ], "nearest": true, "on": "mouseover", "type": "single" } } }, { "encoding": { "opacity": { "condition": { "selection": "selector112", "value": 1 }, "value": 0 }, "x": { "axis": { "title": "Threshold match weight" }, "field": "match_weight", "type": "quantitative" }, "y": { "axis": { "format": "%", "title": "Percentage of unlinkable records" }, "field": "cum_prop", "type": "quantitative" } }, "mark": "point" }, { "encoding": { "x": { "field": "match_weight", "type": "quantitative" } }, "mark": { "color": "gray", "type": "rule" }, "transform": [ { "filter": { "selection": "selector112" } } ] }, { "encoding": { "y": { "field": "cum_prop", "type": "quantitative" } }, "mark": { "color": "gray", "type": "rule" }, "transform": [ { "filter": { "selection": "selector112" } } ] } ], "title": { "subtitle": "Records with insufficient information to exceed a given match threshold", "text": "Unlinkable records" }, "width": 400 }, "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcYAAAHeCAYAAAAB56vfAAAAAXNSR0IArs4c6QAAIABJREFUeF7snQm4TdX7x7/mMUMyNag0kAxFQlSmElGRKbOfSonI+CNzpGSWpH5KoYEoEiUqQpRLJEXKPFzzPA//57v89+04956z99nvPufezbuepyf33v2uvdZnvXt991p7rXelunDhwgVoUgJKQAkoASWgBAyBVCqM6glKQAkoASWgBP4loMKo3qAElIASUAJKIICACqO6gxJQAkpACSgBFUb1ASWgBJSAElACSRPQEaN6xhVLgOvOUqVKdUn9rbVowb8PBSkwj+PHj+PMmTPIkiUL0qZNm2xcI61DshVUb6wEUigBFcYU2jBarOgS6Ny5M4YOHYpevXqhf//+5mYUNooa0759+3D11VeHLcS4cePw3HPPYeDAgejRowfq1q2LadOm4fvvv0fFihXD2i5btgxly5ZFixYt8P7770dU2QoVKmDx4sXYsGEDbrnllkS2DRo0wJQpU/Djjz+C1/o5vfrqq3j55Zfx5ptv4oUXXvBzVbTsPiKgwuijxtKiekfAEsaePXvilVdeSSSMe/fuRa5cucLe8O2338bzzz+PAQMGmM577NixWLlyJTp27IjChQuHtV2yZAnKly+PRo0aYfLkyRFVzBLGv/76C7feemtIYVy4cCHuv//+iPJOaRfzpYNtNGrUKLRr1y6lFU/Lc5kSUGG8TBtWqxWegBNh5MhvzJgxqFGjBn755RfExcWhSpUq6N69O26++WYECyNHnwsWLMDIkSORNWtWPPvssyhTpgyOHTuG2bNno0SJEnjxxRfNaDJYGMePH49PP/0U+fLlw7vvvos1a9aY0SzveeONN5qR33//+1/kyZPH/JsjRo50v/zySxw6dAhPPvkk+vbti/Tp08MaMVrC+N133xnxX7t2LW677TYz8nrqqacSAaJAcxTcpEkTzJs3D3nz5sXo0aMRzp7l5GiZI2DW76GHHjICljFjRhw4cMD8bdasWaaMd911lxldW2L9wAMP4LrrrjP1eeuttzBjxgxjxxeLn376CXfeeaep78SJExOE8bfffjN8Pv/8c+TPnx+VKlVCv379jJ0mJeAVARVGr0hqPr4i4EQYP/roIyNkTAULFsSePXtw5MgR0/FzBBMsjIFTqdmzZ0fJkiWN7VVXXYXcuXPjn3/+we23345169ZdIoxPPPEE6tevb66jwPCaa6+9Frt370bt2rWxbds2I8zWtKsljMy7dOnS5m9MFPE2bdpcIozZsmUzgsT06KOPgmLJOnzyySfmusD0+uuvG/G1UqtWrUxdQ9lXrlzZTOUyP4rdr7/+av7drVs3DBo0CA8++KCZzqW4UZAp5kyrVq1C8eLFE33f/fPPP1GnTh0j4GRBhqw7E3lzdE6xXL9+PTp16oSlS5eaPEeMGIH27dv7yv+0sCmbgApjym4fLV2UCEQijLVq1cJnn32GzZs3G9FiRx8fH+9IGHnt77//jhw5cpjREcWOHTtFllOp119/fULnbwkGv29yBHnNNdeYayZNmmSma/nvRYsWJYwY+W2SYsnRHUdqFKr58+dfIoz81sjvcxSr1q1b45tvvjECY+WVlDAWKVLEjMruu+8+I4yh7DnqbNu2bcJ0MIWta9euuOmmm9CsWTMj2hQ41pkjOv7tjTfewDPPPIN33nknQRg54qOwcfRJ0Sczvjxw1G0xozBSNMmLiflQXMmRLx0PP/xwlDxFs70SCagwXomtrnU2U3wc1XDkMWTIEENk//79Cd8VOfX3wQcfmBEjR1Ls1A8fPmxGMezs+W8nI8Zq1arh66+/Nvnfe++9ZnRHAWD+FKfAtGLFCtx99904f/48unTpgmHDhl3y92BhpBAVKlTokkVDXBXbuHFjs/iGo0NOXVIMg5Ml7kkJI4Wqd+/e5k+PPPJISHtOubKMSY3YOCKlcD7++OP44osvTF5fffUVatasaUaXLJu18vfEiRNGOPkC0LRp00u+u3Jadfjw4QlTqdbPVrkplGwHjoY1KQGvCKgwekVS8/EVAavj5vc+fktk4vdB/szOduvWreb7GoWRwkkBdSOM7LD5jS2UMFKgKB4coXGExW9r/B752GOPmRHRxx9/jB07dpgRYbAwcgUsR1H8DnnPPfck/D3wG+N7772HCRMmmO+VnBqlcHK0SyEqVarUJW1mTaUGCl3Lli1D2nPKkyNDjkZfe+01M/IlM/Lj90ZOpVoseSMrf2vBkSWM1vYSSzhZb46emTjNzG+PHDHWq1fPjL55/ZYtWwxXfmskQzJKkyaNr3xQC5tyCagwpty20ZJFkcDff/+dsKKT4sVvZRRLTvtRQP73v//FRBgpEhRFjvwoLFz8csMNN5gFP9zOwW9+FB1+T+PPFE7rGyNHriwrp3lpy+lFThEHCiNFkIJC8eAoec6cOWYEyJEXt6skNWIMFEbmHcqe+XG6l4kj02+//RY//PCD+TdfJDgly++q1atXN6t0OfJjovDzd8HCyMU6XGjE75Scvs6UKZMZ+TJRGPm9lWxYlz59+pi9opwe5vdfTk+rMEbxgbnCslZhvMIaXKv7LwGKBL/RUQytRBHgCCtz5swJwkgBoZCww+ZiFmsqNdQ+RooDr+Pim1AjRuZVrlw5swKUqy45+uPiHebNTp7TkMyHiWVkmfi3TZs2mdEkF51wmtaaJuVId/r06ciZM2eCMFr7GK29gFYdKTAcSfK7Z2AaPHiwGf0FT42Gs6fYkY2VuDqWU9DcD8oVpKwT62MliyV/DhZG/o7fSzkKJh8mCiRX3nIkyu+ZwWUhE44auVpYkxLwioAKo1ckNR/fEuBo6+DBg2YEmZKW/bNcFFj+FypxepfbQbh1IVziNRs3bjTX2e3PTCqfcPb8G0eGzNsaQVp5cNqT9WA5OSp3wvfUqVPmBYCLeDJkyJCoOPw+y6lUbk0JdY1vnVELniIIqDCmiGbQQigBJaAElEBKIaDCmFJaQsuhBJSAElACKYKACmOKaAYthBJQAkpACaQUAiqMKaUltBxKQAkoASWQIgioMCZzM5w9ezYmRxTxPqlTpzb/JVfiQgwunLBWQ3LlIZfkJ+cRTZGw8Ft5I6mbX69lMAT+5xcfSi7O586dM7eOZEsLn1faRYut132Sl/ldUcKYVCNzuTeXunOJOvdDxTIxAkqxYsXMhmWvEve4Pf3002aJP5exc7M0VwPy9zwiidsDwqVAe6/KxHwYA5R79hj9hZvWuf+MYcG4pJ9L7adOnQoGlU4qcQk/o7lwT56bxFWODJDdsGFDsw0jMDmtb2B5ixYt6qYYrmwYhu7nn382ewmlHIILEOgfrgqXAowY9Wbu3Llmq4qXyTqCLNQJJl7eS5oX97iybwuOpBSYL/d9bt++3ezPDZeC/Y0BHhjsIhrJaZ8UeG9u1/njjz/MFqfgxP277EO8OJ7sihTGwE6YcSkZporLvhlnMpYpGsLIY494QgOXzjP6CI/sYafq1AkD7b1kwVifPFGC0Ur4EHOpPffiMR4nH2xuAA/eV2fdnxu/GTuTG9zdJCvUGzsGBucOTE7qy2gxgeV1Uwa3NtzD16FDBzAggZRDcBkC/cNt+ZLbLlrCyFEo49IyolDwy1Ry1zn4/gzqwLB6DCIfKjkVxkB/44tYShNGBo9gxKWkjmpTYXTpmdaIkQ0euCGYx/cw+gjfjE6fPm3iRFI82VEzZiVHGkwc3TAA8+rVq83GbR7lw71vfIPhqQYMzWVF+ODbGztdNiSPHmIcSIa5YiQRxujklAaDPjNMFkeM/BvfhrjpnKG6+MAzGkpgYngsOipDiHHUxWggDGfGvWEc0TCuJevGqCkUG4oRQ3Ix2DSPJGJHSKfifxy9MdRZ8NQK37ZoTz58+2K+jId58uRJc9wSR6NMLDc3ifPNmtfwPEJGZWGUlpkzZ5prKHzcAM7oLBzB8uF96aWXDEeOVrgBnkct8cFmfE6Wj3yYF/cV0tG56ZwvEGTDUUGo9mH5QpWXRxNxszxHz+TDoNNWclJfbjgPLC+F0kl706fYBmxPvqkz5BtFjhFjOBJhuDnWlYlhz7j5nS8ADIhNjqxr1apVzR5BbpxnFByLA32ZfsQgA7yOI2r6GjfNh2s3q94MERfoH2SUVH6c6g5ODDbAdj169KiZgWBebDcGG2egAh5Xxbx42gb9lGHcknpudu3aZYKHs23Y9vQpxopl4iiZfkEeDGTA+3H/JQWLzx1PPmGUHAYZ5/R8UiNGloHPHafY+LwyMlCwv5MjmfK6O+64w7QRr6GfMgwf78OZFvYB1mwLr+Vzys45VDnZLnwR495RjmjpAwx8EOh7rCevs4LK81nmKSsM/GD5BZ89Kw5sUvXhs9a8eXPDjDaMOMR25dmg3FNK7gyMQGFk/8TAC+TN8nDUFbinlWUN9DfWm37F55fXMuIQy8uXBfYpvJ515FQrf5+UX/Bvofo2vqyH6pNYL/Y37JPZT9K32JcFCiP7CAa55ws3X7C5X5YBJXTEGKFAWsLIh5hxHClIDMTMzp5CwDPh2PlQFPmQstOn09FhOVrhg8jRJTt5PmR0FL6lURw5JcoOj5FIeB4fT1fnQ81OiicyUGD5oHFkyg6RjsnfMeoKy8GHgWViVBJO6/J7lhVj06omHZQjFzo9HZYPLcWHeVEcKTh0Zl5HR6HAsfzsxCmqvJYOzI6CzsPYoMHTl3RW2lOU+BCx7Iw6wtBcPPGB4kZHZ/QVdubciM1zByk4HH3z39bUC2Nfsq50XHLlVCgfdI7g+OLAEGyMbMKzDcn86quvNhwp+DymiJ0mO12OgK2p1FDtwynaUOVlwGp2ruw82cmQoZWc1JfBva3yUrQosE7am23NNqeokSc7KLY325eJIklW9COGOuMLGV+i6Ec8XomdAH2J17MjJjeLA0WXnSA7S5aFv6dfsjMJxSFwcz19M9A/KPxJ5Wcd4mzxYjloxzaikPBFjadlMMg6Ozn6MSPh8N98FuhfST03LD/rypdP2nK2hi8DHBVz6pttzk6PHR1fjhgggO1In6GP8UWKLwT0Bz4zwcLIZ4GdPKfrGQ2IsyZ82eS1gYkjJLKjb7MvYBtYYfrIkaLFfoHiToFj4ksE/+MLYahyUoj4ws1yMjQenzf+ZwVnt8pgXcffs+Nn3egPbHcrxB5nC0LVh/0S/YjM+KJLjuxX+EJClgzKzhcrihbLQ9/iyweFn+XnjJKV2N8F+xv9l8z+85//mJcTvpiwP7QOkOZzxWhJfDFNyi9Yr1B9mxXaMLhPoqDT/1kvvpDQBymCjOTElxhrxEjR51mh7CMo9mxDCqgKo0thTMqMQsYOmm+LfDuy3g4ZJJkPL2M7suEZ5YNTK3xb4kNHMePf+ODyeyUTI5Wwseno7Bis7xR8y2InZwVIZiPy4acw8jsgr2XDM392XOzsAxM7BT5I/Bvfypk/H2J22hxF8g2cJxdQ2NhhBU+lUvz5oDJRTOlY1ggwlFDQ8dgJWOHQODrmQ8o3Wz6ELDMdlW/CHCEkJYx8sPlQ8vsFnZcPMTtp/p7fdVkXCiM7XYo3O0Amii0jnBQoUMB0Xjt37gzZPuxU2ZElVV52NHwzTmoqNVgYk7KnX1jlZUfptL2t46Asn2G7sO04RccOhr7GlxN+s+SDzZEpj1EiK8428LrgqS1LGOkb7JQs4eKIlG3L/ENx4GxGYAr0j1D5BX9f4gsReVrfqjhrwJdI3pdtR570S75AsK0tIQ1+bih4HHnwOeILBJ8BigxFiPfk6JltzxEw82bZ2f4UUfqaFeeVzJiChZF+SmGg+HIkQWHkSxGfn8DEZ4AvYVZYO7748gQVCrcljNYRYXwxoRiznBRR1i1UOfnySgGxfJmCRP8PngJkefgcsa2Z+CzwRdYa4ZER78n+KVR9AqdS2Y6sgyXAfBmgMPMegffhCwU5B5cn2N8ojJwZIAvWly947OsojHw5YptwQV84vwjVt9FXkuqTKORkxxkoJrLmSxj5sW9mf8NRu3WoN+/Nl32+WPNFUYXRpTCyE7ZGjHz4OFRnh8M3D07NBCeKAK/haIcPaWCiANJB+EYZKC588Ck+fBjpSEx8e6PIcgTGxAUpfONjp0CnpchxRMHpHD4YwUfpUPj47ZDTN3R+igFHbXQwOg87K6ujT0oYrUU5vDcdkp1KsBMFC0Xg4gN2UhQ/duTs5Dk1xJcB3osOyU4+UBhZNnY4ToWRo292uhYfi6e16IRiEap9+NLBhzep8nI05lQYk7Jnp2wJI9/inbZ38GIZvgRxaskKps0XKB7JxBcbTn/xbZ0vIBQWjrLCCSNt2UnwhYiJ/sMREgUoFAcKYShhDJVf8MIwTu8Hxj5lflbsWP6bgkhGluDx+UjqueEzyOcqOFFQ+CzxpTE48YWMMyEUYz5fTBwlcUYiWBjZ6XNqkYysF1aOiIOFkULE/KxnzZqqCxRGduxsD84SccaEZeO0pHVWZVLlZP3Yt9AvmfjywOc9uJwsDzt8juiYyJfPNUdprC9nJ3hPinOo+gQKI9uRLyR8EQhMwffhC/jy5csTlSfUixjzoj/yxZs+EfytL5xfhOrbghe+WX0SXyb4nFvHrrH+HCywvDzCjcJI4bdeliiQTBy4sB1VGBM9OuF/EeobI9+e2Oic+uHhqJzftr5B8gHjm701183vS+zceMwNv6ews6FI8W2O19Fp+MbJBuT9AoWRb/Ts3K3z6fi2xpEppyHY2MyL39MoELyGb7rp0qVL1JnRIVgGfjek+NLBOGXKt7lwwhi4KtWpMPLtjSMmJksYKe6MYcn/2PlzitD6XsXpPGv1KAWOI2SnwshRE6ctLT48u5AM2LHxxYVlCdU+XFRDQUiqvJEIY1L2gcLI8jht7+DFC+xI2Z7WA28JI7+R8SWFD/5dd91lZhs4TR5OGDmKD/yeQkHhiwg73lAcwgljqPyssyQtJ7SmxawRCf2eL2z0Q7YXRyxsL/qB9Q2cL5XBzw07Uo54aWsJF58BChDt+CxaQdRpy+eCL4zMi50+p9mZ+CK2d+/eRB08pwr5nZ2zLMyTIszpuWBhpOhRMDlKY2IbcWQYKIwcUVojPrYVv+XT5/kiHaqcFI7AVaDhhDHwOnLhtfz0ESiMnCIPVZ9AYeQMDqfuLT70B864UPAC7+NUGAMX34QTxlB+wZfoUH0bfS6pPok+xRdwfvJisjiwn+SLEPPjSJjPD19QuFaBic85+3IVRo+EkW/qfGvkFCcbmI7Ezonix4eXDwAfdM5902kpQnyY+CbDB4ijEb5NchTFt2UKFB8uTgUECiM7Pj7Y7MD4XYgjRF7PBuf0EL8L8HrrhAF2GsEr4qwDdtkR0qn4xstRJq/lvwOFkZ0VOw5+SwhelSoRRnZYnFZhefkSwO0u7Kw4Cmfny5Ejp2/4M+vmVBhZV/Jh3hQjjj7YUXEEZU0hhmofvmCEEgR2jPzOxM6VnWNgCh4h2wmjJbJO2tupMLIT4Fs+/Y9+w6lafgel//A7Lb/vsFMjd4sDfZYvVuz4yYejRYor29upMAb6R6j8+J0nMPFZ4PdmHh1FQeP9uECHswf8dkWe9AVOwbEjpZgl9dywg6PPkyO/Q/M54AsH24h15XNHP6I9r+HLIn/Ply1Oo3HRFqfk6S8ccQePxJgXX+h4PV+26KOsC6epAxOfWXa8zI/s2Vnzu3ewMPKF1uqArSl5fnMMVU5OcXspjHwJClUfigFH6KwDZyPIhy/tnD2gMHGEz985EcZgf3MqjKH8giPhUH0bn++khJHPONuLPkZ/4ksBZxE4YAkcqXIKldOnrDN9haNs/cYYoSjycmvEyKE9HygrWSeH07k47Ulx4siAibDpcHzT58PCjoCJDywbnQLAh5ZvkOwoOBXGTo6dGTsGDu+tqVR+V+KbK52WidNgFEmOMjmy4lsupxIpuFxwQKcJTtaUGTtRTrPwGyidhm9OTIHTE5aIsiNjx+d0xMgpV5YzqQ6W96UA89uHNZpjOei0fINjfdnJkQUZs9OxhJFTS5zu5TcTjjYtEbK+MVoiZU0rsYNi2cnREgRrRXBw+3BUEaq8LJ/1oT/4O6PFK1x9KdKB5XXa3k6Eke3PUQKnqDhdxsQ3Xk6n8eWMHTwFh6N03tfiwBchdsrkYfkjF2txRB2OQ6A/BfoH759UfvTHwMRFEByR8FusdV++2VOkOZ3OZ4gvinyRpLCz/SmaST03fK6sFd/Mi6MYdoJMfPG0RtYsA1/C6E/8psWXAI4amPiscprbGl1YZaVoWdOt7Jjpl+w0+RISOHJmfhRLjkB4H44O+R9ZB09Js50oToGj6FDlDN4ewRdqjqiDyxl8He/BZ599Q+BIiR1/qPpYsxjWKJYvVnwZ4DPIkSNFOvg+ZM1rgsvDwUCgv4USxuD9hKH8goOMUH1bqJd1jt45aOBLH+vAFy+OeDkbEXhfq39ln8u243oFviDriNGFODox4YdcfuxmJ8O38cDElVucTgk+5oeOQQdl49jte+JKTj7M1ty4lT/fWJkHxdaraBPMk6O6SCJeOGHEt3UKG0diwfVgp803ucDVn07ytK7h6laWN9S+xnDtE+4+nKJheb1IkbS3k/vx5YgvEFyIwranj5EfOwXWly9L1pSjlR9f9ChEvI4jWet8Qyf3C/Q5yz+c5sey8rsYfYCrlp3cN9RzYx1nxW9KwW3DFwVOywcfB0YeZMXp82AmgXVn+djRkynLyPzol4HPAhcM0c94D778UNA5yuL0pNMUqpxO7Z1eF64+fPm2Pm8wP/YjFAs3z2Aof7MrZzi/cNO3se0oepzuDtUf0q/4ws1rgj872ZU33N+vqA3+ElBqqwSUwOVHgKMSbm/gNCtfdjizwpFu8JT75VdzrZEKo/qAElACSiAJAvy8wW9Z/H7LESu/W/H7v6Yrm4COGK/s9tfaKwEloASUQBABFUZ1CSWgBJSAElACAQRUGNUdlIASUAJKQAmoMKoPKAEloASUgBJImkBUR4xc9ss4e8HL7rlEmx+9wy215sZwLr/1epuBOoISUAJKQAkogXAEoiaM3JzLDcoM4ss9NtzgyggkjDTBPVoMPMuNpAyTxmgG/B33/3EjKvccMa4o7e32BGrzKgEloASUgBLwkkBUhNGKbGJFZWccTW7UZQgqBsBmBH1uBuVRN1wazWNLGNKJm2sZSYKbU3k9f9akBJSAElACSiCWBKIijKyAFWWEkQkYl5OhiRjGifFCGb6MEWUomAxfxPiE3ENEQWQkCob1YYgpr6K/xBKo3ksJKAEloAT8TSBqwkgsjMXHWHsMscT4oTz7jEeKcEMtg/Iy4gTD+fB7oxUlnXFLGTuTxzvxWp6/xqOZmBgbj8IZmBgHkQFnNSkBJaAElIASiJQAw3gGp6gJo3UGF4+Q4WkTTDybj/H8GGyYiQJIoWOcQiaOLnkSBIPh8gBQiiIDyvIcs1DxGPntklOxKTUxnmhS4FNKebV8spZQfspPRkBmrf4XHX5REUbrTEKeLhB4Cj0PDeUJFfw9p015vAgDxVorTxlVnpHz+X2R0eR53iCPlKEwhppWVWGMjmPIcvXOWh98GUvlp/xkBGTWfvW/qAgjT6bg0S2BiWcYcpUpz17jVCpXm3JFqnXsDCPU8/BTnjPII4S48IYHc3KVKo9WCpVUGK9Mx5XV2jtrvz743hGQ5aT8lJ+MgMw6lP9FRRjtisqFOTy7L3CPonVmIX8fSVJhjIRW4mu1Y1J+MgIya/U/5ScjILNOUcIoq8ql1iqMMpraMSk/GQGZtfqf8pMRkFmrMMr4ubbWB981OmOo/JSfjIDMWv3vyuSXLFOpMtQ6YvSSnz74MprKT/nJCMis1f+iw0+FUcbV1lod1xZR2AuUn/KTEZBZq/9dmfxUGGXtbmutD5YtIhVGGSLlp/yiSECWtV/7PxVGWbvbWvvVMWwrFqMLlJ8MtPJTfjICMmu/+p8Ko6zdba396hi2FYvRBcpPBlr5KT8ZAZm1X/1PhVHW7rbWfnUM24rF6ALlJwOt/JSfjIDM2q/+p8Ioa3dba786hm3FYnSB8pOBVn7KT0ZAZu1X/1NhlLW7rbVfHcO2YjG6QPnJQCs/5ScjILP2q/+pMMra3dbar45hW7EYXaD8ZKCVn/KTEZBZ+9X/VBhl7W5r7VfHsK1YjC5QfjLQyk/5yQjIrP3qfyqMsna3tfarY9hWLEYXKD8ZaOWn/GQEZNZ+9T8VRlm721r71TFsKxajC5SfDLTyU34yAjJrv/qfCqOs3W2t/eoYthWL0QXKTwZa+Sk/GQGZtV/9T4VR1u621n51DNuKxegC5ScDrfyUn4yAzNqv/qfCKGt3W2u/OoZtxWJ0gfKTgVZ+yk9GQGadXP73v69+NQVv8lBRZEyfNmQlQpVPhVHW7rbWyeUYtgX7/wu0fE5JJX2d8lN+MgIya/W/xPx++2c3erz7vRHEt16qjtw5MqswytzMe2t1XBlT5af8ZARk1up//uJ3/NQZPD90NvYfOYm2dUqjWumCYSugI0ZZ+7q21gfLNTpjqPyUn4yAzFr9z1/8Rnz2M+bHbUTJ2/OhX8sHbQuvwmiLKDoX6IMl46r8lJ+MgMxa/c8//Fas34U+7y9A5gzp8E6XR5E9SwbbwieLMJ49exZHjx5Fjhw5EhVw//79yJIlCzJkSLrwJ0+eRLp06ZAmTZqwlXv99dfRrVs3WwDJdYE+WDLyyk/5yQjIrNX//MHv6InTeG7obBw6dgpdGpbDAyUKOCp4zIVx2LBhGDduHMqUKYPDhw+DAlaoUCFT2M2bN6NYsWL4+uuvcd9992HgwIE4dOgQMmfOjD59+iBVqlRo2rSpsefvwiUVRkftH/IiffCVn4yAzFr9T/nJCFy0HjhxEZau3Y6yRa7Dy00rOM4ypsJ4+vRpMxLkaJGjwgEDBiA+Ph6jR48G/1a/fn1s3LgRY8eORb58+dC5c2dMnz4dLVq0QKdOnbBt2zZzPX+2SyqMdoTC/10sBiNmAAAgAElEQVQ7JuUnIyCzVv9TfjICwPcrN2HYlGVm6vTtTjWQNVN6x1nGVBhZqgMHDiBnzpw4fvw4qlSpgvbt26Nhw4bo2LGj+Zki2bt3bzNirFSpkrmWgrh48WI0adIEkydPRtq0ofefWDVXYXTsA0leqB2T8pMRkFmr/yk/CYH9R07g+aFzwNWoXGzDRTeRpJgLIwu3YsUKtGzZEkWLFsX48ePx1VdfYcaMGfjwww/xyCOPJAjjuXPnEBcXh5IlS2LSpEnImzcv1q1bh4ULF6Jr164oW7ZsyLqqMEbiBomv1Y5J+ckIyKzV/5SfhAD3K3Lf4sOlC6JdndIRZxVzYZw/fz4aNWqEUaNGoUGDBqbA5cqVw+7du5ErVy788ssvuP32283I8J577jF/5+iydevWaNu2LYYOHWpEsU2bNli2bJn57rho0SIzogxO9erVixiIGigBJaAElIB/Cfy4djc+W7wZObOmR496xZA+bWpXlSlYMPFex6hEvrlw4QKyZ8+OefPm4d57700o7NatW8HVpkytWrUyovfYY48lLLAZPHgwqlatar4vLliwAH379kX58uWNMIaaVtURoytfSDDSN3blJyMgs1b/U35uCOzcdxRtR3yNM+fO4fXWVXDHjde4ySbkPumoCOOGDRtw2223XVLQ5s2bY8KECQm/q1mzJnr06GG+MTLt2bPHTK1yQc6ZM2fMwpulS5eaVarNmjULWWkVRlf+oMIow6b8lJ9HBGTZXIkvFhx4dRzzLTZsP4DHy9+Op2ve7RpizKdSIy0pt3QwZcuWLSJTFcaIcCW6+Ep8sGTELrVWfjKayk/5RUpg6g9/4MNvViN/rqx4s8MjSJ82/F73cPmneGGMFI51vQqjW3IX7bRjUn4yAjJr9T/lFwmBTbsOosPoubhwARj6wkO49bqckZg7HhhEZSpVVNIIjVUYIwQWdLl2TMpPRkBmrf6n/JwSOHvuPNqN/Abb9hxGg8pF0OShYk5NQ16nI0YxQncZ6IPvjptlpfyUn4yAzFr9L+Xw++Dr1fhswR+4KV8OjGj3MNKkTiUrXJgZMx0xitGGz0AfLBlg5af8ZARk1up/KYMfF9p0GvMtUqdOhdHtq+H63JGtRQlVCx0xytrXtbU+WK7RGUPlp/xkBGTW6n/Jz+/02XNmawa3aLSoXgJPPlBYVqgAaxVGz1BGlpE+WJHxCr5a+Sk/GQGZtfpf8vMbOyMOs5duMHsVuWcxlXwGNaFSKoyy9nVtrQ+Wa3Q6YpShU37KzwMCsiyk/R/DvTHsG7dkMEB47hzhT1uKtLQqjJES8+h6qWN4VIyQ2Wj5ZISVn/KTEZBZX87+x8Dgzw+djf1HTuL5x0uhRtlbZbCSsFZh9BypswwvZ8d1RkB2lfJTfjICMmv1v+TjN/TTpfjh180oVjAPXn2mkqwgIaxVGKOC1T5TfbDsGYW7QvkpPxkBmbX6X/LwW7Z2OwZMXITMGdJhbKfquPqqTLKCqDBGhZ/rTPXBco3OGCo/5ScjILNW/4s9v0PHTuG5obNx9MRpdKxfFpXuvlFWiDDWOmKMGtrwGeuDJQOv/JSfjIDMWv0v9vz6vL8AK9bvQtki1+HlphVkBbCxVmGMKt7QmeuDJQOv/JSfjIDMWv0vtvzmxW3EyM9+RvYsGcwq1KyZ0ssKoMIYVX6uM9cHyzU6nUqVoVN+ys8DArIsIun/9hw8jjbD5+Dk6bPo1/JBlLw9n+zmDqx1xOgAUjQuicQxonF/uzy1fHaEwv9d+Sk/GQGZ9eXifzwto9u4+fhj815UKXkTOtQrIwPj0FqF0SEory+7XBzXay5O81N+TkklfZ3yU34yAjJrp/43Y/F6/G/WSlx9VUaM7VTDrEaNRVJhjAXlJO7h1DGSqXi66lMIXttXBlD5KT/GQGUsVMZE5X5F7luMVVJhjBXpoPvogy8Dr/yUn4yAzFr9L7r8Lly4gI5jvgVPz6hZ7ja0fqyk7IYRWqswRgjMq8v1wZKRVH7KT0ZAZq3+F11+n8z/HZPnrUH+XFnxZodHTEzUWCYVxljSDriXPlgy8MpP+ckIyKzV/6LHb9Oug+gwei648GboCw/h1utyym7mwlqF0QU0L0z0wZJRVH7KT0ZAZq3+Fx1+Z8+dR7uR32DbnsOoV/EONKtWXHYjl9YqjC7BSc30wZIRVH7KT0ZAZq3+Fx1+783+FZ//uA7X585mplDTpPbwkMUIipwswnj27FkcPXoUOXLkuKSoBw4cQLZs2ZAmTej55JMnTyJdunRhr2Gmr7/+Orp16xYBitheqg+WjLfyU34yAjJr9T/v+XGvIvcspk6VCiPaPYyb8l2qD7I7RmYdc2EcNmwYxo0bhzJlyuDw4cNGwDJlyoQGDRogd+7cSJs2LUqWLImePXti4MCBOHToEDJnzow+ffogVapUaNq0qbHn78IlFcbIHCH4an3wlZ+MgMxa/e/K4seoNoxuwyg3nD7lNGpyppgK4+nTp5EhQwYzWsySJQsGDBiA+Ph45M2bF2fOnEG/fv3AESGFcsmSJXjjjTcwffp0tGjRAp06dcK2bdvM9fzZLqkw2hEK/3ftmJSfjIDMWv3vyuL35ufL8c3Pf5uFNsNeeMgMgpIzxVQYWVFOl+bMmRPHjx9HlSpV0L59ezz++OMGRMaMGTFjxgx07NgRGzZsQOXKlc21FMTFixejSZMmmDx5shlV2iUVRjtCKowyQspP+RWMJgJR3n56seCJGTw5g1sy+F2RWzSSO8VcGFnhFStWoGXLlihatCjGjx9vBJGjyUGDBmHo0KH44osvjCieO3cOcXFxZmp10qRJZmS5bt06LFy4EF27dkXZsmVD8lNhlLmWnx4sWU2jY638ZFyV35XBj2cr8oxFnrXYulZJ1LzvNlnFPbKOuTDOnz8fjRo1wqhRo8x3RSZOn9avXx/p06fH6NGjkT9//kuqx9Fl69at0bZtWyOcFMU2bdpg2bJlZqS5aNEiM6IMTvXq1fMIk2ajBJSAElACXhP439y/8Nvmg7g1/1VoV7Ow19mL8itYMPGMQKoLjMnjcWKW2bNnx7x583Dvvfcm5P7OO+9g1qxZmDlzZpJ3HDx4MKpWrWq+Ly5YsAB9+/ZF+fLljTCGmlbVEaOs8fSNXfnJCMis1f8uf37bjqTFG5/8ZAKDj+1UHVdflUlWaQ+tYzpi5HfD2267dKjcvHlzM+qbMGHCJdVav369uXbPnj3o3bs3xo4daxbocOHN0qVLzSrVZs2ahUShwijzEu2YlJ+MgMxa/e/y5rd67XoM+ux3cCq1Q917UaXUzbIKe2wdU2F0U3Zu6WDi/sZIkgpjJLQSX6sdk/KTEZBZq/9d3vy6jPkaf247ZA4d5uHDKS2leGF0C0yF0S25i3baMSk/GQGZtfrf5cvv65//xpjPlyNrpvR4u1MNZM+SQVbZKFirMEYBqpMs9cF3Qin0NcpP+ckIyKzV/9zzq9dnGrih/+WmFVC2yHXuM4qipQpjFOGGy1ofLBl45af8ZARk1up/7vjFrd+Jvu8vxA3XZMFbnWq6yyQGViqMMYCc1C30wZKBV37KT0ZAZq3+547f2BlxmL10A6qVvBZt693vLpMYWKkwxgCyCqP3kLVjkjFVfspPRsCddfNBM7H/8Al0fKIIKpUp5i6TGFipMMYAsgqj95C1Y5cxVX7KT0YgcmseQMyzFrNlyYD+jYrjliQ20Eeea3QsVBijw9U2V+2YbBGFvUD5KT8ZAZm1+l/k/KZ8vxYT5/6GqqVuRq2S1yCpyDKR5xodCxXG6HC1zVUfLFtEKowyRMpP+UWRQORZdxk7D39u2YceTcojb6bTKoyRI5Rb6D5GGUMVbuUnIyCzVv+7vPgdPnYKTQZ+gTSpU+OTPnWwfetmFUZZE7uzVmF0x82y0o5J+ckIyKzV/y4vfvNXbMKIqctw92350P8/D/o2gEhUgojLmjoyaxXGyHgFX60dk/KTEZBZq/9dXvxem7wEi9dsTThayq/tq8Io80tba786hm3FYnSB8pOBVn7KT0bAufX58xdQv+80nDpzDh90fwxXZ8ukI0bn+Ly9UkeMMp7acSo/GQGZtfrf5cPv1w3x6DX+BxTImx1jOjxiKubX9tURo8wvba396hi2FYvRBcpPBlr5KT8ZAefW785aiZmL16Pug3eg+SPFVRido/P+Sh0xyphqx6n8ZARk1up/lw+/p9+Yhfj9x/B66yooctM1KoyyppVZqzDK+GnHpPxkBGTW6n+XB79tew7j+WFzkCVjOnzcuw5SpbpYL7+2r06lyvzS1tqvjmFbsRhdoPxkoJWf8pMRcGY9feGfeH/OKlS6+0Z0rF82wciv/qfC6KzdXV/lV8dwXWGPDZWfDKjyU34yAs6su7/zHdZs3IOuT5XD/cULqDA6wxa9q3QqVcZWO07lJyMgs1b/8z+/YyfPoNErn5uKfNSrtplOtZJf21dHjDK/tLX2q2PYVixGFyg/GWjlp/xkBOytF67agjc++QnFCubBq89UusTAr/6nwmjf7qIr/OoYokp7aKz8ZDCVn/KTEbC3pihSHP9T4y7Uvr+QCqM9suhfoVOpMsbacSo/GQGZtfqfv/kx2g2nUTmd+nanGrjumqtUGO2a9OzZszh69Chy5MhxyaXHjh1DpkyZkDp16pBZnDx5EunSpUOaNGnC3kaF0a4Vwv9dOyblJyMgs1b/8zc/Lrjhwpu8V2fB/7rUTFQZv7Zv1KZShw0bhnHjxqFMmTI4fPgwKGC5cuVCo0aNkDZtWmzevBldunRBixYtMHDgQBw6dAiZM2dGnz59kCpVKjRt2tTY83fhkgqjvx8su9L79cGyq1es/q78ZKSVX3h+781ehc9//BOPVyiEpx+9S4UxHK7Tp08jQ4YMZrSYJUsWDBgwAPHx8bjuuutw5MgRI4S7du1C/vz5sWbNGvTq1QvTp083ItmpUyds27bNXM+f7ZIKox0hHTHKCCk/5VcwmghEeSe3cHNTPzf3D3y6IorfkleF0a41Dxw4gJw5c+L48eOoUqUK2rdvj3nz5qFq1apo2LAhLly4YKZS//77b7Rq1cpcS0FcvHgxmjRpgsmTJ5uRpV1SYbQjpB27jJDyU34qjEn5QPyBY3h68CyzPYPbNFKn/v9wNwEXJ7dw2/luqPJFbSqVBVqxYgVatmyJokWLYvz48WjWrBnq16+PunXrmvLmzZsXy5Ytww033IC4uDiULFkSkyZNMr9ft24dFi5ciK5du6Js2YuRFBYtWmSEMzjVq1fPrv76dyWgBJSAEvCQwII18Zj+0xbcXfBqtKhyi4c5xzarggUTv/hETRjnz59vvieOGjUKDRo0MDXt378/smXLhg4dOuDcuXNmlHjw4MGERTgcXbZu3Rpt27bF0KFDjSi2adPGiCe/OyaVdMQocyK/vtHJau2dtfKTsVR+/uXXc/wPWLUhHh3rl0Glu29KsiJ+bd+oCCOnSbNnz26mTu+9994EYDNnzsSbb76JuXPnYurUqeACnZ9++inh74MHDzZTrfy+uGDBAvTt2xfly5c3whhqWlWF0b8PlpOS+/XBclK3WFyj/GSUlV/S/HgYMQ8lZl/PoOGB0W4CLfzKLyrCuGHDBtx2222XEG3evDnGjh2LGjVqYO3atThx4gS+/fZbs2qVac+ePejdu7e55syZM2bhzdKlS80qVU7BhkoqjPrgywjIrP364Mtq7Z218pOxTC5+i9dsxWuTl+COG6/B4OeqhKxEcpXPKdVk+cYYqnBbt25Fvnz5zD5FK3FLBxOnWiNJKoyR0Ep8rV8dV1Zr76yVn4yl8vMnvxFTl2H+ik1oXq046la8Q4VR1ozeW6swyphqx6T8ZARk1up//uN34QLQZOAXOHzsFN7s8AhuzJtdhVHWjN5bqzDKmGrHpPxkBGTW6n/+4/fnln3oMnYers6WCR90fyxsBfzavlH5xihr6sisVRgj4xV8tV8dV1Zr76yVn4yl8vMfv4lzf8OU79fi0XK34bnHSqowypowOtYqjDKu2jEpPxkBmbX6n//4tRv5DTbtOoi+LR9AqdvzqzDKmjA61iqMMq7aMSk/GQGZtfqfv/jtP3wCzQfNRIZ0afBJnzpImyb0QRCsmV/bV6dSZX5pa+1Xx7CtWIwuUH4y0MpP+ckIXGo9e+kGjJ0Rh7JFrsPLTSvYZu1X/1NhtG1a2QV+dQxZrb2zVn4ylspP+ckIXGrdd8JCxK3biRefLI2H7rGPIetX/1Nh9NJrksjLr44RZSyOs1d+jlEleaHyU34yAv9aM9pNw37TcfbceUzu+QSyZclgm7Vf/U+F0bZpZRf41TFktfbOWvnJWCo/5Scj8K/1srXbMWDiItx2/dUY9sJDjrL1q/+pMDpqXvcX+dUx3NfYW0vlJ+Op/JSfjMC/1qOn/4K5v/yDxlWLomGVOx1l61f/U2F01LzuL/KrY7ivsbeWyk/GU/kpPxmBf60bD7gY7WZEu4dxy7U5HWXrV/9TYXTUvO4v8qtjuK+xt5bKT8ZT+Sk/GYGL1hu27cdLY7413xX5fdFp8qv/qTA6bWGX1/nVMVxW13Mz5SdDqvyUn4zAReuP5q3Bx/N/R7V7b0Hb2vc4ztKv/qfC6LiJ3V3oV8dwV1vvrZSfjKnyU34yAhetX3pzLjZsP4Beze7HvXdc6zhLv/qfCqPjJnZ3oV8dw11tvbdSfjKmyk/5yQjAfFfk90VGuZna70nbaDeB9/Or/6kwSr3Gxt6vjhFlLI6zV36OUSV5ofJTfjICwDe//IM3p/+CewrlR58WD0SUnV/9z5Uwnj17FmnTpo0IULQu1lipMrJ+dVxZrb2zVn4ylsov5fMb8OEiLPtjO55/vBRqlL01ogL7tX0dCePKlSvRvn17zJ49Gw899BB+//13UJCef/75iCBF42IVRhlVvzqurNbeWSs/GUvll7L5McoNo90w6g3PXuQZjJEkv7avI2EsV66ciZLeq1cvtGvXDiVLlsRff/2F/fv3J/vIUYUxEjdNfK1fHVdWa++slZ+MpfJL2fyWr9uJfhMW4ub8OTDqxWoRF9av7WsrjCdPnkSmTJkwc+ZMjBgxAmvWrMHy5ctRoEABrFq1CsWLF48YlpcGKowymn51XFmtvbNWfjKWyi9l8+NJGjxRo36lImj6cLGIC+vX9rUVRpIoVKgQ7rjjDsyYMQPPPvss8ufPj379+uHo0aPIkiVLxLC8NFBhlNH0q+PKau2dtfKTsVR+KZsfz17kGYxDnq+KQgVyRVxYv7avI2H88MMP0bx5cwOFo8QKFSqgVq1amDx5csSgvDZQYZQR9avjymrtnbXyk7FUfimX36ZdB9Fu5Dcm2s2kl59AqlSRl9Wv7etIGInjwIEDSJMmDbJly4YVK1aY74xOkjWqTBVA9ciRI8iaNSsCfxecF6dw06VLZ+4ZLqkwOmmF0Nf41XFltfbOWvnJWCq/lMtvyvdrMXHub6hS6mZ0qHuvq4L6tX3DCuPTTz+N06dPhwQybtw48/0xqbR7926sXr0aderUwYYNG5AnTx4jrs888wyuuuoq7NmzB3Xr1kWLFi0wcOBAHDp0CJkzZ0afPn2MYDZt2hTMn79TYXTlk46M/Oq4jioXg4uUnwyy8ku5/DqPnYd1W/ahe+PyuK/o9a4K6tf2DSuM4UZ0pHT48GEjckmladOmYfHixRg+fDji4+ONME6YMMFs+ZgyZQqWLFmCli1bYs6cOejcuTOmT59uRLJTp07Ytm2bseHPdklHjHaEwv/dr44rq7V31spPxlL5pUx+jHbTZOAXZpAype+TyJAu/MxdqFr4tX3DCuP58+dNfTmy4whw1KhRuPHGG822jQULFuDXX381053hEsFawrhz507cddddqFy5MhYtWoQOHToYIaxUqRJy5sxpBJFi2qRJE/P90kkQARXGlPlgyUr1r7VfHyyv6i/NR/nJCF6p/Oav2IQRU5fhrlvz4pVWFV1D9Cs/22+MjHJD8eP+RQoj09SpU1G/fn1H2zUChXHu3LlmipQrW5ctW4b06dNj1qxZOHfuHOLi4sx3y0mTJiFv3rxYt24dFi5ciK5du6Js2bLmvhRTCmdwqlevnuuGU0MloASUgBK4lMB78zZg1cYDqFOuAB4smveyxlOwYMFE9bMVRlrceeedWLt2LWrXrm0W33Dak98W//77b7OIxumIkVOnhQsXRrdu3cAFOMzLGk0yj+PHj6N169Zo27Ythg4dakSxTZs2RkRDTevqiFHms359o5PV2jtr5SdjqfxSHr/z5y+gft9pJtrN/7rWRN6c7rfk+bV9HQkjQ8C9+uqr+PLLL42gcSq0Z8+eZgrULgWOGIcMGWIi6IwZMwabN29GmTJlsH379oQp08GDB6Nq1apGLDlV27dvX5QvX94IY6hpVRVGuxYI/3e/Oq6s1t5ZKz8ZS+WX8vj9uiEevcb/gBvyZMNbL1UXFdCv7WsrjJxKffPNN1G0aFFUrFjRLLi5+uqrHcOiMPL7ZO7cuY3gcf8jvzUy9e7d26xSZeIqVf48duxYnDlzxiy8Wbp0qVml2qxZs5D3U2F03BRJXuhXx5XV2jtr5SdjqfxSHr93Z63EzMXrUeeBwmhZvYSogH5tX1thtKZSKZCcTrXbV+iE4o4dO4xQBi7coeAycXo1kqTCGAmtxNf61XFltfbOWvnJWCq/lMfv6TdmIX7/MQx6tjKK3pxbVEC/tq8jYWzcuDE++ugj3H///WbkaH3v43fAjBkzisBJjVUYZQT96riyWntnrfxkLJVfyuK3dfdhtBk+B1kypsNHvWojdWoX4W4CquTX9nUkjFwlyunQ4MRN+ZGO8GRukNhahVFG1K+OK6u1d9bKT8ZS+aUsftMX/on356zCg3fdiM4NLu4GkCS/tq8jYeRq0QsXLiTik9wBxFkgFUaJ28IshkpqubIsV++stXwylspP+UVC4L/jvsPvm/agS8NyeKBEgUhMk7zWr/7nSBi5EnXixIlmOpX/btiwoQkqfu2114rBSTNQYZQR9KvjymrtnbXyk7FUfimH37GTZ/BU/+nmUxmnUTmdKk1+bV9Hwti9e3e89tprJvxb9uzZTYSaIkWKmA3+TqLTSOGGs1dhlNH1q+PKau2dtfKTsVR+KYffgl83Y8inS82CGy688SL5tX1thfHYsWNmEz+3Vbz99ttInTq1ObD4pZdewh9//GE27CdnUmGU0fer48pq7Z218pOxVH4ph98bn/yEhau24D81SqD2/d70635tX1th5PdFfkvkZnvuKWSyQsItX74cpUqVkrWs0FqFUQbQr44rq7V31spPxlL5pQx+jHbT6JXPwenUsR2r4/rckW2bC1ULv7avrTCywjyYmDFKq1evbo6B4skZpUuXDhuqTdbczq1VGJ2zSupKvzqurNbeWSs/GUvllzL4rdm4B93f+Q55r86C/3WpKStUgLVf29eRMDJ8G0PCffzxx2bxDQOIv/zyyyhevLhnAN1mpMLoltxFO786rqzW3lkrPxlL5Zcy+L03exU+//FPPFb+djxT825Zoa4UYWQ9T506hZMnT5oVSxw1JveiG4u9CqPMh7VjUn4yAjJr9b+Uwe/5YXOwbc9hc8QUj5ryKvm1fR2NGOfPn29il7777rsoUKCACR7OmKZ16tTxip/rfFQYXaPTEaMMnfJTfh4QkGXhhfDEHziGpwfPMocR81BiabSbwBp5UT4ZofDWocrnSBj5jfGvv/7C999/bwKIV6tWDatXr8bBgwfN9o3kTCqMMvp+dVxZrb2zVn4ylsov+fnNWLwe/5u1EuWL3oD/Nr5PVqAga7+2r60wWgcV85ipV155xVR7zpw5qFGjhjn9gkdHJWdSYZTR96vjymrtnbXyk7FUfsnPr+f4H7BqQzw61CuDKiVvkhXoShFG1vOWW24xizS4f5F7Gnlu4vr1681RUddcc42nICPNTIUxUmKXXq8dk/KTEZBZq/8lLz9+V+T3RSZGu7kqc3pZga4kYeQ3xjZt2hgxtBLPaHzhhRc8hegmMxVGN9T+tdGOSfnJCMis1f+Sl99rk5dg8ZqtUZlGZc382r62U6lWszGI+IIFC3DgwAE8/PDDZtN/SkgqjLJW8KvjymrtnbXyk7FUfsnHb9Oug2g38hukTpUKY156xLNN/YE18mv7OhJGLrx55JFHjPozbuqaNWvQqFEjE0w8uZMKo6wF/Oq4slp7Z638ZCyVX/Lx6zdhIZav24nKJW/CS/Wis1bEr+3rSBgZ9o3imDt3bjRo0MAEEedpG7oq1d6p/eoY9jWLzRXKT8ZZ+Sm/pAj8tW0/Oo751mzNGN+1Jq7JnlkGKoS1X/3PVhi5sT9jxozgN8UtW7YgTZo0qFevHkqWLIkVK1bg7ru9i5LgpmV0xOiGmn5jlFFTfsrPKwKyfNwKj3XuYvUyt6LNE9GLd+22fDIqzq1F+xizZctmwr/lyJED6dOnNydszJ07F/v27UO6dPIzu5xXI/GVKowSev79OC6rtXfWfn3wvSMgy0n5xZ7frxvi0Wv8D0ifNg3Gd6uJHFkzygoRxtqv7Ws7YmSdGSO1devWJk6qlbinkXsbkzupMMpawK+OK6u1d9bKT8ZS+cWeH6dQOZX6RIVCaPXoXbIC2Fj7tX1thfHcuXP49ddfzSHFrOTGjRtRokQJ3HefswgJR48eNStYGWPVSjzj8fz58ybPUIlxWTka5dRtuKTCKPNrvzqurNbeWSs/GUvlF1t+y/7YjgEfLkLG9GnxXrdanu9bDK6NX9vXVhgpYPnz50eRIkVMSDinaffu3SZsHOOpbtiwAXny5DFByFu1aoVDhw6Z6Vh+n+zXrx8GDhxofsfg5DzzkSLatGlTjBs3zvxOhdEp9civ86vjRl7T6FgoPxlX5Rdbflaw8AaVi6DJQ8VkN3dg7df2tRVG1v3555/H22+/je+A1ocAACAASURBVN69exuBpKgx1a5dO+QpGzyzkWc4Dh8+HPHx8UYYJ0yYgJ9//hlvvfUWuC/y888/R7FixdCtWzdMnz7dBCrv1KmTWfVKG/5sl3TEaEco/N/96riyWntnrfxkLJVf7PgtXLUFb3zyE7JkTIfx3WqZ/0c7+bV9HQlj3rx5wRFgcOIojwtzwiWO/ixhpLAuX74ccXFxuP766zFgwABz+DFP68iZM6cRRIppkyZNMHnyZEdHW6kwylzbr44rq7V31spPxlL5xYYfByLPDpmNXfuPonm14qhb8Q7ZjR1a+7V9HQkjI95w20Zwqly5sq14BQpjy5YtsXDhQsyePRsrV65Ely5dzBYQTtdSLLkFZNKkSaAQr1u3zlzbtWtXlC1bNmQzqDA69NAQl/nVcWW19s5a+clYKr/Y8Pt2+UaMmvaz+aY4oftjZkVqLJJf29eRMEoABgpjx44dkSFDBgwaNMhkSQHkCPHWW281Px8/ftysfm3bti2GDh1qRJExWpctW2a+Oy5atMhcH5y4r1KTElACSkAJJCZw7vwF9P9kNQ4eO4065QrgwaLeHUR8OfAuWLBgomrEVBgZLYeHHc+bN89Mm5YrVw47duxIWHnKUzuqVq1qpl45Su3bty/Kly9vhDFt2rRJtoGOGGWu6dc3OlmtvbNWfjKWyi/6/Gb99BfGzVxhotu82+VRpE1zcY1ILJJf2zcmwsjvkwwnx+nYF1980UylcrUpV6Ra8VZ5hBW/QY4dOxZnzpwxC2943iNXqTZr1ixkG6owytzbr44rq7V31spPxlL5RZff6bPn0Or1WTh49CTa1SmNh0snHh3JShDe2q/t60gY+eH2k08+wQ8//GAWynB/IUd22bNnd8WUJ3Rw0U7gHsXDhw+bvOwW8wTfUIXRVRMkGPnVcWW19s5a+clYKr/o8pu28E9MmLMK+a7OinGdapjYqLFMfm1fR8LYv39/M3Jj4uka/NbHRKG0tm7EEnbgvVQYZeT96riyWntnrfxkLJVf9PidPH0WLQbNxLGTZ9C5QVk8eNeNspu5sPZr+9oKI0eLHBk++uijJpg4N/sXLlwYzZs3NyduWAtnXDDzxESFUYbRr44rq7V31spPxlL5ectv/db9WL9tH9Zv3Yd/dhzE5vhDuDFvdrzZ4RHZjVxa+7V9bYWR3/sYOPy1117D/v37zfRnxYoVUa1aNWzevBkFChRwicwbMxVGGUe/Oq6s1t5ZKz8ZS+Un4/f5d3HYcyw1/tq2D39u2ZdkZj2bVkCZItfJbuTS2q/tayuM5FGzZk189dVXCbFNGUyc4hhJiDiXXG3NVBhtEYW9wK+OK6u1d9bKT8ZS+cn41er+6SUZ5M2ZBQWvzYmC1+ZAwfwX/x+tsxadlNyv7etIGHm81IcffmhO2fjzzz/NhvtRo0aZKdXkTiqMshbwq+PKau2dtfKTsVR+7vkdP3UGDfpON8dGdXmqHG65NmdMwrxFUmK/tm9YYWSl+I0xVLr55pt18Y2Nl/jVMSJx/mheq/xkdJXf5ctv76HjaPnal8idI7M5KSMlJr/6X1hhDDwqKinoTmKlRruxdMQoI+xXx5XV2jtr5Sdjqfzc8+PCmrYjvsZN+bJjdPvkWVxjV3q/tm9YYRwyZAh4HmOo9NJLL5mFOcmZVBhl9P3quLJae2et/GQslZ97fn9s3ouub8/HHTdeg8HPVXGfURQt/dq+jr4xkhsPHOZiGwb85sIbt5v7vW4DFUYZUb86rqzW3lkrPxlL5eeeX9y6neg7YSHuKZQffVo84D6jKFr6tX0dCSOj3jz11FOX4OMZitzCkdxJhVHWAn51XFmtvbNWfjKWys89P+t8xQdKFECXhuXcZxRFS7+2r60wciqVZyVyi8YHH3xgpk5Hjhxp4phu374d1157bRSx2metwmjPKNwVfnVcWa29s1Z+MpbKzz2/r5f9jTFfLMcjZW7BC0/c4z6jKFr6tX1thdGKfNO4cWMT4Jvpyy+/xGOPPabC6MCh/OoYDqoWk0uUnwyz8rt8+U1f+Cfen7MKTz5QGC2ql5BVNErWfvW/sMI4d+5cs/hmzJgxZoM/j43i8U/Dhg0D9zbyMOFQx0FFiXOibHXEKCPtV8eV1do7a+UnY6n83PObOPc3TPl+LZo+XAz1KxVxn1EULf3avrpdI4pOwaz96hhRxuI4e+XnGFWSFyq/y5cfz1jkWYvPPVYKj5a7eNh7Skt+9b+wwshRYrjtGjVq1NARo40n+tUxUsoDpvxkLaH8Ll9+w6cuw3crNqFj/TKodPdNsopGydqv/mf7jZG8Tpw4gffff9+cpsHELRt//PEHpk+fjqxZs0YJqbNsdSrVGadQV/nVcWW19s5a+clYKj/3/AZOXISla7ejZ7MKKHNH8gQJtyu9X9vXkTBaQcSDIXClqgpjeNfwq2PYOXys/q78ZKSV3+XLr8e73+O3f3Zj0LOVUfTm3LKKRsnar/5nK4xnz55FunTpwJHZzJkzzWpUHj310UcfYfny5bALGxcl3gnZ6ohRRtivjiurtXfWyk/GUvm559d+9Fz8s+MARrZ72JyokRKTX9vXVhg5bUohZPi3TJkyYeXKlWCouDvvvNNMpyb3CRsqjLLHwa+OK6u1d9bKT8ZS+bnn9+yQr7Bz31G82+VR5Ls6eT9pXW6famyFkRVu1qyZ2arBY6cCI+DoVKq9U+uDb88o3BXKT/nJCMisU7L/NR7wBQ4fO4WPetXGVZmTN2b1FSmMp06dAvc0VqlSBVOmTEFcXBwaNGiAChUqyLzOA2sdMcogpuQHnzXT8mn7ygjIrFOy/z3eYwrOX7iAma/WT/ZPWlekMA4cOBCMjWpt5t+0aRO6du2K9957Txff2Dx3KfnBUuGRdZrKT/nJCbjL4fTZc3iy12dIlzY1pr9Sz10mMbDya//naCr1hhtuQOnSpTF+/Hi89dZb6Nmzp0Hq5DxGnsqRJUuWRG80+/fvN7/PkCFDks1z8uRJs+iH3zfDJR0xyrzbr44rq7V31spPxlL5ueN38OhJNB04A9kyp8PkXnXcZRIDK7+2ryNhZNDwDh06JGC8/fbbTVi4Rx99NCTa3bt3Y/Xq1ahTpw42bNiAPHnyJFy7efNmFCtWDF9//TXuu+8+cERKkc2cOTP69OljRLRp06YYN26c+Z0KY/Q82K+OGz0ikeWs/CLjFXy18nPHb/veI3hu6GzkyZ4R4//7uLtMYmDl1/YNK4wHDhxIQPfZZ5/h2WefNQLHb4wc7fHUjVBp2rRpWLx4MYYPH474+PgEYTx9+jTq16+PjRs3mqDk+fLlQ+fOnU2wgBYtWqBTp07Ytm2bseHPdklHjHaEwv/dr44rq7V31spPxlL5ueP317b96DjmWxTInQVjOtZ0l0kMrPzavlGPlcrRX6AwduzY0SziGT16NHr37m1GjJUqVTIiS0GkmDZp0gSTJ092FG5OhVHm3X51XFmtvbNWfjKWys8dv1V/x6Pn/35AoeuyYUjb6u4yiYGVX9s3rDC+8sor4AgvVOK3xlDfCC2bQGHkKHLGjBn48MMP8cgjjyQII+OxchRasmRJTJo0CXnz5jUndyxcuNAs8ilbtmzIMqgwyrzbr44rq7V31spPxlL5ueO35PdtGDRpMUrcnBMDnn3YXSYxsPJr+zr6xkh+3LLBg4kD00033YTUqVOHxRsojOXKlQO/PebKlQu//PIL+K2SI8N77rl4yObx48fRunVrtG3bFkOHDjWi2KZNGyxbtsx8d1y0aJEZUQanevVS7qqsGPie3kIJKIErjMCy9Xvx0YKNKFPoGjR64OYrrPbeVrdgwYKJMnQkjPwWyO0a3NAfmJysSg0Uxq1bt4KrTZlatWplRI8h5qwFNoMHD0bVqlXN1OuCBQvQt29flC9f3ghjqHMfdcQocxK/vtHJau2dtfKTsVR+7vjNXLwe785aiYpF86JT44ruMomBlV/b11YYL1y4YBbIMLVs2RLp0/8bYeHll192NJXKUWLu3JcGuWVg8h49ephvjEx79uwxU6sU4TNnzpiFN0uXLjWrVBl5J1RSYZR5t18dV1Zr76yVn4yl8nPH75P5v2PyvDV4pOS1eKHe/e4yiYGVX9vXsTByerNXr15RQ3n48GGTd7Zs2SK6hwpjRLgSXexXx5XV2jtr5Sdjqfzc8Rv/1a/4YtE6PFH2BrR6/OLgIiUmv7avrTASduPGjfHll1+aEzYCt2jUrVvX0crRaDaYCqOMrl8dV1Zr76yVn4yl8nPHb/T0XzD3l3/w1AM3oVH1Mu4yiYGVX9vXkTBylSinQ4OTk2+M0Wavwigj7FfHldXaO2vlJ2Op/Nzxe/2jJVj021a0rHor6lQp5S6TGFj5tX0dCSO/9XFVanDiwphQi2JiwNzcQoVRRtqvjiurtXfWyk/GUvm549f7vQVY+dcutKlRCNXvv8tdJjGw8mv7hhRGnsP4wQcfmL2FPHfxxIkTiTByIz7jmSZnUmGU0fer48pq7Z218pOxVH7u+HV+ax7Wbd2Hjo8XQaWyxdxlEgMrv7ZvSGHkxn5u3n/11VcxYsQInUp16UR+dQyX1fXcTPnJkCq/y5Nfm+FzsHX3YfSoVwzlShaRVTKK1n71v5DCyG0a3ExfoEAB7Ny502yhCE7csG93+kUUmetUqgdw/eq4HlTdkyyUnwyj8nPHr/mgmdh/+AReaXwX7ipayF0mMbDya/uG/cbI8GycUg2VnnrqKZ1KtXEuvzpGDJ4ZR7dQfo4whbxI+V2e/Or1mYaTp8/ijZalUPj2W2WVjKK1X/0v6kHEo8hcR4wewPWr43pQdU+yUH4yjMovcn6czXusxxSkTpUKw5++B0mFNIs81+hY+LV9wwojV6OGGzGWKVNGp1J1xBidJ+r/c/XrgxVVKBFkrvwigJXEpSmR39ETp/FU/89xVeb0GNC4hAqjoIlDta+j7Rq878qVK7FmzZpLitCwYUOdSlVhFLilvWlK7JgCS63ls2/DcFcov8j57T5wDK0Gz0Leq7Ogx5NFVBgjR5hgIRLGp59+GuPHj090e93gb98i+uDbM9KOU8ZI+V1Z/DbuPIgXR32Dm/PnQIeat6kwCprftTDyrERu4udmfgYND9y3WLFiRd3gryNGgVvam+qLhT0jFUYZI7/x+33THvx33HcoenNuPFP1JhVGQfO7Fkbe884770T16tUxZMgQQRGiY6ob/GVcVXiUn4yAzFr9L3J+v/y5A/0/+BH33nEtGle4ToUxcoTeTKXyzEQGEeeBwDxk2ErDhw9HxowZBcWSm6owyhhqx6T8ZARk1up/kfP74dfNGPrpUlS860bULp1HhTFyhN4IowYRd09eH3z37Gip/JSfjIDMOiX63+ylGzB2RhxqlL0V1YrlVGEUNLFoKvXs2bPg3pnglNxxUlkeHTEKvEKFRwZP+Sk/MYHIM/jshz/wwTerUa/iHahwW1YVxsgRejNi7N+/P6yDhAPLMGDAAJ1KtWmUlPjGGVhkLZ/gqVJhlMFTfq74URQpjs2rFUfJAhlUGF1RvGgkGjHqVKp78io87tnpVKqMnfK7PPlxGpXTqc8/XgqF86RWYRQ0s0gY9+3blxAB5/jx4+jSpQs2bdqEn376SSPf6IhR4Jb2pvpiYc8o3BXK7/Ljx4U3XIDTuUFZ3JDtnAqjoIlFwhh834kTJ6JZs2bYsGEDbrnlFkGx5Kb6jVHGUDtO5ScjILNW/4ucH7dqcMtG7+b3I1f6kyqMkSNMsBAJY4UKFczRU1ZiZkxHjhxB1qxZBcWSm6owyhhqx6T8ZARk1up/kfPj5n5u8n+tdWVkOn9EhTFyhN4I48MPP4z4+HiTGaPgcJTIEWPNmjUFRfLGVIVRxlE7JuUnIyCzVv+LnF+7kd9g066DGN2+Gs4f36/CGDlCb4RRcF8cPXoUWbJkQapUqRKyOXDgALJlyxb2++TJkydN+Dm7g5BVGCWto/sEZfSUn/KTEojc/unBsxB/4BjGd6uJo/vjVRgjR5h8wrh7926sXr0aderUMd8i8+TJgy1btqBBgwbInTu3GXmWLFkSPXv2xMCBA8GA5JkzZ0afPn2MiDZt2hTjxo0zvwuXVBgFXqHL5WXwlJ/yExOIPAMeOcWjpz7uXRu7d25TYYwcYfIJ47Rp07B48WIwbBynYSmM3Pd45swZ9OvXDxwRZsqUCUuWLMEbb7yB6dOno0WLFujUqRO2bdtmbPizXVJhtCMU/u86laX8ZARk1up/kfOr1f1TY/TloAYaGSpyfJdYRLz4hqdqfPXVVyhdujRWrFiBEiVK4Prrr4+4GBz9WcJ44sQJMxpkfNUZM2agY8eOZjRZuXJl5MyZ0wgixbRJkyaYPHmyo5M7VBgjbhJHjiHL1Ttr7ThlLJXf5cXPOlkjf66seKfzoyqMsuaNfIP/qVOnjIA9++yz+OKLL4xYlS1b9pJi1K5d21a8AoWRxqdPn8agQYMwdOhQky9FkSIcFxdnplYnTZoEBhRYt24dFi5ciK5duybcd9GiRUY4gxODm2tSAkpACVzuBGbHbcc3K3agQpE8qFf+xsu9ujGpX8GCBRPdJ9WFpIKg/v9lpUqVMqPFUMnJQcWBwsjp0/r16yN9+vQYPXo08ufPf0nWDB7QunVrtG3b1ggnRbFNmzZYtmzZJYt3Ao10xCjzHR1RKD8ZAZm1+l9k/DqPnYd1W/ahZ9MKKFPkOh0xRoYv0dURT6UyB64opTByRPbCCy/g/vvvvyRj/sxFNOFSoDC+8847mDVrFmbOnJmkyeDBg1G1alUz9bpgwQL07dvXHJBMYQx1HxVGmWdox6T8ZARk1up/zvmdPH0W9ftOQ+pUqfBJnzrImD6tCqNzfEle6UoYrZx27NhhNvL/8ssvOHbsmBEvu9Wili2FkStUuRK1ZcuWmDBhwiUFXL9+PW677Tbs2bMHvXv3xtixY80CHS68Wbp0qVmlyj2ToZIKo8wztGNSfjICMmv1P+f8lvy+DYMmLcYdN16Dwc9VMYbKzzm/pK4UCSNHb7Vq1TKRbqw0YsQItG/fXlaqAGvr9A7ub4wkqTBGQivxtfpgKT8ZAZm1+p9zfm99EYc5yzagcdWiaFjlThVG5+hCXulaGM+fP29WpnLUxz2HHCny+yBHj3v37kWuXLk8KJ77LFQY3bPTN04ZO+Wn/OQEnOfwzBtfYdf+oxjyfFUUKnCx39UXC+f8PB0x8mSNa665BqNGjUK7du1M3vPmzcNDDz1kpjrLlCkjK5nQWoVRBlAfLOUnIyCzVv9zxm/voeNo+dqXyJIxHT7uXQdWIDHl54xfqKtcjxi5aDV79uy48847zWZ9jhgZqWbKlCnYtWuX2VqRnEmFUUZfHyzlJyMgs1b/c8bv62V/Y8wXy3Ff0evRvXH5BCPl54yf58LIDEeOHIkOHTpckjenVV955RVZqTywVmGUQdQHS/nJCMis1f+c8Rs0eTGWrNmGF564B4+U+feoP+XnjF9UhJGZMs7p559/Di6SYfxTjiBTQlJhlLWCPljKT0ZAZq3+Z8/vwgXgqf7TcezkGbz/31q4Jvu/8aOVnz2/cFe4nkqV3Tb61iqMMsb6YCk/GQGZtfqfPb91W/eh81vzYIWBC7RQfvb8VBhljKJirY4rw6r8lJ+MgMw6Jfjfp9+txaRvf0ONsrfi+cdLXVKhlFA+N8IjaxXvrHXE6B3LiHJSx40IV6KLlZ/ykxGQWacE//vvuO/A4OEvN62AskWuU2GUNakjfmFjpVo5cGN/586dMXv2bBPLlAHGGSauePHiHhbRXVY6leqOm2WVEh78y/GNU9Yq3llr+8pYJjc/hoFr2G+6qYQVBi6wRsldPju6fi2fI2Fk4O+pU6caBt27dzexS9esWWMW5GTIkMGOTVT/rsIow+tXx5XV2jtr5SdjqfzC8/vlzx3o/8GPl4SBU2GU+ZwTfrbCePbsWaRLl84cLsw4qWnSpEH16tXxwAMPYNWqVck+alRhlDmJdkzKT0ZAZq3+F57fO1+uxJdL1l8SBs5Jxy5rFe+s/dq+tsLIsxJ5skWrVq3M6JD/MSB4jx49cODAAeTIkcM7ii5yUmF0AS3AxK+OK6u1d9bKT8ZS+YXn9/ywOdi25/AlYeBUGGU+54SfrTAyk06dOmHYsGGXlIYHF0+cONG7ErrMSYXRJbj/N9OOSfnJCMis1f9C8zt09CSaDJxhjpea0vfJhDBwTjp2Wat4Z+3X9nUkjAwLFxcXh+nTp+Pvv//G3XffjRdffNHx0VPeYU6ckwqjjK5fHVdWa++slZ+MpfILzW9e3EaM/Oxn3Hfn9eje5N8wcCqMMp9zws+RMDLiDVeiBicGF69QoQIyZszoXUkjzEmFMUJgQZdrx6T8ZARk1up/ofm98clPWLhqS6IwcE46dlmreGft1/Z1JIwMFM5jp5JKJUuWxOLFi5NNHFUYZU7sV8eV1do7a+UnY6n8QvPjNo2kwsCpMMp8zgk/R8LYuHFjLFy40AQN50KcQYMGgQcK8+gp/u7bb79F1apVvSttBDmpMEYAK4lLtWNSfjICMmv1v6T5/bPjANqPnptkGDgnHbusVbyz9mv72gqjtV3jpZdeSliA89prr5n9jNy+kS9fPnMcFVetJkdSYZRR96vjymrtnbXyk7FUfknz+2zBH/jg69VJhoFTYZT5nBN+tsLITDg6ZPSbt99+22zX4MKbTJkymVWp1apVM1Op9913n3eljSAnFcYIYOmIUQZL+Sk/zwkknWHP8T9g1Yb4JMPAOenYY1RM29v49cXHkTAy6g1HhBRHpquuugrjx483R1CNGzfOTLMm1wIcFUZb3wx7gV8dV1Zr76yVn4yl8kvM78zZ82jQbxrOn7+QZBg4FUaZzznh50gYmRE3+i9fvhwnTpxA+fLlTTQcLsjhBv/06dN7V9IIc1JhjBBY0OXaMSk/GQGZtfpfYn4r/9qF3u8tCBkGzknHLmsV76z92r6OhJFi+P777+Ovv/4yxM6fP48//vjD7GvMmjWrdxRd5KTC6AJagIlfHVdWa++slZ+MpfJLzO+92b/i8x/XoVHVoniqSvgD4ZVfdPzPkTDWrFkTX331VaIScGrVThiPHj2KLFmyIFWqVAn2XLTDb5SpU6cOWauTJ0+aUSljs4ZLKozRcQxZrt5Z64MvY6n8/Mev3chvsGnXQbzxfFUULpArbAW0faPTvrbCaK1KpQDNnDkTjz32mBGrjz76yEytBgpeYBE5zbp69WrUqVMHGzZsQJ48ebB37140atTIbPnYvHkzunTpghYtWmDgwIE4dOiQiaTTp08fk2fTpk3N90v+ToVR1vjhrPXBkrFVfspPRuBSaydh4AIt1P9k9EPxsxVGTptSCLldg6O8lStXYsiQIbjzzjvNdGrhwoWTLNm0adPMalVu5YiPjzfCyG0eHGVSCHft2oX8+fOb46t69eplpmUpkozLum3bNmPDn+2SjhjtCIX/uz5Yyk9GQGat/ncpv+9XbsawKUvDhoFTYZT5nBN+tsLITJo1a2a2Znz88cd46qmnEvJ1MpXK0Z8ljE8//bQJBNCwYUMw/iqnUhl7lStec+bMaQSRYsoA5ZMnTzYjS7ukwmhHSIVRRkj5Kb+C0URwSd7Dpy7Ddys2oc0T96B6mVts76svFraIwl7gesTIXPlNcP78+UbUpkyZYgKK161bFw8++KBtqQKFkQce8z/aMjHUHA89vuGGG0yeDC83adIk8/t169aZbSBdu3ZF2bJlzfWLFi0ywhmc6tWrZ1sOvUAJKAElkNIJ9Jy0EkdOnEW/RiWQI0vyrfZP6Zy8LF/BgolffMKOGBk4nCtSS5Uqhb59+6JWrVqmPL///rsJHs7vhAUKFAhbxkBh7N+/vwkW0KFDB7P9g6PEgwcPJizCOX78OFq3bo22bdti6NChRhTbtGljxDPUt0wdMcpcRN84lZ+MgMxa/e9fflviD+GFEV/bhoELJK78ouN/YYVx8ODB6NatW8g7cyRptzgmUBi5eOfNN9/E3LlzwaABPOPxp59+Ssif9+OolFOvCxYsMGLMPZMUxlDTqiqM0XEMWa7eWeuDL2Op/FIuvyPHT2PH3iPYse8odh84hg3b92Pp2u22YeBUGGVt6oRfWGH8/vvv8c033+Ctt95CxYoVUaRIkYQ8y5Qpg9q1a9uWkMLIFaq5c+c2o88aNWpg7dq15t8MPs58mPbs2YPevXtj7NixOHPmjFl4s3TpUrNKld84QyUVRtsmCHuBdpzKT0ZAZn0l+N/OfUfx55Z92LHvCPjvXfuOYuvuwzh+6kyS8Ho2rYAyRa5zBPZK4OcIhMuLRN8Yv/zyS5QoUcJ22tRp2bZu3WqCj3OfopUYXo6JU62RJBXGSGglvlYfLOUnIyCzvpz9L279TsxcvB4r1u9KElLOqzLihjzZcH3ui/9d/PdVuCZ7+C1qTkY8slbxztqv7etoVeqPP/5ovvkFTnsSHVeU2m3w9w5x0jmpMMoI+9VxZbX2zlr5yVhejvwWr9mKid/8hu17L8aWZip5ez7cnC8Hrs+TDTdQCPNkQ5aM/w4M3FK8HPm5ZeHGTjRi5J5FTn9y1WhgXFSuVLX7xuimsJHYqDBGQktHjDJayk/5hSfw9swV+Oqni6Ez8+bMgofuKYiH7rkZV2fL5DU6k58Kowyra2Hk6lEufOG3Pi6GSWlJhVHWIvpgKT8ZAZn15eR/Az5chGV/bEfG9GnxUr0yuK/o9TI4DqwvJ34Oquv5Ja6FkSVp2bIlfv75Z3zyySdmi4WVrrvuupDbKDyvQYgMVRhlpPXBUn4yAjLry8X/fv1rF3q9twDZsmTAK60qomD+HDIwDq0vF34Oq+v5ZSJh5IZ7riwNToxvGuliGa9rpsIoI6oPlvKTEZBZXw7+d+EC0HbEHGzZfRgvPnmvmTqNVboc+MWKVVL3EQnjiBEj5afzbAAAIABJREFUEg4pDsycG/AzZMiQnPWCCqMMvz5Yyk9GQGZ9Ofjf18v+xpgvluPGvNkxun21mM6iXQ78ZB4ksxYJI2+9adMmsyqV4XNy5cqFW2+9VVYij6xVGGUg9cFSfjICMmu/+9+pM+fQ6vUvcejYKfRt+QBK3Z5fBiRCa7/zi7C6nl8uEkbuY+RxU0zdu3c3MUvvvvtujBw50vOCRpqhCmOkxC69Xh8s5ScjILP2u/99u3wjRk37GXfdmtd8W4x18ju/WPMKvp9IGBnkm8dG5ciRw0Sq4SrVV155Bdu3b8e1116brHVTYZTh1wdL+ckIyKz97n+ffrcWk779DQ0qF0GTh4rJYLiw9js/F1X21MS1MDKQeMaMGTFmzBgTNJxnM/I0C+5p5FmK3OOYnEmFUUZfHyzlJyMgs/a7/707a6WJbvNsrZKodd9tMhgurP3Oz0WVPTVxLYwsBcVv7969JowbR4tcocqN/TwaKrmTCqOsBfTBUn4yAjJrv/vfkE+XYsGvm9GlYTk8UCL8SUMyUklb+51fNJhEkqdIGFeuXGk2+PNbo5VmzJiR8N0xkoJ4fa0Ko4yoPljKT0ZAZu13/+v93gKs/GuX+b7I74yxTn7nF2tewfcTCSMzY0g4npTB8xMzZcqEu+66K7nrZO6vwihrBn2wlJ+MgMza7/7XftQ3+GfnQYx8sVrMNvUHEvc7P5n3yK1FwsgzFNu1a4c5c+aYEzY4tco9jBSl5E4qjLIW0AdL+ckIyKz97n8tBs3EvsMnMKH7Y8gVpXio4Qj7nZ/Me+TWImEsVKiQOU/xs88+Q5YsWfDCCy9g4sSJuirVQbuo4zqAFOYS5af8ZARk1nb+V7vnVJw9dx6fD6iHtGlSy27mwtqufC6y9NTEr+WzPXbq7Nmz5tzE5557zhwizDR58mQ0adLEbPgvW7aspyAjzUxHjJESu/R6vzqurNbeWSs/GUs/8+NBww36TkfmDOnwad86MhAurf3Mz2WVPTUTjRgfeOAB8EzGxx9/HFdddRW48CZ79uzYsGGDhoSzaSZ1XJkfKz/lJyMgsw7nf7v2H8Uzb3yFfFdnxbtdHpXdyKW1Ph8uwf2/mUgYN27ciOHDh+PTTz81WzUeffRRdOnSBQ8++KCsVB5Y64hRBlEfLOUnIyCz9rP/rdu6D53fmodCN+TCkDZVZSBcWvuZn8sqe2rmWhg5lcrFN0WLFkXlypVx4sQJ850xpSQVRllL6IOl/GQEZNZ+9r9f/tyB/h/8iNKFr0Xv5vfLQLi09jM/l1X21My1MLIUXIVKgeSWDUa+SUlJhVHWGvpgKT8ZAZm1n/1vXtxGjPzsZ1QtdTPa171XBsKltZ/5uayyp2YiYWzcuDE++ugj3H///WbkyP2MTEOHDjXh4pIzqTDK6OuDpfxkBGTWfva/aQv/xIQ5q1DngcJoWb2EDIRLaz/zc1llT81Ewuj1QcVHjhxB1qxZw55bdvLkSbMa1m6EqsIo8xN9sJSfjIDM2s/+997sVfj8xz+NKFIckyP5mV9y8Aq+p0gYjx8/jgs8pjooRfqt8cCBA3jmmWfMytY9e/agbt26aNGiBQYOHIhDhw6Z+KsMPccRadOmTTFu3Djzu3BJhVHmXvpgKT8ZAZm1n/1vxNRlmL9iEzrUvRdVSt0sA+HS2s/8XFbZUzORMHKE17lzZ8yePRutW7cGT9zgCRvFixePqJATJkwweUyZMgVLlixBy5YtTTQd5j19+nQjkp06dcK2bdsQHx9vfrZLKox2hML/XR8s5ScjILP2s//1m7AQy9ftNAtvuAAnOZKf+SUHL09HjPXr18fUqVNNnjyoeNmyZebIqS1btkS0j3Hnzp0mxipXt/Kw4w4dOhghrFSpEnLmzGkEcfHixSZ4AIMI8CQPu6TCaEdIhVFGSPkpv4JJIuj01rdYv3W/2arBLRvJkVQYZdRdjxityDf9+vXDsWPHzDe/6tWrg5v+V61aFdGoce7cuWaK9NlnnzXimj59esyaNQvnzp1DXFycOeNx0qRJ4DdNHmm1cOFCE5M1XHQdFcboOIYsV++s9cGXsVR+0eP39BuzEL//mNncz03+yZG0fWXUXQsjRYsjt1atWpnRIf9j3NQePXqA3wxz5MjhuGScOi1cuDC6desGTs9my5bNTJnmyZPH5MFvmZyqbdu2rVnxSlFs06aNEVF+d+QokyPK4MRpXU1KQAkogVgS6DohDqfOnMfgFiWRIV3K2sYWSw5+v1fBgolnBGxjpbLSnO4cNmzYJfXndCcDiUeShgwZAir0mDFjsHnzZpQpU8YEIremTAcPHoyqVasasVywYAH69u2L8uXLG2EMNa2qI8ZIWiDxtfrGqfxkBGTWfvU/Bg5nAHEGDmcA8eRKfuWXXLyC7+t6xMiMuCKVU51cIPP333/j7rvvxosvvmi7YjS4EBS8WrVqgd8amXr37m1WqTJxlSp/ZqDyM2fOmIU3S5cuNatUmzVrFpKjCqPMxfTBUn4yAjJrv/rf3kPH0fK1L81RUzxyKrmSX/klFy9PhPHw4cNmBSmPmypSpIg5buqWW24R12nHjh1mOpb7FK3EezFxejWSpMIYCS0dMcpoKT/ld5HA3zsOoMPoueZwYh5SnFxJhVFG3tWIkStQX3vttYQ7ly5dGj///LOsJB5bqzDKgOqDpfxkBGTWfvW/Fet3oc/7C3D3bfnQ/z/Jd5iCX/nJvMY7a1fCyNHhTTfdZEaM/P5HkQxcLONd8dznpMLonh0t9cFSfjICMmu/+t8Pv27G0E+X/l97ZwIeRZG+8RdCOCUQkPtSblRQEBEIq6IoongioCuwKKusiLoLCuvFsYIoLrv+FxXRPVBhXXVF8FbwgA0Y5FJRuW8IhCAQjhAIIf/nLRiYhJnpmf6me6YnXz2PTzCpr/qrX31db1d1dRUuv6gRHu4buzNpvcpPFjXRs7YljFwJys80+O7v66+/Nt8bLlu2zLxjjJekwihrCb2xlJ+MgMzaq/E3e8Ea/P3D5bihc3Pce0Ps+kOv8pNFTfSsbQvjLbfcgltvvRU//fSTGTFSKH3LW/v27VvkPWH03A2/JBXG8FkFyqk3lvKTEZBZezX+3vh8Bd7+6mf0u7o1+l55ngyCwNqr/ARVjqqpbWEM5QX3N410sUxUawVAhVFGVG8s5ScjILP2avy98N4SfPbtetx/c3tce6l8QaJdil7lZ7e+0bazJYzcBu748eNBfenVq1dY27ZFuzL+5akwyujqjaX8ZARk1l6Nv/FvpCPj5+14tF8aOp9fXwZBYO1VfoIqR9XUljBG1QOHClNhlIHVG0v5yQjIrL0afyNe/gIrN+/GhHuvxAXn1pBBEFh7lZ+gylE1VWGMKs7wC9PADZ9VoJzKT/nJCMisg8Xf7yZ9jO27D+ClP/RAg5qRfXst86iotd4fMpoqjDJ+tq01cG2jM4bKT/nJCMisg8Xfr596Dwdyj2LGEzcjpVI52UUE1np/COCF6F/C2itVdmlnrXUqVcZXbyzlJyMgs/Zi/PHM9hsfewulSgGzx/c1P2OVvMgvVqwimZFSYXS4lTRwZYCVn/KTEZBZB4q/fQfz0H/8bDNS5IgxlknvDxl9nUqV8bNtrYFrG51OpcrQKT+H+G3OysHQ5z817xb5jjGWSfsXGX0VRhk/29YauLbRaccuQ6f8HOL3w4ZdePzVr8xqVK5KjWXS/kVGX4VRxs+2tQaubXTascvQKT+H+KX/sBXPvrkQnS+oj0fvTIvCVewXof2LfXa0VGGU8bNtrYFrG5127DJ0ys8hfh9nrMOU2UvR49KmGHLzxVG4iv0itH+xz06FUcZOZK2BK8Knn2vI8Ck/B/hxj1TulXrbFa3wm+5thFeQmWv/4gw/XZUq42pprYFriShkBuWn/GQEZNaB4u9fn3yPmfNXYWCPC9HrspayCwit9f6QAdSpVBk/29YauLbR6VSgDJ3yc4Dfjl8O4pEpc5Fz6AgeuPUSXHNJ4yhcxX4R2r/YZ6dTqTJ2ImsNXBE+nQqU4VN+UeT34Tdr8daXP4PfMTapm4rx93RFpfLJwivIzLV/cYafTqXKuFpaa+BaItKpVBki5ecwv735FTD1g2XgaJGpRYPqGHv35TEXxVAjHgeRRFS0V/s/FcaImjnyzF4NjMhr6oyF8pNxVX4yfjO/WIp/zV1nCqlWuTz6d2+DbhefKys0itbavjKY+o5Rxs+2tQaubXTGUPkpPxkB+9art/6Ch1+aawq4uUsL9O/eGmXLJNkv0AFLvT9kUONKGA8dOmQOQK5cuXLQWuXl5SE5ORlJSaEDUTcRdyYwZKVGz1pvfBlL5Wefn+8w4qvbN8aDvS6xX5CDltq+MrhxIYwUu0GDBiEnJwelS5dG27ZtMXbsWIwfP978rmLFihg9ejRKlSqF/v37Y+rUqeZ3oZIKozOBISs1etZ648tYKj97/Hbn5GLQsx+iEIV4/bGbUPWs8vYKcthK21cGOC6Ecdq0afj222/x0ksvobCwEO+99x5at26NkSNHYubMmRg4cCCGDx+Obdu2ISsry/y/VVJhtCIU+u96Yyk/GQGZdbzGn+9bxYvOTcVT914jq6SD1vHKz1dlr/rn6uKbUaNGYcmSJVi6dCnq16+PcePGoUePHujatStSU1ONIC5YsAD9+vXDjBkzUKZMGcuQUmG0RBQyg1cDV1br6FkrPxnLeOSXf+w4+o2bhdwj+XiwZ0tcnXahrJIOWscjP//qetU/V4Xxrrvuwvz58/Hxxx9j+fLleOSRR7BlyxbzvpFi2a5dO0yfPh21atXC6tWrTd4RI0agY8eOQUNLhVF213k1cGW1jp618pOxjEd+ny3egBdmLkbDmikYflMLNG4c24/4QxGOR34qjBHeE8OGDUO5cuUwYcIEY0kB5AixadOm5v9zc3MxePBgDB06FJMmTTKiOGTIECxatMi8d0xPTzf5i6fevXtH6IlmVwJKQAmcSWBN5n68/b9NyN5/BHdcdi46tjhbMSU4gUAPPq6OGN944w28+uqrmDt3rpk27dSpEzIzM0+tPJ04cSK6detm3i/OmzcPY8aMQVpamhHGYNOqOmKURa0+cSo/GQGZdbzE31fLN2FW+hpsyNxrKnRWhbJ47dEbsW3rZh0xCpo4Xto3WBXiYvHNkSNH8OCDD5qpVK425YrU22+/3ficnZ0NvoOcMmUK8vPzzcKbjIwMs0p1wIABOpUqCE6dinEInn5nKQYbDx3nF0s34vn/fnuqLtd2aGJOzqiVWkm/oxW2cDy0r53+z9URo8/BvXv3IiUlpcg3ivv37zd/5u8jSTpijITWmXm9GriyWkfPWvnJWMYDv+ffWYQvlm1C9w5N0P/qC1DF79OMePDPTscua5XoWXuVX0yEMXrYARVGGU2vBq6s1tGzVn4ylvHAb8DTs7H3QB4mP3QtzqldpUiF4sE/FUZZjNnhp8LoHHNTst5YMsDKT/nJCIS23rnnIO557iPzTvHNUbeckVnjT0bfq/xUGGXtbmnt1cCwrJhLGZSfDLTyC85v3bY9yFi53Rwl9as2DTHijk4qjLJwSxh+KoxRDoTixWnHJAOs/JSfjACwPfsANmXtw+adOdi4Yx+27NqPzN0HihQ79NZL0D3AocMafzL6XuWnwihrd0trrwaGZcVcyqD8ZKBLOr9HX/kSP27MDgixWf1q4H88dDitdYOA5yuWdH6y6PPuqyQVRmnLW9jrjSUDrPyUn10Cr332A/779Upj3qJhdTSvXw3n1klF03qpOLdO1bCK1fgLC1PQTF7lp8Ioa3dLa68GhmXFXMqg/GSgSyK//YeOYOoHyzD/+y0G3oR7r8QF59awBbIk8rMFKoiRV/mpMEYzCgKU5dXAcBhL2MUrv7BRBcxY0vhxlemwF+fgQO5RVCyXjPtvaY/LLmxoG2JJ42cblApjtNHJytPvGGX89MZXfjICMutox9/vJ3+O9Zl7cUnLunio1yVFPta342m0/bPjQygb9U9GNBg/HTHKuFpaa+BaIgqZQfkpv3AI8ESMf89ZgT0H8sximucfiM4Zihp/4dAPnser/FQYZe1uae3VwLCsmEsZlJ8MdKLzO3qsAH+c+iXWbttjQF3YtBb+0PtSVE+pIAN30jrR+UUFUohCvMpPhdHhyPBqYDiMJezilV/YqAJmTHR+L85agk8XrUflimUx6Pq2uKrdOTJgxawTnV9UYQUozKv8VBgdjgyvBobDWMIuXvmFjarECeOWrBzc//ynpt5/HXqN+Qwj2knjT0bUq/xUGGXtbmnt1cCwrJhLGZSfDHQi81v40zZMmL4AnS+oj0fvTJOBCmKdyPwcAZYgI24VRoejQ28sGWDlp/yCEXh3/ipM++R73JTWHL/t2VYGSoVR+fkRUGF0JBxOF6oduwyw8lN+wQg8959vzEf8g29sh56dmslAqTAqPxVGR2IgYKHasctYKz/lF4jAnv2HcfezH6CwEHh1xPWoWbWSDJQKo/JTYXQkBlQYHcCqwiiDmqj8XvlgOT5YuCbocVEyajrjU9L56VRqtCJAnzgdIZmoHbsjsAIUmoj8vlu7E0+9ng5+wzj5oe44p3Z4G4LbYZ6I/OxwsGvjVX4qjHZbPEw7rwZGmNVzPJvykyFONH6LV2XiT6/9z0C5oXMz3HtDOxkgC+tE4+corAR6MFNhdDhS9MaSAVZ+JZvfkfwCbN99AJt37sPM+auxaec+A+RXbRpixB2dZHDCsNb4CwNSiCxe5afCKGt3S2uvBoZlxVzKoPxkoOOd34qf16BCytnYtS8Xu/Ydwq69/C8Xew4cRubuAziUl18EABfZ3JDWDDd3aSEDE6Z1vPNT/8JsyCDZgvGLmTDu2bMHlSpVQrly5QK6nJeXh+TkZCQlJYWsuZ6u4UxgyEqNnrXe+DKW8cKPZyTyg/wNmfv8BPAQOCK0SrVSK6FBzRRc3b6x+ZjfzRQv/ILVWf2TRUNcCePmzZvRunVrfPrpp+jcuTPGjx+PnJwcVKxYEaNHj0apUqXQv39/TJ061fwuVFJhdCYwZKVGz1pvfBnLWPLLOZiH9BXbkL5iC37cmB2wIslJpVGzWiXzuUXNVP6siBpVK4FiWDO1Is6uEvr+l9Gxto4lP2vvAPUvHErB88SNMB49ehR9+vTBxo0bMWXKFNSuXRsPP/wwZs6ciYEDB2L48OHYtm0bsrKyzP9bJRVGK0Kh/643lvKTETjTOv2HrZi7bCOWrt5x6o887aJLm4aonlL+lPDVqFoRe3ZlonHjxtF2IWrl6f0hQ+lVfq5PpQ4bNgxXXXUVJk+ejFGjRpkRY9euXZGammoEccGCBejXrx9mzJiBMmXKWLaKCqMlopAZvBq4slpHz1r5nWC5PfsAPl+yAV8s24icg0fM76qcVQ5XXHQOftWmAVo0qB4QuvKTxaLyc4afq8L47rvvYvbs2Xj99ddx7bXXnhLGgoICLF26FO3atcP06dNRq1YtrF69GvPnz8eIESPQsWNHU/v09HQjnMVT7969ZXTUWgkogYgJHD12HMs37EHGqmxsyDpo7EuVAlrUq4LOLWugdaOqKF26VMTlqoEScJNAoBkLV4WxU6dO2LVrF6pXr47FixejefPmZmTYvn17wyE3NxeDBw/G0KFDMWnSJCOKQ4YMwaJFi8x7x0BJR4yyENInTuUXKYE1W/dgzpINmPf9Zhw+csyYV0upgKvbn4trOzSJ6L2gxl+k9IvmV37O8HNVGLdu3QquNmUaNGiQEb0bb7zx1AKbiRMnolu3bub94rx58zBmzBikpaUZYQw2rarC6ExgyEqNnrXe+DKW0eT34cK1+OTb9eA5iL50cYs6uO7SpujQqq4tR6Ppny0HLIzUPxlVr/JzVRj9Effs2ROPPfaYecfIlJ2dbaZWuSAnPz/fLLzJyMgwq1QHDBgQtHVUGEtm4MpqHT1rr974kRJ49t8Lkb5i66nRYY8OTdC9QxOkVi4faVFF8pcUfiJIIYyVn4xs3KxKDVaN/fv3mz+lpKREVFMVxohwnZFZbyzlZ0Xg5feX4aNv1qJS+WQ8cnsncJQYraTxJyOp/JzhF7MRo6w6p61VGGUk9cZSfqEIvP3Vz3jj8xUoWyYJE+69Es0bVJMBK2at8SfDqfyc4afCKONqaa2Ba4koZAblFzt+Xy3fjL+8nWFWlo76za9wcfPojRR9tdL2jV37yq4cnrVX21eFMbz2tZ3Lq4Fhu8JRNlR+MqB2+S1auR1PT1+A48cLMaxPR3Rt20jmSBBru/454kyAQtU/GWmv8lNhlLW7pbVXA8OyYi5lUH4y0Hb4/bxpNx7/+1c4VnAcd/W4ELde1lLmRAhrO/455owKY9TRerV9VRijHgpFC/RqYDiMJezilV/YqAJmjJTf5qwcPDJlrvk+kSdYDLr+IpkDFtaR+ueoMyqMUcfr1fZVYYx6KKgwRhOpV2+saDKQlBUJv+x9uRj24hzsO5hnpk45hep0isQ/p30JVL76J6PuVX4qjLJ2t7T2amBYVsylDMpPBjpcfgdyj2L4S3Ow45eDZpENF9u4sZ1buP7JKNi3Vv/ss6OlV/mpMMra3dLaq4FhWTGXMig/Gehw+OUdPYaRU7/Ehsy95nMMfpbBzzPcSOH454Yfwa6h/snoe5WfCqOs3S2tvRoYlhVzKYPyk4G24ldwvBCj/jkPP6zPQr0alTFpyNXmQ363kpV/bvmhwugMaa+2rwqjM/FwqlSvBobDWMIuXvmFjSpgxlD8CguBiW+e2OqNZyP++b5uZjNwN5O2r4y28nOGnwqjjKultQauJaKQGZRfdPkdyS/Ajxt2YfnanfhxYzbWZ+5F5YpljSjWPbuy7GI2rLV9bUDzM1F+zvBTYZRxtbTWwLVEpMIoQ2TJLz+5Cr5fl4Xv1mVhxYZdRfLXSq2EP96Zhqb1Uh30InjRen/IsCs/Z/ipMMq4Wlpr4FoisuzYAx0kKis1etbx2L787GLZ2p1YvmYnlq/dgdyTZyay1uXLlkGbJjXRrlkdtG1WKyajRH/68chP/Uvs+yOc9lVhjF4MBCxJb3wZYOVnzY8f43+3bqcZEXKKlJ9c+BLP925arxraNqtt/mvV6GwklQ586Lf1laKfQ9tXxlT5OcNPhVHG1dJaA9cSkY4YbSCi+H2xbKOZIl215ZciJVQ9q7wRwQ4t66Jq8mFc0Kq5jSu4Y6L3h4yz8nOGnwqjjKultQauJSIVRhuIJs9cjM8XbzhlSSG8uHltXNSsNhrVqnLq9xp/NuD6mSi/kslPhVHW7pbWemNZIlJhtIHo/uc/xRbua3p7J1x2YcOgJWj82YCrwiiDlgD8VBijFgKBC9KOSQZY+Z3JL/dIPvqOmYkySaXxzthe5mewpPw0/mQEZNZejT8VRlm7W1p7NTAsK+ZSBuV3Jujv1u7Ek/+ch5YNq+O5+7rpiNvBWNT4k8H1Kj8VRlm7W1p7NTAsK+ZSBuV3Juj/fPkTZsz5ETd1aYHfWhwLpfxkgar8SiY/FUZZu1ta641liUhHPBEiGjNtPpau3oERd3TCr9oEf7/IYjX+IoRbLLvyK5n8VBhl7W5prTeWJSIVxggR8f0i3zP+Y2RP1KxaSflFyC+S7Hr/RkLrzLxe5RcTYdy7dy9SUlKQlBT8aJu8vDwkJyeHzMNmePbZZzFy5EhZ6zlo7dXAcBBJREUrv6K4tu8+gN9N+hgplcphxhM3W7JUfpaI9MFChigh+bkqjFu2bEHfvn1Ro0YNlClTBu3atcMTTzyB8ePHIycnBxUrVsTo0aNRqlQp9O/fH1OnTjW/C5VUGGVRrR2nt/h9uWwT/vrOInQ8rx4e79/F0nltX0tECdmxy2odPWuvxp+rwjhu3Djk5+dj7Nix4IiwQoUKWLhwIZ577jnMnDkTAwcOxPDhw7Ft2zZkZWWZ/7dKKoxWhEL/3auBK6t19Kzd5jf+jXRk/LwdA3tciF6XtbSsiNv+WTpULIP6FymxovmVnzP8XBXGw4cPm9Fg+fLlMXv2bAwbNgzr1q3DlVdeidTUVCOICxYsQL9+/TBjxgwzqrRKKoxWhFQYZYTih9/RYwW4Y+x74M+pw68LawNw7Thlra/8SiY/V4WRiI8ePYoJEyZg0qRJmDVrlhHFgoICLF261EytTp8+HbVq1cLq1asxf/58jBgxAh07djStk56eboSzeOrdu7es9dRaCXiAwOK1v2D61xvQtE5lPNDTerTogSqpi0og5gQCnd7jqjBy+rRPnz4oW7YsJk+ejDp16hSBkpubi8GDB2Po0KFGOCmKQ4YMwaJFi8xIM1DSEaMsrvSJOL75/bA+C0tW78DiVTuwLXu/cfbhvh1x+UWNwnJc2zcsTEEzKb+Syc9VYXzllVfw4Ycf4v333w9Ie+LEiejWrZt5vzhv3jyMGTMGaWlpRhiDTauqMJbMwJXVOnrW0e44cw4dwbcrt2PJqh1Yvm4neKSUL1Uqn4yO59fH72/rEHYFou1f2BcOM6P6FyaoINmUnzP8XBXGu+66C9OmTStSkzVr1qBZs2bIzs7GqFGjMGXKFLNAhwtvMjIyzCrVAQMGBK29CqMzgSErNXrWiX7jFxYCa7b9YkaFS1ZlYn3mXvB3vtSwZgouaVkXHVrVQ6tG1YPOnAQjnuj8ohdpgUtSfjLCXuXnqjCGQrx//4lpIn7fGElSYYyE1pl5vRq4slpHz9oOv0N5+Vi25sT0KH9ylOhL3BD8/HNrmLMUO55fz/IDfqua2PHPqsxo/l39k9FUfs7wixthtFs9FUa75E7Y6Y3lDr9NO3OwZHWmmSJduWU3jh8/PSysXLEs2rfgqLAu2reog/JlrVdjh+u1tm+4pHTEKCOVWPxUGJ2IBr8ytWOSAfYqvyP5Bfh+XRYWnxTD3Tm5RUBIp0jDpepVfuHWz+l8yk9G2Kv8VBhl7W5p7dXAsKyYSxm8xC9r7yF19zBnAAAW+klEQVQsXplpxHDFhl3IP3bcsSnScPF7iV+4dXIzn/KT0fYqPxVGWbtbWns1MCwr5lIGL/BbuuUIvli6EdzH1D/5VpG2b14H7VtGd4o0XPxe4BfoO7Jw6+d0PuUnI+xVfiqMsna3tPZqYFhWzKUM8c5vwdKf8Mx/fzxFo0HNFFx6Xj3zrvD8c2q4RCn4ZeKdn/onCxHl5ww/FUYZV0trDVxLRCEzxDu/sf+YiyXrfkHaBQ0wsEcb1K52lqzCUbaOd37qn6zBlZ8z/FQYZVwtrTVwLRF5WhjvfmY2snPy8PKw61CvRmVZZR2w1viTQVV+JZOfCqOs3S2t9cayRORpYbzp8bfNpxezxvdBUunA2xbKCMisNf6Un4yAzNqr8afCKGt3S2uvBoZlxVzKEM/8+GF+v3GzUKVSOUwP49Bgl5AVuUw886Oj6p8sKpSfM/xUGGVcLa01cC0ReXbEuCFzLx6a/Dka103F/z1wjayiDllr/MnAKr+SyU+FUdbultZ6Y1ki8qwwLlq5HeNeT8elrerhiQFdZBV1yFrjTwZW+ZVMfiqMsna3tNYbyxKRZ4Xxo2/W4eX3l+K6jk1x300XyyrqkLXGnwys8iuZ/FQYZe1uaa03liUizwrja5/9gP9+vRIDurdB7ytaySrqkLXGnwys8iuZ/FQYZe1uaa03liUizwrjpLcy8PV3mzGsz6Xo2vYcWUUdstb4k4FVfiWTnwqjrN0trfXGskTkWWF89JUv8ePGbDx9T1e0blxTVlGHrDX+ZGCVX8nkp8Ioa3dLa72xLBF5Vhjvee4j7NxzEK88fD3qVI+vHW98UDX+Ejf+WDNtX2faV4VRxtXSWgPXElHcC+Omnfuwddd+bM7KwZasEz8z/TYMf/ep21C2TJKsog5Za/zJwCq/kslPhVHW7pbWemNZIoobYaTYbc3ej4079hkBpBhSFEOlxrXPwv89dL2skg5aa/zJ4Cq/kslPhVHW7pbWemNZInJdGLP35Z4c/eWYn5t35hhBPJpfENAXjga5D2rDWlXAA4b5k6do1K1+FjZu3Ag9Nsl+G+v9YZ+dTqXK2IXip8IoZ+t6xx5NlxO5Y9p7IA9bdp0QvhPToCemQnOP5AcWwOQk1D/7hABS+Bqd/Mn3h6VKBd4HNZH5RTPOgpWl/GSUlZ8z/FQYZVwtrUta4B49VmA21S44Xnjy53Hz74LjxwP/vqAQxwtP/7040B07dqBOnTqWnH0ZFq3MxPrte4wQHsg9GtSuSd3UIuJHMbSzgKaktW/YDRFmRuUXJqgg2ZSfM/xUGGVcjTWn4E507oUoKPATgsJCbNq0BfXq1TfCcEIgigrG8ZO/94mJL99pcTlt5y8uwf/N8k/YFC/TZ+P/+4OHclGuXPlT/pm/FRw/XZ+TolZAAWO5hSfLLTghZkeCTD9GAau4iLLJSWhQg1OfKWhYs8qpkaAdAdQRj7g5AhagHbuMq/Jzhl9MhPHQoUOoUKECSpcuHbRWeXl5SE5ORlJS6NV+zz77LEaOHCmjY9N68szF+HzxBpvWiWtGQeIRTEmlS5/8Wcq0NX9X2u/3J/59Ip/v3758vt/nHc5F5cpnFbFLSiqN0qV8tifLTiplflexfPIpMXTj0GDtmGRxrPyUn4yAzDpY/LkqjLt378avf/1rlClTBps3b8YjjzyCgQMHYvz48cjJyUHFihUxevRo8z6nf//+mDp1qvldqBRLYXxx1hLMWbwhQGd/uqM/frwA5cuWPSUKxTv+pJMdur84FP13KRghoKgYMTgtNv4CYYQl6aT4+Odj+RQen5AknS6Dv9u1Kwv16tU9LVDMZ/KcFixj62dXVMhOlycL0cDW2nHKqCo/5ScjILP2avy5KozPPPMMDhw4YIRw586d5t3Rjz/+iCeffBIzZ840Ijl8+HBs27YNWVlZ5v+tUiyF0co3/t2rgRFO3dzIo/xklJWf8pMRkFl7Nf5cFcbf/va36NatG26//XYUFhaakcz69esxaNAgpKamGkFcsGAB+vXrhxkzZpiRpVVSYbQiFPrvXg1cWa2jZ638ZCyVn/KTEZBZx8VUap8+fcD/brvtNlObWrVqYdGiRWjQoAGWLl2Kdu3aYfr06eb3q1evxvz58zFixAh07NjR5E9PTzfC6Z/4HjI/P/DyexkytVYCSkAJKIFEJlCjRg3cfffdZ1TR1RHjn/70J6SkpOD3v/89CgoKzChx3759pxbh5ObmYvDgwRg6dCgmTZpkRHHIkCFGPIN9RxbvI0b1T3ZbKT/lJyMgs9b4K5n8XBXG999/Hy+88AI+//xzvPPOO/jLX/6Cb7755hT5iRMnmqlWvl+cN28exowZg7S0NCOMwaZVNXBLZuDKah09a40/GUvlp/xkBGTWweLPVWE8fPgwrrvuOvz888/gv+fMmYNLL73U1Cw7OxujRo3ClClTzNQoF95kZGSYVaoDBgwIWnu9sZwJDFmp0bPW9pWxVH7KT0ZAZu3V+HNVGH2It27ditq1a5vvFH1p//795p+cao0keRV8JHV0Mq/yk9FVfspPRkBmrfHnDL+YCKOsKkWtuSCnS5cu0SwyqmWpfzKcyk/5yQjIrDX+SiY/zwujrNnUWgkoASWgBJRAUQIqjBoRSkAJKAEloAT8CCS8MPITECarreXciAru+lO5cmU3LmXrGr/88guqV69uy9Yto8zMTNStW9ety0V0HfLjO3L/d+cRFeBwZm67WKVKFYevElnx3BOZ+yH7M+Nm9VycV6lSpcgKcyB3IP+OHTuGgwcPomrVqg5cMbIiA/nHEujj3r17we/0YpkC+cfNXfiZHj/XczqxnRhH/p/7sR8+66zgR8nRp4QVRt/KVu7JyoNk+d3kP/7xD5QvX97ptjijfG5ewI9IGzVqZPaIpR/t27d33Y9gF/zyyy9x3333mQ0WuME7dyi68cYb48Y/nyMffvghbrjhBnPKR7DvWmPhNNv0zjvvNBtV7Nq1Cw888ABuvvnmWLgS8Jrffvstnn76adOR07+xY8fikksuial/jLMVK1aYb5X/8Ic/4JZbbjH+/Otf/8Lzzz+PevXqmc6dO2DFonMP5h8/MeMezlxNzwWDXPzSokUL11kG88/nCLfWJF9+GheLFMw/+vPoo4/i/PPPd5Qf4/yHH37ArbfeinXr1qFmzZrmQeGee+4xgxN+BcGNZoJtO5qwwshdcx577DGzWw5T165dzWYBvXv3djxOJkyYYHb4adKkibnWNddcg4cfftj8fPfdd82NFauApT/F/SObxx9/3HxDSm4MHu48FKtU3D/6sWnTJuPjv//975gLY3H/eHP16NEDffv2BUeNvCHJNFapuH933XUXOnToYB5+2JFz68XJkye76t4HH3xgNvK4/vrrzXW///57vPbaa5g9ezb+/Oc/G2GkEHLkyNEER7YPPvig2U+ZHanTKRz/jh49inLlypnRIkch48aNM99cu8EyHP98jPi9+Msvv2x4utXPhOsfdzXjQKF+/fpYuXKluZcpktLEe3DatGmnimE/y13S/vrXv5o2ojDy7x9//DHefvttLFy4ELwvgvVzCSuMDAoO4zlk5hN969atzYblDRs2lLaBpT1HW9ycgCMwJo4k2BD8uXz5clx77bWmsWKVivvHqQVONXNKa9iwYWYai9+TxioV9+/IkSPm+1d2pGQY6xFjcf86depk/Pr0009xxRVXgBtVtGzZMlb4zGjfP/44I8AR7E033WSEiJ2lb5tFt5xkB0VhfOihh4pckk/tHG1TGDdu3Ggezrh/MhMF57vvvjMzLE6ncPyjDxx1cAqQr2iuuuoqUx/u/ex0Ctc/7v35u9/9Dk888YQRbreEMRz/yIwPFGzruXPnmj2xea+wj5YmvsLwffLnXxZnlnzCyEPPL7roIlx55ZVmwMQd2DiyDpQ8L4wUv0CN3717d/N099Zbb5kREDsKdvpOpi+++MJMRXKqisdrNWvWzEyZspPkkwmffhm4l19+Ofgtp9spmH98Z8fguf/++7F27VrwiZPTvm6nYP5xlMOA5hMeAz1WwhjMv7Zt25o2fe6550xnzhHP3//+d7fxIZh/r7zyitlpiiNaTk3ee++9QTuEaDvN0TNH+++9954pmp3iOeecgzZt2pj/9xdG5uWMju8p/o033jA7YDnJMhL/fGyWLVtmYvGCCy5w/PVMJP6xL+RMBR8kGIPs85wWxkj845Qm+0Tez5zi5AweN2+hQNpJ/n2/76GP5fj6fv7bXxjJgscZMv65m1rZsmXB1zMJKYwMAO6vWjyxM3jxxRdNR8AbzDd6s9MA4dpQeLdv346vv/7a3DRnn322EWOeO8knqosvvhhLliwB94yl+LidgvlHweYNxSDloc+xeA9LFoH84/QfffO9E1u8eLH5N5/4GNhupmD8+P6Y04GcTuVop3PnzjGZEQjmH0eQFCb6xdEj34H+9NNPrqD75z//ic8++8zM1nBbRz4kkpPv3Y6/MHKmgjMXvgcf3jNMfAfpVIrEP/rAhw8+9P7tb38zDxpOp0j8Y8dPUeD9wYVWa9asMSLAVzdOpUj843m8fF9M3zjCe/311008+k+BRuKnf9/P6VG+vmJi3+9bZOYvjHyYYfyxj+MsGX3wjSaLX9fzI8ZgILntHDtUBofbK/GKT2VxuE6R5EIDvmvk1MFTTz0VSQxENW9x/+644w5zMzk9og63Ev7+cQUbnzR9qXnz5li1ahX4M1YLcIrz47FpfHfCBS48HYZCwIexWKXi/nEamg89nPLjgyKnU9mRuJnCmWqjPxdeeCFeeukl8+qDnTxnX/hu3ukUjn+MRfYlnAbkO1s3Uzj+caqSD+ZMnILmQiHOFPB9ntMpHP/Ij6+yKFx8ncQBDRcx8f6RpnCmUvnwyhk7Dpj4eo3XJq9A+3AnrDDyKYRPCP6JTzfFfydtkED2xTsm3yiCebkykBunV6tWzYlLh1Vmcf/YqXMVly/xRXU8vQP1r1Qsp1J9fhTnx5uN00IchbETYsfO6dVYpeL+cQaDU0gcjbETePXVV83o0c0UquPkVJpvFS9nUnxTa1yow8VWbjwAheOfbyrQn9tvfvMb2yOeSPiH459/eVyJzPeMTk+l+q4Zrn98qOB7Wb4PZAxyLUM0+sJQwsi+jSNV9mlc1c53jUzcm5uv2QKlhBXGSILOjbxcDMRv8LhIw40b3Y066TWKEuBUEWcG4jHxaZ2rURl/8Z448uF0G6f4NSUeAcYiFzFFQxDt0GE/TKEM9b2xCqMdsmqjBJSAElACCUtAhTFhm1YrpgSUgBJQAnYIqDDaoaY2SkAJKAElkLAEVBgTtmm1YkpACSgBJWCHgAqjHWpqowSUgBJQAglLQIUxYZtWK6YEvEeAH/kzVahQwXvOq8cJQ0CFMWGaUisSjAA/JOZHxcUTv9fkt609e/Y0H+VH+0PyPXv2mGO8nnzySbPbUbiJm2Y/88wz4OcfxY8B44fl3HYw2jvX8KNnfmPLfUuDJbv1CbfezMcP/LnVV6hN7EP5wc88XnjhBbNnbVpaWiSX1rxK4BQBFUYNhoQn8Oabb5pNFfgdH7dG4/aA7DS5GTS36eM+i9x1n1uVRTPxpA1+1/jHP/7RnGgSbmJ+7ifJo3GKfxdJf3m6Q7RPP+H3ja1atQr5Qbjd+oRbb+bjXrP85jfUNnCh/NiyZYvZ55e8yVGTErBDQIXRDjW18SQBnnBCQeTmyqNHjzZ14E4rFEbuDMNjapg4WuNm1hQnHqfDLdVmzZplxJUiyyN9uKUUd2jh1n48NYInGXC7NXbq3Hlm0qRJ5iNmChsFl6M/2nBfVW4dx00euIclTxfgzjk87YIjHQqfvzDyI2TuFPLJJ58Y37lXLHewKS6M3HOU0488iYR+8kxNnpbCa3D7PG5Rx83iudMIr0N/uPE5zz7kJt3kwXPquC0g+fjycfTG+tNnHrsUqj6+oOAInPvYzpw50+wuwlNHuJk/t99i+bwWzzDkDkFkxn/TDz6wsB14Hebn6G/o0KFGrDkC5Ifh3MeXu6wE84PbyM2ZM8fsQMT9TH3nPHoyYNXpmBFQYYwZer2w2wRCCSM7Uu7ZyP0lucEwD7rmqIUiwkQRpNiwg+YG2BQhigdFlBvGUww4XcoNsCksFNCrr7761IiPZ4NSXLmHL7froghxhMY9amnLfRy50wsFz18YKRw8Hofbo1GAKdQUuuLCyLM/KbDc4ooiytHxeeedZ3zjvqgsk9vWca9Kige3xqLo8D9ux0YB4fUpVjz9omnTpkZcKPL8HU8O4XaKvhFs8fr4H3xMTjzhgVOeFHzuV0wePLmFQskHBG6lRhZkzC3qODrmVl2XXXbZqRExmXADfpbBPUo53U3xpFAG84MPN76Dj/nQEo2z/tyOU71e7AmoMMa+DdQDlwiEEka+g6SosPPnPrvcy5GdP4WRnS33deSpFPx/7s7P0RhHURQEik6vXr3MYagcHXIESrFjGezAeRoDBY6nCVAQKFwc+VEkeOoEO29O/fF6PNKIoyXfVCpHrhQQ7vPI0wQ4/RtMGLk5PQ8Aprhz43qOHCm+fHfIkSp9p6jwGtz3k/uQchTKk0r8p1I5IuPIjoLGY4LoL8/Ro5/B6sNNoX2Jh8RyBPuf//zHbFzOI6Z4diHrQfFjefXq1TOCx4cH7uX6v//9z/yeh1H7poq7dOli8vMhgA8cPEbOXxgDceXDgE6lunRDJfBlVBgTuHG1akUJhBJG3ztGjqo4EqQIcRNmigkFjiM8Ls7hNB2F0bfPIoWC51hyRERB++ijj8xF/UdYvneM/DuFgtfi8UWcbuXhvByh8ToUXp4/yL/7hJGbkXODZN9iG274TqELNGKkMFMMuZCGI0EKIIWNvnM0zCOIeKKAb8qSwkzRKi6MFEWKI/eU5CiSJ62zvhRl/3em/vXxfz/r2weTwke/+QDBUSHP7uM0KsWQokdR5gk4vsRjnPjg4BNG32iPZXCKmntrsj6+EWMgrrRRYdQ7X0pAhVFKUO09Q8CuMLKj5oiJq0v5LpHvBflejP/mNCTffVFEubCH4sQO33dmXzAhYSdP4WA+jq7YyVOM+R6OZfmEkeLJ6Vr+5GpUilmwEaOVMPJ6HBXzfSPf1XE06j9ipADz4FZOyXbr1s38nQfK8lgyTufSr3CEkQHBKVGOAjlFywcJXpspIyPDiBtX27IeXBXMkSUPjuVxWBxp+4SRU6GsL0fgvC6nZ/1HjKGEkdPdnIotvqrXM8GqjsaUgApjTPHrxd0k4BNGnvHHDpfJt/iG03icDuToj+/U/EeMPmHku0dOUXL0w8TRFt/58SgnvrPjCI2Jnfdrr71mRlvs0Pn5BUdNPBuPh6lyRMhrcXTKhTy+xHeBFCP/zzU4WqXI8J0hhY/TpbxeqBEj/Wc96A/FhzYcMXK6lr6xLP6eozDWjaNWvhflKJdCRqGiIPpGv6wn+XAqM1h9iq/o9b0f5E/6wneyHLlyNMlpaHLjNX3MKPwcqXLxERf8cJTI9470iYw4Zc3pZx40S1+C+cHRKD/54LQtR+G8tiYlECkBFcZIiWn+Ek+AxyJxYQnfk/kfIcZFJRxZRnL0FFde8rw4Tv9xJWegRAHjSJLXC3SoaiQNwtGtr6zix59RkPguj6temXbu3GlWuXLBjlNHpfHzCvLyXdO/LnxXSdHmYh7mo+BxxBvom1R/O9aBDxQsMxjTSJhp3pJHQIWx5LW51lgJeIIA3yVyBa1/4jtYjqA1KQEnCagwOklXy1YCSkBEgB/z8/MWTr9yxMjFR5qUgNMEVBidJqzlKwEloASUgKcIqDB6qrnUWSWgBJSAEnCagAqj04S1fCWgBJSAEvAUgf8HSGbHBmt4rdwAAAAASUVORK5CYII=", "text/plain": [ "\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": [ "linker.unlinkables_chart()" ] }, { "cell_type": "markdown", "id": "ca2570b7-9f68-4fb2-b7a9-527233a8fcd4", "metadata": {}, "source": [ "In the above chart, we can see that about 1.3% of records in the input dataset are unlinkable at a threshold match weight of 6.11 (correponding to a match probability of around 98.6%)" ] }, { "cell_type": "markdown", "id": "fd531cd2", "metadata": {}, "source": [ "## Next steps\n", "\n", "Now we have trained a model, we can move on to using it predict matching records." ] }, { "cell_type": "markdown", "id": "83fd8e7f", "metadata": {}, "source": [ "\n", "## Further reading\n", "\n", "Full documentation for all of the ways of estimating model parameters can be found [here](https://moj-analytical-services.github.io/splink/linkerest.html)." ] } ], "metadata": { "kernelspec": { "display_name": "splink_demos", "language": "python", "name": "splink_demos" }, "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": "3b53fa520a31e303a9636a08ff10a3bbc14893ee50cb37445791fa59628fc75b" } } }, "nbformat": 4, "nbformat_minor": 5 }