Fix bug with labelling tool where it didn't work offline

[RobinL]

Type of PR

• BUG
• FEAT
• MAINT
• DOC

Closes #1645

Give a brief description for the solution you have provided

Updates consistent with #1639 to make sure it works offline

Here's a script to produce a tool to ensure it works offline

script to test

import duckdb
import pandas as pd

import splink.duckdb.duckdb_comparison_library as cl
from splink.duckdb.duckdb_linker import DuckDBLinker

df = pd.read_parquet(
    "https://github.com/moj-analytical-services/splink_demos/raw/master/data/historical_figures_with_errors_50k.parquet"
)

# df = pd.read_parquet("synthetic_50k_clean.parquet")
# df["postcode_fake"] = df["postcode"]
ids = list(df["unique_id"].sample(1))
# ids.append("Q52155443-1")

# ids = ["Q52155443-1"]

settings = {
    "probability_two_random_records_match": 0.01,
    "link_type": "dedupe_only",
    "blocking_rules_to_generate_predictions": [
        "l.first_name = r.first_name",
        "l.surname = r.surname",
    ],
    "comparisons": [
        cl.jaro_winkler_at_thresholds(
            "first_name", [0.9, 0.7], term_frequency_adjustments=True
        ),
        cl.jaro_winkler_at_thresholds(
            "surname", [0.9, 0.7], term_frequency_adjustments=True
        ),
        cl.levenshtein_at_thresholds("dob", [1, 2], term_frequency_adjustments=True),
        cl.levenshtein_at_thresholds(
            "postcode_fake", 2, term_frequency_adjustments=True
        ),
        cl.exact_match("birth_place", term_frequency_adjustments=True),
        cl.exact_match("occupation", term_frequency_adjustments=True),
    ],
    "retain_matching_columns": True,
    "retain_intermediate_calculation_columns": True,
    "max_iterations": 2,
    "em_convergence": 0.01,
}


linker = DuckDBLinker(df, settings)


linker.estimate_probability_two_random_records_match(
    [
        "l.first_name = r.first_name and l.surname = r.surname and l.dob = r.dob",
        "substr(l.first_name,1,2) = substr(r.first_name,1,2) and l.surname = r.surname and substr(l.postcode_fake,1,2) = substr(r.postcode_fake,1,2)",
        "l.dob = r.dob and l.postcode_fake = r.postcode_fake",
    ],
    recall=0.6,
)

linker.estimate_u_using_random_sampling(target_rows=1e5)

for this_id in ids:
    linker.labelling_tool_for_specific_record(
        this_id,
        out_path=f"try_labels/labelling_tool_dedupe_only.html",
        show_splink_predictions_in_interface=True,
        view_in_jupyter=False,
        overwrite=True,
        match_weight_threshold=-10,
    )

[ADBond]

Great, this works perfectly 👍

[RobinL]

Thanks!