What's Changed

• Explicit selection by @ADBond in #2484
• Fix clustering in debug mode by @ADBond in #2485
• Less caching in debug mode by @ADBond in #2488
• Update changelog by @RobinL in #2497
• remove unnecessary import by @lubrst in #2500
• Spark test session handling by @ADBond in #2504
• Fix count_comparisons_from_blocking_rule by @RobinL in #2503
• Streamline docs by @RobinL in #2505
• Test and fix debug mode by @ADBond in #2481
• Improve compare two records by @RobinL in #2498
• Bug - get columns of DuckDB frame even when table is empty by @ADBond in #2510
• Update CONTRIBUTING.md with correct link by @zmbc in #2513
• Constrain dev pandas version by @ADBond in #2518
• Update lockfile + fixes for latest package versions by @ADBond in #2514
• Avoid bug with checkpointing by switching to parquet by @RobinL in #2525
• Clustering allows match weight args not just match probability by @RobinL in #2454
• Explicit tf columns select by @ADBond in #2527
• Make Settings._columns_used_by_comparisons unquoted by @ADBond in #2532
• Pairwise string distance comparison by @zmbc in #2517
• Bias blog 2 by @RossKen in #2408
• 4.0.6 release by @RobinL in #2537

New Contributors

• @lubrst made their first contribution in #2500

Full Changelog: v4.0.5...v4.0.6

What's Changed

• add EMA use case by @RobinL in #2468
• Change name of second __splink__cluster_count_row_numbered query, prevent table name conflict by @browo097302 in #2447
• Add iteration number to neighbours_filtered table by @ADBond in #2470
• Fix docs examples by @ADBond in #2471
• Docs - correct heading and link text by @ADBond in #2472
• Simplify Altair import by @ADBond in #2479
• Specify version range for pytest-cov in CI by @ADBond in #2489
• Compare two records - allow dataframes to be registered by @RobinL in #2493
• 4.0.5 release by @RobinL in #2495

Full Changelog: v4.0.4...v4.0.5

What's Changed

• Handle threshold_match_probablity 0 in predict() #2420 by @browo097302 in #2425
• Take converged clusters out of play by @RobinL in #2436
• Fix clustering in linky jobs with source dataset column on Postgres by @ADBond in #2444
• Cluster multiple thresholds v2 by @RobinL in #2437
• Used .blocking_rule_sql property match_weights_interactive_history_chart() by @browo097302 in #2446
• restore pretty print of SplinkDataFrame by @RobinL in #2450
• 2440 add docstring to customrule by @RobinL in #2452
• Cluster multiple add stats by @RobinL in #2453
• Score missing intra-cluster edges by @ADBond in #2442
• Fix cluster studio docstring by @ADBond in #2455
• Docs cleanup by @Thomas-Hirsch in #2460
• Fix profile charts issue by @RobinL in #2466
• 4.0.4 release by @RobinL in #2467

New Contributors

• @browo097302 made their first contribution in #2425

Full Changelog: v4.0.3...v4.0.4

What's Changed

• fix dead links by @RobinL in #2430
• Cluster without linker by @RobinL in #2412
• Better autocomplete for dataframes by @RobinL in #2434
• v4.0.3 release by @RobinL in #2435

Full Changelog: v4.0.2...v4.0.3

What's Changed

• Fix performance issue with exploding blocking rules by @RobinL in #2385
• Add cookbook to examples by @RobinL in #2388
• fix docs by @RobinL in #2389
• Create llm prompt by @RobinL in #2366
• 2351 fix spark sampling by @aymonwuolanne in #2390
• Improve number formatting and descriptions on match weight charts by @RobinL in #2392
• add labelling tool by @RobinL in #2393
• Fix ColumnsReversedLevel by @RobinL in #2395
• Add is_in_level and compute_comparison_vector_value testing functions to internals by @RobinL in #2396
• Migrate tests of comparisons and comparison levels to new testing framework by @RobinL in #2397
• Add AbsoluteDifferenceLevel by @RobinL in #2398
• TimeDifference docstring by @RobinL in #2400
• More levels docstrings by @RobinL in #2401
• add dates docs by @RobinL in #2402
• Better docstrings by @RobinL in #2404
• Add cosine similiarity comparison level and comparison by @RobinL in #2405
• add gov transformation mag link by @RobinL in #2406
• Add cosine similarity tests and allow schemad data by @RobinL in #2407
• Consistency in usage of sql_dialect, sql_dialect_str, sqlglot_dialect by @RobinL in #2391
• ArraySubset comparison level by @RobinL in #2416
• Interactive comparison notebook by @RobinL in #2417
• 4.0.2 release by @RobinL in #2418

Full Changelog: v4.0.1...v4.0.2

What's Changed

• Bias blog by @ericakane-moj in #2279
• Fix bug in Postgres example by @fhightower in #2352
• Added new use case to index.md by @AnthonyTacquet in #2363
• Fixing issue with reaonly filesystems by @RossHammer in #2357
• Update changelog by @ADBond in #2370
• avoid attempting to cast Infinity to double for spark backend by @bkitej-rw in #2372
• Fix Spark 'InfinityD' bug by @ADBond in #2374
• Support duckdbpyrelation as input type by @RobinL in #2375
• Bump actions/download-artifact from 3 to 4.1.7 in /.github/workflows by @dependabot in #2377
• Splink datasets - simplify + restructure by @ADBond in #2378
• Fix docs reference for renamed class by @ADBond in #2380
• Update upload-artifact version in docs CI by @ADBond in #2381
• Allow a specific m and u probabilities to be fixed during training by @RobinL in #2379
• Allow all charts to be generated as a dict by @RossHammer in #2361
• Splink 401 release by @RobinL in #2386

New Contributors

• @probjects made their first contribution in #2172
• @DavidFrenchSG made their first contribution in #2204
• @astimoore made their first contribution in #2229
• @dkaufman-rc made their first contribution in #2240
• @ericakane-moj made their first contribution in #2277
• @bnm3k made their first contribution in #2342
• @fhightower made their first contribution in #2352
• @AnthonyTacquet made their first contribution in #2363
• @RossHammer made their first contribution in #2357
• @bkitej-rw made their first contribution in #2372

Full Changelog: v4.0.0...v4.0.1

See
https://moj-analytical-services.github.io/splink/blog/2024/07/24/splink-400-released.html
for release announcement

What's Changed

• Comparison that has tf adjustments = True properly accounts for column expressions by @RobinL in #2267
• Adjust package top level imports by @ADBond in #2269
• Evaluation docstrings by @RobinL in #2271
• Remove broken EM training options by @ADBond in #2272
• Restore lat-long SQL test by @ADBond in #2273
• Consistent db_api argument name by @ADBond in #2278
• Turn off previously configured options by @ADBond in #2276
• Remove jan 1st option from date of birth comparison by @RobinL in #2281
• update release blog by @RobinL in #2284
• Small fixes by @ADBond in #2285
• Update Splink 4 docs by @ADBond in #2283
• update version by @RobinL in #2286

Full Changelog: v4.0.0.dev8...v4.0.0.dev9

What's Changed

• Docs links by @RobinL in #2237
• Cherrypick various patches to master by @RobinL in #2241
• Update docstrings splink4 by @RobinL in #2246
• as spark dataframe in docs by @RobinL in #2247
• More docstrings by @RobinL in #2248
• Docstrings 3 by @RobinL in #2250
• Restore spark test mark by @ADBond in #2253
• add note about excludedocs by @RobinL in #2256
• Del accidentally committed testing script by @RobinL in #2258
• Splink 4 release blog v1 by @RobinL in #2235
• Find biggest block by @RobinL in #2260
• Blocking tutorial by @RobinL in #2262
• prevent integer overflow by @RobinL in #2263
• Remove clustering pairwise output format by @ADBond in #2264
• improve blocking below thres by @RobinL in #2265
• splink 4 dev8 release by @RobinL in #2266

Full Changelog: v4.0.0.dev7...v4.0.0.dev8

What's Changed

• Update docs for Splink4 by @RobinL in #2203
• Update comparison template library by @RobinL in #2214
• Further splink4 docs work by @RobinL in #2215
• Move comparison helpers by @RobinL in #2216
• Restore dev guides by @RobinL in #2217
• add back tags by @RobinL in #2218
• Splink4 docs: fix more links by @RobinL in #2225
• Athena linker splink4 migration by @RobinL in #2226
• Athena linker migration 2 by @RobinL in #2227
• Restore Athena example to docs by @RobinL in #2228
• Block to IDs by @RobinL in #2231
• dev7 release by @RobinL in #2236

Full Changelog: v4.0.0.dev6...v4.0.0.dev7