v2.5.0

What's Changed

• Add method to facilitate using pytest by @AdriMarteau in #1701
• GSK-2495 Convert backend errors to python exceptions by @andreybavt in #1724
• [GSK-2533] Add a flag to handle control characters inside LLM response decoding by @pierlj in #1731
• GSK-2545 Update Suite.upload() method to upload modifications to existing suite by @kevinmessiaen in #1736

New Contributors

• @AdriMarteau made their first contribution in #1701

Full Changelog: v2.4.0...v2.5.0

v2.4.0

What's Changed

• Add the test metadata in the test result page by @kevinmessiaen in #1691
• Updated ground truth tests to have new test details by @kevinmessiaen in #1711
• Put all BaseMode.meta fields into @property methods by @kevinmessiaen in #1700
• updating tests catalog by @rabah-khalek in #1718
• Update index.rst by @rabah-khalek in #1721
• Fixed error from contribution push when model is text generation by @kevinmessiaen in #1716
• GSK-2471 GSK-2458 Fixed issue when loading code from uninstalled dep by @kevinmessiaen in #1714
• GSK-2470 Debugging a test with kwargs doesn't work by @kevinmessiaen in #1728
• Renamed borderline with underconfidence by @kevinmessiaen in #1712
• Docs: replace space duplication instructions by docker template creation instructions by @luca-martial in #1704
• Added OCR typo transformation by @goyalpramod in #1697
• Fix mistake in underconfidence issue metric name [GSK-2546] by @mattbit in #1733
• GSK-2437 Get logs from a running job by @andreybavt in #1727
• Feature: Llm hub refactoring by @kevinmessiaen in #1695

Full Changelog: v2.3.2...v2.4.0

v2.3.2

What's Changed

• Fix the docstring by @Hartorn in #1715
• Regenerating pdm.lock by @Hartorn in #1689

Full Changelog: v2.3.1...v2.3.2

v2.3.1

Full Changelog: v2.3.0...v2.3.1

v2.3.0

What's Changed

• Bump actions/upload-artifact from 3 to 4 by @dependabot in #1705
• Bump SonarSource/sonarcloud-github-action from 2.0.2 to 2.1.1 by @dependabot in #1706
• [GSK-2203] Moved non ml worker related code outside from ml worker by @kevinmessiaen in #1683
• Fixed import of Query, QueryBasedSliceFunction by @kevinmessiaen in #1713
• GSK-1943 - make jobs cancellable from the Hub by @andreybavt in #1698

Full Changelog: v2.2.0...v2.3.0

v2.2.0

What's Changed

• Add speech-to-text typos transformation by @goyalpramod in #1686
• [GSK-2035, GSK-2003, GSK-2263] Complete refactoring of prompt injection by @rabah-khalek in #1602
• Still upload suite when some artifacts fail to be uploaded by @kevinmessiaen in #1699

New Contributors

• @goyalpramod made their first contribution in #1686

Full Changelog: v2.1.3...v2.2.0

v2.1.3

What's Changed

• Fixed newline issue when reading env file by @kevinmessiaen in #1688
• Fix run start instructions by @kevinmessiaen in #1690
• [GSK-2329] Use object instead of string for documentation by @kevinmessiaen in #1675
• Add slack message on release by @Hartorn in #1693
• [GSK-2406] Metric name is not shown in the Hub by @Googleton in #1692
• Added a LLM-as-judge per row evaluator by @kevinmessiaen in #1672
• [GSK-2288] Add more tracking to Python Lib by @Googleton in #1663
• Add data quality tests by @Kranium2002 in #1651

New Contributors

• @Kranium2002 made their first contribution in #1651

Full Changelog: v2.1.2...v2.1.3

v2.1.2

What's Changed

• GSK-2368 Recreate docker container on start by @andreybavt in #1676
• GSK-2080 Fix wording by @Googleton in #1678
• Fix CI (breaking because of pdm and windows-2019) by @Hartorn in #1679
• Text transformations used by EthicalBiasDetector are not deterministic [GSK-2332] by @pierlj in #1674
• [GSK-1279] Fisher's exact test and permutation test for slice metrics significance testing by @luca-rossi in #1671
• [GSK-2378] Only persist/read cache when explicitly asked by @kevinmessiaen in #1680
• Add ipcc model and dataset creation for e2e tests by @Hartorn in #1684

Full Changelog: v2.1.1...v2.1.2

v2.1.1

Full Changelog: v2.1.0...v2.1.1

v2.1.0

What's Changed

• Added a small description to specify what language are supported for LLM scan by @kevinmessiaen in #1657
• Add examples to AVID report [GSK-2282] by @mattbit in #1652
• [GSK-2184] Fix error when uploading dataset with bool column by @Inokinoki in #1658
• Add docs for advanced scan usage by @mattbit in #1660
• [GSK-2292 GSK-2293] Add args kwargs to the public methods of the model and all by @kevinmessiaen in #1661
• [GSK-2323] Set content-type to json actively for dumped str by @Inokinoki in #1662
• Fixing doc links in scan widget [GSK-2327] by @mattbit in #1665
• Fixed doc format of QueryBasedSliceFunction by @kevinmessiaen in #1666
• fixed typo in TestResult by @rabah-khalek in #1668
• [GSK-2294] Fix doc for slicing and transformations by @Hartorn in #1670
• Added is_default_value property to TestInputDTO by @kevinmessiaen in #1659
• Add huggingface report template by @cy-moi in #1656
• Add language support in LLM generators [GSK-2152] by @pierlj in #1664
• Added test_llm_ground_truth_similarity by @kevinmessiaen in #1669
• [GSK-2346] Avoid copying whole dataset when doing slicing by @Hartorn in #1673
• [GSK-2326] Handle environment variable when starting hub using cli by @kevinmessiaen in #1667

New Contributors

• @pierlj made their first contribution in #1664

Full Changelog: v2.0.7...v2.1.0