v2.12.1

What's Changed

• Add trace-based evaluation in agent evaluation metrics to get correct score by @AnilSorathiya in #478
• notebook: Edited Run dataset-based tests by @validbeck in #480
• Add optional model doc slug field by @mdeyell-valid-mind in #479
• Add library version by @mdeyell-valid-mind in #481
• [SC-14933] Fix-missing-values-threshold-and-rename-parameter by @juanmleng in #482
• Fix unclosed ul tag in document_agentic_ai.ipynb by @nrichers in #483
• fix: use PAT instead of GITHUB_TOKEN for release tag workflow by @eggshell in #484

New Contributors

• @mdeyell-valid-mind made their first contribution in #479

Full Changelog: v2.11.3...v2.12.1

v2.11.3

What's Changed

• notebooks: Added updated copyright footer by @validbeck in #466
• Create and add scorer store in the list_tests interface by @AnilSorathiya in #465
• notebook templates & scripts: Append copyright & copyright verification/update by @validbeck in #467
• Update scorer metics path from scorer to scorers by @AnilSorathiya in #468
• [SC-14192] Pin aiohttp[speedups] to avoid backports.zstd dependency by @juanmleng in #470
• Fix langgraph version in the agentic demo notebook by @AnilSorathiya in #471
• [SC-14222] Incorrect plotly figures displayed by @juanmleng in #473
• notebook: Document an agentic AI system by @validbeck in #472
• notebooks & notebook templating: Removed references to selecting use cases & added missing templates by @validbeck in #474
• notebooks: Reorganization in preparation for editing by @validbeck in #475
• notebooks: Quick update for the "Learn more" section by @validbeck in #477
• feat: release from tags instead of branch promotion by @eggshell in #446

Full Changelog: v2.11.1...v2.11.3

v2.11.1

What's Changed

• [SC-XXX] Expose log_test_result function by @nibalizer in #461
• Fix integration tests by @johnwalz97 in #464
• Add agentic metrics from deepeval as scorers by @AnilSorathiya in #460

Full Changelog: v2.11.0...v2.11.1

v2.11.0

What's Changed

• Enable release notes checks to post comment with instructions by @nrichers in #432
• Update release_notes_check.py by @nrichers in #431
• Update API host URL to prod in reference by @validbeck in #433
• Manually refresh Quarto docs & revert workflow experiment by @nrichers in #435
• Cutting an RC for Dependency Work by @johnwalz97 in #414
• Hotfix/john/dont require metadata decorators by @johnwalz97 in #437
• Remove .qmd files from the .gitignore and regenerate Python API source by @nrichers in #439
• Remove sentry use from lib by @AnilSorathiya in #438
• [SC 12739] Resolve the Warning from agent_dataset.py by @AnilSorathiya in #436
• [SC 12254] Add new deepeval tests in lib by @AnilSorathiya in #426
• notebooks & notebook templating: Updates to model registration UX by @validbeck in #442
• Nrichers/sc 12743/fix quarto docs workflow issue by @nrichers in #440
• prod -> main: pull hotfixes back into main by @johnwalz97 in #444
• Hotfix: Use correct prod version and repush the fix for tags and tasks by @johnwalz97 in #445
• [SC 12707] Add G-eval test in lib by @AnilSorathiya in #434
• [SC-13093] Fix plotly express figures not displaying in UI by @juanmleng in #447
• [SC-13147] Add disk cleanup step to validmind-library CI by @juanmleng in #449
• [SC 13105] Minor correction to dataset test cases and tools in the agentic AI use case demo by @AnilSorathiya in #448
• [SC-13270] Include both figures and tables in test result descriptions by @juanmleng in #450
• [SC-13271] Pass test parameters to LLM test result description generation by @juanmleng in #451
• [SC-13352] Small fixes to TrainingTestDegradation test by @juanmleng in #452
• notebooks: Findings renamed to artifacts by @validbeck in #453
• Provide fallback if tiktoken cannot be imported by @cachafla in #456
• [SC-13583] Fix CumulativePredictionProbabilities test docstring by @juanmleng in #457
• Fix warnings on some tests by @cachafla in #459
• Pure HTML widgets by @cachafla in #458
• Prod 2.10.1 to main by @cachafla in #462
• Merge Main into Prod (prod-2.11.0) by @github-actions[bot] in #463

Full Changelog: v2.9.5...v2.11.0

v2.10.0

What's Changed

• sc-11551: Improve Library dependencies by @johnwalz97 in #404
• SC-2886: PII Detection for Uploaded Test Results by @johnwalz97 in #410
• notebook: Enable PII detection in tests by @validbeck in #416
• Bring Hotfix To main by @johnwalz97 in #419
• notebooks: Quick update for finding types by @validbeck in #420
• John/make-torch-import-optional by @johnwalz97 in #423
• Support for the Deepeval dataset (LLMTestCase) for LLM tests by @AnilSorathiya in #401
• [SC-12193] Enable custom test result description output structure by @juanmleng in #425
• [SC-12472] Small edits to custom_test_results_description notebook by @juanmleng in #427
• [SC 11945] Demo notebook for agent testing documentation by @AnilSorathiya in #422
• fix: trust env for proxies in asyncio by @nibalizer in #429
• Enable release notes checks to post comment with instructions by @nrichers in #432
• Update release_notes_check.py by @nrichers in #431
• Update API host URL to prod in reference by @validbeck in #433
• Manually refresh Quarto docs & revert workflow experiment by @nrichers in #435
• Cutting an RC for Dependency Work by @johnwalz97 in #414
• Remove .qmd files from the .gitignore and regenerate Python API source by @nrichers in #439
• Remove sentry use from lib by @AnilSorathiya in #438
• [SC 12739] Resolve the Warning from agent_dataset.py by @AnilSorathiya in #436
• [SC 12254] Add new deepeval tests in lib by @AnilSorathiya in #426

Full Changelog: v2.9.1...v2.10.0

v2.9.3

What's Changed

• Merge Main into Prod (prod-2.8.28) by @github-actions[bot] in #397
• Merge Main into Prod (prod-2.8.29) by @github-actions[bot] in #399
• Merge Main into Prod (prod-2.9.1) by @github-actions[bot] in #413
• sc-11551: Improve Library dependencies by @johnwalz97 in #404
• SC-2886: PII Detection for Uploaded Test Results by @johnwalz97 in #410
• notebook: Enable PII detection in tests by @validbeck in #416
• Hotfix: Remove LLM is_configured() Check from Test Description Generation by @johnwalz97 in #418
• Bring Hotfix To main by @johnwalz97 in #419
• notebooks: Quick update for finding types by @validbeck in #420
• Nibz/manual prod ii by @nibalizer in #421

Full Changelog: v2.9.1...v2.9.3

v2.10.0-rc.1

ValidMind 2.10.0 Release Candidate 1

🚀 Key Updates

• Python 3.12 Support: Added compatibility with Python 3.12
• Lighter Core: Streamlined core dependencies for faster installation
• Graceful Dependencies: Better handling of missing optional packages
• Enhanced Testing: Added tox framework for dependency validation

�� Available Extras

• all, huggingface, llm, nlp, pytorch, stats, xgboost, explainability, credit_risk, datasets

---

This is a Release Candidate. Test thoroughly before final release.

v2.9.1

What's Changed

External

• Update load_datasets_predictions notebook by @AnilSorathiya in #402
• Add generalize plots and statistical tests by @AnilSorathiya in #395
• [SC-11665] Remove scorecard with GLM notebook by @juanmleng in #408
  #411
• Support new interface assign_scores to VM dataset by @AnilSorathiya in #407
• new notebook: Document an Excel-based application scorecard model by @LoiAnsah in #400

Internal

• notebooks: Edits for document types UI update by @validbeck in #396
• Edit how_to notebooks: Explore tests by @LoiAnsah in #391
• Remove Use of /genai Route by @johnwalz97 in #405
• Add .cursorrules by @cachafla in #406
• Fix kaleido version by @cachafla in #409
• Let TableOutputHandler take precedence by @cachafla in #412

Full Changelog: v2.8.29...v2.9.1

v2.8.29

v2.8.29 is created to sync pypi and git. Below is the diff from v2.8.26

What's Changed

• Merge Main into Prod (prod-2.8.10) by @github-actions[bot] in #321
• Merge Main into Prod (prod-2.8.12) by @github-actions[bot] in #332
• Release to Prod [2.8.20] by @johnwalz97 in #353
• Merge Main into Prod (prod-2.8.22) by @github-actions[bot] in #362
• Merge Main into Prod (prod-2.8.26) by @github-actions[bot] in #388
• 2.8.27 version sync by @johnwalz97 in #389
• Merge Main into Prod (prod-2.8.27) by @github-actions[bot] in #390
• [SC-10335] Repo Sync repo-sync/7325 (2025-06-27) by @eggshell in #387
• [SC-11248] Hotfix: Test results not being logged to API when no tables present by @juanmleng in #392
• Merge Prod hotfixes into Main (prod-2.8.28-to-main) by @github-actions[bot] in #393
• Edit how_to notebooks: Explore test suites by @LoiAnsah in #385
• Juan/sc 10379/list return types in the list tests output by @johnwalz97 in #383
• Add Support for LLM Agentic Model in VM Library by @AnilSorathiya in #384
• Extend the predict_fn in the init_model to support multiple output columns by @AnilSorathiya in #394

Full Changelog: v2.8.26...v2.8.28

v2.8.28

What's Changed

• Merge Main into Prod (prod-2.8.10) by @github-actions[bot] in #321
• Merge Main into Prod (prod-2.8.12) by @github-actions[bot] in #332
• Release to Prod [2.8.20] by @johnwalz97 in #353
• Merge Main into Prod (prod-2.8.22) by @github-actions[bot] in #362
• Merge Main into Prod (prod-2.8.26) by @github-actions[bot] in #388
• 2.8.27 version sync by @johnwalz97 in #389
• Merge Main into Prod (prod-2.8.27) by @github-actions[bot] in #390
• [SC-10335] Repo Sync repo-sync/7325 (2025-06-27) by @eggshell in #387
• [SC-11248] Hotfix: Test results not being logged to API when no tables present by @juanmleng in #392
• Merge Prod hotfixes into Main (prod-2.8.28-to-main) by @github-actions[bot] in #393
• Edit how_to notebooks: Explore test suites by @LoiAnsah in #385
• Juan/sc 10379/list return types in the list tests output by @johnwalz97 in #383
• Add Support for LLM Agentic Model in VM Library by @AnilSorathiya in #384
• Extend the predict_fn in the init_model to support multiple output columns by @AnilSorathiya in #394

Full Changelog: v2.8.26...v2.8.28