Unitxt 1.17.2

What's Changed

• Feature/add global mmlu cards by @eliyahabba in #1561
• Add quality dataset by @eliyahabba in #1563
• Add CollateInstanceByField operator to group data by specific field by @sarathsgvr in #1546
• Fix prompts table benchmark by @ShirApp in #1565
• Create new IntersectCorrespondingFields operator by @pklpriv in #1531
• Add granite documents format by @elronbandel in #1566
• Revisit huggingface cache policy by @elronbandel in #1564
• Add global mmlu lite sensitivity cards by @eliyahabba in #1568
• Update version to 1.17.2 by @elronbandel in #1569

Full Changelog: 1.17.1...1.17.2

Unitxt 1.17.1

What's Changed

Non backward compatible change

• Renamed criterias in LLM-as-a-Judge metrics to criteria - Breaking change by @tejaswini in #1545

New features

• Add Replicate inference support by @elronbandel in #1544
• Add text2sql tasks by @perlitz in #1414
• Add deduplicate operator by @elronbandel in #1549

New Assets

• Add more granite llm as judge artifacts by @martinscooper in #1516
• Add mtrag benchmark by @elronbandel in #1548

Documentation

• End of year summary blog post by @elronbandel in #1530
• Update notification banner styles and add 2024 summary blog link by @elronbandel in #1538
• Updated documentation and examples of LLM-as-Judge by @tejaswini in #1532
• Eval assist documentation by @tejaswini in #1537

Bug Fixes

• Fix Australian legal qa dataset by @elronbandel in #1542
• Set use 1 shot for wikitq in tables_benchmark by @yifanmai in #1541
• Bugfix: indexed row major serialization fails with None cell values by @yifanmai in #1540
• Solve issue of expired token in Unitxt Assistant by @eladven in #1543
• add a filter to wikitq by @ShirApp in #1547
• Fix the authentication problem by @eladven in #1550
• Attach assitant answers to their origins with url link by @elronbandel in #1528
• Update end of year summary blog by @elronbandel in #1552
• Add data classification policy to CrossProviderInferenceEngine initialization based on selected model by @elronbandel in #1539
• Fix recently broken rag metrics by @elronbandel in #1554
• Finqa hash to top by @elronbandel in #1555
• Refactor safety metric to be faster and updated by @elronbandel in #1484
• Improve assistant by @elronbandel in #1556

New Contributors

• @tejaswini made their first contribution in #1532

Full Changelog: 1.17.0...1.17.1

Unitxt 1.17.0 - New LLM as Judges!

Importnat Changes

write abstract for update talk about unitxt covering the following topics:

• Criteria based LLM as Judges - Improved class of llm as judges with customizable judging criteria (read more)
• Unitxt assistant - A textual assistant expert in unitxt to help developers (read more)
• New benchmarks: Tables, Vision - Benchmarks for table understanding and image understanding compiled by the community and collaborators (read more)
• Support for all major inference providers - Inference for evaluation or llm as judges can be channel to any inference provider such as: azure, aws and watsonx (read more)

Detailed Changes

• Fix typing notation for python 3.8 by @elronbandel in #1453
• Instance_metric and apply_metric keep only one instance at a time in mem, at the expense of repeated passes over input stream (2 times for instance_metric, #metrics for apply_metric) by @dafnapension in #1448
• simplify class parameter listing on web page by @dafnapension in #1454
• Bring code coverage tests back to life by @elronbandel in #1455
• Fix coverage tests by @elronbandel in #1456
• make demos_pool a local var rather than a separate stream by @dafnapension in #1436
• Adding upper case and last non empty line processor by @antonpibm in #1458
• performance by bluebench by @dafnapension in #1457
• Add UNITXT_MOCK_INFERENCE_MODE environment variable to performance workflow by @elronbandel in #1461
• remove redundant lines from performance.yml by @dafnapension in #1462
• Benjams/add bioasq miniwiki datasets by @BenjSz in #1460
• Add SocialIQA dataset by @elronbandel in #1468
• Add parallelization to RITS inference by @arielge in #1441
• Fix the type handeling for tasks to support string types by @elronbandel in #1470
• Update version to 1.16.1 by @elronbandel in #1472
• extend choices arrangement functionality with ReorderableMultipleChoi… by @eliyahabba in #1464
• Add GPQA dataset by @elronbandel in #1474
• Add simple QA dataset by @elronbandel in #1475
• Add LongBench V2 dataset by @elronbandel in #1476
• Adding typed recipe test by @antonpibm in #1473
• Add place_correct_choice_position to set the correct choice index and… by @eliyahabba in #1481
• Add MapReduceMetric a new base class to integrate all metrics into by @elronbandel in #1459
• Add multi document support and FRAMES benchmark by @elronbandel in #1477
• Update version to 1.16.2 by @elronbandel in #1483
• Add Azure support and expand OpenAI model options in inference engine by @elronbandel in #1485
• Benjams/fix bioasq card by @BenjSz in #1486
• add separator to csv loader by @BenjSz in #1488
• Fix bug in metrics loading in tasks by @elronbandel in #1487
• Update version to 1.16.3 by @elronbandel in #1489
• Fix bootstrap condition to handle cases with insufficient instances by @elronbandel in #1490
• Update version to 1.16.4 by @elronbandel in #1491
• Simplify artifact link [Non Backward Compatible!] by @elronbandel in #1494
• Added NER example by @yoavkatz in #1492
• Add example for evaluating tables as images using Unitxt APIs by @elronbandel in #1495
• Mm updates by @alfassy in #1465
• Fix wrong saving of artifact initial dict by @elronbandel in #1499
• Accelerate and improve RAG Metrics by @elronbandel in #1497
• Make clinc preparation faster by @elronbandel in #1501
• Fix templates lists in vision cards by @elronbandel in #1500
• Add vision benchmark example by @elronbandel in #1502
• Update vis bench by @elronbandel in #1505
• Add Balance operator by @elronbandel in #1507
• Fix for demos_pool with images. by @elronbandel in #1509
• Remove new balance operator and use existing implementation by @elronbandel in #1510
• Fixes and adjustment in rag metrics and related inference engines by @lilacheden in #1466
• Tables bench by @ShirApp in #1506
• Keep metadata over main unitxt stages by @eladven in #1512
• Fix: Improved handling of place_correct_choice_position for flexibl… by @eliyahabba in #1511
• Fixes in LLMJudge by @lilacheden in #1498
• Verify metrics prediction_type without loading metric by @elronbandel in #1519
• Add Unitxt Assistant beta by @elronbandel in #1513
• Ensure fusion do not call streams before use by @elronbandel in #1518
• Minor llm as judge fix/changes by @martinscooper in #1467
• Fix: Selected option for supporting negative indexes in place_correct… by @eliyahabba in #1522
• Refactor rag metrics and judges by @lilacheden in #1515
• Add Llama 3.1 on Vertex AI to CrossProviderInferenceEngine by @yifanmai in #1525
• fix external_rag example by @lilacheden in #1526
• Add search to assistant for much faster response by @elronbandel in #1524
• fixed division by 0 in compare performance results by @dafnapension in #1523
• Add two criteria based direct llm judges by @lilacheden in #1527
• Update version to 1.17.0 by @elronbandel in #1535

New Contributors

• @eliyahabba made their first contribution in #1464

Full Changelog: 1.16.0...1.17.0

Unitxt 1.16.4

What's Changed

• Fix bootstrap condition to handle cases with insufficient instances by @elronbandel in #1490

Unitxt 1.16.3

What's Changed

• Add Azure support and expand OpenAI model options in inference engine by @elronbandel in #1485
• Benjams/fix bioasq card by @BenjSz in #1486
• add separator to csv loader by @BenjSz in #1488
• Fix bug in metrics loading in tasks by @elronbandel in #1487

Unitxt 1.16.2

What's Changed

• extend choices arrangement functionality with ReorderableMultipleChoi… by @eliyahabba in #1464
• Add GPQA dataset by @elronbandel in #1474
• Add simple QA dataset by @elronbandel in #1475
• Add LongBench V2 dataset by @elronbandel in #1476
• Adding typed recipe test by @antonpibm in #1473
• Add place_correct_choice_position to set the correct choice index and… by @eliyahabba in #1481
• Add MapReduceMetric a new base class to integrate all metrics into by @elronbandel in #1459
• Add multi document support and FRAMES benchmark by @elronbandel in #1477

New Contributors

• @eliyahabba made their first contribution in #1464

Unitxt 1.16.1

• Fix typing notation for python 3.8 by @elronbandel in #1453
• Instance_metric and apply_metric keep only one instance at a time in mem, at the expense of repeated passes over input stream (2 times for instance_metric, #metrics for apply_metric) by @dafnapension in #1448
• simplify class parameter listing on web page by @dafnapension in #1454
• Bring code coverage tests back to life by @elronbandel in #1455
• Fix coverage tests by @elronbandel in #1456
• make demos_pool a local var rather than a separate stream by @dafnapension in #1436
• Adding upper case and last non empty line processor by @antonpibm in #1458
• performance by bluebench by @dafnapension in #1457
• Add UNITXT_MOCK_INFERENCE_MODE environment variable to performance workflow by @elronbandel in #1461
• remove redundant lines from performance.yml by @dafnapension in #1462
• Benjams/add bioasq miniwiki datasets by @BenjSz in #1460
• Add SocialIQA dataset by @elronbandel in #1468
• Add parallelization to RITS inference by @arielge in #1441
• Fix the type handeling for tasks to support string types by @elronbandel in #1470

1.16.0

Main Changes

What's Changed

Usability

• Add error message when saving artifacts that got changed by @elronbandel in #1417
• A simple way to create and evaluate given a 'task' in the catalog and python data structure by @yoavkatz in #1413
• Evaluation results class for easier access to results by @elronbandel in #1326
• Eval Assist integration by @martinscooper in #1409

Documentation

• Update to new logo by @elronbandel in #1427
• Indentation within docstrings to improve appearance on web pages, on the way - eliminating two red lines from "make docs-server" by @dafnapension in #1429
• Add catalog search with tags filtering by @elronbandel in #1430
• Update catalog search engine by @elronbandel in #1431
• Add custom titles to catalog items by @elronbandel in #1432
• Change card to dataset in the catalog search tags by @elronbandel in #1433
• Updated documentation to show use of installed version and chat api by @yoavkatz in #1435
• Fix documentation for task registration example by @Etelis in #1443

Bug Fixes

• fix mistral format used in llmaj (when not using chat_api) by @lilacheden in #1425
• Fix LMMSEval Inference Engine to work with chat api and fix examples by @elronbandel in #1440
• metadata is set only once in recipe by @dafnapension in #1437
• verify only fresh artifacts are fetched by @dafnapension in #1444
• add data_classification_policy_to_clapnq by @BenjSz in #1451

CI/CD

• eliminate exceeding line_limit errors, and many red lines from "make docs-server" by @dafnapension in #1434

New Contributors

• @Etelis made their first contribution in #1443

Full Changelog: 1.15.10...1.16.0

1.15.10

What's Changed

• Fix arenahard bluebench template by @perlitz in #1405
• Fixed formal types of infer() and also added runtime check by @yoavkatz in #1406
• not using "score" as metric main_score by @lilacheden in #1407
• Fix model strings for Llama 3 on Together AI by @yifanmai in #1411
• Adjust binary llmaj to new engines and add rits support by @lilacheden in #1408
• Granite Guardian RAG metrics by @arielge in #1393
• Solved many red lines in 'make docs-server' by @dafnapension in #1418
• Fix artifact dict assignment bug by @elronbandel in #1419
• Remove top level imports from guerdian metric (as it adds dependencis to unitxt) by @elronbandel in #1420
• Make types compatible with python 3.8 by @elronbandel in #1423
• Benjams/loaders fix separator by @BenjSz in #1424
• Update version to 1.15.10 by @elronbandel in #1426

Full Changelog: 1.15.9...1.15.10

Unitxt 1.15.9

Main changes

• Artifacts in the catalog can now be links to other artifacts and can also be marked deprecated.

What's Changed

• artifact link by @dafnapension in #1363
• Add processors also as operators by @antonpibm in #1397
• added 'add_link_to_catalog' for easily adding artifact_links with/without deprecation msg by @dafnapension in #1398
• Safety updates by @bnayahu in #1391
• Reduce error message clutter by @yoavkatz in #1401
• Update version to 1.15.9 by @yoavkatz in #1404

Full Changelog: 1.15.8...1.15.9