v0.2.0

What's Changed

• Add MockLLM to resolve restart evaluate bug by @bwook00 in #453
• Add 'rank' strategy as selecting best module strategy by @vkehfdl1 in #451
• Delete ragas context precision & Enable dict input of retrieval metrics by @vkehfdl1 in #455
• Can't use passage compressor when there is no retrieval gt values in QA datset by @vkehfdl1 in #452
• delete pull_request_target on test.yml by @vkehfdl1 in #457
• Feat/dashboard by @hongsw in #432
• dump version 0.2.0 by @vkehfdl1 in #458

New Contributors

• @hongsw made their first contribution in #432

Full Changelog: v0.1.12...v0.2.0

v0.1.12

What's Changed

• [Hotfix] delete assert top_k at passage_augmenter/base by @bwook00 in #435
• [Hotfix] add **kwargs at pass_passage_augmenter by @bwook00 in #437
• Add gpt-4o support at openai_llm by @vkehfdl1 in #441
• [Hotfix] Respond when the input 'scores_list' is np.ndarray at threshold_cutoff_pure by @bwook00 in #442
• Resolve Korean encoding unicode error by @vkehfdl1 in #444
• dump version 0.1.12 by @vkehfdl1 in #446

Full Changelog: v0.1.11...v0.1.12

v0.1.11

What's Changed

• Fix error at UPR and Colbert Reranker on GPU by @vkehfdl1 in #424
• [HotFix] empty cache and return only float value at vllm.py by @vkehfdl1 in #426
• Refactor prompt maker run.py for preventing long processing time and OOM by @vkehfdl1 in #427
• Add Ragas context precision metric by @vkehfdl1 in #431
• Add threshold cutoff module by @bwook00 in #428
• Add percentile cutoff module by @bwook00 in #429
• dump version v0.1.11 by @vkehfdl1 in #433

Full Changelog: v0.1.10...v0.1.11

v0.1.10

What's Changed

• Add various option for tokenize BM25 by @vkehfdl1 in #409
• Mock cohere and JinaAI test codes by @vkehfdl1 in #420
• Add Metrics documentation by @bwook00 in #419
• dump version 0.1.10 by @vkehfdl1 in #423

Full Changelog: v0.1.9...v0.1.10

v0.1.9

What's Changed

• Add retrieval_ndcg_metric by @bwook00 in #386
• Add mrr retrieval metric by @bwook00 in #390
• Add MAP (Mean Average Precision) retrieval metric by @bwook00 in #401
• Make new openai_llm generator module by @vkehfdl1 in #404
• Parallel Processing of Ko Reranker by @bwook00 in #403
• Fix mock error by @vkehfdl1 in #406
• dump version 0.1.9 by @vkehfdl1 in #408

Full Changelog: v0.1.8...v0.1.9

v0.1.8

What's Changed

• add library setuptools-scm and build by @vkehfdl1 in #387
• [Hotfix] Add pass augmenter at support by @bwook00 in #389
• Show logging at module level by @vkehfdl1 in #395
• Resolve minor error at colbert and long_context_reorder by @vkehfdl1 in #397
• Resolve summary.csv error at recency filter threshold paramter + dump version 0.1.8 by @vkehfdl1 in #392

Full Changelog: v0.1.7...v0.1.8

v0.1.7

What's Changed

• Parallel processing at UPR reranker by @bwook00 in #334
• dump version 0.1.7 by @vkehfdl1 in #385

Full Changelog: v0.1.6...v0.1.7

v0.1.6

What's Changed

• test for large embeddings and version control for latest chromaDB by @vkehfdl1 in #375
• Fix retrieval and retrieval token metrics to use unaswerable by @bwook00 in #371
• add pypi deploy CD by @vkehfdl1 in #374
• Parallel Processing of Colbert Reranker by @vkehfdl1 in #378
• Minor fix at publish.yml by @vkehfdl1 in #381
• Resolve error Colbert Reranker on CUDA environment by @vkehfdl1 in #379
• Add test code if QA data was created from a corpus by @bwook00 in #382
• Fix error at publish.yml by @vkehfdl1 in #383

Full Changelog: v0.1.5...v0.1.6

v0.1.5

What's Changed

• Create 'Passage Augmenter' node and its first module 'prev_next_augmenter' by @bwook00 in #347
• [Hotfix] resolve Tart reranker error by @bwook00 in #363
• [Hotfix] resolve colbert, flag embedding reranker batch error by @bwook00 in #366
• Add pass passage augmenter module by @bwook00 in #352
• [Hotfix] recency filter change threshold parameter type str to datetime by @bwook00 in #369

Full Changelog: v0.1.4...v0.1.5

v0.1.4

What's Changed

• �Add Recency filter module by @bwook00 in #312
• Resolve error when put larger than corpus_df at 'make_single_content_qa' content_size by @vkehfdl1 in #324
• Parallel processing at sentence transformer reranker by @vkehfdl1 in #325
• Add refine module by @bwook00 in #323
• fix error by sorting in index in 'generate_qa_llama_index_by_ratio by @vkehfdl1 in #344
• Mock openai chat completion calls for preventing tons of openai usage. by @vkehfdl1 in #349
• generate sitemap.xml automatically for SEO by @vkehfdl1 in #351
• Parallel processing at colbert reranker by @bwook00 in #326
• parallel processing of g_eval by @vkehfdl1 in #350
• Parallel processing at �Tart reranker by @bwook00 in #329
• Parallel processing at MonoT5 reranker by @bwook00 in #330
• trigger test at pull_request_target for access github repo secret fro… by @vkehfdl1 in #354
• Created Contributors file by @gauravsaini02 in #348
• Parallel processing at Flag embedding reranker and Flag embedding llm reranker by @bwook00 in #328

New Contributors

• @gauravsaini02 made their first contribution in #348

Full Changelog: v0.1.3...v0.1.4