v0.2.11

What's Changed

• add support for gpt-4o-mini by @vkehfdl1 in #581
• Impute relevance score when there is no score at the hybrid fusion. by @vkehfdl1 in #580
• dump version v0.2.11 by @vkehfdl1 in #582

Full Changelog: v0.2.10...v0.2.11

v0.2.10

What's Changed

• add example yaml code at G-eval docs to use specific G-eval metrics by @bwook00 in #572
• Refactoring hybrid retrieval with more advanced optimizations by @vkehfdl1 in #574
• dump version v0.2.10 by @vkehfdl1 in #575

Full Changelog: V0.2.9...v0.2.10

V0.2.9

What's Changed

• Feature/#550 by @vkehfdl1 in #551
• make sacrebleu parameter available in upper stream by @Eastsidegunn in #549
• update docs 'batch' parameter by @vkehfdl1 in #552
• delete sacrebleu extra dependency for Korean and Japanese by @Eastsidegunn in #554
• Add window_replacement module at prompt_maker node by @bwook00 in #559
• resolve error that make_single_content_qa generate duplicate qid by @vkehfdl1 in #562
• Auto-save dataset file to the I/O at make_single_content_qa by @vkehfdl1 in #565
• refactor the documentation about using local model by @vkehfdl1 in #567
• dump version 0.2.9 by @vkehfdl1 in #568

Full Changelog: v0.2.8...V0.2.9

v0.2.8

What's Changed

• add emoji at README by @bwook00 in #534
• Fix typo at query decompose and edit documentation of it by @vkehfdl1 in #537
• delete whitespace and empty string at expanded queries by @vkehfdl1 in #541
• Resolve error when using strategy at query expansion by @vkehfdl1 in #543
• Add decorator that converts np.array and pd.Series to python list by @vkehfdl1 in #542
• Resolve error at FlagEmbedding Rerankers by @vkehfdl1 in #546
• dump version 0.2.8 by @vkehfdl1 in #547

Full Changelog: v0.2.7...v0.2.8

v0.2.7

What's Changed

• change the embedding model type list to the latest version by @vkehfdl1 in #523
• Delete python 3.8 tag and add 3.11, 3.12 tag at pyproject.toml by @bwook00 in #525
• Fix vllm to mockllm at add more llm models docs by @bwook00 in #524
• Openai token limit feature at passage augmenter and passage filter. by @vkehfdl1 in #526
• temporal not using numpy 2.0.0 by @vkehfdl1 in #527
• Delete ray dependency in AutoRAG package by @vkehfdl1 in #528
• load YAML file as utf-8 codec by @vkehfdl1 in #531
• dump version v0.2.7 by @vkehfdl1 in #532

Full Changelog: v0.2.6...v0.2.7

v0.2.6

What's Changed

• dump version v0.2.6 by @vkehfdl1 in #509
• window read_parquet error by @Eastsidegunn in #513
• Add meta tag at each docs by @vkehfdl1 in #514
• change package name to 'evaluation' by @vkehfdl1 in #516

New Contributors

• @gnekt made their first contribution in #492

Full Changelog: v0.2.5...v0.2.6

v0.2.5

What's Changed

• add reset_index at cast_qa_dataset and cast_corpus_dataset by @bwook00 in #475
• Adjust embedding batch size at YAML file by @vkehfdl1 in #481
• Directly watch each module result at dashboard by @vkehfdl1 in #480
• release 0.2.4 by @vkehfdl1 in #482
• Resolve error that tokenized tokens and logprob tokens length sometimes mismatched by @vkehfdl1 in #483
• add score fusion additional explanation at dbsf and rsf docs by @bwook00 in #485
• dump version 0.2.5 by @vkehfdl1 in #486

Full Changelog: v0.2.3...v0.2.5

v0.2.3

What's Changed

• add funding.yml by @vkehfdl1 in #471
• HotFix at vllm.py by @vkehfdl1 in #473

Full Changelog: v0.2.2...v0.2.3

v0.2.2

What's Changed

• reset index at split dataframe by @vkehfdl1 in #469

Full Changelog: v0.2.1...v0.2.2

v0.2.1

What's Changed

• Implement longllmlingua compressor module by @bwook00 in #459
• Use generator module at hyde and decompose instead of LLMPredictorType from LlamaIndex by @bwook00 in #464
• Add normalize mean strategy by @bwook00 in #465
• implement multi_query_expansion module by @bwook00 in #461
• dump version v0.2.1 by @vkehfdl1 in #466

Full Changelog: v0.2.0...v0.2.1