v0.27.0
PromptingTools v0.27.0
Added
- Added a keyword-based search similarity to RAGTools to serve both for baseline evaluation and for advanced performance (by having a hybrid index with both embeddings and BM25). See
?RT.KeywordsIndexer
and?RT.BM25Similarity
for more information, to build usebuild_index(KeywordsIndexer(), texts)
or convert an existing embeddings-based indexChunkKeywordsIndex(index)
.
Updated
- For naming consistency,
ChunkIndex
in RAGTools has been renamed toChunkEmbeddingsIndex
(with an aliasChunkIndex
for backwards compatibility). There are now two main index types:ChunkEmbeddingsIndex
andChunkKeywordsIndex
(=BM25), which can be combined into aMultiIndex
to serve as a hybrid index.
Commits
Merged pull requests:
- Add BM25 Index (#157) (@svilupp)
- Implement E2E Hybrid retrieval (BM25 + embeddings) (#159) (@svilupp)
Closed issues:
- [FR] RAG: Add HybridChunkIndex with BM25 support (#119)