github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-07-07 21:50:25 +00:00
Code Issues Packages Projects Releases Wiki Activity

Minor update to ensemble retriever to handle a mix of Documents or str (#14552)

Browse Source
This commit is contained in:
Lance Martin 2023-12-12 15:16:49 -08:00 committed by GitHub
parent ca7da8f7ef
commit 282362382c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 14 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
libs/langchain/langchain/retrievers/ensemble.py
View File

@ -103,6 +103,13 @@ class EnsembleRetriever(BaseRetriever):
for i, retriever in enumerate(self.retrievers) for i, retriever in enumerate(self.retrievers)
] ]
# Enforce that retrieved docs are Documents for each list in retriever_docs
for i in range(len(retriever_docs)):
retriever_docs[i] = [
Document(page_content=doc) if not isinstance(doc, Document) else doc
for doc in retriever_docs[i]
]
# apply rank fusion # apply rank fusion
fused_documents = self.weighted_reciprocal_rank(retriever_docs) fused_documents = self.weighted_reciprocal_rank(retriever_docs)
@ -130,6 +137,13 @@ class EnsembleRetriever(BaseRetriever):
for i, retriever in enumerate(self.retrievers) for i, retriever in enumerate(self.retrievers)
] ]
# Enforce that retrieved docs are Documents for each list in retriever_docs
for i in range(len(retriever_docs)):
retriever_docs[i] = [
Document(page_content=doc) if not isinstance(doc, Document) else doc
for doc in retriever_docs[i]
]
# apply rank fusion # apply rank fusion
fused_documents = self.weighted_reciprocal_rank(retriever_docs) fused_documents = self.weighted_reciprocal_rank(retriever_docs)