AI/dify
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/langgenius/dify.git synced 2025-08-12 20:39:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

multi-retrival use dataset's top-k (#8416)

Browse Source
This commit is contained in:
Jyong 2024-09-14 21:48:44 +08:00 committed by GitHub
parent 65162a87b6
commit 7e611ffbf3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
api/core/rag/retrieval/dataset_retrieval.py
View File

@ -426,7 +426,7 @@ class DatasetRetrieval:
retrieval_method=retrieval_model["search_method"], retrieval_method=retrieval_model["search_method"],
dataset_id=dataset.id, dataset_id=dataset.id,
query=query, query=query,
top_k=top_k, top_k=retrieval_model.get("top_k") or 2,
score_threshold=retrieval_model.get("score_threshold", 0.0) score_threshold=retrieval_model.get("score_threshold", 0.0)
if retrieval_model["score_threshold_enabled"] if retrieval_model["score_threshold_enabled"]
else 0.0, else 0.0,

7
api/core/tools/tool/dataset_retriever/dataset_multi_retriever_tool.py
View File

@ -165,7 +165,10 @@ class DatasetMultiRetrieverTool(DatasetRetrieverBaseTool):
if dataset.indexing_technique == "economy": if dataset.indexing_technique == "economy":
# use keyword table query # use keyword table query
documents = RetrievalService.retrieve( documents = RetrievalService.retrieve(
retrieval_method="keyword_search", dataset_id=dataset.id, query=query, top_k=self.top_k retrieval_method="keyword_search",
dataset_id=dataset.id,
query=query,
top_k=retrieval_model.get("top_k") or 2,
) )
if documents: if documents:
all_documents.extend(documents) all_documents.extend(documents)
@ -176,7 +179,7 @@ class DatasetMultiRetrieverTool(DatasetRetrieverBaseTool):
retrieval_method=retrieval_model["search_method"], retrieval_method=retrieval_model["search_method"],
dataset_id=dataset.id, dataset_id=dataset.id,
query=query, query=query,
top_k=self.top_k, top_k=retrieval_model.get("top_k") or 2,
score_threshold=retrieval_model.get("score_threshold", 0.0) score_threshold=retrieval_model.get("score_threshold", 0.0)
if retrieval_model["score_threshold_enabled"] if retrieval_model["score_threshold_enabled"]
else 0.0, else 0.0,