AI/dify
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/langgenius/dify.git synced 2025-08-11 16:08:58 +08:00
Code Issues Packages Projects Releases Wiki Activity

add error msg for hit test (#4704)

Browse Source
This commit is contained in:
Jyong 2024-05-28 14:54:53 +08:00 committed by GitHub
parent e6f6a59f3b
commit 1b2d862973
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 81 additions and 64 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
api/core/rag/datasource/retrieval_service.py
View File

@ -33,6 +33,7 @@ class RetrievalService:
return []
all_documents = []
threads = []
exceptions = []
# retrieval_model source with keyword
if retrival_method == 'keyword_search':
keyword_thread = threading.Thread(target=RetrievalService.keyword_search, kwargs={
@ -40,7 +41,8 @@ class RetrievalService:
'dataset_id': dataset_id,
'query': query,
'top_k': top_k,
'all_documents': all_documents
'all_documents': all_documents,
'exceptions': exceptions,
})
threads.append(keyword_thread)
keyword_thread.start()
@ -54,7 +56,8 @@ class RetrievalService:
'score_threshold': score_threshold,
'reranking_model': reranking_model,
'all_documents': all_documents,
'retrival_method': retrival_method
'retrival_method': retrival_method,
'exceptions': exceptions,
})
threads.append(embedding_thread)
embedding_thread.start()
@ -69,7 +72,8 @@ class RetrievalService:
'score_threshold': score_threshold,
'top_k': top_k,
'reranking_model': reranking_model,
'all_documents': all_documents
'all_documents': all_documents,
'exceptions': exceptions,
})
threads.append(full_text_index_thread)
full_text_index_thread.start()
@ -77,6 +81,10 @@ class RetrievalService:
for thread in threads:
thread.join()
if exceptions:
exception_message = ';\n'.join(exceptions)
raise Exception(exception_message)
if retrival_method == 'hybrid_search':
data_post_processor = DataPostProcessor(str(dataset.tenant_id), reranking_model, False)
all_documents = data_post_processor.invoke(
@ -89,8 +97,9 @@ class RetrievalService:
@classmethod
def keyword_search(cls, flask_app: Flask, dataset_id: str, query: str,
top_k: int, all_documents: list):
top_k: int, all_documents: list, exceptions: list):
with flask_app.app_context():
try:
dataset = db.session.query(Dataset).filter(
Dataset.id == dataset_id
).first()
@ -104,12 +113,15 @@ class RetrievalService:
top_k=top_k
)
all_documents.extend(documents)
except Exception as e:
exceptions.append(str(e))
@classmethod
def embedding_search(cls, flask_app: Flask, dataset_id: str, query: str,
top_k: int, score_threshold: Optional[float], reranking_model: Optional[dict],
all_documents: list, retrival_method: str):
all_documents: list, retrival_method: str, exceptions: list):
with flask_app.app_context():
try:
dataset = db.session.query(Dataset).filter(
Dataset.id == dataset_id
).first()
@ -139,12 +151,15 @@ class RetrievalService:
))
else:
all_documents.extend(documents)
except Exception as e:
exceptions.append(str(e))
@classmethod
def full_text_index_search(cls, flask_app: Flask, dataset_id: str, query: str,
top_k: int, score_threshold: Optional[float], reranking_model: Optional[dict],
all_documents: list, retrival_method: str):
all_documents: list, retrival_method: str, exceptions: list):
with flask_app.app_context():
try:
dataset = db.session.query(Dataset).filter(
Dataset.id == dataset_id
).first()
@ -168,3 +183,5 @@ class RetrievalService:
))
else:
all_documents.extend(documents)
except Exception as e:
exceptions.append(str(e))