pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2026-02-02 08:35:08 +08:00
Code Issues Packages Projects Releases Wiki Activity

refactor: introduce common normalize method in rerank base class (#12550)

Browse Source 
### What problem does this PR solve?

introduce common normalize method in rerank base class

### Type of change

- [x] Refactoring
This commit is contained in:
Stephen Hu
2026-01-12 11:07:11 +08:00
committed by GitHub
parent ff11e3171e
commit 638c510468
1 changed files with 18 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
rag/llm/rerank_model.py
View File

@ -36,6 +36,22 @@ class Base(ABC):
def similarity(self, query: str, texts: list): def similarity(self, query: str, texts: list):
raise NotImplementedError("Please implement encode method!") raise NotImplementedError("Please implement encode method!")
@staticmethod
def _normalize_rank(rank: np.ndarray) -> np.ndarray:
"""
Normalize rank values to the range 0 to 1.
Avoids division by zero if all ranks are identical.
"""
min_rank = np.min(rank)
max_rank = np.max(rank)
if not np.isclose(min_rank, max_rank, atol=1e-3):
rank = (rank - min_rank) / (max_rank - min_rank)
else:
rank = np.zeros_like(rank)
return rank
class JinaRerank(Base): class JinaRerank(Base):
_FACTORY_NAME = "Jina" _FACTORY_NAME = "Jina"
@ -121,15 +137,7 @@ class LocalAIRerank(Base):
except Exception as _e: except Exception as _e:
log_exception(_e, res) log_exception(_e, res)
# Normalize the rank values to the range 0 to 1 rank = Base._normalize_rank(rank)
min_rank = np.min(rank)
max_rank = np.max(rank)
# Avoid division by zero if all ranks are identical
if not np.isclose(min_rank, max_rank, atol=1e-3):
rank = (rank - min_rank) / (max_rank - min_rank)
else:
rank = np.zeros_like(rank)
return rank, token_count return rank, token_count
@ -215,15 +223,7 @@ class OpenAI_APIRerank(Base):
except Exception as _e: except Exception as _e:
log_exception(_e, res) log_exception(_e, res)
# Normalize the rank values to the range 0 to 1 rank = Base._normalize_rank(rank)
min_rank = np.min(rank)
max_rank = np.max(rank)
# Avoid division by zero if all ranks are identical
if not np.isclose(min_rank, max_rank, atol=1e-3):
rank = (rank - min_rank) / (max_rank - min_rank)
else:
rank = np.zeros_like(rank)
return rank, token_count return rank, token_count