diff --git a/backend/apps/data_training/curd/data_training.py b/backend/apps/data_training/curd/data_training.py index c99b02ee..78c533b8 100644 --- a/backend/apps/data_training/curd/data_training.py +++ b/backend/apps/data_training/curd/data_training.py @@ -198,7 +198,7 @@ FROM ( 1 - (embedding <=> :embedding_array) ) AS similarity FROM data_training AS child ) TEMP -WHERE similarity > {settings.EMBEDDING_SIMILARITY} and oid = :oid and datasource = :datasource +WHERE similarity > {settings.EMBEDDING_DATA_TRAINING_SIMILARITY} and oid = :oid and datasource = :datasource ORDER BY similarity DESC LIMIT {settings.EMBEDDING_DATA_TRAINING_TOP_COUNT} """ diff --git a/backend/apps/terminology/curd/terminology.py b/backend/apps/terminology/curd/terminology.py index 6e9bd414..7d0ef1c8 100644 --- a/backend/apps/terminology/curd/terminology.py +++ b/backend/apps/terminology/curd/terminology.py @@ -300,7 +300,7 @@ FROM ( 1 - (embedding <=> :embedding_array) ) AS similarity FROM terminology AS child ) TEMP -WHERE similarity > {settings.EMBEDDING_SIMILARITY} and oid = :oid +WHERE similarity > {settings.EMBEDDING_TERMINOLOGY_SIMILARITY} and oid = :oid ORDER BY similarity DESC LIMIT {settings.EMBEDDING_TERMINOLOGY_TOP_COUNT} """ diff --git a/backend/common/core/config.py b/backend/common/core/config.py index e5311307..962a8a84 100644 --- a/backend/common/core/config.py +++ b/backend/common/core/config.py @@ -89,7 +89,9 @@ class Settings(BaseSettings): LOCAL_MODEL_PATH: str = '/opt/sqlbot/models' DEFAULT_EMBEDDING_MODEL: str = 'shibing624/text2vec-base-chinese' EMBEDDING_ENABLED: bool = True - EMBEDDING_SIMILARITY: float = 0.4 + EMBEDDING_DEFAULT_SIMILARITY: float = 0.4 + EMBEDDING_TERMINOLOGY_SIMILARITY: float = EMBEDDING_DEFAULT_SIMILARITY + EMBEDDING_DATA_TRAINING_SIMILARITY: float = EMBEDDING_DEFAULT_SIMILARITY EMBEDDING_DEFAULT_TOP_COUNT: int = 5 EMBEDDING_TERMINOLOGY_TOP_COUNT: int = EMBEDDING_DEFAULT_TOP_COUNT EMBEDDING_DATA_TRAINING_TOP_COUNT: int = EMBEDDING_DEFAULT_TOP_COUNT