feat(ChatKnowledge): Support Financial Report Analysis (#1702)

Co-authored-by: hzh97 <2976151305@qq.com>
Co-authored-by: Fangyin Cheng <staneyffer@gmail.com>
Co-authored-by: licunxing <864255598@qq.com>

dbgpt/rag/assembler/base.py

@@ -52,7 +52,7 @@ class BaseAssembler(ABC):
         with root_tracer.start_span("BaseAssembler.load_knowledge", metadata=metadata):
             self.load_knowledge(self._knowledge)
 
-    def load_knowledge(self, knowledge: Optional[Knowledge] = None) -> None:
+    def load_knowledge(self, knowledge: Knowledge) -> None:
         """Load knowledge Pipeline."""
         if not knowledge:
             raise ValueError("knowledge must be provided.")
@@ -66,14 +66,14 @@ class BaseAssembler(ABC):
         """Return a retriever."""
 
     @abstractmethod
-    def persist(self) -> List[str]:
+    def persist(self, **kwargs: Any) -> List[str]:
         """Persist chunks.
 
         Returns:
             List[str]: List of persisted chunk ids.
         """
 
-    async def apersist(self) -> List[str]:
+    async def apersist(self, **kwargs: Any) -> List[str]:
         """Persist chunks.
 
         Returns:

dbgpt/rag/assembler/bm25.py

@@ -183,7 +183,7 @@ class BM25Assembler(BaseAssembler):
             chunk_parameters=chunk_parameters,
         )
 
-    def persist(self) -> List[str]:
+    def persist(self, **kwargs) -> List[str]:
         """Persist chunks into elasticsearch.
 
         Returns:
@@ -212,7 +212,7 @@ class BM25Assembler(BaseAssembler):
         self._es_client.indices.refresh(index=self._index_name)
         return ids
 
-    async def apersist(self) -> List[str]:
+    async def apersist(self, **kwargs) -> List[str]:
         """Persist chunks into elasticsearch.
 
         Returns:

dbgpt/rag/assembler/db_schema.py

@@ -85,7 +85,7 @@ class DBSchemaAssembler(BaseAssembler):
         """Return chunk ids."""
         return self._chunks
 
-    def persist(self) -> List[str]:
+    def persist(self, **kwargs: Any) -> List[str]:
         """Persist chunks into vector store.
 
         Returns:

dbgpt/rag/assembler/embedding.py

@@ -124,7 +124,7 @@ class EmbeddingAssembler(BaseAssembler):
             retrieve_strategy,
         )
 
-    def persist(self) -> List[str]:
+    def persist(self, **kwargs) -> List[str]:
         """Persist chunks into store.
 
         Returns:
@@ -132,7 +132,7 @@ class EmbeddingAssembler(BaseAssembler):
         """
         return self._index_store.load_document(self._chunks)
 
-    async def apersist(self) -> List[str]:
+    async def apersist(self, **kwargs) -> List[str]:
         """Persist chunks into store.
 
         Returns:

dbgpt/rag/assembler/summary.py

@@ -118,7 +118,7 @@ class SummaryAssembler(BaseAssembler):
         """Generate summary."""
         return await self._extractor.aextract(self._chunks)
 
-    def persist(self) -> List[str]:
+    def persist(self, **kwargs: Any) -> List[str]:
         """Persist chunks into store."""
         raise NotImplementedError