Fix: LangChain Dependences

app/rag/rag_chain.py

@@ -5,10 +5,14 @@ from langchain.chains.combine_documents import create_stuff_documents_chain
 
 
 def create_rag_chain(llm, retriever):
-    contextualize_q_system_prompt = """Given a chat history and the latest user question \
-    which might reference context in the chat history, formulate a standalone question \
-    which can be understood without the chat history. Do NOT answer the question, \
-    just reformulate it if needed and otherwise return it as is."""
+    contextualize_q_system_prompt = """
+    Given a chat history and the latest user question \
+    which might reference context in the chat history,
+    formulate a standalone question \
+    which can be understood without the chat history.
+    Do NOT answer the question, \
+    just reformulate it if needed and otherwise return it as is.
+    """
     contextualize_q_prompt = ChatPromptTemplate.from_messages(
         [
             ("system", contextualize_q_system_prompt),
@@ -21,12 +25,13 @@ def create_rag_chain(llm, retriever):
     )
 
     # ___________________Chain con el chat history_______________________-
-    qa_system_prompt = """You are an assistant for question-answering tasks.  \
+    qa_system_prompt = """
+    You are an assistant for question-answering tasks.  \
     Use the following pieces of retrieved context to answer the question. \
     If you don't know the answer, just say that you don't know. \
-    The length of the answer should be sufficient to address what is being asked, \
+    The length of the answer should be sufficient to address
+    what is being asked, \
     meaning don't limit yourself in length.\
-    
     {context}"""
     qa_prompt = ChatPromptTemplate.from_messages(
         [
@@ -37,4 +42,5 @@ def create_rag_chain(llm, retriever):
     )
     question_answer_chain = create_stuff_documents_chain(llm, qa_prompt)
 
-    return create_retrieval_chain(history_aware_retriever, question_answer_chain)
+    return create_retrieval_chain(
+        history_aware_retriever, question_answer_chain)

app/rag/retriever.py

@@ -4,7 +4,9 @@ from langchain_chroma import Chroma
 def create_retriever(embeddings, persist_directory: str):
     # Cargamos la vectorstore
     # vectordb = Chroma.from_documents(
-    #     persist_directory=st.session_state.persist_directory,  # Este es el directorio del la vs del docuemnto del usuario que se encuentra cargado en la session_state.
+    #     persist_directory=st.session_state.persist_directory,
+    # Este es el directorio del la vs del docuemnto del usuario
+    # que se encuentra cargado en la session_state.
     #     embedding_function=embeddings,
     # )
     vectordb = Chroma(

app/rag/vectorstore.py

@@ -13,3 +13,5 @@ def create_verctorstore(docs_split: list, embeddings, file_name: str):
             documents=docs_split,
             embedding=embeddings,
         )
+
+        return vectordb