Update functions.py

functions.py

@@ -540,6 +540,38 @@ def format_google_results(google_results):
     return formatted_documents
 
 
+def QA_chain(llm):
+    """
+    Creates a question-answering chain using the provided language model.
+
+    Args:
+        llm: The language model to use for generating answers.
+
+    Returns:
+        An LLMChain configured with the question-answering prompt and the provided model.
+    """
+    # Define the prompt template
+    prompt = PromptTemplate(
+        template="""You are an assistant for question-answering tasks. 
+
+        Use the following pieces of retrieved documents to answer the question. If you don't know the answer, just say that you don't know.
+        Do not repeat yourself!
+        Be informative and concise.
+
+        Question: {question} 
+
+        Documents: {documents} 
+
+        Answer:
+        """,
+        input_variables=["question", "documents"],
+    )
+
+    # Create and return the question-answering chain
+    return LLMChain(prompt=prompt, llm=llm, output_parser=StrOutputParser())
+
+
+
 def grade_generation_v_documents_and_question(state,hallucination_grader,answer_grader ):
     """
     Determines whether the generation is grounded in the document and answers the question.
@@ -627,7 +659,7 @@ def generate(state):
     """
     question = state["question"]
     documents = state["documents"]
-    generation = rag_chain.invoke({"documents": documents, "question": question})
+    generation = QA_chain.invoke({"documents": documents, "question": question})
     steps = state["steps"]
     steps.append("generate_answer")
     generation_count = state["generation_count"]