Update app.py

app.py

@@ -42,8 +42,12 @@ def get_vector_store(chunks):
 
 # Function to get conversational chain
 def get_conversational_chain():
-    prompt_template = """
-    Answer the question based on the provided context. If the answer isn't in the context, please say, "Answer is not available in the context." Don't make up information.
+    from langchain.prompts import PromptTemplate
+
+    # Map prompt
+    map_prompt = PromptTemplate(
+        template="""
+    Answer the following question based on the given context. If the answer isn't in the context, please say, "Answer is not available in the context." Do not provide information not included in the context.
 
     Context:
     {context}
@@ -52,11 +56,34 @@ def get_conversational_chain():
     {question}
 
     Answer:
-    """
+    """,
+        input_variables=["context", "question"]
+    )
+
+    # Combine prompt
+    combine_prompt = PromptTemplate(
+        template="""
+    You are provided with several answers to a question from different parts of a document. Your task is to combine these answers into a single, coherent, and comprehensive response. If the answer isn't available, state "Answer is not available in the context."
+
+    Answers:
+    {summaries}
+
+    Question:
+    {question}
+
+    Final Answer:
+    """,
+        input_variables=["summaries", "question"]
+    )
+
     try:
         model = ChatGoogleGenerativeAI(model="gemini-pro", client=genai, temperature=0.3)
-        prompt = PromptTemplate(template=prompt_template, input_variables=["context", "question"])
-        chain = load_qa_chain(llm=model, chain_type="map_reduce", prompt=prompt)
+        chain = load_qa_chain(
+            llm=model,
+            chain_type="map_reduce",
+            question_prompt=map_prompt,
+            combine_prompt=combine_prompt
+        )
         return chain
     except (GoogleAPIError, InvalidArgument):
         raise RuntimeError("Error creating conversational chain. Please try again in a minute.")
@@ -72,7 +99,7 @@ def user_input(user_question):
     try:
         embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")  # type: ignore
         new_db = FAISS.load_local("faiss_index", embeddings, allow_dangerous_deserialization=True)
-        docs = new_db.similarity_search(user_question, k=4)  # You can adjust k as needed
+        docs = new_db.similarity_search(user_question, k=4)  # Adjust k if needed
         chain = get_conversational_chain()
         response = chain({"input_documents": docs, "question": user_question}, return_only_outputs=True)
         return response
@@ -81,6 +108,7 @@ def user_input(user_question):
 
 
 
+
 # Main function to run the Streamlit app
 def main():
     st.set_page_config(