Update app.py

app.py

@@ -43,29 +43,26 @@ def rag_from_pdf(question, pdf_file, api_key):
         vectorstore = initialize_vectorstore(documents, api_key)
         retriever = vectorstore.as_retriever(search_type="similarity", search_kwargs={"k": 3})
         
-        # Initialize the LLM
-        llm = ChatOpenAI(model="gpt-3.5-turbo")
-        
-        # Create a prompt template for combining context and question
-        prompt_template = """
-        You are a helpful assistant answering questions based on the provided PDF document.
-        Only use the given context to answer the question.
-        Question: {question}
-        Context: {context}
-        """
-        prompt = ChatPromptTemplate.from_template(prompt_template)
-        
         # Retrieve relevant documents
         retrieved_docs = retriever.get_relevant_documents(question)
         context = "\n".join([doc.page_content for doc in retrieved_docs])
         
-        # Generate response using the LLM
+        # Check if there's relevant context
         if not context.strip():
             return "No relevant information found in the document to answer the question."
         
-        formatted_prompt = prompt.format(question=question, context=context)
-        response = llm(completion=formatted_prompt)
-        return response.strip()
+        # Initialize the LLM
+        llm = ChatOpenAI(model="gpt-3.5-turbo")
+        
+        # Create the chat prompt
+        messages = [
+            {"role": "system", "content": "You are a helpful assistant answering questions based on the provided PDF document."},
+            {"role": "user", "content": f"Question: {question}\n\nContext: {context}"}
+        ]
+        
+        # Generate response
+        response = llm(messages=messages)
+        return response["choices"][0]["message"]["content"].strip()
     except Exception as e:
         return f"An error occurred: {str(e)}"