switched to chat engine

backend.py

@@ -12,6 +12,7 @@ from huggingface_hub import hf_hub_download
 from llama_cpp import Llama
 import spaces
 from huggingface_hub import login
+from llama_index.core.memory import ChatMemoryBuffer
 
 
 huggingface_token = os.getenv("HUGGINGFACE_TOKEN")
@@ -78,17 +79,15 @@ def handle_query(query_str, chathistory):
 
     try:
         # Create a streaming query engine
-        query_engine = index.as_query_engine(text_qa_template=text_qa_template, streaming=False, similarity_top_k=1)
+        """query_engine = index.as_query_engine(text_qa_template=text_qa_template, streaming=False, similarity_top_k=1)
         
         # Execute the query
         streaming_response = query_engine.query(query_str)
         
         r = streaming_response.response
         cleaned_result = r.replace("<end_of_turn>", "").strip()
-        yield cleaned_result
-        
-        
-        
+        yield cleaned_result"""
+      
         # Stream the response
         """outputs = []
         for text in streaming_response.response_gen:
@@ -96,6 +95,18 @@ def handle_query(query_str, chathistory):
             outputs.append(str(text))
             yield "".join(outputs)"""
           
+        memory = ChatMemoryBuffer.from_defaults(token_limit=1500)
+        chat_engine = index.as_chat_engine(
+        chat_mode="context",
+        memory=memory,
+        system_prompt=(
+            "Sei un assistente italiano di nome Ossy che risponde solo alle domande o richieste pertinenti. "
+          ),
+        )
+        
+        response = chat_engine.chat(query_str)
+        yield response   
+          
         
     except Exception as e:
         yield f"Error processing query: {str(e)}"