Update app.py

app.py

@@ -4,6 +4,8 @@ import gradio as gr
 from llama_index import GPTVectorStoreIndex, SimpleDirectoryReader, ServiceContext
 from llama_index import OpenAIEmbedding
 import os
+from llama_index.memory import ChatMemoryBuffer
+
 logging.basicConfig(stream=sys.stdout, level=logging.INFO)
 logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))
 os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY")
@@ -16,11 +18,16 @@ def build_index(documents):
     return GPTVectorStoreIndex.from_documents(documents)
 
 def query_index(index, similarity_top_k=3, streaming=True):
-    query_engine = index.as_query_engine(
-        similarity_top_k=similarity_top_k,
-        streaming=streaming
+    memory = ChatMemoryBuffer.from_defaults(token_limit=1500)
+    chat_engine = index.as_chat_engine(
+    chat_mode="context",
+    memory=memory,
+    system_prompt=(
+        "You are a chatbot, able to have normal interactions, as well as talk"
+        " about Yoga and anything related to it."
+    ),
     )
-    return query_engine  # return the query engine instance
+    return chat_engine  # return the query engine instance # return the query engine instance
 
 data = load_data()
 index = build_index(data)
@@ -28,7 +35,7 @@ query_engine = query_index(index)  # initialize the query engine
 
 def get_response(text,history=None):
     # Use the initialized query engine to perform the query
-    response = str(query_engine.query(text))
+    response = str(query_engine.chat(text))
     return response
 
 t = gr.ChatInterface(get_response, analytics_enabled=True)