update

app.py

@@ -1,8 +1,12 @@
 from langchain import HuggingFaceHub, PromptTemplate
 from langchain.memory import ConversationBufferMemory
-from langchain.chains import ConversationChain
+from langchain.agents import AgentType, initialize_agent
+from langchain.chains import AgentChain
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 import gradio as gr
+import os
+
+os.environ["HUGGINGFACEHUB_API_TOKEN"] = "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"
 
 template = """Question: {history}
 ------------------
@@ -11,19 +15,23 @@ Answer: Let's think step by step."""
 prompt = PromptTemplate(template=template, input_variables=["history"])
 
 # Create a memory module with a maximum capacity of 1000 items
-memory = ConversationBufferMemory()
+memory = ConversationBufferMemory(max_capacity=1000)
 # Callbacks support token-wise streaming
 callbacks = [StreamingStdOutCallbackHandler()]
-# Instantiate the LLMChain with the model and tokenizer
+# Instantiate the Hugging Face model
 llm = HuggingFaceHub(repo_id="gpt2")
 
-conversation = ConversationChain(llm=llm, memory=memory, callbacks=callbacks, prompt=prompt)
+# Define the tools
+tools = []
+
+# Initialize the agent chain
+agent_chain = initialize_agent(tools, llm, agent=AgentType.CONVERSATIONAL_REACT_DESCRIPTION, verbose=True, memory=memory)
+
+conversation = AgentChain(agent_chain, callbacks=callbacks, prompt=prompt)
 
 # Define the Gradio interface
 def chatbot_interface(input_text):
     response = conversation.predict(input_text)
-    memory.chat_memory.add_user_message(input_text)
-    memory.chat_memory.add_ai_message(response)
     return response
 
 # Define the Gradio app