Question

My LangChain app stopped working after upgrading to v0 2 What changed in agent execution

Answer 1

The v0.2 upgrade broke a lot of people's agents. Here's what actually changed:

The biggest thing — AgentExecutor isn't the default anymore
from langchain.agents import initialize_agent, AgentType

agent = initialize_agent(

    tools=tools,

    llm=llm,

    agent=AgentType.OPENAI_FUNCTIONS,

    verbose=True

)

agent.run("What's the weather in Delhi?")

In the new version ,initialize_agent  this function is deprecated. The new way uses LCEL
 

from langchain import hub

from langchain.agents import create_openai_functions_agent, AgentExecutor

prompt = hub.pull("hwchase17/openai-functions-agent")

agent = create_openai_functions_agent(llm, tools, prompt)

agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

agent_executor.invoke({"input": "What's the weather in Delhi?"})