How can we use create_pandas_dataframe_agent in Multi-agent Collaboration?

[ya0002]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain.agents.agent_types import AgentType
from langchain.agents import tool, create_tool_calling_agent, AgentExecutor
from langchain_experimental.agents.agent_toolkits import create_pandas_dataframe_agent

data_agent = AgentExecutor(...)##is an Agent Executor object, this creates the dataframe

pandas_agent = create_pandas_dataframe_agent(
    ChatOpenAI(temperature=0),
    df,
    verbose=True,
    agent_type=AgentType.OPENAI_FUNCTIONS,
    agent_executor_kwargs={"handle_parsing_errors":True}
)

Description

I'm following Basic Multi-agent Collaboration.

I have an agent that creates a dataframe and was wondering how could I use hat dataframe with the agent we get from create_pandas_dataframe_agent since it requires the dataframe to be directly passed to it.

System Info

Running everything on Google colab

[dosubot[bot]]

Hey there, @ya0002! 👋 I'm here to assist you with any questions or issues you may have while waiting for a human maintainer. Feel free to ask me anything!

To use create_pandas_dataframe_agent in a multi-agent collaboration where the dataframe is created by another agent, you can pass the dataframe created by the first agent to the create_pandas_dataframe_agent function. Here is an example:

from langchain_openai import ChatOpenAI
from langchain_experimental.agents import create_pandas_dataframe_agent
import pandas as pd

# Assume agent1 creates a dataframe
df = pd.read_csv("titanic.csv")

# Create a language model instance
llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)

# Use the dataframe created by agent1 to create a pandas dataframe agent
agent_executor = create_pandas_dataframe_agent(
    llm,
    df,
    agent_type="tool-calling",
    verbose=True
)

In this example, the dataframe df is created by another agent (e.g., by reading a CSV file), and then it is passed to the create_pandas_dataframe_agent function to create a new agent that can work with this dataframe ^[1].

Details

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.

[amitkadu]

@dosu review above response. df is created using read_csv and not by using agent.

[mraguth]

@dosu My objective is to integrate a pandas DataFrame into a multi-agent collaboration or agentic workflow (link enclosed below). How can I pass the pandas DataFrame to agents within that agentic workflow? Do you have some example code that could help me with this?

https://langchain-ai.github.io/langgraph/tutorials/multi_agent/multi-agent-collaboration/