Migrating from Langchain to LangGraph with Pandas DataFrame - No Idea and No Documentation

[mraguth]

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

import json
from pydantic import BaseModel
from langchain.agents import create_tool_calling_agent
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain.agents import AgentExecutor
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_experimental.tools import PythonAstREPLTool
from langchain.agents import AgentExecutor, OpenAIFunctionsAgent

from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_experimental.tools import PythonAstREPLTool
from langchain.agents import AgentExecutor, OpenAIFunctionsAgent
from pydantic import BaseModel, Field, model_validator
from langchain_core.tools import BaseTool
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder

import ast
import re
import sys
from contextlib import redirect_stdout
from io import StringIO
from typing import Any, Dict, Optional, Type
from langchain_core.callbacks.manager import (
    AsyncCallbackManagerForToolRun,
    CallbackManagerForToolRun,
)
from langchain_core.runnables.config import run_in_executor
from langchain_core.tools import BaseTool
from pydantic import BaseModel, Field, model_validator
from langchain_experimental.utilities.python import PythonREPL

class PythonInputs(BaseModel):
    """Python inputs."""

    query: str = Field(description="code snippet to run")


template = """
You are working with a pandas dataframe in Python. The name of the dataframe is `df`.\n"
It is important to understand the attributes of the dataframe before working with it. This is the result of running `df.head().to_markdown()`.\n"

<df>
{dhead}
</df>

You are not meant to use only these rows to answer questions - they are meant as a way of telling you about the shape and schema of the dataframe.
You also do not have use only the information here to answer questions - you can run intermediate queries to do exploratory data analysis to give you more information as needed.
Convert your curiosity or questions into executable Python code using Pandas to learn as needed.

"""  # noqa: E501

template = template.format(dhead=df.head().to_markdown())

# Define the prompt
prompt = ChatPromptTemplate.from_messages(
    [
        ("system", template),
        MessagesPlaceholder(variable_name="agent_scratchpad"),
        ("human", "{input}"),
        MessagesPlaceholder("chat_history", optional=True),
    ]
)

repl = PythonAstREPLTool(      ##### This is the function brining df in to the agent_executor
    locals={"df": df},
    name="python_repl",
    description="A Python shell. Use this to execute python commands. Input should be a valid python command. When using this tool, sometimes output is abbreviated - make sure it does not look abbreviated before using it in your answer.",
    args_schema=PythonInputs,
    prompt = prompt,
)


tools = [repl]

agent = create_openai_functions_agent(
    llm=llm, prompt=prompt, tools=tools
)
agent_executor = AgentExecutor(
    agent=agent, 
    tools=tools, 
    max_iterations=10, 
    early_stopping_method="force",    ##### Changed from generate due to Generate error.
    allow_dangerous_code=True,
    allow_dangerous_requests=True,
) | (lambda x: x["output"])

# Typing for playground inputs
class AgentInputs(BaseModel):
    input: str

agent_executor = agent_executor.with_types(input_type=AgentInputs)

# Define the workflow
output_query_str = "Top 5 brands by total BVs"

def process_input(input_data):
    messages = agent_executor.invoke({"input":output_query_str})
    return messages

input_data = output_query_str
output = process_input(input_data)
print(output)

Description

There is a lot of information available, but it's unclear how to proceed with the migration when using a pandas DataFrame in Langchain to achieve the same results in LangGraph. Any thoughts? Above my code works perfect in Langchain. Why I need LangGraph if no path is defined? These are real life stuctured data RAG in real corporations. No path, no Go to production :-(

System Info

Name: langchain
Version: 0.3.2

Name: langgraph
Version: 0.2.34