Issues with CSV agent and Pandas agent ( Completions.create() got an unexpected keyword argument 'tools')

[ghost]

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_experimental.agents.agent_toolkits.csv.base import create_csv_agent
from langchain_openai import OpenAI
from langchain_community.embeddings import HuggingFaceInferenceAPIEmbeddings
from langchain_experimental.agents.agent_toolkits import create_pandas_dataframe_agent

agent = create_csv_agent(
      llm ,
    'train.csv',
    agent_type=AgentType.ZERO_SHOT_REACT_DESCRIPTION,
)
agent.run("how many rows are there?")¨
f=pd.read_csv('train.csv',delimiter=';',encoding='Latin-1')
print(df.head())
agent = create_pandas_dataframe_agent(llm, df, agent_type="openai-tools", verbose=True)
agent.invoke(
    {
        "input": "What's the correlation between age and fare? is that greater than the correlation between fare and survival?"
    }
)
**## Output ##
  return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
TypeError: Completions.create() got an unexpected keyword argument 'tools'** 

**and when changing the LLM or agent type i get this error: in create_react_agent
    raise ValueError(f"Prompt missing required variables: {missing_vars}")
ValueError: Prompt missing required variables: {'tools', 'tool_names'}**

Description

I am trying to use the CSV agent to query my CSV but I keep getting the error: TypeError: Completions.create() got an unexpected keyword argument 'tools'. I tried different agent types but the error is still occuring ( i followed the example from the documentation)
if anyone has an idea how to fix this or encountered this issue before please reach out!

System Info

System Information

OS: Windows
OS Version: 10.0.19045
Python Version: 3.11.8 (tags/v3.11.8:db85d51, Feb 6 2024, 22:03:32) [MSC v.1937 64 bit (AMD64)]

Package Information

langchain_core: 0.1.22
langchain: 0.1.6
langchain_community: 0.0.19
langsmith: 0.0.87
langchain_experimental: 0.0.50
langchain_openai: 0.0.5

[dosubot[bot]]

🤖

Hey @Khawlao! Great to see you back here, diving into another adventure with LangChain. Hope you're ready for some troubleshooting action! 🚀

Based on the error message you're seeing, it seems like the Completions.create() function is receiving an unexpected keyword argument 'tools'. This is likely because the 'tools' argument is not expected by the create_pandas_dataframe_agent() function, which is called within the create_csv_agent() function.

In the provided code, the create_csv_agent() function accepts **kwargs: Any as an argument and passes it to create_pandas_dataframe_agent(). If 'tools' is included in **kwargs, it will be passed to create_pandas_dataframe_agent(). However, if create_pandas_dataframe_agent() does not accept 'tools' as an argument, it will raise a TypeError.

To resolve this issue, you need to ensure that the arguments you are passing to create_csv_agent() are expected by the create_pandas_dataframe_agent() function. If 'tools' is not a valid argument for create_pandas_dataframe_agent(), you should remove it from the arguments you are passing to create_csv_agent().

Here is the relevant code:

def create_csv_agent(
    llm: LanguageModelLike,
    path: Union[str, IOBase, List[Union[str, IOBase]]],
    pandas_kwargs: Optional[dict] = None,
    **kwargs: Any,
) -> AgentExecutor:
    ...
    return create_pandas_dataframe_agent(llm, df, **kwargs)

In this code, **kwargs is passed to create_pandas_dataframe_agent(). If 'tools' is included in **kwargs, it will be passed to create_pandas_dataframe_agent(). If 'tools' is not a valid argument for create_pandas_dataframe_agent(), a TypeError will be raised.

You can find the source code for this function in the LangChain repository.

As for the create_pandas_dataframe_agent() function, it expects the following arguments:

1. llm: Language model to use for the agent.
2. df: Pandas dataframe or list of Pandas dataframes.
3. agent_type: One of "openai-tools", "openai-functions", or "zero-shot-react-description". Defaults to "zero-shot-react-description". "openai-tools" is recommended over "openai-functions".
4. callback_manager: DEPRECATED. Pass "callbacks" key into 'agent_executor_kwargs' instead to pass constructor callbacks to AgentExecutor.
5. prefix: Prompt prefix string.
6. suffix: Prompt suffix string.
7. input_variables: DEPRECATED. Input variables automatically inferred from constructed prompt.
8. verbose: AgentExecutor verbosity.
9. return_intermediate_steps: Passed to AgentExecutor init.
10. max_iterations: Passed to AgentExecutor init.
11. max_execution_time: Passed to AgentExecutor init.
12. early_stopping_method: Passed to AgentExecutor init.
13. agent_executor_kwargs: Arbitrary additional AgentExecutor args.
14. include_df_in_prompt: Whether to include the first number_of_head_rows in the prompt. Must be None if suffix is not None.
15. number_of_head_rows: Number of initial rows to include in prompt if include_df_in_prompt is True.
16. extra_tools: Additional tools to give to agent on top of a PythonAstREPLTool.
17. **kwargs: DEPRECATED. Not used, kept for backwards compatibility.

You can find the source code for this function in the LangChain repository.

I hope this helps! If you have any other questions, feel free to ask.

Sources

• libs/experimental/langchain_experimental/agents/agent_toolkits/csv/base.py
• libs/experimental/langchain_experimental/agents/agent_toolkits/pandas/base.py

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.

If you want to continue the conversation, start your reply with @dosu-bot.

[ghost]

@dosu-bot In my code I don't have an argument 'tools' I followed the exact example given here : https://python.langchain.com/docs/integrations/toolkits/csv and I keep getting the error : in create_react_agent
raise ValueError(f"Prompt missing required variables: {missing_vars}")
ValueError: Prompt missing required variables: {'tools', 'tool_names'}**

[baskaryan]

@Khawlao make sure you have openai>=1 installed, believe you just have an out of date openai dependency

[ghost]

@baskaryan Hi! I have openai version 1.12.0 so I don't think that's the issue :')

[baskaryan]

you're right, apologies! believe this was the issue, should be fixed in next release of langchain-experimental (0.0.51) #17442

[shivamsh314]

hey @Khawlao , what is the llm that you are using?

[HelloWorldLTY]

I am using Claude and I face this error.