Similar create_csv_agent but for .xls or .xlsx?

[VadeanFlaviuAlexandru]

long story short, and thanks to google,python and streamlit, i can upload .csv files and use chatGPT to talk with them:

xls = st.file_uploader("Incarca XLS:", type=["xlsx", "xls"])

if xls is not None:
    with NamedTemporaryFile(mode='w+b', suffix='.xls', delete=False) as f:
        f.write(xls.getvalue())
        f.flush()
        llm = OpenAI(temperature=0)
        user_question = st.text_input("Intreaba-ma orice!")
        # agent = create_csv_agent(llm, f.name, verbose=False)
        buttonPressed = st.button("test")
        if buttonPressed:
            with st.spinner("Se citeste... 📄🤔... o secunda, te rog..."):
                if user_question is not None and user_question != "":
                    # st.write(agent.run(user_question))
                    pass

My question is: how can i replace the agents if i only want to upload .xlsx and .xls files? i commented out the parts i think need to be changed, but i did not found any concrete examples. Can someone help?

[dosubot[bot]]

🤖

Salut,

Pentru a modifica codul existent pentru a încărca fișiere .xlsx și .xls în loc de fișiere .csv, poți utiliza clasa UnstructuredExcelLoader din modulul langchain.document_loaders.excel. Iată codul modificat:

from typing import Any, List, Optional, Union

from langchain.agents.agent import AgentExecutor
from langchain.agents.agent_toolkits.pandas.base import create_pandas_dataframe_agent
from langchain.schema.language_model import BaseLanguageModel
from langchain.document_loaders.excel import UnstructuredExcelLoader


def create_excel_agent(
    llm: BaseLanguageModel,
    path: Union[str, List[str]],
    pandas_kwargs: Optional[dict] = None,
    **kwargs: Any,
) -> AgentExecutor:
    """Create excel agent by loading to a dataframe and using pandas agent."""
    try:
        import pandas as pd
    except ImportError:
        raise ImportError(
            "pandas package not found, please install with `pip install pandas`"
        )

    _kwargs = pandas_kwargs or {}
    if isinstance(path, str):
        loader = UnstructuredExcelLoader(path, mode="elements")
        df = pd.DataFrame(loader.load())
    elif isinstance(path, list):
        df = []
        for item in path:
            if not isinstance(item, str):
                raise ValueError(f"Expected str, got {type(path)}")
            loader = UnstructuredExcelLoader(item, mode="elements")
            df.append(pd.DataFrame(loader.load()))
    else:
        raise ValueError(f"Expected str or list, got {type(path)}")
    return create_pandas_dataframe_agent(llm, df, **kwargs)

În acest cod, funcția create_excel_agent este creată pentru a înlocui create_csv_agent. Clasa UnstructuredExcelLoader este utilizată pentru a încărca fișiere .xlsx și .xls. Restul codului rămâne la fel.

Sper că acest lucru te ajută! Dacă ai alte întrebări, nu ezita să întrebi.

Sources

• libs/langchain/langchain/agents/agent_toolkits/csv/base.py
• libs/langchain/langchain/document_loaders/excel.py

---

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.