LLMWorkbook

"Effortlessly harness the power of LLMs on Excel and DataFrames—seamless, smart, and efficient!"

LLMWorkbook is a Python package designed to seamlessly integrate Large Language Models (LLMs) into your workflow with tabular data, be it Excel, CSV, DataFrames/Arrays. This package allows you to easily configure an LLM, send prompts row-wise from any tabular datasets, and store responses back in the DataFrame with minimal effort.

Documentation Website

Visit our complete documentation site →

For comprehensive guides, examples, and API reference, visit our dedicated documentation website.

Features

• Easily map LLM responses to a specific column in a pandas DataFrame, Excel, CSV.
• Run list of prompts easily.
• Get started with easy to follow Examples

LLMWorkbook v1.4.3 🦦:

✔ New OpenAI Responses Endpoint

Installation

Install the package from GitHub:

    pip install llmworkbook

Quick Start

Wrapper Utilities for LLM Preparation

LLMWorkbook provides wrapper utilities to prepare various data formats for LLM consumption. These utilities transform input data into a format suitable for LLM processing, ensuring consistency and compatibility. These wrapper methods can handle popular data sources like Excel (xlsx), CSV, Pandas DataFrames, multi dimensional arrays.

See Examples for details. - Github - Examples

Providers Supported -

• OpenAI
• Gpt4All
• Ollama

1. Import the Package

import pandas as pd
from llmworkbook import LLMConfig, LLMRunner, LLMDataFrameIntegrator

2. DataFrame

# Provide a dataframe, the usual
df = pd.DataFrame(data)

3. Configure the LLM

config = LLMConfig(
    provider="openai",
    system_prompt="Process these Data rows as per the provided prompt",
    options={
        "model": "gpt-4o-mini",
        "temperature": 1,
        "max_tokens": 1024,
    },
)

4. Create a Runner and Integrate

runner = LLMRunner(config)
integrator = LLMDataFrameIntegrator(runner=runner, df=df)

5. Add LLM Responses to DataFrame

updated_df = integrator.add_llm_responses(
    prompt_column="prompt_text",
    response_column="llm_response",
    async_mode=False  # Set to True for asynchronous requests
)

Example code is available in the Git Repository for easy reference.

Future Roadmap

• Add support for more LLM providers (Google VertexAI, Cohere, Groq, MistralAI).
• Add an interface frontend for low code applications.
• Implement rate-limiting and token usage tracking.
• Summarized history persisted across session to provide quick context for next session.

Extended Documentation

Detailed documentation for each module is available in the Documentation file.

• Wrapping Data file.
• Providers - OpenAI Gpt4All Ollama
• CLI Usage file.
• LLMDataFrameIntegrator - Row/Batch Processing

Links

Homepage Repository Documentation Examples Bug Tracker Issues