Bavisitter: Babysitting LLMs for Visualization

Bavisitter is visual analytics-centered LLM client.

Get Started

!pip install bavisitter

import os
from bavisitter import Bavisitter

os.environ["OPENAI_API_KEY"] = "sk-"

Bavisitter(df, model="gpt-4o", color_mode="dark")

Using Multiple LLM Provider

By leveraging LiteLLM, you can use diverse LLMs by simply changing the model string. Check LiteLLM document for supported providers

Development

Install pnpm and Hatch to set up the development environment.

git clone https://github.com/jiwnchoi/bavisitter.git && cd bavisitter
pnpm install
pnpm dev # Runs Vite dev server and Jupyter Lab

Reference

Bavisitter: Integrating Design Guidelines into Large Language Models for Visualization Authoring

Jiwon Choi, Jaeung Lee, and Jaemin Jo

Proceedings of Conference on 2024 IEEE Visualization & Visual Analytics (IEEE VIS), Tampa Bay, USA

@inproceedings{choi2024bavisitter,
  title={Bavisitter: Integrating Design Guidelines into Large Language Models for Visualization Authoring},
  author={Choi, Jiwon and Lee, Jaeung and Jo, Jaemin},
  booktitle={2024 IEEE Visualization and Visual Analytics (VIS)},
  year={2024},
  organization={IEEE}
}