Skip to content

The repository contains all datasets and evaluations for "StatBot.Swiss: Bilingual Open Data Exploration in Natural Language" paper.

License

Notifications You must be signed in to change notification settings

dscc-admin-ch/statbot.swiss

Repository files navigation

Statbot.swiss

The repository contains all datasets and evaluations for the statbot.swiss benchmark.

Install

  • conda create -n statbot python=3.11.4

  • conda activate statbot

  • pip install -r requirements.txt

Usage

To run langchain with ChatGPT:

  1. provide openai-api-key in src/config.json file.

  2. reproduce results based on gpt-3.5-turbo-16k run the command below.

  3. For random shot selection, choose random and for the selection based on similarity choose similarity for --shot-selection-strategy.

  • Command to run:

    python src/main.py --shot-selection-strategy <random|similarity>

Aknowlegement

This work is the output of the INODE4StatBot.swiss project, funded by the Swiss Federal Statistical Office.
It is a collaboration between the Institute of Computer Science at ZHAW, the Competence Center for Data Science of the Federal Statistical Office, the Cantonal Statistical Office Zurich, and the Swiss Data Science Center.

License

This work is licensed under the MIT license. Third-party software and data are subject to their respective licenses.

Citation

Please cite the paper as follows:

@article{statbot2024,
  title={StatBot. Swiss: Bilingual Open Data Exploration in Natural Language},
  author={Nooralahzadeh, Farhad and Zhang, Yi and Smith, Ellery and Maennel, Sabine and Matthey-Doret, Cyril and de Fondville, Rapha{\"e}l and Stockinger, Kurt},
  journal={arXiv preprint arXiv:2406.03170},
  year={2024}
}

About

The repository contains all datasets and evaluations for "StatBot.Swiss: Bilingual Open Data Exploration in Natural Language" paper.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published