ferret documentation#
ferret is Python library for benchmarking interpretability techniques on Transformers.
Use any of the badges above to test our live demo, view a video demonstration, or explore our technical paper in detail.
Installation#
To install our latest stable release in default mode (which does not include the depenencies for the speech XAI functionalities), run this command in your terminal:
pip install -U ferret-xai
If the speech XAI functionalities are needed, then run:
pip install -U ferret-xai[speech]
At the moment, the speech XAI-related dependencies are the only extra ones, so installing with ferret-xai[speech] or ferret-xai[all] is equivalent.
These are the preferred methods to install ferret, as they will always install the most recent stable release.
If you don’t have pip installed, this Python installation guide can guide you through the process.
Citation#
If you are using ferret for your work, please consider citing us!
@inproceedings{attanasio-etal-2023-ferret,
title = "ferret: a Framework for Benchmarking Explainers on Transformers",
author = "Attanasio, Giuseppe and Pastor, Eliana and Di Bonaventura, Chiara and Nozza, Debora",
booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations",
month = may,
year = "2023",
publisher = "Association for Computational Linguistics",
}
Also, ferret’s Speech XAI functionalities are based on
@misc{pastor2023explaining,
title " Explaining Speech Classification Models via Word-Level Audio Segments and Paralinguistic Features",
author= "Pastor, Eliana and Koudounas, Alkis and Attanasio, Giuseppe and Hovy, Dirk and Baralis, Elena",
month = september,
year = "2023",
eprint = "2309.07733",
archivePrefix = "arXiv",
primaryClass = "cs.CL",
publisher = "",
}