---

There is an increasing need for transparency and fairness in Machine Learning (ML) models predictions. Consider for example a banker who has to explain to a client why his/her loan application is rejected, or a healthcare professional who must explain what constitutes his/her diagnosis. Some ML models are indeed very accurate, but are considered to be hard to explain, relatively to popular linear models.

Source of figure: James, Gareth, et al. An introduction to statistical learning. Vol. 112. New York: springer, 2013.

We do not want to sacrifice this high accuracy to explainability. Hence: ML explainability. There are a lot of ML explainability tools out there, in the wild.

The teller is a model-agnostic tool for ML explainability. Agnostic, as long as the input ML model possesses methods fit and predict, and is applied to tabular data. The teller relies on:

• Finite differences to explain ML models predictions: a little increase in model's explanatory variables + a little decrease, and we can obtain approximate sensitivities of its predictions to changes in these explanatory variables.
• Conformal prediction (so far, as of october 2022) to obtain prediction intervals for ML methods

Installation

• From Pypi, stable version:

pip install the-teller

• From Github, for the development version:

pip install git+https://github.com/Techtonique/teller.git

Package description

These notebooks will be some good introductions:

• Heterogeneity of marginal effects
• Significance of marginal effects
• Model comparison
• Classification
• Interactions
• Prediction intervals for regression

Contributing

Your contributions are welcome, and valuable. Please, make sure to read the Code of Conduct first.

If you're not comfortable with Git/Version Control yet, please use this form.

In Pull Requests, let's strive to use black for formatting:

pip install black
black --line-length=80 file_submitted_for_pr.py

API Documentation

https://techtonique.github.io/teller/

Dependencies

• Numpy
• Pandas
• Scipy
• scikit-learn

Citation

@misc{moudiki2019teller,
	author={Moudiki, T.},
	title={\code{teller}, {M}odel-agnostic {M}achine {L}earning explainability},
	howpublished={\url{https://github.com/thierrymoudiki/teller}},
	note={BSD 3-Clause Clear License. Version 0.x.x.},
	year={2019--2020}
	}

References

For sensitivity analysis:

• Press, W. H., Teukolsky, S. A., Vetterling, W. T., & Flannery, B. P. (1992). Numerical recipes in C (Vol. 2). Cambridge: Cambridge university press.
• Jones E, Oliphant E, Peterson P, et al. SciPy: Open Source Scientific Tools for Python, 2001-, http://www.scipy.org/ [Online; accessed 2019-01-04]
• Scikit-learn: Machine Learning in Python, Pedregosa et al., JMLR 12, pp. 2825-2830, 2011.

For prediction intervals:

• Romano, Y., Patterson, E., & Candes, E. (2019). Conformalized quantile regression. Advances in neural information processing systems, 32.

License

BSD 3-Clause © Thierry Moudiki, 2019.