Last week, I introduced the teller, a model-agnostic tool for ML explainability. The teller’s philosophy is to rely on Taylor series 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.

This new version of the package (v0.2.0) improves the interface, and introduces Jackknife Student t-tests for the significance of marginal effects. For these tests, you’ll need to have at least 30 observations in the testing set.

Installation

Currently from Github, for the development version:

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

Significance of marginal effects

The following notebook will give you an introduction to this functionality:

A summary of the teller's results

Contributions/remarks are welcome as usual, you can submit a pull request on Github.

Note: I am currently looking for a gig. You can hire me on Malt or send me an email: thierry dot moudiki at pm dot me. I can do descriptive statistics, data preparation, feature engineering, model calibration, training and validation, and model outputs’ interpretation. I am fluent in Python, R, SQL, Microsoft Excel, Visual Basic (among others) and French. My résumé? Here!

Comments powered by Talkyard.