-
Effector: an eXplainability library for global and regional effects | by Vasilis Gkolemis | Medium
Global and regional effects offer a simple way to explain any black-box machine learning model trained on tabular data. Effector is a Python package that makes computing global and regional effects effortless.
-
Github Actions for connecting two Github repositories
In this guide, we'll leverage GitHub Actions to automate the publication of the documentation for my `Python` package. The documentation needs to be first pushed to a distinct GitHub repository, which is repsponsible for serving the GitHub Pages.
-
Aleatoric and epistemic uncertainty in classification problems
Discuss how categorical entropy can be used to quantify the aleatoric and epistemic uncertainty in Bayesian modeling.