Dear scientific python community
I am very glad to announce the release of model-diagnostics 1.2 with its new plot_marginal and compute_marginal.
Model-diagnostics helps you to assess calibration and performance of most supervised model - be it machine learning or statistical - for point predictions (like the mean, a quantile, or the probability for binary classification). The focus is on visualization and user-friendliness while well backed by statistical theory.
The new plot_marginal, for instance, gives a great overview of the calibration as well as the model effect by a single feature, see 1.2 release notes:
Best,
Christian