Notebooks and explainers with code

Predicting diabetes with pycaret

We use the Predictive Power Score to evaluate which features have the highest predictive power with respect to Outcome.

Predicting house prices with pycaret

We use the Predictive Power Score to evaluate which features have the highest predictive power with respect to SalePrice.

Demonstration of overfitting and underfitting

This notebook illustrates why we require a train/test split (or cross-validation) to balance overfitting vs. underfitting.

Deconfounding explained

A demonstration howcorrelations ‘magically’ disappear if confounders are added to your model.