GBM Model Evaluation — Beyond Prediction AccuracyModel explainability, interpretable model benchmarking, weakness detection, benign overfitting, and robustness tests.Feb 10Feb 10
Model Diagnostics: Prediction UncertaintyToday’s PiML tutorial is about quantification of prediction uncertainty for a pre-trained machine learning model. We take a deep dive into…Oct 30, 2023Oct 30, 2023
Model Diagnostics: Overfitting & RobustnessOverfitting models have poor generalisation and limited real-world applicability. They also tend to be less robust to covariate drift. Oct 21, 2023Oct 21, 2023
Model Diagnostics Trilogy — Part 1PiML was born as an educational toolbox for interpretable machine learning in Python community. It is designed for both model development…Oct 9, 2023Oct 9, 2023