PIML Toolbox

The PiML Toolbox is an open-source Python platform facilitating the development and analysis of physics-informed machine learning (PIML) models. PIML integrates physical laws and domain knowledge into machine learning algorithms, improving model accuracy, interpretability, and data efficiency compared to purely data-driven approaches. Current research emphasizes applications in areas like system identification, control, and condition monitoring, often employing models such as GAMs, GAMI-Nets, and XGBoost, alongside model-agnostic explainability techniques. This toolbox empowers researchers and practitioners to build and rigorously evaluate more reliable and understandable machine learning models across various scientific and engineering disciplines.