Getting Quantum-Chemical Insights using Machine Learning

A Machine Learning project that investigates the relation between molecular geometry and atomization energy. Custom Ridge Regression model was used which was trained on both linear and complex feature representations of the molecules of the 'QM7' datataset. The predictions were also explained using explanability methods, linking them to existing chemical litertaure.