The code for Silva et al. "Physically Regularized Machine Learning Emulators of Aerosol Activation"
Code in the repository demonstrates running the Pyrcel model to generate the output used in the paper, as well as sample machine learning training scripts for the DNN, Ridge, and XGBoost regression schemes.
The pyrcel model run script and the DNN training script is written in python, with the DNN training an iPython Notebook.
The Ridge and XGBoost regression training scripts are in R.
The datasets used for training and thier description are in the "pyrcelOutput" directory.