This repo contains resources to design and run a machine learning model that we developed for Santa Cruz
In simple terms our objective was to create a machine learning surrogate model of the popular SWAN wave model.
The folder build_design_matrices provides resources to create the ML
design matrices (X and Y or features or labels) from standard SWAN model inputs
and outputs.
The folder ml_model_forecast contains resources to use these design matrices
to train and run a machine learning model
More details regarding the implementation available from our paper, A machine learning framework to forecast wave conditions