AutoRank is a post processing pipeline for numerical model grids. Its goal is to find the best observable tracers of any unobservable quantity. Given a dataset of models exploring the range of possible physical conditions, and containing both the unobservable target quantity and the observable quantities (e.g., line intensities), it will produce a ranking of the tracers and provide trained random forest models to predict the target quantity from any of the observable tracers.
See Bron et al. (2020) for a complete description of the method and an example of possible application.
Scientists
Emeric Bron