mrIML

This goal of mrIML is to enable users to build and interpret multivariate machine learning models by harnessing the power and flexibility of the tidyverse (tidy model syntax in particular). This package builds off ideas from Gradient Forests ( Ellis et al 2012), and ecological genomics ( Fitzpatrick and Keller, 2014).

This package can be of use for any multi-response machine learning problem, but was designed to handle data common to community ecology (site by species data) and ecological genomics (individual or population by SNP loci). An introduction to the package and some worked examples are also provided by Fountain-Jones et al, 2021