Fits a tidymodels workflow classifying stations as inside or outside a
high-abundance patch. Replaces the biomod2 model in
original/buildZoopModel.R: cross-validation via rsample::vfold_cv() instead
of biomod2's CV.nb.rep/data.split.perc, and a workflows::workflow() object
instead of biomod2's on-disk model directory and string-pasted run names.
Arguments
- dat
labeled modeling data from
label_patch()with covariates attached- config
a config list, as returned by
load_config()
Value
a list with workflow (the fitted workflow), metrics (cross-validated
performance), importance (permutation variable importance), type (the
model type fitted), model_data (what it was fitted on), predictors, and
threshold
References
Kuhn M, Wickham H (2020). Tidymodels: a collection of packages for modeling and machine learning using tidyverse principles. https://www.tidymodels.org
Thuiller W, Lafourcade B, Engler R, Araújo MB (2009). BIOMOD - a platform for ensemble forecasting of species distributions. Ecography 32(3), 369-373. doi:10.1111/j.1600-0587.2008.05742.x — what this replaces
See model_types() for the reference behind each model, and
evaluation_table() for the metrics.