We previously had dictated which covariates should be used for which taxa through focalCovariates. @RRTogunov rightly pointed out that this was a clunky process - this is an attempt at making things a little smoother.
data/external/focalTaxa.csv - extra columns added for the focalCovariates we care about. Covariates used are dictated by results of our domain expert interviews.
functions/modelPreparation.R - Model now takes 'env' from the focalTaxa table
pipeline/import/environmentalImport.R - Selected parameters dictated from focalTaxa
Additional changes
pipeline/import/environmentalImport.R - Buffer here was massive and unnecessary - has been reduced
data/external/focalCovariates.csv - prediction datasets have full punctuation now, allowing for proper selection
Why have changes been made?
We previously had dictated which covariates should be used for which taxa through focalCovariates. @RRTogunov rightly pointed out that this was a clunky process - this is an attempt at making things a little smoother.
What changes have been made?
Additional changes