My random forest is being used to calculate predictions in a Shiny app, but the forest object is very large. I'm using the save.memory=T option to save space, but an alternative pathway would be to store the covariate combinations and terminal-node quantities (from save.memory=F) for rapid prediction in a SQL database. (My organization has used the SQL-database pathway before for Shiny apps that need to access large datasets.)
I'm not sure how the terminal-node quantities are stored on the backend in rfsrc, but is there a dataset-like object that could be extracted from the forest object and modified for use in this way? Thanks in advance!
My random forest is being used to calculate predictions in a Shiny app, but the forest object is very large. I'm using the
save.memory=Toption to save space, but an alternative pathway would be to store the covariate combinations and terminal-node quantities (fromsave.memory=F) for rapid prediction in a SQL database. (My organization has used the SQL-database pathway before for Shiny apps that need to access large datasets.)I'm not sure how the terminal-node quantities are stored on the backend in
rfsrc, but is there a dataset-like object that could be extracted from the forest object and modified for use in this way? Thanks in advance!