Generate models and a plot — model.report • moleculaR

Screen for models, cross validate them and plot. Designed for interactive work.

Usage

model.report(
  dataset,
  min = 2,
  max = floor(dim(mod_data)[1]/5),
  leave.out = "",
  predict = F
)

Arguments

dataset: a dataframe with outcome column (must be named 'output')
min: minimum # of features (default = 2)
max: max # of features (defaults = # of observations / 5)
leave.out: name of observations to leave out (e.g. 'p_Br')
predict: if leave.out is not empty, should a prediction of it be computed?

Value

models list, CV results for k=3/5/LOO and a plot of choice. interactive.