This is the print() method for summary objects created by summary.vsel().
It displays a summary of the results from a varsel() or cv_varsel() run.
Arguments
- x
An object of class
vselsummary.- digits
Passed to
print.data.frame()(for the table containing the submodel performance evaluation results) andprint.default()(for the vector containing the reference model performance evaluation results).- ...
Arguments passed to
print.data.frame()(for the table containing the submodel performance evaluation results) andprint.default()(for the vector containing the reference model performance evaluation results).
Value
The output of summary.vsel() (invisible).
Details
In the submodel predictive performance table printed at (or towards)
the bottom, column ranking_fulldata contains the full-data predictor
ranking and column cv_proportions_diag contains the main diagonal of the
matrix returned by cv_proportions() (with cumulate as set in the
summary.vsel() call that created x). To retrieve the fold-wise
predictor rankings, use the ranking() function, possibly followed by
cv_proportions() for computing the ranking proportions (which can be
visualized by plot.cv_proportions()).