plotModelCV: Plot cross-validated variable importance

In environmentalinformatics-marburg/gpm: Geospatial predictive modeling using parameterized and unparameterized models.

Description

This function plots variable imporatnce for cross-validated rfe variable selection classes or train classes. It uses the k cross-validations to compute the mean +/- sd error or standard deviation metric.

Usage

plotModelCV(model, metric = model$metric, tuningValue = "Variables",
  xlim = "minmax", ylim = "minmax", sderror = FALSE, grid = TRUE)

Arguments

model	A rfe or train object. See rfe and train
metric	the metric to be used. Note this needs to be the metric used to calculate the rfe or train model
tuningValue	The tuning value which is depicted on the x axis. For rfe models default is "Variables", the number of variables.
xlim	the xlim for the plot
ylim	the ylim for the plot
sderror	If TRUE then standard error is calculated. If FALSE then standard deviations are used
grid	Print grid or not

Value

a trellis object

Note

if rfe is used as model, then returnResamp = "all" must be set in rfe training

environmentalinformatics-marburg/gpm documentation built on July 11, 2020, 11:12 a.m.