plot_ffs {CAST}

R Documentation

Plot results of a Forward feature selection or best subset selection

Description

A plotting function for a forward feature selection result. Each point is the mean performance of a model run. Error bars represent the standard errors from cross validation. Marked points show the best model from each number of variables until a further variable could not improve the results. If type=="selected", the contribution of the selected variables to the model performance is shown.

Usage

plot_ffs(
  ffs_model,
  plotType = "all",
  palette = rainbow,
  reverse = FALSE,
  marker = "black",
  size = 1.5,
  lwd = 0.5,
  pch = 21,
  ...
)

Arguments

ffs_model	Result of a forward feature selection see ffs
plotType	character. Either "all" or "selected"
palette	A color palette
reverse	Character. Should the palette be reversed?
marker	Character. Color to mark the best models
size	Numeric. Size of the points
lwd	Numeric. Width of the error bars
pch	Numeric. Type of point marking the best models
...	Further arguments for base plot if type="selected"

Author(s)

Marvin Ludwig and Hanna Meyer

See Also

ffs, bss

Examples

## Not run: 
data(iris)
ffsmodel <- ffs(iris[,1:4],iris$Species)
plot_ffs(ffsmodel)
#plot performance of selected variables only:
plot_ffs(ffsmodel,plotType="selected")

## End(Not run)

---