MboPlotDependencies: MboPlotDependencies

In PhilippScheller/VisBayesOpt: Visualizing model based optimization runs from package mlrMBO

Description

This class generates plots for the pairwise dependencies of the features in the mbo run. It only covers the numeric search space (i.e. discrete params not considered).

Super class

VisBayesOpt::MboPlot -> MboPlotDependencies

Methods

Public methods

• MboPlotDependencies$new()

• MboPlotDependencies$plot()

• MboPlotDependencies$clone()

Inherited methods

• VisBayesOpt::MboPlot$set_param_vals()

---

Method new()

Creates a new instance of this R6 class.

Usage

MboPlotDependencies$new(opt_state)

Arguments

opt_state

(OptState).

---

Method plot()

Plots prior distributions of mbo run specified in the set of parameters.

Usage

MboPlotDependencies$plot(
  color_y = self$param_vals$color_y,
  search_space_components = getParamIds(self$opt_state$opt.path$par.set)[1:2]
)

Arguments

color_y

(logical(1) | TRUE)
Specifies which dimension the color of the scatter represents. If TRUE then color corresponds to value of 'y'. If FALSE color corresponds to 'iteration'. Default is TRUE.

search_space_components

(list())
Specifies the search space components which should be plotted.

Returns

(ggplot).

---

Method clone()

The objects of this class are cloneable with this method.

Usage

MboPlotDependencies$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

PhilippScheller/VisBayesOpt documentation built on Sept. 14, 2020, 12:47 p.m.