renderExampleRunPlot: Renders plots for exampleRun objects, either in 1D or 2D, or... In mlrMBO: Bayesian Optimization and Model-Based Optimization of Expensive Black-Box Functions

Description

The graphical output depends on the target function at hand. - For 1D numeric functions the upper plot shows the true function (if known), the model and the (infill) points. The lower plot shows the infill criterion. - For 2D mixed target functions only one plot is displayed. - For 2D numeric only target functions up to four plots are presented to the viewer: - levelplot of the true function landscape (with [infill] points), - levelplot of the model landscape (with [infill] points), - levelplot of the infill criterion - levelplot of the standard error (only if learner supports standard error estimation). - For bi-criteria target functions the upper plot shows the target space and the lower plot displays the x-space.

Usage

 1 2 3 4 5 6 7 8 9 10 11 12 13 14 renderExampleRunPlot( object, iter, densregion = TRUE, se.factor = 1, single.prop.point.plots = FALSE, xlim = NULL, ylim = NULL, point.size = 3, line.size = 1, trafo = NULL, colors = c("red", "blue", "green"), ... )

Arguments

 object [function] MBOExampleRun or MBOExampleRunMultiObj object. iter [integer] Selected iteration of object to render plots for. densregion [logical(1)] Should the background be shaded? Default is TRUE. Only used if learner supports computation of standard error. se.factor [numeric(1)] If the model provides local standard error estimation, in addition to the mean response yhat(x) +- se.factor * se(x) is plotted above and below. Default is 1. single.prop.point.plots [logical(1)] Parameter for MOI-MBO Multi-point proposal: Should every proposed point be displayed in a single plot - or one plot per Iteration? Default is FALSE indicating single plots per proposed points. xlim [numeric(2)] For 1D: xlim parameter for first and second plot. Default is range of x-values evaluated in run object object. ylim [numeric(2)] For 1D: ylim parameter for first plot, for the second plot ylim is always set automatically, depending on the range of the evaluated infill criterion. Default for the first plot is a heuristic to have the true function and yhat(x) +- se.factor2 * se(x) both in the plot. Note that this heuristic might change the ylim setting between plot iterations. point.size [numeric(1)] Point size for plotted points. Default is 3. line.size [numeric(1)] Line width of the graphs of plotted functions. trafo [list] List of transformation functions of type MBOTrafoFunction for the different plots. For 1D: The list elements should be named with “y” (applied to objective function and model) or “crit” (applied to the criterion). Only applied to plots with numeric parameters. For 2D: The list should contain at least one element “y”, “yhat”, “crit” or “se”. This way one can specify different transformations for different plots. If a single function is provided, this function is used for all plots. colors [character(3)] Specify colors for point in the plots. Must be a vector of length 3, each element a color for the type design, prop and seq respectively. Default is red for the initial design, blue for already proposed points and green for the actual iteration. ... [any] Currently not used.

Value

[list]. List containing separate ggplot object. The number of plots depends on the type of MBO problem. See the description for details.

mlrMBO documentation built on Jan. 13, 2021, 7:57 a.m.