Acquisition Function OptimizerSource:
Logging level during the infill optimization. Can be
"trace". Default is
"warn", i.e., only warnings are logged.
Should the infill optimization be warm-started by evaluating the best point(s) present in the bbotk::Archive of the actual bbotk::OptimInstance? This is sensible when using a population based infill optimizer, e.g., local search or mutation. Default is
integer(1) | "all"
Number of best points selected from the bbotk::Archive that are to be used for warm starting. Can also be "all" to use all available points. Only relevant if
warmstart = TRUE. Default is
It can happen that the candidate resulting of the infill optimization was already evaluated in a previous iteration. Should this candidate proposal be ignored and the next best point be selected as a candidate? Default is
Should errors during the infill optimization be caught and propagated to the
loop_functionwhich can then handle the failed infill optimization appropriately by, e.g., proposing a randomly sampled point for evaluation? Default is
Id used when printing.
Set of hyperparameters.
Creates a new instance of this R6 class.
AcqOptimizer$new(optimizer, terminator, acq_function = NULL)
Helper for print outputs.
Optimize the acquisition function.
data.table::data.table() with 1 row per optimum and x as columns.