Selector that selects the top n_select individuals based on the fitness value, breaking ties randomly. When n_select is larger than the number
of individuals, the selection wraps around: All nrow(values) individuals are selected at least floor(nrow(values) / n_select)
times, with the top nrow(values) %% n_select individuals being selected one more time.
shuffle_selection :: logical(1)
Whether to shuffle the selected output. When this is TRUE, selected individuals are returned in random order, so when this
operator is e.g. used in mies_generate_offspring(), then subsequent recombination operators effectively operate on pairs
(or larger groups) of random individuals. Otherwise they are returned in order, and recombination operates on the first
batch of n_indivs_in returned individuals first, then the second batch etc. in order. Initialized to TRUE (recommended).
This Selector can be created with the short access form sel()
(sels() to get a list), or through the the dictionary
dict_selectors in the following way:
# preferred:
sel("best")
sels("best") # takes vector IDs, returns list of Selectors
# long form:
dict_selectors$get("best")miesmuschel::MiesOperator -> miesmuschel::Selector -> miesmuschel::SelectorScalar -> SelectorBest
new()Initialize the SelectorBest object.
SelectorBest$new(scalor = ScalorSingleObjective$new())scalor(Scalor)Scalor to use to generate scalar values from multiple objectives, if multi-objective optimization is performed.
Initialized to ScalorSingleObjective: Doing single-objective optimization normally, throwing an error if used
in multi-objective setting: In that case, a Scalor needs to be explicitly chosen.
sb = sel("best")
p = ps(x = p_dbl(-5, 5))
# dummy data; note that SelectorBest does not depend on data content
data = data.frame(x = rep(0, 5))
fitnesses = c(1, 5, 2, 3, 0)
sb$prime(p)
sb$operate(data, fitnesses, 2)
#> [1] 2 4
sb$param_set$values$shuffle_selection = FALSE
sb$operate(data, fitnesses, 4)
#> [1] 2 4 3 1