predict_front: Predicted Pareto front

In coldfir3/moko: Multi-Objective Kriging Optimization

Description

This function creates a predicted pareto front based on the mean of Kriging models. The predicted mean of each objective and constraint is passed to the nsga2 algorithm that builds .

Usage

predict_front(model, lower, upper, control = NULL, modelcontrol = NULL)

Arguments

model	Object of class mkm.
lower	Vector of lower bounds for the variables to be optimized over (default: 0 with length model@d).
upper	Vector of upper bounds for the variables to be optimized over (default: 1 with length model@d).
control	An optional list of control parameters that controlls the optimization algorithm. One can control: popsize (default: 200); generations (default: 30); cdist (default: 1/model@d); mprob (default: 15); mdist (defult: 20).
modelcontrol	An optional list of control parameters to the mkm function (default: object@control).

Value

object of class ps containing the predicted Pareto front

Examples

# ------------------------
# The Nowacki Beam
# ------------------------
n <- 100
doe <- cbind(sample(0:n,n),sample(0:n,n))/n
res <- t(apply(doe, 1, nowacki_beam))
model <- mkm(doe, res, modelcontrol = list(objective = 1:2, lower=c(0.1,0.1)))
pf <- predict_front(model, c(0,0), c(1,1))
plot(nowacki_beam_tps$set)
points(pf$set, col='blue')

coldfir3/moko documentation built on May 13, 2019, 8:49 p.m.