selwscale: Window Scaling

In adana: Adaptive Nature-Inspired Algorithms for Hybrid Genetic Optimization

Window Scaling

Description

Window Scaling is a method based on subtracting the worst fitness value from the other fitness values. In this case, since the scaled values of the worst fit individuals will be 0, these individuals will not be given a chance to be selected.

Usage

selwscale(fitvals, ns, fmin, ...)

Arguments

fitvals	Vector of fitness values belonging to individuals
ns	Number of individuals to be selected
fmin	The number to subtract from all fitness values.
...	Further arguments passed to or from other methods.

Value

The indices of randomly selected individuals are returned.

Author(s)

Zeynel Cebeci & Erkut Tekeli

See Also

select, selrand, selrswrp, selrws, selrws2, selsus, seldet, selrss, selsscale, selsscale2, sellscale, selrscale, selrscale2, selpscale, selescale, seltour, seltour2

Examples

fitvals = c(6.1, 3.5, 6.2, 4.4, 5.2)
fmin = min(fitvals)
cnames = paste0("C",1:length(fitvals))
matpool = selwscale(fitvals, fmin=fmin)
cat(cnames[matpool],"\n")
fitvals = fitvals[matpool]
fitvals
matpool = selwscale(fitvals, fmin=fmin)
cat(cnames[matpool],"\n")
fitvals = fitvals[matpool]
fitvals
fmin = min(fitvals)
matpool = selwscale(fitvals, fmin=fmin)
cat(cnames[matpool],"\n")

adana documentation built on March 18, 2022, 6:03 p.m.