Generates an interleaved lattice-based design with low fill and high separation distance properties.
InterleavedFillSepD(p,n,w=rep(1,p),pfrom=p,a=1/2,nmin=floor(n*.8),nmax=ceiling(n*1.2),
coefF=-4,coefS=1,msC=0,NL=10,NP=100,NJ=10,NS=100);The value returned from the function is a matrix containing the generated design. Remark that no qualified design might be found if (a) both nmax and nmin are to close to n, (b) both n and msC are small, or (c) both pfrom and msC are small while p is much bigger than pfrom.
Number of dimensions.
Targeted number of points, must be an integer greater than one.
Optional, weights of the dimensions.
Optional, number of dimensions designs are generated and supplemented from, no more than p and no more than 8.
Optional, translation parameter with a=0 for uniform design and a=1 for pushing the points to the boundary.
Optional, minimal acceptable number of points, no less than n.
Optional, maximal acceptable number of points, no greater than n.
Optional, coefficient of r_F in the criterion.
Optional, coefficient of r_S in the criterion.
Optional, maximal allowed sum of projuct weights for pairs of binary aliased dimensions.
Optional, maximum number of lattices to try.
Optional, maximum number of dimension permutations to try.
Optional, maximum number of discretization choices for each lattice and s vector combination.
Optional, maximum number of choices to supplement from.
This function generates an interleaved lattice-based design with low fill and high separation distance properties in p dimensions and around n points, following the algorithm provided in the paper "Efficient Kriging using interleaved lattice-based designs with low fill and high separation distance properties".
He, Xu (2024). "Efficient Kriging using interleaved lattice-based designs with low fill and high separation distance properties", SIAM/ASA Journal on Uncertainty Quantification, 12(4): 1113-1134.
InterleavedFillSepD(p=2,n=20);
Run the code above in your browser using DataLab