sim.dpp.modal {demu} | R Documentation |
sim.dpp.modal()
uses the DPP-based design emulator of Pratola et al. (2018)
to draw a sample of the n
-run optimal design for a Gaussian process
regression model with stationary correlation function r(x,x^\prime)
, where the
entries of R
are formed by evaluating r(x,x^\prime)
over a grid of candidate
locations.
sim.dpp.modal(R,n=0,eigs=NULL)
R |
A correlation matrix evaluated over a grid of candidate design sites. |
n |
Size of the design to sample. |
eigs |
One can alternatively pass the pre-computed eigendecomposition of the correlation matrix instead of |
For more details on the method, see Pratola et al. (2018). Detailed examples demonstrating the method are available at http://www.matthewpratola.com/software.
A vector of indices to the sampled design sites.
Pratola, Matthew T., Lin, C. Devon, and Craigmile, Peter. (2018) Optimal Design Emulators: A Point Process Approach. arXiv:1804.02089.
demu-package
sim.dpp.modal.fast
sim.dpp.modal.seq
library(demu)
# candidate grid
ngrid=20
x=seq(0,1,length=ngrid)
X=as.matrix(expand.grid(x,x))
l.d=makedistlist(X)
# draw design from DPP mode
n=21
rho=0.01
R=rhomat(l.d,rep(rho,2))$R
pts=sim.dpp.modal(R,n)
# Could plot the result:
# plot(X,xlim=c(0,1),ylim=c(0,1))
# points(X[pts,],pch=20)