update.const_C: C update with new observations

In activegp: Gaussian Process Based Design and Analysis for the Active Subspace Method

C update with new observations

Description

Update Constantine's C with new point(s) for a GP

Usage

## S3 method for class 'const_C'
update(object, Xnew, Znew, ...)

Arguments

object	A const_C object, the result of a call to the C_GP function.
Xnew	matrix (one point per row) corresponding to the new designs
Znew	vector of size nrow(Xnew) for the new responses at Xnew
...	not used (for consistency of update method)

Value

The updated const_C object originally provided.

See Also

C_GP to generate const_C objects from mleHomGP objects; update_C2 for an update using faster expressions.

Examples

 
################################################################################
### Active subspace of a Gaussian process
################################################################################
library(hetGP); library(lhs)
set.seed(42)

nvar <- 2
n <- 100

# theta gives the subspace direction
f <- function(x, theta, nugget = 1e-3){
  if(is.null(dim(x))) x <- matrix(x, 1)
  xact <- cos(theta) * x[,1] - sin(theta) * x[,2]
  return(hetGP::f1d(xact) + 
    rnorm(n = nrow(x), sd = rep(nugget, nrow(x))))
}

theta_dir <- pi/6
act_dir <- c(cos(theta_dir), -sin(theta_dir))

# Create design of experiments and initial GP model
design <- X <- matrix(signif(maximinLHS(n, nvar), 2), ncol = nvar)
response <- Y <- apply(design, 1, f, theta = theta_dir)
model <- mleHomGP(design, response, known = list(beta0 = 0))

C_hat <- C_GP(model)

print(C_hat)
print(subspace_dist(C_hat, matrix(act_dir, nrow = nvar), r = 1))

# New designs
Xnew <- matrix(runif(2), 1)
Znew <- f(Xnew, theta_dir)

C_new <- update(C_hat, Xnew, Znew)
print(C_new)
subspace_dist(C_new, matrix(act_dir, nrow = nvar), r = 1)

activegp documentation built on May 29, 2024, 9:22 a.m.