gp_energy: Energy of a GP model

In gplite: General Purpose Gaussian Process Modelling

Energy of a GP model

Description

Returns the energy (negative log marginal likelihood) of a fitted GP model with the current hyperparameters. The result is exact for the Gaussian likelihood and dependent on the approx for other cases.

Usage

gp_energy(gp, include_prior = TRUE)

Arguments

gp	The fitted GP model.
include_prior	Whether to add log density of the prior to the result (in which case the result is -(log marginal likelihood + log prior))

Value

The energy value (negative log marginal likelihood).

References

Rasmussen, C. E. and Williams, C. K. I. (2006). Gaussian processes for machine learning. MIT Press.

Examples

# Generate some toy data
set.seed(1242)
n <- 500
x <- matrix(rnorm(n * 3), nrow = n)
f <- sin(x[, 1]) + 0.5 * x[, 2]^2 + x[, 3]
y <- f + 0.5 * rnorm(n)
x <- data.frame(x1 = x[, 1], x2 = x[, 2], x3 = x[, 3])

# Basic usage
gp <- gp_init(cf_sexp(), lik_gaussian())
gp <- gp_fit(gp, x, y)
e <- gp_energy(gp)

gplite documentation built on Aug. 24, 2022, 9:07 a.m.