Daniel@0: Daniel@0: Daniel@0: Daniel@0: Netlab Reference Manual gpcovar Daniel@0: Daniel@0: Daniel@0: Daniel@0:

gpcovar Daniel@0:

Daniel@0:

Daniel@0: Purpose Daniel@0:

Daniel@0: Calculate the covariance for a Gaussian Process. Daniel@0: Daniel@0:

Daniel@0: Synopsis Daniel@0:

Daniel@0:
Daniel@0: cov = gpcovar(net, x)
Daniel@0: [cov, covf] = gpcovar(net, x)
Daniel@0: 
Daniel@0: Daniel@0: Daniel@0:

Daniel@0: Description Daniel@0:

Daniel@0: Daniel@0:

cov = gpcovar(net, x) takes Daniel@0: a Gaussian Process data structure net together with Daniel@0: a matrix x of input vectors, and computes the covariance Daniel@0: matrix cov. The inverse of this matrix is used when calculating Daniel@0: the mean and variance of the predictions made by net. Daniel@0: Daniel@0:

[cov, covf] = gpcovar(net, x) also generates the covariance Daniel@0: matrix due to the covariance function specified by net.covarfn Daniel@0: as calculated by gpcovarf. Daniel@0: Daniel@0:

Daniel@0: Example Daniel@0:

Daniel@0: In the following example, the inverse covariance matrix is calculated Daniel@0: for a set of training inputs x and is then Daniel@0: passed to gpfwd so that predictions (with mean ytest and Daniel@0: variance sigsq) can be made for the test inputs Daniel@0: xtest. Daniel@0:
Daniel@0: 
Daniel@0: cninv = inv(gpcovar(net, x)); 
Daniel@0: [ytest, sigsq] = gpfwd(net, xtest, cninv);
Daniel@0: 
Daniel@0: Daniel@0: Daniel@0:

Daniel@0: See Also Daniel@0:

Daniel@0: gp, gppak, gpunpak, gpcovarp, gpcovarf, gpfwd, gperr, gpgrad
Daniel@0: Pages: Daniel@0: Index Daniel@0:
Daniel@0:

Copyright (c) Ian T Nabney (1996-9) Daniel@0: Daniel@0: Daniel@0: Daniel@0: