GP-COV:  Print covariance matrix for a Gaussian process at given points.

GP-cov prints the covariance matrix for a Gaussian process at a given
set of input points.

Usage: 
 
    gp-cov log-file index [ extra ] [ / train-inputs ]

If no extra argument is given, the covariance matrix for latent values
is used, without the noise variance being added to the diagonal.  If
the extra argument is the string "+noise", the noise is added.  This
is valid only for a regression model, and only if the noise variances
do not vary on a case-by-case basis.  If there is more than one target
variable, the noise added is that for the first target.  If the extra
argument is a positive number, that number is added to the diagonal.

By default, the covariance matrix is for the set of training inputs,
as specified with data-spec, but this can be overridden by specifying
a file of training inputs at the end of the command line.  Target
values are not needed.

The covariance matrix is written to standard output, to 18 digit
precision, one row per line.  The full matrix is written even though
it is symmetrical.

            Copyright (c) 1995-2003 by Radford M. Neal