NET-MC:  Do Markov chain simulation to sample networks.

The net-mc program is the specialization of xxx-mc to the task of
sampling from the posterior distribution for a neural network model,
or from the prior distribution, if no training set is specified.  See
xxx-mc.doc for the generic features of this program.

The following applications-specific sampling procedures are implemented:

   sample-hyper   Does Gibbs sampling for the hyperparameters controlling
                  the distributions of parameters (weights, etc.).

   sample-noise   Does Gibbs sampling for the noise variances.

   sample-sigmas  Does both sample-hyper and sample-noise.

Default stepsizes are set by a complicated heuristic procedure that is
described in Appendix A of the thesis.

Tempering methods are not currently supported.

            Copyright (c) 1995 by Radford M. Neal