NOTES ON THE VERSION OF 1999-02-20

Changes in this version:

1) Examples of neural network models for survival analysis have been
   added.  Some bugs in this part of the software have been fixed too.


Bug fixes.

Some bugs regarding the survival analysis models have been fixed.  A
problem that arose when automatic standardization was used for a variable
that always had the same value in the training data has been fixed.  The
computation of the likelihood for binary models is now more bullet-proof,
and should avoid floating point overflow even in extreme cases.


Known bugs and other deficiencies.

1) The facility for plotting quantities using "plot" operations in xxx-mc
   doesn't always work for the first run of xxx-mc (before any
   iterations exist in the log file).  A work-around is to do a run of
   xxx-mc to produce just one iteration before attempting a run of
   xxx-mc that does any "plot" operations.

2) The CPU time features (eg, the "k" quantity) will not work correctly
   if a single iteration takes more than about 71 minutes.

3) The latent value update operations for Gaussian processes may recompute 
   the inverse covariance matrix even when an up-to-date version was 
   computed for the previous Monte Carlo operation.

4) Covariance matrices are stored in full, even though they are symmetric,
   which sometimes costs a factor of two in memory usage.

5) The "p" option of net-pred isn't allowed for survival models with
   piecewise-constant hazard.  There's no big reason for this - I just
   haven't gotten around to it.