Class UntrainableDistribution

All Implemented Interfaces:
Serializable, Distribution, Changeable

A distribution which does not interact with the training framework. This class behaves in exactly the same manner as SimpleDistribution, except that it has a no-op registerWithTrainer method. It is useful for building Markov models where you wish to train only a subset of the Distributions.
Since:
1.3
Author:
Thomas Down
See Also: