Interface TransitionTrainer


public interface TransitionTrainer
An object that can be used to train the transitions within a MarkovModel.
Author:
Matthew Pocock
  • Method Summary

    Modifier and Type
    Method
    Description
    void
    addCount(State from, State to, double count)
    Add 'count' to the transition from->to.
    void
    Clears all of the counts to zero.
    void
    train(double nullModel, double weight)
    Trains the transition, given an expected probability, and a weight for that probability.
  • Method Details

    • addCount

      void addCount(State from, State to, double count) throws IllegalSymbolException, IllegalTransitionException
      Add 'count' to the transition from->to.

      This method may be called multiple times with the same from,to pair in which case, the times should be summed.

      Parameters:
      from - the source state
      to - the destination state
      count - the number of counts to add
      Throws:
      IllegalSymbolException
      IllegalTransitionException
    • train

      void train(double nullModel, double weight) throws IllegalSymbolException
      Trains the transition, given an expected probability, and a weight for that probability.

      This is equivalent to adding a count of nullModel * weight to each transition and then training with a weight of 0.

      Parameters:
      nullModel - the nullModel to use
      weight - how many lots of the null model to add
      Throws:
      IllegalSymbolException
    • clearCounts

      void clearCounts()
      Clears all of the counts to zero.