[Biojava-l] Why does Baum-welch not work?

[fabian schreiber]

Hello!
I've a question.
I want to use the Baum-Welch-algorithm to train my Markov-Modell.
First i set all the transition- and emission-probabilities, so that
i have a kind of basis modell.
Now i want to train my modell with lots of sequences.
After the training had finished, i looked up the probabilities of
the markov modell to find changes.
But all the probabilities has changed to "NaN", which is no good.

Can someone tell me, where the mistake is?
Here is my code:

//Create Markov Modell - The method createCasino generates an Alphabet
and sets //the probabilities for the transitions and emissions
MarkovModel casino = createCasino();

DP dp=DPFactory.DEFAULT.createDP(casino);


BaumWelchTrainer bwtrainer = new BaumWelchTrainer(dp);


SequenceDB seqDB = new HashSequenceDB("hashdb");
// here the DB is filled with the sequences --> this works

//Set the stopper
  StoppingCriteria stopper= new StoppingCriteria()
             {public boolean isTrainingComplete(TrainingAlgorithm ta)
             {return (ta.getCycle() > 10);}};
//Train the modell
bwtrainer.train(seqDB, 1.0, stopper);

Thats it.
Can somebody help me, why
all probabilitites changed from 0.xx to NaN?
Thanks for your support.

Fabian

_______________________________________________
Biojava-l mailing list  -  Biojava-l at biojava.org
http://biojava.org/mailman/listinfo/biojava-l