When we use HMM for classification, we need to train one HMM per class. My question is: How to find the matrices A,B,\pi?? What is the meaning of them?

To clarify: A =[aij] transition matrix, aij probability for moving from state I to state j When I have just one state as I denote above how would I calculate them out?

  • $\begingroup$ Why do you think you need to train one HMM per state? HMMs, by construction, deal with multiple possible states simultaneously... $\endgroup$ – jbowman Oct 1 '18 at 20:16
  • $\begingroup$ I read this step in this Question <br/> stats.stackexchange.com/questions/91290/…> $\endgroup$ – OmarMH87 Oct 1 '18 at 20:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.