tops-ime-usp

ToPS (Toolkit for Probabilistic Models of Sequences) is an objected-oriented framework implemented using C++ that facilitates the integration of probabilistic models for sequence data over a user defined alphabet. ToPS contains the implementation of five distinct models to analyze discrete sequences: (i) multinomial distribution; (ii) variable-length Markov chain (VLMC); (iii) inhomogeneous Markov chain; (iv) hidden Markov model; (v) and generalized hidden Markov model (GHMM). The user can implement models either by manual description of the probability values in a configuration file, or by using training algorithms provided by the system. The ToPS framework also includes a set of programs that implement bayesian classiers, sequence samplers, and sequence decoders. Finally, ToPS is an extensible and portable system that facilitates the implementation of other probabilistic models, and the development of new programs. ToPS was developed at

simple_tops_example
casino_hmm
casino_markov_chain
hmm
multinomial
vlmc
wam