mlpy.stats.dbn.hmm.DiscreteHMM.score_samples

DiscreteHMM.score_samples(obs)

Compute the log probability of the evidence.

Compute the log probability of the evidence (likelihood) under the model and the posteriors.

Parameters:

obs : array_like, shape (n, len, nfeatures)

Sequence of nfeatures-dimensional data points. Each row corresponds to a single point in the sequence.

Returns:

logp : float

Log likelihood of the sequence obs.

posteriors : array_like, shape (n, ncomponents)

Posterior probabilities of each state for each observation