chainsaddiction.utils

chainsaddiction.utils.local_decoding(lcsp) → numpy.ndarray

Parameters:

lcsp (numpy.ndarray) – Log of conditional state probabilities

Return type:

numpy.ndarray

Compute the most likely state for each observation individually.

chainsaddiction.utils.global_decodung(lgamma, ldelta, lcsp) → numpy.ndarray

Parameters:

• lgamma (numpy.ndarray) – Log of transition probability matrix

• ldelta (numpy.ndarray) – Log of initial distribution

• lcsp (numpy.ndarray) – Log of conditional state probabilities

Return type:

numpy.ndarray

Compute the most likey sequence of hidden states given the sequence of observations and the estimated transition probailities.