osl_dynamics.simulation.HMM_Dirichlet#

class osl_dynamics.simulation.HMM_Dirichlet(n_samples, trans_prob, concentration, n_states=None, n_channels=None, stay_prob=None, window_size=1)#

Bases: osl_dynamics.simulation.base.Simulation

Simulate an HMM with Dirichlet distribution as the observation model.

Parameters:
n_samplesint

Number of raw samples to draw. Must be divisible by window_size.

trans_probnp.ndarray or str

Transition probability matrix as a numpy array or a str (‘sequence’, ‘uniform’) to generate a transition probability matrix.

concentrationnp.ndarray or str

Concentration vector for each state, shape (n_states, n_channels), or ‘random’.

n_statesint, optional

Number of states when concentration=’random’.

n_channelsint, optional

Number of channels when concentration=’random’.

window_sizeint, optional

Raw samples per hidden-state step (default 1). State transitions occur between windows; samples within each window share a state. Both time_series and state_time_course have n_samples rows.

stay_probfloat

Used to generate the transition probability matrix is trans_prob is a str.

window_size = 1#
obs_mod#
n_states#
n_channels#
hmm#
state_time_course#
time_series#
property n_modes#
property mode_time_course#