def run_HMM(time_series, **kwargs):
# Directly use knowledge of the number of modes
K = kwargs['num_gt_labels']
D_obs = kwargs['data_dim']
obs_hypparams = {
'mu_0': np.zeros(D_obs),
'sigma_0': np.eye(D_obs),
'kappa_0': 0.10,
'nu_0': D_obs + 2
}
obs_distns = [
pyhsmm.distributions.Gaussian(**obs_hypparams) for _ in range(K)
]
hmm = HMM(alpha=kwargs['alpha'],
init_state_concentration=kwargs['init_state_concentration'],
obs_distns=obs_distns)
return inference(time_series, hmm, **kwargs)
alpha=6.,
gamma=6.,
obs_distns=GMMs,
dur_distns=true_dur_distns)
data, truelabels = truemodel.generate(1000)
#################
# remove data #
#################
# using an HMM of course
deletion_hmm = HMM(
init_state_concentration=10.,
alpha=6.,
gamma=6.,
obs_distns=GMMs[:2], # placeholders, not really used
)
_, todelete = deletion_hmm.generate(data.shape[0])
data[todelete == 0] == np.nan
#####################################
# set up FrozenMixture components #
#####################################
# list of all Gaussians
component_library = [c for m in GMMs for c in m.components]
library_size = len(component_library)