This is the repository that includes my python implementation of IO-HMM

My implementation is easy to use. Please see examples and code in the notebook.

Example usage:

SHMM = UnSupervisedIOHMM(num_states=2, max_EM_iter=1000, EM_tol=1e-4)
SHMM.setData([speed])
SHMM.setModels(model_emissions = [LM(), MNLD()], model_transition=MNLP(solver='lbfgs'))
SHMM.setInputs(covariates_initial = [], covariates_transition = [], covariates_emissions = [[],[]])
SHMM.setOutputs([['rt'],['corr']])