hmm.add_transition(s33, s333, 0.01)
hmm.add_transition(s333, s1, 0.01)
hmm.add_transition(s333, s11, 0.01)
hmm.add_transition(s333, s111, 0.01)
hmm.add_transition(s333, s2, 0.01)
hmm.add_transition(s333, s22, 0.01)
hmm.add_transition(s333, s222, 0.01)
hmm.add_transition(s333, s3, 0.01)
hmm.add_transition(s333, s33, 0.01)
hmm.add_transition(s333, s333, 0.92)
hmm.bake()
hmm.fit(X) # , weights=w) hmm does not support weights in pomegranate
preds = hmm.predict(X)
probs = hmm.predict_proba(X)
data_thr['preds'] = pd.Series(preds).astype("category")
color_key = ["red", "blue", "yellow", "grey", "black", "purple", "pink",
"brown", "green", "orange"] # Spectral9
color_key = color_key[:len(set(preds))+2]
covs = np.array([np.array(hmm.states[m].distribution.parameters[1])
for m in range(9)])
means = np.array([np.array(hmm.states[m].distribution.parameters[0])
for m in range(9)])
# transform cov for non-standardizeed data:
covs = np.array([np.dot(np.diag(np.sqrt(scaler.var_)),
np.dot(covs[j], np.diag(np.sqrt(scaler.var_))))
