model = LibraryHSMMIntNegBinVariant(init_state_concentration=10.,
alpha=6.,
gamma=6.,
obs_distns=obs_distns,
dur_distns=dur_distns)
model.add_data(data, left_censoring=True)
##################
# infer things #
##################
for i in progprint_xrange(50):
model.resample_model()
likes = model.Viterbi_EM_fit()
plt.figure()
truemodel.plot()
plt.gcf().suptitle('truth')
plt.figure()
plt.plot(likes)
plt.gcf().suptitle('likes')
plt.figure()
model.plot()
plt.gcf().suptitle('inferred')
plt.show()
##################
# infer things #
##################
train_likes = []
test_likes = []
for i in progprint_xrange(5):
# for i in progprint_xrange(50):
model.resample_model()
train_likes.append(model.log_likelihood())
test_likes.append(model.log_likelihood(test_data, left_censoring=True))
model.truncate_num_states(10, destructive=True, mode='random')
model.Viterbi_EM_fit()
# print 'training data likelihood when in the model: %g' % model.log_likelihood()
# print 'training data likelihood passed in externally: %g' % sum(model.log_likelihood(data,left_censoring=True) for data in training_datas)
# plt.figure()
# truemodel.plot()
# plt.gcf().suptitle('truth')
# plt.figure()
# model.plot()
# plt.gcf().suptitle('inferred')
# plt.figure()
# plt.plot(train_likes,label='training')
# plt.plot(test_likes,label='test')