def _do_backward_pass(self, framelogprob):
n_observations = framelogprob.shape[0]
state_combinations = [tuple(x) for x in list(itertools.product(np.arange(self.n_states), repeat=self.n_chains))]
bwdlattice = np.zeros((n_observations, self.n_states ** self.n_chains))
fhmmc._backward(n_observations, self.n_chains, self.n_states, state_combinations, self.log_startprob,
self.log_transmat, framelogprob, bwdlattice)
return bwdlattice
def _do_backward_pass(self, framelogprob):
n_observations = framelogprob.shape[0]
state_combinations = [
tuple(x) for x in list(
itertools.product(np.arange(self.n_states),
repeat=self.n_chains))
]
bwdlattice = np.zeros((n_observations, self.n_states**self.n_chains))
fhmmc._backward(n_observations, self.n_chains, self.n_states,
state_combinations, self.log_startprob,
self.log_transmat, framelogprob, bwdlattice)
return bwdlattice
def test_backward_with_hmmlearn(self):
r = np.random.randn
obs = [np.array([[-600 + r(), 100 + r()], [-300 + r(), 200 + r()], [0 + r(), 300 + r()]]) for _ in xrange(10)]
hmm = GaussianHMM(n_components=3)
hmm.fit(obs)
# Calculcate bwdlattice using hmmlearn algorithm
framelogprob = hmm._compute_log_likelihood(obs[0])
start = timeit.default_timer()
bwdlattice1 = hmm._do_backward_pass(framelogprob)
print('hmmlearn took %fs' % (timeit.default_timer() - start))
# Calculate bwdlattice using fhmm algorithm with #chains = 1. This should yield the exact same results
start = timeit.default_timer()
bwdlattice2 = np.zeros(bwdlattice1.shape)
fhmmc._backward(obs[0].shape[0], 1, hmm.n_components, [(x,) for x in xrange(hmm.n_components)],
hmm._log_startprob.reshape(1, 3), hmm._log_transmat.reshape(1, 3, 3), framelogprob, bwdlattice2)
print('fhmm took %fs' % (timeit.default_timer() - start))
self.assertTrue(np.allclose(bwdlattice1, bwdlattice2))