def test_compute_beta(hmm_discreet):
sample_data = np.array([[1, 1, 1, 0, 0, 0, 0, 0], [0, 1, 0, 0, 0, 1, 0, 0],
[0, 0, 0, 0, 0, 1, 0, 1], [0, 0, 1, 0, 1, 1, 1, 1],
[0, 0, 0, 0, 1, 0, 1, 0], [0, 1, 0, 0, 1, 1, 0, 1],
[0, 0, 0, 1, 1, 1, 0, 0]])
beta = np.array([[0.00670425, 0.0081531], [0.01313852, 0.01440698],
[0.02967516, 0.02299752], [0.060588, 0.04698],
[0.12366, 0.09612], [0.252, 0.198], [0.51, 0.42],
[1., 1.]])
beta1 = beta.round(decimals=8)
possible_observations = Counter(sample_data[i][j]
for i in range(len(sample_data))
for j in range(len(sample_data[i])))
M = 2
V = len(possible_observations)
hmm_discreet = hmm_discreet(M, sample_data, V)
hmm_discreet.pi = np.array([0.5, 0.5])
hmm_discreet.A = np.array([[0.7, 0.3], [0.4, 0.6]])
hmm_discreet.B = np.array([[0.6, 0.4], [0.3, 0.7]])
hmm_discreet.compute_beta(0)
beta2 = hmm_discreet.beta.round(decimals=8)
np.testing.assert_array_equal(beta1, beta2)
def test_compute_alpha(hmm_discreet):
sample_data = np.array([[1, 1, 1, 0, 0, 0, 0, 0], [0, 1, 0, 0, 0, 1, 0, 0],
[0, 0, 0, 0, 0, 1, 0, 1], [0, 0, 1, 0, 1, 1, 1, 1],
[0, 0, 0, 0, 1, 0, 1, 0], [0, 1, 0, 0, 1, 1, 0, 1],
[0, 0, 0, 1, 1, 1, 0, 0]])
alpha = np.array([[0.2, 0.35], [0.112, 0.189], [0.0616, 0.1029],
[0.050568, 0.024066], [0.0270144, 0.008883],
[0.01347797, 0.00403024], [0.006628, 0.00193846],
[0.00324899, 0.00094544]])
alpha1 = alpha.round(decimals=8)
possible_observations = Counter(sample_data[i][j]
for i in range(len(sample_data))
for j in range(len(sample_data[i])))
M = 2
V = len(possible_observations)
hmm_discreet = hmm_discreet(M, sample_data, V)
hmm_discreet.pi = np.array([0.5, 0.5])
hmm_discreet.A = np.array([[0.7, 0.3], [0.4, 0.6]])
hmm_discreet.B = np.array([[0.6, 0.4], [0.3, 0.7]])
hmm_discreet.compute_alpha(0)
alpha2 = hmm_discreet.alpha.round(decimals=8)
np.testing.assert_array_equal(alpha1, alpha2)
def test_compute_p_from_beta(hmm_discreet):
p1 = 0.004194434664
sample_data = np.array([[1, 1, 1, 0, 0, 0, 0, 0], [0, 1, 0, 0, 0, 1, 0, 0],
[0, 0, 0, 0, 0, 1, 0, 1], [0, 0, 1, 0, 1, 1, 1, 1],
[0, 0, 0, 0, 1, 0, 1, 0], [0, 1, 0, 0, 1, 1, 0, 1],
[0, 0, 0, 1, 1, 1, 0, 0]])
possible_observations = Counter(sample_data[i][j]
for i in range(len(sample_data))
for j in range(len(sample_data[i])))
M = 2
V = len(possible_observations)
hmm_discreet = hmm_discreet(M, sample_data, V)
hmm_discreet.pi = np.array([0.5, 0.5])
hmm_discreet.A = np.array([[0.7, 0.3], [0.4, 0.6]])
hmm_discreet.B = np.array([[0.6, 0.4], [0.3, 0.7]])
hmm_discreet.compute_beta(0)
p2 = hmm_discreet.compute_p_from_beta(0)
p2 = p2.round(decimals=12)
assert p1 == p2