def test_kalman_filter_update():
kf = KalmanFilter(
data.transition_matrix,
data.observation_matrix,
data.transition_covariance,
data.observation_covariance,
data.transition_offsets,
data.observation_offset,
data.initial_state_mean,
data.initial_state_covariance)
# use Kalman Filter
(x_filt, V_filt) = kf.filter(X=data.observations)
# use online Kalman Filter
n_timesteps = data.observations.shape[0]
n_dim_state = data.transition_matrix.shape[0]
x_filt2 = np.zeros((n_timesteps, n_dim_state))
V_filt2 = np.zeros((n_timesteps, n_dim_state, n_dim_state))
for t in range(n_timesteps - 1):
if t == 0:
x_filt2[0] = data.initial_state_mean
V_filt2[0] = data.initial_state_covariance
(x_filt2[t + 1], V_filt2[t + 1]) = kf.filter_update(
x_filt2[t], V_filt2[t], data.observations[t + 1],
transition_offset=data.transition_offsets[t]
)
assert_array_almost_equal(x_filt, x_filt2)
assert_array_almost_equal(V_filt, V_filt2)
def test_kalman_filter():
kf = KalmanFilter(
data.transition_matrix,
data.observation_matrix,
data.transition_covariance,
data.observation_covariance,
data.transition_offsets,
data.observation_offset,
data.initial_state_mean,
data.initial_state_covariance)
(x_filt, V_filt) = kf.filter(X=data.observations)
assert_array_almost_equal(
x_filt[:500],
data.filtered_state_means[:500],
decimal=7
)
assert_array_almost_equal(
V_filt[:500],
data.filtered_state_covariances[:500],
decimal=7
)
