本文整理汇总了Python中pykalman.KalmanFilter.sample方法的典型用法代码示例。如果您正苦于以下问题：Python KalmanFilter.sample方法的具体用法？Python KalmanFilter.sample怎么用？Python KalmanFilter.sample使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类pykalman.KalmanFilter的用法示例。
在下文中一共展示了KalmanFilter.sample方法的3个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: test_kalman_sampling
# 需要导入模块: from pykalman import KalmanFilter [as 别名]
# 或者: from pykalman.KalmanFilter import sample [as 别名]
def test_kalman_sampling():
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, z) = kf.sample(100)
assert_true(x.shape == (100, data.transition_matrix.shape[0]))
assert_true(z.shape == (100, data.observation_matrix.shape[0]))示例2: KalmanFilter
# 需要导入模块: from pykalman import KalmanFilter [as 别名]
# 或者: from pykalman.KalmanFilter import sample [as 别名]
transition_offset = [-0.1, 0.1]
observation_matrix = np.eye(2) + random_state.randn(2, 2) * 0.1
observation_offset = [1.0, -1.0]
transition_covariance = np.eye(2)
observation_covariance = np.eye(2) + random_state.randn(2, 2) * 0.1
initial_state_mean = [5, -5]
initial_state_covariance = [[1, 0.1], [-0.1, 1]]
kf = KalmanFilter(
transition_matrix, observation_matrix, transition_covariance,
observation_covariance, transition_offset, observation_offset,
initial_state_mean, initial_state_covariance,
random_state=random_state
)
states, observations = kf.sample(
n_timesteps=50,
initial_state=initial_state_mean
)
# sample from model
kf = KalmanFilter(
transition_matrix, observation_matrix, transition_covariance,
observation_covariance, transition_offset, observation_offset,
initial_state_mean, initial_state_covariance,
random_state=random_state,
em_vars=[
'transition_matrices', 'observation_matrices',
'transition_covariance', 'observation_covariance',
'observation_offsets', 'initial_state_mean',
'initial_state_covariance'
]
)示例3: KalmanFilter
# 需要导入模块: from pykalman import KalmanFilter [as 别名]
# 或者: from pykalman.KalmanFilter import sample [as 别名]
observation_matrix = np.eye(2) + random_state.randn(2, 2) * 0.1
observation_offset = [1.0, -1.0]
initial_state_mean = [5, -5]
n_timesteps = 50
# sample from model
kf = KalmanFilter(
transition_matrices=transition_matrix,
observation_matrices=observation_matrix,
transition_offsets=transition_offset,
observation_offsets=observation_offset,
initial_state_mean=initial_state_mean,
random_state=0
)
states, observations_all = kf.sample(
n_timesteps, initial_state=initial_state_mean
)
# label half of the observations as missing
observations_missing = np.ma.array(
observations_all,
mask=np.zeros(observations_all.shape)
)
for t in range(n_timesteps):
if t % 5 != 0:
observations_missing[t] = np.ma.masked
# estimate state with filtering and smoothing
smoothed_states_all = kf.smooth(observations_all)[0]
smoothed_states_missing = kf.smooth(observations_missing)[0]