wddtw_pairwise_distance#

wddtw_pairwise_distance(X: ndarray, y: ndarray = None, window: float = None, g: float = 0.05) → ndarray[source]#

Compute the wddtw pairwise distance between a set of time series.

Parameters:

X: np.ndarray, of shape (n_instances, n_channels, n_timepoints) or: (n_instances, n_timepoints)

A collection of time series instances.
y: np.ndarray, of shape (m_instances, m_channels, m_timepoints) or: (m_instances, m_timepoints) or (m_timepoints,), default=None

A collection of time series instances.
window: float, default=None: The window to use for the bounding matrix. If None, no bounding matrix is used.
g: float, defaults=0.05: Constant that controls the level of penalisation for the points with larger phase difference. Default is 0.05.

Raises:

ValueError: If X is not 2D or 3D array when only passing X. If X and y are not 1D, 2D or 3D arrays when passing both X and y. If n_timepoints is less than 2.

Examples

>>> import numpy as np
>>> from aeon.distances import wddtw_pairwise_distance
>>> # Distance between each time series in a collection of time series
>>> X = np.array([[[1, 2, 3]],[[49, 58, 61]], [[73, 82, 99]]])
>>> wddtw_pairwise_distance(X)
array([[ 0.        , 20.86095125, 49.37503255],
       [20.86095125,  0.        ,  6.04844149],
       [49.37503255,  6.04844149,  0.        ]])

>>> # Distance between two collections of time series
>>> X = np.array([[[19, 12, 39]],[[40, 51, 69]], [[79, 28, 91]]])
>>> y = np.array([[[110, 15, 123]],[[14, 150, 116]], [[9917, 118, 29]]])
>>> wddtw_pairwise_distance(X, y)
array([[1.03345029e+03, 4.17910276e+03, 2.68408251e+07],
       [1.60419481e+03, 3.21952986e+03, 2.69227971e+07],
       [2.33574763e+02, 6.64390438e+03, 2.66663693e+07]])

>>> X = np.array([[[10, 22, 399]],[[41, 500, 1316]], [[117, 18, 9]]])
>>> y_univariate = np.array([[100, 11, 199],[10, 15, 26], [170, 108, 1119]])
>>> wddtw_pairwise_distance(X, y_univariate)
array([[  7469.9486745 ],
       [159295.70501427],
       [  1590.15378267]])