pairwise_distance¶

pairwise_distance(x: ndarray, y: ndarray | None = None, metric: str | Callable[[ndarray, ndarray, Any], float] | None = None, **kwargs: Unpack) → ndarray[source]¶

Compute the pairwise distance matrix between two time series.

Parameters:

Xnp.ndarray

A collection of time series instances of shape (n_cases, n_timepoints): or (n_cases, n_channels, n_timepoints).

ynp.ndarray or None, default=None

A single series or a collection of time series of shape (m_timepoints,) or (m_cases, m_timepoints) or (m_cases, m_channels, m_timepoints)

metricstr or Callable

The distance metric to use. A list of valid distance metrics can be found in the documentation for aeon.distances.get_distance_function or by calling the function aeon.distances.get_distance_function_names.

kwargsAny

Extra arguments for metric. Refer to each metric documentation for a list of possible arguments.

Returns:

np.ndarray (n_cases, n_cases): pairwise matrix between the instances of X.

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 metric is not a valid string or callable.

Examples

>>> import numpy as np
>>> from aeon.distances import pairwise_distance
>>> # Distance between each time series in a collection of time series
>>> X = np.array([[[1, 2, 3]],[[4, 5, 6]], [[7, 8, 9]]])
>>> pairwise_distance(X, metric='dtw')
array([[  0.,  26., 108.],
       [ 26.,   0.,  26.],
       [108.,  26.,   0.]])

>>> # Distance between two collections of time series
>>> X = np.array([[[1, 2, 3]],[[4, 5, 6]], [[7, 8, 9]]])
>>> y = np.array([[[11, 12, 13]],[[14, 15, 16]], [[17, 18, 19]]])
>>> pairwise_distance(X, y, metric='dtw')
array([[300., 507., 768.],
       [147., 300., 507.],
       [ 48., 147., 300.]])

>>> X = np.array([[[1, 2, 3]],[[4, 5, 6]], [[7, 8, 9]]])
>>> y_univariate = np.array([11, 12, 13])
>>> pairwise_distance(X, y_univariate, metric='dtw')
array([[300.],
       [147.],
       [ 48.]])