petitjean_barycenter_average

petitjean_barycenter_average(X: ndarray, distance: str = 'dtw', max_iters: int = 30, tol=1e-05, init_barycenter: ndarray | str = 'mean', weights: ndarray | None = None, precomputed_medoids_pairwise_distance: ndarray | None = None, verbose: bool = False, random_state: int | None = None, **kwargs) → ndarray

Compute the barycenter average of time series using a elastic distance.

This implements an adapted version of ‘petitjean’ (original) DBA algorithm [1].

Parameters:

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

(n_cases, n_timepoints)

A collection of time series instances to take the average from.

distance: str, default=’dtw’

String defining the distance to use for averaging. Distance to compute similarity between time series. A list of valid strings for metrics can be found in the documentation form aeon.distances.get_distance_function.

max_iters: int, default=30

Maximum number iterations for dba to update over.

tolfloat (default: 1e-5)

Tolerance to use for early stopping: if the decrease in cost is lower than this value, the Expectation-Maximization procedure stops.

init_barycenter: np.ndarray or, default=None

The initial barycenter to use for the minimisation. If a np.ndarray is provided it must be of shape (n_channels, n_timepoints). If a str is provided it must be one of the following: [‘mean’, ‘medoids’, ‘random’].

weights: Optional[np.ndarray] of shape (n_cases,), default=None

The weights associated to each time series instance, if None a weight of 1 will be associated to each instance.

precomputed_medoids_pairwise_distance: np.ndarray (of shape (len(X), len(X)),

default=None

Precomputed medoids pairwise.

verbose: bool, default=False

Boolean that controls the verbosity.

random_state: int or None, default=None

Random state to use for the barycenter averaging.

**kwargs

Keyword arguments to pass to the distance method.

Returns:

np.ndarray of shape (n_channels, n_timepoints)

Time series that is the average of the collection of instances provided.

References

[1]

F. Petitjean, A. Ketterlin & P. Gancarski. A global averaging method for dynamic time warping, with applications to clustering. Pattern Recognition, Elsevier, 2011, Vol. 44, Num. 3, pp. 678-693