DTWDistance

class DTWDistance(points_distance: typing.Callable[[float, float], float] = CPUDispatcher(<function simple_dist>), trim_series: bool = False)[source]

Bases: etna.clustering.distances.base.Distance

DTW distance handler.

Init DTWDistance.

Parameters
  • points_distance (Callable[[float, float], float]) – function to be used for computation of distance between two series’ points

  • trim_series (bool) – True if it is necessary to trim series, default False.

Notes

Specifying manual points_distance might slow down the clustering algorithm.

Inherited-members

Parameters
  • points_distance (Callable[[float, float], float]) –

  • trim_series (bool) –

Methods

get_average(ts, **kwargs)

Get series that minimizes squared distance to given ones according to the Distance.

set_params(**params)

Return new object instance with modified parameters.

to_dict()

Collect all information about etna object in dict.