AggrDist

Panel distance from tabular distance aggregation.

Quickstart

from sktime.dists_kernels.compose_tab_to_panel import AggrDist

estimator = AggrDist(transformer, aggfunc=None, aggfunc_is_symm=False)

Parameters(3)

transformer: pairwise transformer of BasePairwiseTransformer scitype

aggfunc: aggregation function (2D np.array) -> float or None, optional

default = None = np.mean

aggfunc_is_symm: bool, optional, default=False

whether aggregation function is symmetric (should be set according to aggfunc) i.e., invariant under transposing argument, it always holds that aggfunc(matrix) = aggfunc(np.transpose(matrix))

Examples

Mean pairwise euclidean distance between between time series
>>> from sktime.dists_kernels import AggrDist, ScipyDist
>>> mean_euc_tsdist = AggrDist (ScipyDist ()) Mean pairwise Gaussian kernel between time series
>>> from sklearn.gaussian_process.kernels import RBF
>>> mean_gaussian_tskernel = AggrDist (RBF ())