Zurück zu den Modellen
Transformer

EAgglo

API-ReferenzQuellcode
MultivariatReihen ungleicher Länge

Hierarchical agglomerative estimation of multiple change points.

E-Agglo is a non-parametric clustering approach for multivariate timeseries[R7d39cf41170f-1]_, where neighboring segments are sequentially merged_ to maximize a goodness-of-fit statistic. Unlike most general purpose agglomerative clustering algorithms, this procedure preserves the time ordering of the observations.

This method can detect distributional change within an independent sequence, and does not make any distributional assumptions (beyond the existence of an alpha-th moment). Estimation is performed in a manner that simultaneously identifies both the number and locations of change points.

Schnellstart

python
from sktime.detection.eagglo import EAgglo

estimator = EAgglo(member=None, alpha=1.0, penalty=None)

Tags

Fähigkeiten

  • Multivariat
  • Reihen ungleicher Länge
  • Inverse Transformation: Nicht unterstützt
  • Fehlende Werte: Nicht unterstützt
  • Entfernt fehlende Werte: Nicht unterstützt
  • Gleicht Reihenlängen an: Nicht unterstützt

Eigenschaften

Eingabetypscitype:transform-input
Series
Ausgabetypscitype:transform-output
Series
Label-Typscitype:transform-labels
None
Fit ist leerfit_is_empty
Nein
Behält den Zeitindextransform-returns-same-time-index
Nein
Benötigt Xrequires_X
Ja
Benötigt yrequires_y
Nein
X und y brauchen denselben IndexX-y-must-have-same-index
Nein

Parameter(3)

memberarray_like (default=None)

Assigns points to the initial cluster membership, therefore the first dimension should be the same as for data. If None it will be initialized to dummy vector where each point is assigned to separate cluster.

alphafloat (default=1.0)

Fixed constant alpha in (0, 2] used in the divergence measure, as the alpha-th absolute moment, see equation (4) in [1].

penaltystr or callable or None (default=None)

Function that defines a penalization of the sequence of goodness-of-fit statistic, when overfitting is a concern. If None not penalty is applied. Could also be an existing penalty name, either len_penalty or mean_diff_penalty.

Beispiele

>>> from sktime.detection.datagen import piecewise_normal_multivariate
>>> X = piecewise_normal_multivariate (means = [[1, 3 ], [4, 5 ]], lengths = [3, 4 ],
... random_state = 10)
>>> from sktime.detection.eagglo import EAgglo
>>> model = EAgglo ()
>>> model. fit_transform (X) array([0, 0, 0, 1, 1, 1, 1])

Referenzen

[1] (1,2)

Matteson, David S., and Nicholas A. James. “A nonparametric approach for

multiple change point analysis of multivariate data.” Journal of the American Statistical Association 109.505 (2014): 334-345.

[2]

James, Nicholas A., and David S. Matteson. “ecp: An R package for

nonparametric multiple change point analysis of multivariate data.” arXiv preprint arXiv:1309.3295 (2013).