Back to models
Transformer

ClustererAsTransformer

API referenceSource
MultivariateMissing valuesRemoves missing valuesUnequal-length series

Use a clusterer as a transformer.

This adapter is used in coercions, when passing a clusterer to a transformer slot.

The transformation is series-to-primitives, transforming a time series into its cluster assignment.

The adapter dispatches BaseTransformer.transform to BaseClusterer.predict, and requires a clusterer that is able to make cluster assignments via predict, see the capability:predict tag for clusterers.

Quickstart

python
from sktime.clustering.compose import ClustererAsTransformer

estimator = ClustererAsTransformer(clusterer)

Tags

Capabilities

Properties

Input typescitype:transform-input
Series
Output typescitype:transform-output
Primitives
Label typescitype:transform-labels
None
Fit is emptyfit_is_empty
No
Keeps the time indextransform-returns-same-time-index
Yes
Requires Xrequires_X
Yes
Requires yrequires_y
No
X and y need the same indexX-y-must-have-same-index
No

Parameters(1)

clusterersktime clusterer, i.e., estimator inheriting from BaseClusterer

this is a “blueprint” clusterer, state does not change when fit is called

Examples

>>> from sktime.clustering.compose import ClustererAsTransformer
>>> from sktime.clustering.dbscan import TimeSeriesDBSCAN
>>> from sktime.dists_kernels import AggrDist
>>> from sktime.datasets import load_unit_test
>>> X, _ = load_unit_test (split = "train")
>>> clusterer = TimeSeriesDBSCAN (AggrDist. create_test_instance ())
>>> cluster_assign_trafo = ClustererAsTransformer (clusterer)
>>> cluster_assign_trafo. fit (X) ClustererAsTransformer(
... )
>>> cluster_assignment = cluster_assign_trafo. transform (X)