Clustering

The aeon.clustering module contains algorithms for time series clustering.

All clusterers in aeon can be listed using the aeon.registry.all_estimators utility, using estimator_types=”clusterer”, optionally filtered by tags. Valid tags can be listed using aeon.registry.all_tags.

Clustering Algorithms

TimeSeriesKMeans([n_clusters, init, ...])	Time series K-means clustering implementation.
TimeSeriesKMedoids([n_clusters, init, ...])	Time series K-medoids implementation.
TimeSeriesKShape([n_clusters, init, n_init, ...])	Kshape algorithm: wrapper of the tslearn implementation.
TimeSeriesKernelKMeans([n_clusters, kernel, ...])	Kernel K Means [R3e7c374b18c1-1]: wrapper of the tslearn implementation.
TimeSeriesCLARA([n_clusters, init, ...])	Time series CLARA implementation.
TimeSeriesCLARANS([n_clusters, init, ...])	Time series CLARANS implementation.
ElasticSOM([n_clusters, distance, init, ...])	Elastic Self-Organising Map (SOM) clustering algorithm.
KSpectralCentroid([n_clusters, max_shift, ...])	K-Spectral Centroid clustering implementation.

Deep learning

AEFCNClusterer([estimator, ...])	Auto-Encoder based Fully Convolutional Network (FCN), as described in [R4f194d8f8b22-1].
AEResNetClusterer([estimator, ...])	Auto-Encoder with Residual Network backbone for clustering.
AEDCNNClusterer([estimator, ...])	Auto-Encoder based Dilated Convolutional Networks (DCNN), as described in [1]_.
AEDRNNClusterer([estimator, ...])	Auto-Encoder based Dilated Recurrent Neural Network (DRNN).
AEAttentionBiGRUClusterer([estimator, ...])	Auto-Encoder based the Attention Bidirectional GRU Network.
AEBiGRUClusterer([estimator, ...])	Auto-Encoder based Bidirectional GRU Network.

Feature-based

Catch22Clusterer([features, catch24, ...])	Canonical Time-series Characteristics (catch22) clusterer.
SummaryClusterer([summary_stats, estimator, ...])	Summary statistic clusterer.
TSFreshClusterer([default_fc_parameters, ...])	Time Series Feature Extraction based on Scalable Hypothesis Tests clusterer.

Compose

ClustererPipeline(transformers, clusterer[, ...])

Pipeline of transformers and a clusterer.

Averaging

elastic_barycenter_average(X[, distance, ...])	Compute the barycenter average of time series using a elastic distance.
mean_average(X, **kwargs)	Compute the mean average of time series.
petitjean_barycenter_average(X[, distance, ...])	Compute the barycenter average of time series using a elastic distance.
subgradient_barycenter_average(X[, ...])	Compute the stochastic subgradient barycenter average of time series.
shift_invariant_average(X[, initial_center, ...])

Dummy

DummyClusterer([strategy, n_clusters, ...])

Dummy clustering for benchmarking purposes.

Base

BaseClusterer()

Abstract base class for time series clusterers.

BaseDeepClusterer([estimator, batch_size, ...])

Abstract base class for deep learning time series clusterers.