get_clusterer_by_name

tsml_eval.experiments.get_clusterer_by_name(clusterer_name, random_state=None, n_jobs=1, fit_contract=0, checkpoint=None, data_vars=None, row_normalise=False, **kwargs)

Return a clusterer matching a given input name.

Basic way of creating a clusterer to build using the default or alternative settings. This set up is to help with batch jobs for multiple problems and to facilitate easy reproducibility through run_clustering_experiment.

Generally, inputting a clusterer class name will return said clusterer with default settings.

Parameters:

clusterer_namestr

String indicating which clusterer to be returned.

random_stateint, RandomState instance or None, default=None

Random seed or RandomState object to be used in the clusterer if available.

n_jobs: int, default=1

The number of jobs to run in parallel for both clusterer fit and predict if available. -1 means using all processors.

fit_contract: int, default=0

The number of data points to use in the clusterer fit if available.

checkpoint: str, default=None

Checkpoint to save model

data_vars: list, default=None

List of arguments to load the dataset using tsml_eval.utils.experiments import load_experiment_data.

row_normalise: bool, default=False

Whether to row normalise the data if it is loaded using data_vars.