skactiveml.pool.k_greedy_center#

skactiveml.pool.k_greedy_center(X, y, batch_size=1, random_state=None, missing_label=nan, mapping=None, n_new_cand=None)[source]#

An active learning method that greedily forms a batch to minimize the maximum distance to a cluster center among all unlabeled datapoints.

Parameters:#

Xarray-like of shape (n_samples, n_features): Training data set, usually complete, i.e., including the labeled and unlabeled samples.
ynp.ndarray of shape (n_selected_samples, ): Index of datapoints already selected.
batch_sizeint, optional (default=1): The number of samples to be selected in one AL cycle.
random_stateNone or int or np.random.RandomState, default=None: Random state for candidate selection.
missing_labelscalar or string or np.nan or None, default=np.nan: Value to represent a missing label.
mappingNone or np.ndarray of shape (n_candidates,), default=None: Index array that maps candidates to X (candidates = X[mapping]).
n_new_candint or None, default=None: The number of new candidates that are additionally added to X. Only used for the case, that in the query function with the shape of candidates is (n_candidates, n_feature).