Changelog

Version 0.3.0

• Fix bug with random initialization of KMedoids [#129].

• KMedoids supports array-like init method [#137].

• Add a stopping criterion and parameter tuning heuristic for Huber robust mean

  estimator.

• Add CLARA (Clustering for Large Applications) which extends k-medoids to

  be more scalable using a sampling approach. [#83].

• Fix _estimator_type for robust estimators. Fix misbehavior of scikit-learn’s cross_val_score and GridSearchCV for RobustWeightedClassifier.

Version 0.2.0

April 14, 2021

• Add RobustWeightedClassifier, RobustWeightedRegressor and RobustWeightedKMeans estimators that rely on iterative reweighting of samples to be robust to outliers. [#42].

• Added Common Nearest-Neighbors clustering estimator CommonNNClustering [#64]

• Added PAM algorithm to KMedoids with method="pam" parameter which produces better solutions but at higher computational cost [#73]

• Binary wheels were uploaded to PyPi, making the installation possible without a C compiler [#66]

List of contributors (in alphabetical order)

Christos Aridas, Jan-Oliver Joswig, Timothée Mathieu, Roman Yurchak