G-Means

Implementation of the G-Means algorithm for learning k in a k-means clustering. Paper published in NIPS 2003

Parameters

• min_obs: the minimum number of observations that a cluster can have
• max_depth: the maximum number of times that a cluster can be split before giving up (just set this to be high, e.g. 200 or so)
• random_state: the random seed that sklearn.MiniBatchKMeans() uses
• strictness: how strict should the Anderson-Darling test for normality be - 0 is least strict, 4 is most strict (best to be either 3 or 4, since the test is run so many times)

Usage

gmeans = GMeans(min_obs=100,
	max_depth=500,
	random_state=1010,
	strictness=3)
gmeans.fit(iris)
gmeans.labels_