`scipy.cluster._vq:update_cluster_means`

Signature

def update_cluster_means ( obs , labels , nc )

Summary

The update-step of K-means. Calculate the mean of observations in each cluster.

Parameters

obs : ndarray: The observation matrix. Each row is an observation. Its dtype must be float32 or float64.
labels : ndarray: The label of each observation. Must be a 1d array.
nc : int: The number of centroids.

Returns

cb : ndarray: The new code book.
has_members : ndarray: A boolean array indicating which clusters have members.

Notes

The empty clusters will be set to all zeros and the corresponding elements in has_members will be False. The upper level function should decide how to deal with them.

Aliases

scipy.cluster._vq.update_cluster_means