Fast EM Clustering

Submitted by on Jul 09 2014 } Suggest Revision
By: Andrew Moore, Paul Hsiung, Peter Sand
From: Auton Lab, CMU
Paper:
Resource Type:
Code
License:
Other/Unknown
Language:
Data Format:

Description

EM is a popular method for finding mixture models. This implementation is based on multiresolution kd-trees, which dramatically reduces the cost of EM-based clustering, with savings rising linearly with the number of datapoints.
Post comment
Cancel