[Scipy-tickets] [SciPy] #782: cluster.hierarchy.fcluster() returns strange results

SciPy scipy-tickets@scipy....
Thu Nov 6 10:18:14 CST 2008


#782: cluster.hierarchy.fcluster() returns strange results
---------------------------+------------------------------------------------
 Reporter:  vladan         |       Owner:  somebody
     Type:  defect         |      Status:  new     
 Priority:  normal         |   Milestone:  0.7.0   
Component:  scipy.cluster  |     Version:  devel   
 Severity:  normal         |    Keywords:          
---------------------------+------------------------------------------------
 cluster.hierarchy.fcluster() returns results that appear incorrect to me.
 This happens for 'inconsistent', 'maxclust' and 'distance' criteria (I
 didn't try 'monocrit'). The results are consistent, which might indicate
 that the criteria for the flat cluster generation are ok (type problem?).
 See the attached example.

 I'm using python 2.5, numpy 1.2.1 on a x86_64 machine.

-- 
Ticket URL: <http://scipy.org/scipy/scipy/ticket/782>
SciPy <http://www.scipy.org/>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list