[Scipy-tickets] [SciPy] #1774: Jaccard

SciPy Trac scipy-tickets@scipy....
Fri Dec 14 17:30:56 CST 2012


#1774: Jaccard
----------------------------+-----------------------------------------------
  Reporter:  vinigracindo   |       Owner:  somebody                
      Type:  defect         |      Status:  closed                  
  Priority:  high           |   Milestone:  Unscheduled             
 Component:  scipy.spatial  |     Version:  0.9.0                   
Resolution:  invalid        |    Keywords:  jaccard, distance, error
----------------------------+-----------------------------------------------
Changes (by pv):

  * status:  needs_info => closed
  * resolution:  => invalid


Comment:

 Ok, we are following here the conventional definition used e.g. in Octave
 and Matlab for the Jaccard distance of non-boolean arrays. I think the
 present behavior is intentional, even though it does not agree with the
 unordered set based definition of Jaccard distance.

 http://permalink.gmane.org/gmane.comp.python.scientific.devel/17098

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1774#comment:3>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list