[Scipy-tickets] [SciPy] #1854: scipy.stats.gaussian_kde.integrate functions are slow

SciPy Trac scipy-tickets@scipy....
Fri Mar 1 08:54:24 CST 2013


#1854: scipy.stats.gaussian_kde.integrate functions are slow
-------------------------------------------------+--------------------------
 Reporter:  itissid                              |       Owner:  rgommers   
     Type:  enhancement                          |      Status:  new        
 Priority:  normal                               |   Milestone:  Unscheduled
Component:  scipy.stats                          |     Version:  0.11.0     
 Keywords:  gaussian_kde, numerical integration  |  
-------------------------------------------------+--------------------------

Comment(by josefpktd):

 gaussian_kde is not designed to taking shortcuts. It always uses the full
 sample. This works well if there are few points, or the dimension is
 larger and points are spread out, but not for a sample like yours with a
 very large number of densely located points.

 I think, it would not be a good design to add optimized code for the 1-D
 case directly to the implementation of gaussian_kde.

 I don't know if there are smaller changes that could be made directly in
 integrate_box_1d to get a faster approximate solution, but it would still
 leave the basic problem that evaluating the density will be slow in large
 samples.

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1854#comment:4>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list