[pymvpa] searchlight demo clarification

[Yaroslav Halchenko]

If you add following lines after call to searchlight, to print out
statistics of the map:

    # run searchlight on example dataset and retrieve error map
    sl_map = sl(dataset)

    import scipy
    print scipy.stats.describe(sl_map)

what does it spit out as output to the console

should be smth like
[SLC] DBG:  Doing 530 spheres: 530 (1 features) [100%]
(530, (0.092592592592592601, 0.62037037037037035), 0.48485324947589076, 0.0040453295402856134, array(-2.9434484598708481), 14.248522373545246)
Running searchlight with radius: 5 ...
[SLC] DBG:  Doing 530 spheres: 530 (3 features) [100%]
(530, (0.0092592592592592639, 0.63888888888888895), 0.40936408106219407, 0.015214333420764333, array(-1.2377745803617177), 1.0966798101276449)
Running searchlight with radius: 10 ...
[SLC] DBG:  Doing 530 spheres: 530 (8 features) [100%]
(530, (0.013888888888888895, 0.63888888888888895), 0.32408280922431854, 0.01858597917627762, array(-0.45317547552735754), -0.44109126548914901)

if it is similar, then problem just in plotting...

On Mon, 08 Dec 2008, Dornisch, John wrote:

>    Yes...
>    On 12/8/08 9:30 AM, "Yaroslav Halchenko" <[1]debian at onerussian.com>
>    wrote:

>      weird... btw -- did you mean searchlight_2d.py?

> Ссылки

>    1. file://localhost/tmp/debian@onerussian.com

-- 
Yaroslav Halchenko
Research Assistant, Psychology Department, Rutgers-Newark
Student  Ph.D. @ CS Dept. NJIT
Office: (973) 353-1412 | FWD: 82823 | Fax: (973) 353-1171
        101 Warren Str, Smith Hall, Rm 4-105, Newark NJ 07102
WWW:     http://www.linkedin.com/in/yarik