[pymvpa] pprocess and number of CPU cores?

Matthias Ekman Matthias.Ekman at nf.mpg.de
Tue Apr 6 13:10:59 UTC 2010

Yaroslav Halchenko wrote:
> On Tue, 06 Apr 2010, Matthias Ekman wrote:
>>> if you believe that GNB (gaussian naive bayes classifier) is good for
>>> you (i.e. effects aren't strongly multivariate), you can do full brain
>>> searchlight cross-validation in a matter of a few seconds.
>> wow -- this pretty beast is really fast ;-) ... no need to buy a new
>> server anymore. Thanks for saving us 5000$ ;-)
> Sorry for shopping demotivation.
> But keep in mind -- this is GN(aive)B, so you might find results inferior
> to using generic searchlight with not-so-naive multivariate learners ;)

thanks! I was just kidding anyway ;-)


More information about the Pkg-ExpPsy-PyMVPA mailing list