[pymvpa] Hyperalignment and Sensitivity Maps

Michael Hanke michael.hanke at gmail.com
Thu Dec 6 18:14:02 UTC 2012


On Thu, Dec 6, 2012 at 6:43 PM, Kimberly Zhou <kyqzhou at gmail.com> wrote:

> Hi All,
> My apologies in advance for the noob questions.
> I've been looking at hyperalignment in PyMVPA and was wondering if anyone
> else has been using this and is clear on how it works. Is there a way to
> get a sensitivity map after hyperalignment and more importantly, would that
> be a valid thing to do?

Hyperalignment will provide you with a transformation for each input
dataset that will align the feature space of the respective dataset with
the computed "common" space. In a sense the transformation matrix provides
you with information what original features contribute how much to each
feature in the common space. You could think of it as a sensitivity

Take a look at the Neuron paper -- it should give you some idea what you
can do with this tool.


GPG key:  1024D/3144BE0F Michael Hanke
Jabber: michael.hanke at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/attachments/20121206/0d289d6e/attachment.html>

More information about the Pkg-ExpPsy-PyMVPA mailing list