[pymvpa] PCA transformation prior to SVM classification (Jakob Scherer)

Brian Murphy brian.murphy at unitn.it
Mon Nov 29 12:47:13 UTC 2010


Dear Jakob
> On Thu, Nov 25, 2010 at 5:40 PM, Andreas Mueller
> <amueller at ais.uni-bonn.de> wrote:
>   
>> Hi there.
>> First of: I don't think this is a PyMVPA question.
>> There are some machine learning communities like
>> http://metaoptimize.com/qa/ and http://agbs.kyb.tuebingen.mpg.de/km/bb/.
>>     
>
> I'll post questions like these there in the future. Thanks for
> pointing me towards these lists.
I think the PyMVPA list is the perfect place to discuss these issues! 
Many of the people using PyMVPA are new to machine learning, and may 
find it hard to express their problem in terms that machine learning 
experts understand and/or to interpret the answers. Also neuroscientists 
tend to ask slightly different questions than machine learners or BCI 
people - often our primary interest is understanding why we were able to 
decode, rather than just improving the raw prediction accuracy. And we 
also use some rather peculiar algorithms (e.g. a search for CSP comes up 
empty on metaoptimize). So I think we should take advantage of our 
particular community to exchange ideas and expertise,

Brian



More information about the Pkg-ExpPsy-PyMVPA mailing list