[Mne_analysis] pyriemann.classification.TSclassifier

[JR KING]

        External Email - Use Caution        

Hi

CSP is a spatial filter, so it removes information. When there are many
classes, this can be a bad strategy.

HTH
JR

On Thu, 7 Mar 2019 at 16:32, Alexandre Gramfort <alexandre.gramfort at inria.fr>
wrote:

>         External Email - Use Caution
>
> hi,
>
> it's not using the same input features for the logistic regression.
> It's not surprising that pyriemann offers improved performance.
> See published literature on this topic.
>
> HTH
> Alex
>
> On Wed, Mar 6, 2019 at 11:43 PM Julian Long <julianlong988 at gmail.com>
> wrote:
>
>>         External Email - Use Caution
>>
>> Hi all,
>> Why pyriemann TS Classifier from pyriemann.classification.TSclassifier
>> which is logistic regression classifier performs better than the csp(common
>> spatial patterns)+logistic regression classifier especially in the case of
>> multi-classification?
>> _______________________________________________
>> Mne_analysis mailing list
>> Mne_analysis at nmr.mgh.harvard.edu
>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
>
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20190307/b7d63627/attachment-0001.html