Fw: [mvpa-toolbox] Digest for mvpa-toolbox@googlegroups.com - 1Message in 1 Topic

[clair...@gmail.com]

Sorry for the message below - it was a mistake.
Claire

---

From: "Claire (Jung Eun) Han" <clair...@gmail.com>

Date: Tue, 20 Sep 2011 11:41:55 -0400

To: <mvpa-t...@googlegroups.com>

Subject: Re: [mvpa-toolbox] Digest for mvpa-t...@googlegroups.com - 1 Message in 1 Topic

can't believe you are still enlisted in this...!

On Mon, Sep 19, 2011 at 10:27 PM, <mvpa-t...@googlegroups.com> wrote:

  Today's Topic Summary

Group: http://groups.google.com/group/mvpa-toolbox/topics

• Erratic Classification [1 Update]

 Topic: Erratic Classification

MS Al-Rawi <raw...@yahoo.com> Sep 19 02:40AM -0700 ^

 
In addition to Greg's comments 
 
 
> Perhaps the data is somehow being overfitted?
 
In NNs, if the training error goes very low, you might have overfitting, you may try another classifier, e.g. Logistic Regression. Have you done detrending and z-scoring to your data? 
 
Regards,
Rawi


 

--
You received this message because you are subscribed to the Google Groups "Princeton MVPA Toolbox for Matlab" group.
To post to this group, send email to mvpa-t...@googlegroups.com.
To unsubscribe from this group, send email to mvpa-toolbox...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/mvpa-toolbox?hl=en.