Cross-validation query

[Chris Day]

Hi,

Hope someone can answer this quick question...

Using the tutorial as an example: when the cross-validation.m function
starts, does it use 10 different classifiers for each run, or does it
produce 10 instances of the same classifer for each run?

Or in other words, are the neural network weights set the same for all
runs before training/testing begins?

Hope I've explained myself clearly there!

Regards,

Chris

[MS Al-Rawi]

In cross-validation, a classifier (instance) is used for each of the n folds. 

---

From: Chris Day <loi...@gmail.com>
To: Princeton MVPA Toolbox for Matlab <mvpa-t...@googlegroups.com>
Sent: Friday, September 16, 2011 11:09 AM
Subject: [mvpa-toolbox] Cross-validation query

--
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+unsub...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/mvpa-toolbox?hl=en.