[prev in list] [next in list] [prev in thread] [next in thread] 

List:       wekalist
Subject:    Re: [Wekalist] Help to a newbie
From:       Peter Reutemann <fracpete () waikato ! ac ! nz>
Date:       2005-11-24 21:19:10
Message-ID: 43862E4E.2030008 () waikato ! ac ! nz
[Download RAW message or body]

> Now I want the program to randomly choose randomly which
> trainingexamples to choose for training, I want say 5 trees, so 5
> randomly chosen sets of training examples.

You could do something like that in the Experimenter. Setup the Vote 
meta classifier with your base classifiers and use 5 iterations. The 
results are averaged over those 5 runs (normally one performs 10 
iterations with 10fold cross-validation). In each iteration/fold the 
training set (and the therefore also the test set) is generated randomly 
and presented to the meta classifier.

HTH

BTW don't forget to put the wekalist in the list of recipients...

Cheers, Peter
-- 
Peter Reutemann, Dept. of Computer Science, University of Waikato, NZ
http://www.cs.waikato.ac.nz/~fracpete/     +64 (7) 838-4466 Ext. 5174

_______________________________________________
Wekalist mailing list
Wekalist@list.scms.waikato.ac.nz
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
[prev in list] [next in list] [prev in thread] [next in thread]