Re: Help: model output
2010-08-31 23:59:03 GMT
The three commands should output the same model. See page 38 of the manual (the manual is refering to using explorer, but the same principle should apply):
"Note: No matter which evaluation method is used, the model that is output is
always the one build from all the training data"
If you're still curious, read the output of performing your commands. There should be a representation of the tree in human readable format under a heading "J48 pruned tree".
Or, the explanation of the -d option may help:
The model after training can be saved via this parameter. Each classifier has a different binary format for the model, so it can only be read back by the exact same classifier on a compatible dataset. Only the model on the training set is saved, not the multiple models generated via cross-validation.
Best of luck,
2010/8/31 TanAmy <amytanxy <at> hotmail.com>:
> Hi All,
> I was a little confused about the model output when I was using weka
> building classifiers.
> I generated J48 classifiers by using the following three commands. I'd like
> to know whether model 1,2,3 are the same or not. What are the differences
> between them. Thanks!
> weka.classifiers.trees.J48 -t input.arff -d 1.model
> weka.classifiers.trees.J48 -t input.arff -d 2.model -T test.arff
> weka.classifiers.trees.J48 -t input.arff -d 3.model -split-percentage 66
> Best Regards!
> Wekalist mailing list
> Send posts to: Wekalist <at> list.scms.waikato.ac.nz
> List info and subscription status:
> List etiquette:
_______________________________________________ Wekalist mailing list Send posts to: Wekalist <at> list.scms.waikato.ac.nz List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html