Cross validation in Weka_Split the data set or not?

Good afternoon,
I have a confusion about the CV application in Weka. I will appreciate it
if you explain which method is right?

1) Should I first split my data as training and test set in Weka (I used
the stratified folds option to do that to create 80-20% sets) then apply 5
fold CV on my training set to tune parameters and then evaluate the
Classifier's performance using the test data by choosing 5 fold CV option?

2) Should I directly use the entire data set in Weka, choose the CV
option, tune parameters and use the entire data set again while presenting
the evaluation results?

3) Should I manually create different folds (e.g. 5 folds), get each
result one by one, and calculate the average?

Thanks a lot,



_______________________________________________
Wekalist mailing list -- [email protected]
Send posts to [email protected]
To unsubscribe send an email to [email protected]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html