So AutoWekaClassifier also output a model that used RandomForest with *default* options?
Anyway, when you run AutoWekaClassifier in a cross-validation, it may not actually choose
the same learning algorithm or parameter combination for each training fold. The model
that is output (RandomForest in your case) is the one AutoWekaClassifier chooses for the
*full* dataset as loaded into the Preprocess panel. This is the same for all WEKA
classifiers, including AutoWekaClassifier.
Note that with only 80 instances, the variance in accuracy estimates obtained using
cross-validation will be very large. I doubt that any of your observed differences in
accuracy are statistically significant.
On 11/06/2018, at 2:44 AM, Abdrahman0x
Thank you for your Reply.
My data is around 80 instances and 3000 attributes.
For the parameters, I used for the Random Forest (manually) the default
Sent from: http://weka.8497.n7.nabble.com/
Wekalist mailing list
Send posts to: Wekalist(a)list.waikato.ac.nz
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html