Weka explorer feature selection ZeroR learning scheme

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Weka explorer feature selection ZeroR learning scheme

nestor
After performing a feature selection atributte wrapper random searcher
with:

Evaluator:    weka.attributeSelection.WrapperSubsetEval -B
weka.classifiers.rules.ZeroR -F 5 -T 0.01 -R 1 --
Search:weka.attributeSelection.RandomSearch -F 25.0 -seed 1

The explorer returns the following:

Attribute Subset Evaluator (supervised, Class (numeric): 5 Score):
     Wrapper Subset Evaluator
     Learning scheme: weka.classifiers.rules.ZeroR
     Scheme options:
     Subset evaluation: RMSE
     Number of folds for accuracy estimation: 5

How does it work as to perform an RMSE ZeroR learning scheme? I'm not
able to get the same features by simple calculation.

Best Regards

Nestor
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
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
Reply | Threaded
Open this post in threaded view
|

Re: Weka explorer feature selection ZeroR learning scheme

andria lan
Nestor, 

So you try to be cleaver by asking here?!! Dr Samer already answered your question, but you just ignore, and you don't even thank him. 

You are so bad!!!!




On Wed, Feb 5, 2020 at 11:14 AM nestor <[hidden email]> wrote:
After performing a feature selection atributte wrapper random searcher
with:

Evaluator:    weka.attributeSelection.WrapperSubsetEval -B
weka.classifiers.rules.ZeroR -F 5 -T 0.01 -R 1 --
Search:weka.attributeSelection.RandomSearch -F 25.0 -seed 1

The explorer returns the following:

Attribute Subset Evaluator (supervised, Class (numeric): 5 Score):
     Wrapper Subset Evaluator
     Learning scheme: weka.classifiers.rules.ZeroR
     Scheme options:
     Subset evaluation: RMSE
     Number of folds for accuracy estimation: 5

How does it work as to perform an RMSE ZeroR learning scheme? I'm not
able to get the same features by simple calculation.

Best Regards

Nestor
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
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

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
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
Reply | Threaded
Open this post in threaded view
|

Re: Weka explorer feature selection ZeroR learning scheme

Eibe Frank-3
In reply to this post by nestor
ZeroR does not use any predictor attributes and simply considers the distribution of class values for prediction (i.e., for classification, it implements the majority class classifier). If you apply wrapper-based subset selection with ZeroR as the base learning scheme (which is actually the default in WrapperSubsetEval), it will not select any attributes because the ZeroR model cannot be improved by adding predictor attributes: they are ignored anyway!

Cheers,
Eibe

On Wed, Feb 5, 2020 at 4:14 PM nestor <[hidden email]> wrote:
After performing a feature selection atributte wrapper random searcher
with:

Evaluator:    weka.attributeSelection.WrapperSubsetEval -B
weka.classifiers.rules.ZeroR -F 5 -T 0.01 -R 1 --
Search:weka.attributeSelection.RandomSearch -F 25.0 -seed 1

The explorer returns the following:

Attribute Subset Evaluator (supervised, Class (numeric): 5 Score):
     Wrapper Subset Evaluator
     Learning scheme: weka.classifiers.rules.ZeroR
     Scheme options:
     Subset evaluation: RMSE
     Number of folds for accuracy estimation: 5

How does it work as to perform an RMSE ZeroR learning scheme? I'm not
able to get the same features by simple calculation.

Best Regards

Nestor
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
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

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
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