Orange Forum • View topic - SVM feature based scoring

SVM feature based scoring

A place to ask questions about methods in Orange and how they are used and other general support.

SVM feature based scoring

Postby avico510 » Wed Mar 14, 2012 3:28

I am trying to run the script from the example page (http://orange.biolab.si/doc/reference/O ... ation.svm/) for

SVM Based feature selection and scoring

I keep getting an error for the last line of the script " data_subset_of_features = rfe(table, 5)"

File "C:\Python27\Lib\site-packages\pythonwin\pywin\framework\scriptutils.py", line 325, in RunScript
exec codeObject in __main__.__dict__
File "C:\Python27\svm1.py", line 7, in <module>
data_subset_of_features = rfe(table, 5)
File "C:\Python27\lib\site-packages\orange\Orange\classification\svm\__init__.py", line 658, in __call__
scores = self.getAttrScores(data, progressCallback=progressCallback)
File "C:\Python27\lib\site-packages\orange\Orange\classification\svm\__init__.py", line 631, in getAttrScores
weights = getLinearSVMWeights(self.learner(data), sum=False)
TypeError: 'int' object is not callable

any ideas would be greatly appreciated. Thank you kindly

Re: SVM feature based scoring

Postby Ales » Wed Mar 14, 2012 11:29

The example works fine for me.
You didn't by any chance copy the example by hand, and changed the parameter to RFE from 'l' (as in learner) to '1'.


Return to Questions & Support