Orange Forum • View topic - Orange + SVM Classifier error

Orange + SVM Classifier error

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

Orange + SVM Classifier error

Postby w4k1ng » Sat May 24, 2014 14:03

Hi I have used Orange to classify text with a basic dataset and now I'm using a much bigger set to classify into different classes. The other classifier's have worked to some degree but SVM is the one I want. Every time I connect it up it brings up this log:

KernelException Traceback (most recent call last): File "C:\Python27\lib\site-packages\Orange\OrangeCanvas\scheme\widgetsscheme.py", line 643, in process_signals_for_widget widget.handleNewSignals() File "C:\Python27\lib\site-packages\Orange\OrangeWidgets\Classify\OWSVM.py", line 190, in handleNewSignals self.applySettings() File "C:\Python27\lib\site-packages\Orange\OrangeWidgets\Classify\OWSVM.py", line 213, in applySettings self.classifier=self.learner(self.data) File "C:\Python27\lib\site-packages\Orange\classification\svm__init__.py", line 197, in call return self.learn_classifier(examples) File "C:\Python27\lib\site-packages\Orange\classification\svm__init__.py", line 201, in learn_classifier data = self._normalize(data) File "C:\Python27\lib\site-packages\Orange\classification\svm__init__.py", line 272, in _normalize return data.translate(newdomain) KernelException: bad allocation

I have no idea what it means but presume there's something wrong in the dataset. Can anyone tell me what might be the problem and/or how to tackle it?

Thanks

Re: Orange + SVM Classifier error

Postby Ales » Wed May 28, 2014 10:47

This means the process is out of memory. See this thread for a possible solution.

Re: Orange + SVM Classifier error

Postby w4k1ng » Sat May 31, 2014 0:58

Thanks Ales.


Return to Questions & Support



cron