Hi All,
I have a collection of data that comprises of 2 classes, however, the number of samples for one class is much higher than the other.
I found that this will cause an error during the training of the SVM as written here
I know how to use the SVM vi's in LabVIEW, however I dont understand how can I weight the samples of one class more than the samples of the other class in IMAQ train SVM?
How do I use the parameter Weighted cost?
Do I just give a higher value of weighted cost for one class than the other? If yes, then how can I figure out how much should that higher
value be? Or is the weighted cost value set based on trial and error method?
Thanks,
Regards
Aveo