Got an error when re-training SVM.
I have already trained and saved a SVM classifier file called svm.yml. But when I tried to add one more category to that, this error came out: Bad argument (There is only a single class) in cvPreprocessCategoricalResponse
Does the svm train support a "re-training" in which I append or add additional training vectors and labels to it? I really need to know that. Any help is appreciated. Thanks.
"Does the svm train support a "re-training ?" - no, it clears all internal data, when you call train()
Do you know why this error appears? I have same problem. And i don't think it is related to re-training.
@waschbaer, please do not post answers, if you have a question or comment.
also, have a sharp look at it , again, it is not the same error !)