Machine Learning Labels

asked 2014-10-25 21:14:41 -0600

xiaox
1

Hi, there. I want to ask whether should I label my classes as +1, -1 for SVM/AdaBoost or it is not so important if I label my classes as 0, +1 or some other labels that can distinguish each class.

edit retag flag offensive close merge delete

add a comment

1 answer

Sort by » oldest newest most voted

answered 2014-10-31 09:19:20 -0600

ej
26 ●1 ●3

Not important.

From the User guide of CvStatModel:

Responses are usually stored in a 1D vector (a row or a column) of CV_32SC1 (only in the classification problem) or CV_32FC1 format, one value per input vector. Although, some algorithms, like various flavors of neural nets, take vector responses.

For classification problems, the responses are discrete class labels.