How do I include multi-dimensional data in fitcsvm for SVM training

Hey chaps,

So I have 3-dimensional data at the moment i.e. x=(x_1, x_2, x_3). Each data point vector x=(x_1, etc.) has a label -1 or 1. When I try and include the 1-dimensional data (0nly one x) the training algorithm "fitcsvm" works fine. If I try and use [x_1, x_2, x_3] (3-dimensional), I get the following output:

"Error using classreg.learning.FullClassificationRegressionModel.prepareDataCR (line 137) X and Y do not have the same number of observations."

any ideas guys/gals? thanks