GPCLASS Procedure

The GPCLASS procedure performs classification—a common step in data exploration—on quantitative data in SAS Viya.

This procedure performs binary classification by using Gaussian process classification (GPC), which is a method of probabilistic modeling that assumes that an underlying function is generated by a Gaussian process (GP) that determines the class labels of the data. As a nonparametric method, GPC allows the model parameters to adapt to the input data. To be specific, the Gaussian process adaptively learns the parameters of the latent function that lies in its corresponding reproducing kernel Hilbert space (RKHS), which is defined by the kernel of the Gaussian process, thus enabling good prediction performance.