smile-kotlin/smile.regression/gpr

gpr

object gpr(source)

Gaussian Process for Regression.

Functions

Link copied to clipboard
fun <T> approx(    x: Array<T>,     y: DoubleArray,     t: Array<T>,     kernel: MercerKernel<T>,     noise: Double,     normalize: Boolean = true): GaussianProcessRegression<T>

Fits an approximate Gaussian process model with a subset of regressors.

Link copied to clipboard
fun <T> nystrom(    x: Array<T>,     y: DoubleArray,     t: Array<T>,     kernel: MercerKernel<T>,     noise: Double,     normalize: Boolean = true): GaussianProcessRegression<T>

Fits an approximate Gaussian process model with Nystrom approximation of kernel matrix.