Predict from a fitted psvr_fit model — predict.psvr_fit • psvr

Predict from a fitted psvr_fit model

Usage

# S3 method for class 'psvr_fit'
predict(object, newdata, ...)

Arguments

object: An object of class "psvr_fit" from psvr().
newdata: Numeric matrix of new inputs, one observation per row (M × p).
...: Ignored.

Value

Numeric vector of length M with predicted values.

Examples

set.seed(1)
X <- matrix(rnorm(40), 20, 2)
y <- rlnorm(20)
fit <- psvr(X, y, loss = "rmspe", kernel = make_kernel("rbf", sigma = 1),
            gamma = 100)
predict(fit, X[1:3, , drop = FALSE])
#> [1] 0.9189556 0.7369731 0.8524386