R/cossonet.predict.R

In cossonet: Sparse Nonparametric Regression for High-Dimensional Data

#' The function `cossonet.predict` predicts predictive values for new data based on an object from the `cossonet` function.
#'
#' @param model The fitted cossonet object.
#' @param testx The new data set to be predicted.
#'
#' @return A list of predicted values for the new data set.
#'
#' @examples
#' \donttest{
#' set.seed(20250101)
#' tr = data_generation(n = 200, p = 20, SNR = 9, response = "continuous")
#' tr_x = tr$x
#' tr_y = tr$y
#'
#' te = data_generation(n = 1000, p = 20, SNR = 9, response = "continuous")
#' te_x = te$x
#' te_y = te$y
#'
#' # Fit the model
#' fit = cossonet(tr_x, tr_y, family = 'gaussian', gamma = 0.95, kernel = "spline", scale = TRUE,
#'       lambda0 = exp(seq(log(2^{-4}), log(2^{0}), length.out = 20)),
#'       lambda_theta = exp(seq(log(2^{-8}), log(2^{-6}), length.out = 20))
#'       )
#'
#' # Predict new dataset
#' pred = cossonet.predict(fit, te_x)
#' }
#' @export

cossonet.predict = function(model, testx)
{
  family = model$family
  if(family == "gaussian") obj = gaussian()
  if(family == "binomial") obj = binomial()
  if(family == "poisson") obj = poisson()

  nbasis = length(model$data$basis.id)
  te_n <- dim(testx)[1]

  if(class(testx)[1] == "data.frame") testx = matrix(unlist(testx), nrow = te_n)
  testx = apply(testx, 2, rescale)

  K = make_anovaKernel(testx, model$data$x[model$data$basis.id, ], model$data$kernel, model$data$kparam)
  d = K$numK

  R = array(NA, c(te_n, nbasis, d))
  for(j in 1:d){
    R[, , j] = K$K[[j]]
  }

  Rtheta <- combine_kernel(R, model$theta_step$theta.new/(model$data$wt^2))

  if(family != "Cox"){
    f.new = c(Rtheta %*% model$c_step$c.new + model$c_step$b.new)

    out = list(f.new = f.new, mu.new = obj$linkinv(f.new))
  }

  if(family == "Cox"){
    f.new = c(Rtheta %*% model$c_step$c.new)

    out = list(f.new = f.new)
  }

  rm(K)
  rm(R)
  rm(Rtheta)

  return(out)
}