save.bigKRLS: save.bigKRLS

In bigKRLS: Optimized Kernel Regularized Least Squares

Description

save function, recommended when bigKRLS output contains big matrices (once N > 2,500 the kernel is stored this way). Base R data will be stored in a list in an .RData file, big matrices will be stored in .txt files. Call load.bigKRLS() to retrieve.

Usage

save.bigKRLS(object, model_subfolder_name, overwrite.existing = FALSE,
  noisy = TRUE)

Arguments

object	bigKRLS output (regression, prediction, and crossvalidation). Use load.bigKRLS(model_subfolder_name), not load().
model_subfolder_name	A name of a folder where the file(s) will be written.
overwrite.existing	Logical. If TRUE, write over folders with the same name. Default == FALSE.
noisy	Logical. If TRUE display progress, additional instructions. Default == TRUE.

Examples

# y <- as.matrix(ChickWeight$weight)
# X <- matrix(cbind(ChickWeight$Time, ChickWeight$Diet == 1), ncol = 2)

# out <- bigKRLS(y, X, Ncores=1)
# save.bigKRLS(out, "exciting_results") 

# don't use save() unless out$has.big.matrices == FALSE
# load.bigKRLS("/path/to/exciting_results") 

bigKRLS documentation built on Aug. 3, 2019, 1:02 a.m.