optim_control: Set up and control optimization parameters

In tramME: Transformation Models with Mixed Effects

Set up and control optimization parameters

Description

Set up and control optimization parameters

Usage

optim_control(
  method = c("nlminb", "BFGS", "CG", "L-BFGS-B"),
  scale = TRUE,
  trace = FALSE,
  ntry = 5,
  ok_warnings = "NA/NaN function evaluation",
  ...
)

Arguments

method	Optimization procedure.
scale	Logical; if TRUE rescale the fixed effects design matrix to improve convergence.
trace	Logical; print trace of the optimization.
ntry	Number of restarts with new random initialization if optimization fails to converge.
ok_warnings	Control for what warnings will be reported during optimization. If TRUE, no warnings will be reported, if FALSE all warnings are displayed. In case of a character vector, the matching warnings are treated as unimportant, and not reported. See also Notes.
...	Optional arguments passed to auglag, nlminb or optim as a list of control parameters.

Note

Irrespective of the value of the ok_warnings argument, all warnings are collected in the opt$warnings element of the tramME object.

tramME documentation built on July 2, 2024, 5:07 p.m.