fit_opt: fit_opt

In gkane26/modelfitr: Toolbox to fit models to data

fit_opt

Description

nonlinear optimization: wrapper for optimx and nloptr

Usage

fit_opt(
  objective,
  start,
  lower = NULL,
  upper = NULL,
  hessian = FALSE,
  package = "optimx",
  method = NULL,
  restart = FALSE,
  consecutive = 0L,
  max_runs = Inf,
  sigma = 0.1,
  tol = 1e-05,
  verbose = FALSE,
  opt_args = list(),
  obj_args = list(),
  ...
)

Arguments

objective	function; the objective function to minimize
start	numeric vector; starting parameters
lower	numeric vector; lower bounds for parameters
upper	numeric vector; upper bounds for parameters
hessian	logical; if TRUE, calculate hessian at solution
package	string; package to use, see Details. Default = "optimx"
method	string; method to use from package
restart	logical; if TRUE, restart optimization until no more improvement. Default = FALSE
max_runs	integer; maximum number of optimx runs. Default = Inf
sigma	numeric; standard deviation for random perturbation of starting values
verbose	logical; if TRUE, print result summary after each iteration.Default=FALSE
opt_args	list; list of arguments to be passed to optimx
obj_args	list; list of arguments to be passed to objective function
...	further arguments objective
conescutive	integer; stop after this number of consecutive runs without improvement. Default = 0
tolerance	numeric; new fit must change objective by 'tolerance' to be considered an improvement

Details

Package options = "optimx", "optim", "nloptr", "pracma"

gkane26/modelfitr documentation built on March 21, 2022, 10:52 a.m.