Function to fit a given DDRep to a given DDModel

Fit_DDModel(
  model = NULL,
  data = NULL,
  DL_model = NULL,
  DL_scale = list(INPUT = list(Center = FALSE, Stddev = FALSE), OUTPUT = list(Center =
    FALSE, Stddev = FALSE)),
  grid_path = NULL,
  trials = 10000L,
  s_sampling = FALSE,
  simplex_struc = c(0L),
  simplex_coef = list(alpha = 1, beta = 0.5, gamma = 2, sigma = 0.5, tol = 1e-07, nfunc
    = 120L, nshrink = 3L)
)

Arguments

model	`DDModel` object
data	`DDRep` object or list of `DDRep` objects
DL_model	(optional) `Model` in the form of a keras neural network model
DL_scale	(optional) `list` containing mean and sd of the transformation used in the deep learning model while training.
grid_path	(optional) `path` to a directory containing a .GRID fileset. If NULL the model will be fitted using 20 randomly drawn startparametersets from the model-DOMAIN.
trials	`integer` indicating the number of trials used while fitting (s_sampling = FALSE) or the maximum number of trials used while super sampling (s_sampling = TRUE)
s_sampling	`bool` indicating super sampling while fitting
simplex_struc	`numeric vector` containing the number of simplex iterations per sorting cycle.
simplex_coef	`list` containing named parameters for the SIMPLEX

Value

DDFit object

Details

This is a rather complex function that is very flexible depending on your input!

Random Fitting:: Exclude all optional parameters
Grid Fitting:: Specify a grid_path
Deep Learning Fitting: Specify DL_model and DL_scale