General purpose model fit function, called by higher order pr_fit_comaprison() and pr_cross_validate() functions. In itself a wrapper around the pr_fit_parameter() function.
pr_fit(
model = "TT",
data = phenor::phenocam_DB,
method = "GenSA",
control = list(max.call = 2000),
par_ranges = system.file("extdata", "parameter_ranges.csv", package = "phenor",
mustWork = TRUE),
plot = TRUE,
...
)Arguments
- model
the model name to be used in optimizing the model
- data
dataset generated by the format_phenocam() or format_modis() routines, or adhering to the general model optimization input format.
- method
optimization method to use (default = GenSA) - GenSA : Generalized Simulated Annealing algorithm - genoud : GENetic Optimization Using Derivatives - BayesianTools: various bayesian based optimization tools
- control
list of control parameters to be passed to the optimizer
- par_ranges
a vector of starting parameter values (function specific) defaults to the parameter ranges as provided with the current models and set forth by Basler (2016)
- plot
TRUE / FALSE, plot model fit
- ...
additional control parameters to be passed
Details
This function together with pr_predict() is in many ways the core of the package.
Examples
if (FALSE) { # \dontrun{
pr_fit(model,
par_ranges = "parameter_ranges.csv")
} # }