Perform a quick, automated, TPC fit — quickfit

Performs a simple TPC fit using nls_multstart. This function tries to use a sensible default configuration, however if you want to use the more custom elements of nls_multstart then you will need to construct your own.

Usage

quickfit_tpc(data, model_name, temp, trait, start_adjusts = 0, iter, ...)

Arguments

data: the data to fit a model to
model_name: the model name as a string
temp: the column name (as a string) containing the temperature data
trait: the column name (as a string) containing the temperature data
start_adjusts: any adjustments to make to the lower and upper starting bounds. If 0 < start_adjusts < 1, this will be interpreted as a proportion of the base starting values.
iter: number of combinations of starting parameters which will be tried . If a single value is provided, then a shotgun/random-search/lhs approach will be used to sample starting parameters from a uniform distribution within the starting parameter bounds. If a vector of the same length as the number of parameters is provided, then a gridstart approach will be used to define each combination of that number of equally spaced intervals across each of the starting parameter bounds respectively. Thus, c(5,5,5) for three fitted parameters yields 125 model fits. Supplying a vector for iter will override convergence_count.
...: additional arguments to pass to nls_multstart.

Value

The nls model object of the fit model

Author

Francis Windram

Examples

if (FALSE) { # \dontrun{
data("chlorella_tpc")

d <- subset(chlorella_tpc, curve_id == 1)

quickfit_tpc(
  d,
  "briere1_1999",
  "temp",
  "rate",
  start_adjusts = 10,
  iter = 150
)

quickfit_tpc(
  d,
  "briere1_1999",
  "temp",
  "rate",
  start_adjusts = 0.8,
  iter = c(5,5,5)
)
} # }