Update a bmm model

Update an existing bmm mode. This function calls brms::update.brmsfit(), but it applies the necessary bmm postprocessing to the model object before and after the update.

Usage

# S3 method for class 'bmmfit'
update(
  object,
  formula.,
  newdata = NULL,
  recompile = NULL,
  file = NULL,
  file_compress = TRUE,
  ...
)

Arguments

object

An object of class bmmfit

formula.

A bmmformula(). If missing, the original formula is used. Currently you have to specify a full bmmformula

newdata

An optional data frame containing the variables in the model

recompile

Logical, indicating whether the Stan model should be recompiled. If NULL (the default), update tries to figure out internally, if recompilation is necessary. Setting it to FALSE will cause all Stan code changing arguments to be ignored.

file

Either NULL or a character string. If a string, the updated model is saved via saveRDS in a file named after the string, as in bmm(). update() never writes to the file the original fit was read from: pass file explicitly to save the updated fit.

file_compress

Logical or a character string, specifying one of the compression algorithms supported by saveRDS when saving the updated model object.

...

Further arguments passed to brms::update.brmsfit()

Value

An updated bmmfit object refit to the new data and/or formula. If file is given, it names the newly written file. If not, the file field is carried over from the original fit, so the updated object still points at the file it came from – but that file is not rewritten: it still holds the fit as it was before this update. Pass file to save the updated fit.

Details

When updating a brmsfit created with the cmdstanr backend in a different R session, a recompilation will be triggered because by default, cmdstanr writes the model executable to a temporary directory. To avoid that, set option "cmdstanr_write_stan_file_dir" to a nontemporary path of your choice before creating the original bmmfit.

For more information and examples, see brms::update.brmsfit()

Examples

if (FALSE) { # isTRUE(Sys.getenv("BMM_EXAMPLES"))
# generate artificial data from the Signal Discrimination Model
# generate artificial data from the Signal Discrimination Model
dat <- data.frame(y = rsdm(2000))

# define formula
ff <- bmf(c ~ 1, kappa ~ 1)

# fit the model
fit <- bmm(
  formula = ff,
  data = dat,
  model = sdm(resp_error = "y"),
  cores = 4,
  backend = "cmdstanr"
)

# update the model
fit <- update(fit, newdata = data.frame(y = rsdm(2000, kappa = 5)))
}