Skip to contents

The method run launches LibBi with a particular set of command line arguments. Normally, this function would not be run by the user, but instead one of the client functions sample, filter, or optimise, or rewrite, which pass any options on to run. Note that any options specified here are stored in the libbi object and do not have to be specified again if another command is run on the object.

Usage

# S3 method for libbi
run(
  x,
  client,
  proposal = c("model", "prior"),
  model,
  fix,
  config,
  log_file_name = character(0),
  init,
  input,
  obs,
  time_dim = character(0),
  coord_dims = list(),
  thin,
  output_every,
  chain = TRUE,
  seed = TRUE,
  debug = FALSE,
  ...
)

Arguments

x

a libbi object; if this is not given, an empty libbi object will be created

client

client to pass to LibBi

proposal

proposal distribution to use; either "model" (default: proposal distribution in the model) or "prior" (propose from the prior distribution)

model

either a character vector giving the path to a model file (typically ending in ".bi"), or a bi_model object; by default, will use any model given in x

fix

any variable to fix, as a named vector

config

path to a configuration file, containing multiple arguments

log_file_name

path to a file to text file to report the output of LibBi; if set to an empty vector (character(0)) or an empty string (""), which is the default, a temporary log file will be generated

init

initialisation of the model, either supplied as a list of values and/or data frames, or a (netcdf) file name, or a libbi object which has been run (in which case the output of that run is used). If the object given as x has been run before, it will be used here with init-np set to the last iteration of the previous run, unless init is given explicitly.

input

input of the model, either supplied as a list of values and/or data frames, or a (netcdf) file name, or a libbi object which has been run (in which case the output of that run is used as input)

obs

observations of the model, either supplied as a list of values and/or data frames, or a (netcdf) file name, or a libbi object which has been run (in which case the output of that run is used as observations)

time_dim

The time dimension in any R objects that have been passed (init, input) and obs); if NULL (default), will be guessed from the given observation

coord_dims

The coord dimension(s) in any obs R objects that have been passed; if NULL (default), will be guessed from the given observation file given

thin

any thinning of MCMC chains (1 means all will be kept, 2 skips every other sample etc.); note that LibBi itself will write all data to the disk. Only when the results are read in with bi_read will thinning be applied.

output_every

real; if given, noutputs will be set so that there is output every output_every time steps; if set to 0, only generate an output at the final time

chain

logical; if set to TRUE and x has been run before, the previous output file will be used as init file, and init-np will be set to the last iteration of the previous run (unless target=="prediction"). This is useful for running inference chains.

seed

Either a number (the seed to supply to LibBi), or a logical variable: TRUE if a seed is to be generated for RBi, FALSE if LibBi is to generate its own seed

debug

logical; if TRUE, print more verbose messages and write all variables to the output file, irrespective of their setting of 'has_output'

...

list of additional arguments to pass to the call to LibBi. Any arguments starting with `enable`/`disable` can be specified as boolean (e.g., `assert=TRUE` or `cuda=TRUE`). Any `dry-` options can be specified with a `"dry"` argument, e.g., `dry="parse"`. Any options that would be specified with `with`/`without` can be specified as character vector to an option named `with`/`without`, respectively, e.g. with="transform-obs-to-state".

Value

an updated libbi object, except if client is 'rewrite', in which case invisible NULL will be returned but the rewritten model code printed

See also