Runs R6Experiment in parallel — R6Experiment

This function is most useful to simulate the posterior distribution for a single model in parallel in one machine. This function is not used when calibrating the model and not useful for parallelization across multiple nodes.

Usage

R6Experiment_run(
  self,
  n_cores,
  parallel,
  cluster_eval_script = NULL,
  model_from_cluster_eval,
  cluster_type = "PSOCK",
  packages = NULL,
  ...
)

Arguments

self: experiment object
n_cores: number of cores to use
parallel: whether to evaluate run in parallel
cluster_eval_script: path to script that instantiates necessary functions. this will often mean sourcing functions external to the package and loading dependencies for the model. needed if parallel = T
model_from_cluster_eval: T if model is instantiated in the cluter eval scripts, F otherwise. Use T if using models that need compilation (like odin) and F otherwise.
cluster_type: either "FORK" or "PSOCK".
packages: character vector of packages to be loaded before running the model in paralle.
...: additional parameters to be passed to the model simulation function.

Value

results data.frame from all simulations in parallel