Skip to contents

Save a model run configuration to a documented YAML file

Source: R/0-save_config.R
save_config.Rd

Round-trips a full run configuration, the model_config() structure plus the estimation controls and fit_control() bundle, to a git-diffable YAML file, with each field's documentation emitted as a comment and a spec-tree + provenance header. Only fields that differ from their defaults are written, so two configurations diff to just their real differences. The parameter values (inits/map/bounds) are NOT stored; pair the config with a saved fit for those.

Usage

save_config(x, file = "Rceattle_config.yaml", ...)

Arguments

x

A fitted Rceattle object, a data list holding $model_config, an Rceattle_run_config, or an Rceattle_model_config.

file

Output path for the .yaml file.

...

Estimation controls / fit_control to record (passed to run_config()).

Value

Invisibly, the Rceattle_run_config that was written.

See also

load_config(), run_config(), model_config(), fit_mod().

Examples

cfg <- model_config(msmMode = 1, initMode = "OffsetEquilibrium")
f <- file.path(tempdir(), "run.yaml")
save_config(cfg, f)
identical(load_config(f)$model_config$msmMode, 1)
#> [1] TRUE