By default it will pick the most recent error from NONMEM
as prompt, but can also ask questions.
Usage
luna_help(prompt = NULL, model = "gpt-4.1-mini", verbose = TRUE, ...)
Arguments
- prompt
question to ask LLM. By default NULL, which means it will
look at the log and extract the last event, e.g. an error from NONMEM
- model
LLM, e.g. "gpt-4.1-mini"
- verbose
verbose output?
- ...
optional parameters passed onto ellmer::chat()
Examples
if (FALSE) { # \dontrun{
luna_help() # will get help from LLM on last NONMEM error
luna_help("How do I write a model with non-linear elimination?")
} # }