Pre-load all neural network models into the cache for faster subsequent predictions. This is useful when you want to minimize latency for the first prediction call.
Examples
if (FALSE) { # \dontrun{
# Pre-load models at startup
preload_all_models()
# Now predictions will be faster
results <- run_minter_scenarios(...)
} # }