Skip to contents

Creates candidate starting values evenly spread across the parameter bounds. Low-signal data gets more starts with the slope biased small.

Usage

generate_start_lists(model_constraints)

Arguments

model_constraints

Output of compute_model_constraints().

Value

Named list of lists: for each model, a list of named numeric vectors suitable as start for nlsLM().