Skip to contents

Fetch the last gradient created by running an likelihood. This errors if the last call to dust_likelihood_run did not use adjoint = TRUE. The first time you call this (after a particular set of parameters) it will trigger running the reverse model.

Usage

dust_likelihood_last_gradient(obj)

Arguments

obj

A dust_filter object, created by dust_filter_create or a dust_unfilter object created by dust_unfilter_create

Value

A vector (if ungrouped) or a matrix (if grouped).