For now, the target of a penalty involving scaled variables should be scaled by the user directly.
It would be easy to implement it for minimize_states, controls and parameters, but it is more complex for other penalties.
But we should still find a way to do it for the user, as it is super unintuitive.
For now, the target of a penalty involving scaled variables should be scaled by the user directly. It would be easy to implement it for minimize_states, controls and parameters, but it is more complex for other penalties. But we should still find a way to do it for the user, as it is super unintuitive.