Hello, I would like to ask how to set different optimizers for other complex nonlinear systems. For example, I am currently trying to use a Koopman neural network for linearized learning on the Duffing oscillator system, and using the LBFGS optimizer for optimization. The first optimization is very slow, and the results obtained are not very good. Then I switched to Adam for optimization, but the learned Koopman operator results are also not good. How should I adjust it?