What does the "maxiters" solver option do?

[Volker]

@ChrisRackauckas

like a long tspan

I have a question. How I can simulate/solve the ODE i.e. for quite long simulation duration? I.e. 2000s.

I ask, because I got this warning as well. Not during the optimization, but later when I just solved the ODE for a long tspan. I tried to increase the maxiter parameter, but then the solving didn´t finish (I let it run in the background for some hours). I could imagine, that simply stacking smaller solvings would work, but I think there is perhaps a more comfortable way, which I don´t know.

I got the warning in the context of the problem, which is described under the following link.

Fitting a dynamic system with an exogenous input (nonhomogenous neural ode) via DiffEqFlux

I wouldn’t use ReverseDiff or inplace here. Another mistake that you made was that, notice that the solution is a row vector for each output component, so you were broadcasting a row vector against a column vector in the loss function to generate a matrix. That’s not what you wanted. Here’s a code that trains: using DifferentialEquations, Flux, Optim, DiffEqFlux, DiffEqSensitivity, Plots tspan = (0.1f0, Float32(10.0)) tsteps = range(tspan[1], tspan[2], length = 100) t_vec = collect(tsteps) ex…