I'm solving a system of stochastic differential equations in Julia:
#define drift fuction
F!(du,u,t,p)=...
#define noise function
G!(du,u,t,p)=...
#constructing and solving the problem
u_0=zeros(100,100)#initial condition
tspan=(0.0,10.0)
problem=SDEProblem(F!,G!,u_0,tspan,p)
solution=solve(problem,SOSRA(),reltol=1e-3,abstol=1e-3,save_everystep = false)
Now, I read here that you can use the keyword argument "progress" to get updates on the remaining number of seconds to complete the run, and you can get this information with a frequency set by the argument "progress_steps":
solution=solve(problem,SOSRA(),reltol=1e-3,abstol=1e-3,save_everystep = false,progress=true,progress_steps=1000)
However, I would be interested in getting more detailed updates from the solver, like current chosen $dt$ (for adaptive algorithm), current time $t$, and current maximum value of the solution $u_{max}$. I read here that this should be possible with "progress_message" at least for ODE's. However I don't know the precise syntax to make it work. Also, I would like to possibly print these updates into a .txt file rather than having them in the terminal. Is there a way to do that? Thanks!
I've tried
solution=solve(problem,SOSRA(),reltol=1e-3,abstol=1e-3,save_everystep = false, progress=true,progress_message=true)
But it gives an error
You likely want to use the lower level init and step! interface.
Specifically,
integrator = init(problem,SOSRA(),reltol=1e-3,abstol=1e-3)
step!(integrator) # takes 1 timestep
step!(integrator, .4) # steps until t==0.4