finmod

These are the ones used in the Robust and efficient parameter estimation in dynamic models of biological systems paer and additional files 3 and 4. The coding and data files...

@KirillZubov Can I suggest a PINO-based spatiotemporal MFG example and test https://www.mdpi.com/2227-7390/12/6/803 @ChrisRackauckas Also the excellent Sophon.jl https://yichengdwu.github.io/Sophon.jl/dev/ @YichengDWu for interation with MTK and other SciML packages

Now it is published as updated but runtests genetic_algorithms still give an error.

@ChrisRackauckas Please advise on the sequence: - PR to put https://github.com/finmod/DiffEqParamEstim.jl/blob/master/test/lorenz%20true%20test.ipynb on DiffEqBenchmarks and then insert CMAE-ES in the family of available optimizers or - sort out first the best...

@ChrisRackauckas I notice that CMAE-ES is also available at Evolutionary.jl. It is in one of your runtests somewhere?

@ChrisRackauckas The benchmark for parameter estimation of the lorenz system from a dataset is available here: https://github.com/finmod/DiffEqParamEstim.jl/blob/master/test/parameter%20estimation%20benchmark%20for%20lorenz%20system.ipynb Can you please check that the set up of the data and model...

EM isn't possible for this because of the nonlinear transformation. SAEM (a stochastic approximation of the likelihoods) is, and should be considered (good paper). MAP is a good idea too...

In fact, I am trying to run cmaes.jl on the bench. For this input: optimizers = [ OptimRestart(NelderMead()), CMAESoptim, NelderMead(), BlackBoxOptimMethod(:adaptive_de_rand_1_bin_radiuslimited), BlackBoxOptimMethod(:adaptive_de_rand_1_bin), #NLoptOptimMethod(:LN_BOBYQA), # fails too #NLoptOptimMethod(:LN_PRAXIS), #this guy often...

Thank you for the prompt reply. For now, I will wait until the dust settles. It is clear that svae has a vocation to be integrated as an autograd example/model...

https://github.com/SciML/ModelingToolkit.jl/issues/860#event-7316091627. Will `expand_derivatives` solve this issue?