search

ChrisRackauckas / universal_differential_equations

Repository for the Universal Differential Equations for Scientific Machine Learning paper, describing a computational basis for high performance SciML
https://arxiv.org/abs/2001.04385
MIT License
219 stars 59 forks source link

OptimizationFunction Definition #49

Closed ugccelik closed 2 years ago

ugccelik commented 2 years ago

Hello,

I was investigating the LotkaVolterra problem scripts and i saw that the optimization function is defined as follows;

optf = Optimization.OptimizationFunction((x,p)->loss(x), adtype)

i am wondering why loss function is given to optimization-function definition as : (x,p)->loss(x) in here while it is only depending on ''p'' ? Since, it isn't working when it is written as loss(p).

ChrisRackauckas commented 2 years ago

That p is the hyperparameters of the optimization, not the parameters of the model.

ugccelik commented 2 years ago

Thank you for your response. Is there any documentation about these hyper-parameters of the optimization, where can i find them. Since documentation about OptimizationFunction don't mention anything about it.

ChrisRackauckas commented 2 years ago

Good point. https://github.com/SciML/SciMLBase.jl/pull/243