search

SciML / SciMLSensitivity.jl

A component of the DiffEq ecosystem for enabling sensitivity analysis for scientific machine learning (SciML). Optimize-then-discretize, discretize-then-optimize, adjoint methods, and more for ODEs, SDEs, DDEs, DAEs, etc.
https://docs.sciml.ai/SciMLSensitivity/stable/
Other
328 stars 70 forks source link

train a neural network for Parameter Estimation of Ordinary Differential Equations #1049

Closed yqlai1993 closed 4 months ago

yqlai1993 commented 4 months ago

is it possible to train a neural network for for Parameter Estimation of Ordinary Differential Equations? The neural network is not used in the example, but it seems to be used in the last image. Would you provide the full code？

ChrisRackauckas commented 4 months ago

I'm not sure what example you're referring to?

yqlai1993 commented 4 months ago

the first example in tutorials https://docs.sciml.ai/SciMLSensitivity/dev/tutorials/parameter_estimation_ode/

ChrisRackauckas commented 4 months ago

There are no neural networks in that example. Is there another tutorial you're thinking of?

yqlai1993 commented 4 months ago

There are no neural networks in that example. Is there another tutorial you're thinking of?

Flux.jl is used in the last image. how does it work?

ChrisRackauckas commented 4 months ago

That's an old picture, we should update it .

yqlai1993 commented 4 months ago

That's an old picture, we should update it .

thx