Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

ModelingToolkitNeuralNets #29

Open
AlCap23 opened this issue May 14, 2024 · 4 comments
Open

ModelingToolkitNeuralNets #29

AlCap23 opened this issue May 14, 2024 · 4 comments
Labels
enhancement New feature or request

Comments

@AlCap23
Copy link
Member

AlCap23 commented May 14, 2024

ModelingToolkitNeuralNetworks

This would be nice to include in the near future ⌛
@AlCap23 AlCap23 added the enhancement New feature or request label May 14, 2024
@AlCap23
Copy link
Member Author

AlCap23 commented May 14, 2024

@ChrisRackauckas is the API of MTKNN able to produce gradients and hessians w.r.t. to the parameters right now?

@ChrisRackauckas
Copy link

It's just a model builder. If the code it builds is differentiable then it is, and it just uses Lux and SciML solvers the standard way. There's examples of the generated models being fit in the docs.

@AlCap23
Copy link
Member Author

AlCap23 commented May 14, 2024

Ah, sorry I wasn't clear. Symbolic gradients and hessian 😅. I saw the wrapper , but was still unsure about the symbolic derivative part.

We've been exper with this using FastDiff right now, but it seems a customised chain rules / diff rule for the full network works better.

@ChrisRackauckas
Copy link

No matrix calculus in symbolics right now

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@ChrisRackauckas @AlCap23