Replies: 1 comment 1 reply
-
yes, there is - The |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi,
I have a PDE model which features quite a complicated set of features, but I want to regress only a handful of unknown coefficients on to the exact known model.
If I define a general library which contains all the different terms in the PDE, and allows for all inputs, this results in a library with lots of terms and constraining the library explicitly after construction becomes too difficult as there are 1000s of terms. I really only need 10 or so in each equation.
It would be easier to limit the number of terms in each equation first using the generalized Library. So for instance, a term of the form x*y**0.5 would appear with the correct inputs only.
Is it possible to use the generalized library feature with the weak formulation? Or is there another method for constructing the exact known model?
Is it possible to construct different libraries for each equation?
Thanks in advance
Beta Was this translation helpful? Give feedback.
All reactions