Skip to content

Parrays 0.5 #178

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.

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

Draft
wants to merge 4 commits into
base: parrays-0.5
Choose a base branch
from
Draft

Parrays 0.5 #178

wants to merge 4 commits into from

Conversation

lamBOOO
Copy link

@lamBOOO lamBOOO commented Aug 4, 2025

Hi @JordiManyer,

I spent the last two days trying to make some progress, but I have to give up for now. Although I apparently fixed the PeriodicBCTests and the FESpaces (with heavy LLM usage 😅), I'm stuck at the MultiFieldTests and don't know what to do anymore.

I think that I would need way more time to understand the architecture properly instead of just trying to fix some errors blindly.

The progress in these four commits could also be complete BS, but I can't really tell without a deeper dive into the data structures. I just wanted to ping you and save this pseudo-progress somewhere.

@JordiManyer
Copy link
Member

Im glad to see LLMs will not take my job anytime soon 🤣

I will not commit the changes as is, but I will use them as learning to get a fix. Can you briefly explain what the issues for the periodic models was?

@lamBOOO
Copy link
Author

lamBOOO commented Aug 6, 2025

Yeah, I only had good experience with LLMs for coding if I knew exactly what to do for the promting. 😀

The issue with the periodic case was, that get_free_dof_values(u), used in Gridap.FESpaces.solve! didn't respect the "periodic structure" in the vector. I could fix it by overwriting the Gridap.FESpaces.solve! by using the similar function as done here: c1049e5#diff-8f937e22a7913b08b6cc1767736c1fd408cb88da2cd2ae15a5335f8ca5833fe4R851

Since overwriting functions produced a warning, the next commit tried to overload the get_free_dof_values directly for distributed FEFunctions. But I'm not 100% sure if that's the best way to deal with that.

Does that help?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants