-
Notifications
You must be signed in to change notification settings - Fork 199
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
Strength matrix #1234
base: master
Are you sure you want to change the base?
Strength matrix #1234
Conversation
* New routines to build an auxillliary M-matrix from A and used to construct S. * New user routine to access this capability with various options.
…e entries weighted by entries in A
…s to stored complexity info.
Hi @oseikuffuor1 and @waynemitchell . This branch is now up to date with master. I removed the cogent proxy from the Makefile (it doesn't build currently, and we probably don't want it in hypre anyway) and I added a one line description of the strength option to The code doesn't work correctly for the matrices I am testing, but I'm pretty sure it's because the matrices are negative definite and the current code looks for positive off-diagonal entries instead of same-sign-as-diagonal. I'll try to fix that. |
Thanks Rob!
…-Daniel
Daniel Osei-Kuffuor
Center for Applied Scientific Computing
Lawrence Livermore National Laboratory
Livermore, CA 94550
Mail Code: L-561
Phone: 925-422-7683
Email: ***@***.******@***.***>
________________________________
From: Rob Falgout ***@***.***>
Sent: Monday, February 10, 2025 1:48 PM
To: hypre-space/hypre ***@***.***>
Cc: Osei-Kuffuor, Daniel ***@***.***>; Mention ***@***.***>
Subject: Re: [hypre-space/hypre] Strength matrix (PR #1234)
Hi @oseikuffuor1<https://urldefense.us/v3/__https://github.com/oseikuffuor1__;!!G2kpM7uM-TzIFchu!xSkLDougQ-EqEZF3Jor9tfaYqzr3qs3KPXN-Qz9s30ihjC1TwxfGOiobKyhSuO7CYrjmN_9GjbP4obmMWSK1brpShnyUoQ$> and @waynemitchell<https://urldefense.us/v3/__https://github.com/waynemitchell__;!!G2kpM7uM-TzIFchu!xSkLDougQ-EqEZF3Jor9tfaYqzr3qs3KPXN-Qz9s30ihjC1TwxfGOiobKyhSuO7CYrjmN_9GjbP4obmMWSK1brqO9iejBQ$> . This branch is now up to date with master. I removed the cogent proxy from the Makefile (it doesn't build currently, and we probably don't want it in hypre anyway) and I added a one line description of the strength option to ij.c (it will need to be updated at some point).
The code doesn't work correctly for the matrices I am testing, but I'm pretty sure it's because the matrices are negative definite and the current code looks for positive off-diagonal entries instead of same-sign-as-diagonal. I'll try to fix that.
—
Reply to this email directly, view it on GitHub<https://urldefense.us/v3/__https://github.com/hypre-space/hypre/pull/1234*issuecomment-2649316543__;Iw!!G2kpM7uM-TzIFchu!xSkLDougQ-EqEZF3Jor9tfaYqzr3qs3KPXN-Qz9s30ihjC1TwxfGOiobKyhSuO7CYrjmN_9GjbP4obmMWSK1broWWDIAgQ$>, or unsubscribe<https://urldefense.us/v3/__https://github.com/notifications/unsubscribe-auth/AFMCRS7JXICSP2YFMGXEP3D2PENBXAVCNFSM6AAAAABW3PH236VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBZGMYTMNJUGM__;!!G2kpM7uM-TzIFchu!xSkLDougQ-EqEZF3Jor9tfaYqzr3qs3KPXN-Qz9s30ihjC1TwxfGOiobKyhSuO7CYrjmN_9GjbP4obmMWSK1brrdE_EQWQ$>.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
Okay, this seems to fix the negative diagonal case. The strength measure doesn't really help with the matrices I am testing though. :( This deserves a much closer investigation, of course. |
This branch implements a new strength measure idea.