Timeline for How to speed up NDSolve when there are ~400k equations?
Current License: CC BY-SA 4.0
35 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 24, 2019 at 7:11 | vote | accept | Lammond W | ||
| Dec 24, 2019 at 4:50 | comment | added | Lammond W | @xzczd Haha actually I didn't know that (the reason I needed to delete it is that my supervisor didn't want the codes to be public). I've post a new version of codes. | |
| Dec 24, 2019 at 4:44 | history | edited | Lammond W | CC BY-SA 4.0 | added 932 characters in body |
| Dec 23, 2019 at 3:23 | comment | added | xzczd♦ | Well, then do you know you code is still visible in the revision history? | |
| Dec 21, 2019 at 15:25 | comment | added | Lammond W | @xzczd Sorry! I'm trying to simplify the code as not to involve my research purposes. You know. My supervisor disliked it. I'll make it as soon as possible. | |
| Dec 21, 2019 at 14:02 | comment | added | xzczd♦ | Why do you remove the code sample? Indeed, it's not a good sample (not sufficiently simplified), but a question with no code sample is worse. | |
| Dec 21, 2019 at 7:08 | history | edited | Lammond W | CC BY-SA 4.0 | deleted 5496 characters in body |
| Dec 21, 2019 at 0:00 | history | tweeted | twitter.com/StackMma/status/1208175410523054082 | ||
| Dec 20, 2019 at 13:50 | comment | added | Michael E2 | It's possible you ran up against a time constraint to Solve for the derivatives. It can be set with "NDSolveOptions" -> "DefaultSolveTimeConstraint" -> time. See mathematica.stackexchange.com/a/131428/4999 for details. | |
| Dec 20, 2019 at 12:45 | answer | added | Alex Trounev | timeline score: 5 | |
| Dec 20, 2019 at 12:44 | comment | added | Chris K | I've had similar problems with large systems before. One thing you might try to identify the problem is breaking up the NDSolve into stages as described here. I even went so far as trying to rewrite NDSolve``ProcessEquations here; sometimes this helped, but never enough. When I'm really stuck, I end up resorting to VODE/FORTRAN, which is very fast but ugly. | |
| Dec 20, 2019 at 12:03 | comment | added | xzczd♦ | Don't miss those posts linked to that post, for example this one: mathematica.stackexchange.com/a/208770/1871 | |
| Dec 20, 2019 at 10:48 | comment | added | Lammond W | @xzczd {Sys,Bc} = PrepareEquations[tMin,tMax,NoC,NoN]; takes 8 mins to prepare the ~320k equations. Besides, I've read the post before but it didn't make sense to me then. I'll give it a careful look and thinking again. Thanks! | |
| Dec 20, 2019 at 10:25 | comment | added | xzczd♦ | How much time does {Sys,Bc} = PrepareEquations[tMin,tMax,NoC,NoN]; take? "So, what does the option Method->{"EquationSimplification"->"Residual"} really means?" We already have this post: mathematica.stackexchange.com/a/158519/1871 | |
| Dec 20, 2019 at 8:31 | history | edited | Lammond W | CC BY-SA 4.0 | added 47 characters in body |
| Dec 20, 2019 at 8:18 | history | edited | Lammond W | CC BY-SA 4.0 | deleted 74 characters in body |
| Dec 20, 2019 at 7:49 | history | edited | Lammond W | CC BY-SA 4.0 | added 171 characters in body |
| Dec 20, 2019 at 7:37 | history | edited | Lammond W | CC BY-SA 4.0 | added 2 characters in body |
| Dec 20, 2019 at 7:36 | comment | added | Lammond W | @xzczd Thanks for your reminding me. I've simplified the code to make it easy to understand for those who aren't familiar with material computing, so there may be some mistakes. Anyway, I've corrected the error. Thanks again! | |
| Dec 20, 2019 at 7:12 | comment | added | xzczd♦ | You code involves things like Cinit = 0.439673 % inside a Module, which is obviously incorrect. Please notice % doesn't represent percentage in Mathematica, it's the short form of Out. | |
| Dec 20, 2019 at 7:00 | history | edited | Lammond W | CC BY-SA 4.0 | deleted 16 characters in body |
| Dec 20, 2019 at 6:38 | history | edited | Lammond W | CC BY-SA 4.0 | added 3 characters in body |
| Dec 20, 2019 at 5:59 | history | edited | Lammond W | CC BY-SA 4.0 | added 4165 characters in body |
| Dec 20, 2019 at 2:57 | comment | added | Lammond W | @AlexTrounev Sorry to mislead you, and I will re-edit the description to make it clear right now. | |
| Dec 19, 2019 at 21:26 | comment | added | Alex Trounev | @LammondW The code contains many typos. I recommend formulating the problem in the form of equations. | |
| Dec 19, 2019 at 16:10 | history | edited | Lammond W | edited tags | |
| Dec 19, 2019 at 15:26 | history | edited | Lammond W | CC BY-SA 4.0 | added 1 character in body |
| Dec 19, 2019 at 14:22 | history | edited | Lammond W | CC BY-SA 4.0 | added 27 characters in body |
| Dec 19, 2019 at 14:15 | history | edited | Lammond W | CC BY-SA 4.0 | added 27 characters in body |
| Dec 19, 2019 at 14:05 | history | edited | Lammond W | CC BY-SA 4.0 | added 27 characters in body |
| Dec 19, 2019 at 13:50 | history | edited | Lammond W | CC BY-SA 4.0 | added 70 characters in body |
| Dec 19, 2019 at 13:28 | history | edited | Lammond W | CC BY-SA 4.0 | added 313 characters in body |
| Dec 19, 2019 at 13:22 | history | edited | Lammond W | CC BY-SA 4.0 | added 313 characters in body |
| Dec 19, 2019 at 13:00 | review | First posts | |||
| Dec 19, 2019 at 13:34 | |||||
| Dec 19, 2019 at 12:57 | history | asked | Lammond W | CC BY-SA 4.0 |