Dear all, Thank you for your ongoing support. It's quite amusing that every time I post a question here, I end up solving it myself—perhaps writing it out helps me think more clearly.
In any case, I've successfully managed to serialize and deserialize solutions. I serialized the block vector one block at a time using `<< solution.block(i);` for each block, and deserialized in the same manner with `>> saved_solution.block(i);` for each block. It is time now to use it and see how it goes. Feel free to reach out if you have any questions on this. Best regards, Najwa On Wednesday, December 4, 2024 at 8:37:04 AM UTC+3 Najwa Alshehri wrote: > Thank you for your response. > > For those who might face something similar, I found that the saved file > does not have the solution saved and the last thing that was saved is the > dof_handler (assuming that I can read the saved file :) ). It looks like > (<< solution; ) was not doing the job. > > It turned out that if I saved the solution block after the other, it > worked. "Worked" in the sense that I could see the solution in the saved > file. ( << solution.block(0); << solution.block(1); ). The solution in the > saved file matches the solution I wanted to save. > > I am not sure if this is the correct way to do it, but it seems like it is > because when I deserialize, I get a saved_solution with the correct size > and number of blocks. > > Unfortunately, I still have saved_solution, which is all zeros when I > deserialize, but I am working on it and will update you for the benefit of > others if I find a solution to it. > > Best, > Najwa > > > > > On Wednesday, December 4, 2024 at 7:57:59 AM UTC+3 Wolfgang Bangerth wrote: > >> On 12/3/24 02:32, Najwa Alshehri wrote: >> > >> > The exact same steps I used for elliptic problem with FE_Q and solution >> that >> > is written in vector. It was working perfectly. >> > >> > New things here, we are using fe system and block vector, others should >> be the >> > same. >> >> Najwa: >> This does not trigger any "oh, I know what Najwa's problem is" reaction >> in me. >> As I mentioned, create a minimal test case that illustrates the problem >> and >> nothing else -- then we can look at it. >> >> Best >> W. >> >> -- >> ------------------------------------------------------------------------ >> Wolfgang Bangerth email: bang...@colostate.edu >> www: http://www.math.colostate.edu/~bangerth/ >> >> >> -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. To view this discussion visit https://groups.google.com/d/msgid/dealii/7cdf7ffd-ffcc-4353-9856-1c9f71879d3cn%40googlegroups.com.
