Forum Message

 

 

We have moved the forum to https://forum.ngsolve.org . This is an archived version of the topics until 05/05/23. All the topics were moved to the new forum and conversations can be continued there. This forum is just kept as legacy to not invalidate old links. If you want to continue a conversation just look for the topic in the new forum.

Notice

The forum is in read only mode.

Data Race Condition while using AddElementMatrixSymmetric

More
5 years 3 months ago #1369 by jhauser
With this I get the error:
‘class std::shared_ptr<ngla::SparseMatrixSymmetric<double> >’ has no member named ‘SetZero’

I used
Code:
for(int i=0; i<ndof_y;i++) { FlatVector<double> mat_rowvals = maty->GetRowValues(i); for (int k=0; k<mat_rowvals.Size(); k++) { mat_rowvals[k]=0.0; } }
as a work around, but I still get a similar wrong matrix. Different every time I run the program.
At least with the python script. (I use python 3 to execute it) If I use "MyMaxwellAssembleSMALL" in jupyter notebook, I get the correct matrix. (which is printed in my terminal, when I use this function)
More
5 years 3 months ago #1370 by joachim
Code:
matrix -> SetZero()
if matrix is a shared pointer
More
5 years 3 months ago #1371 by jhauser
Oh, you're absolutely right. Thank you!
However the problem still remains. This doesn't change the every changing entries of the matrix. It's still different every time I run it.
More
5 years 3 months ago #1373 by christopher
With maty->SetZero() I get consistent results, can you repost the new version of your code?

(which is printed in my terminal, when I use this function)

This is because jupyter pipes only the python stdout and not the c++ one. You can use
Code:
pybind11::print(maty)
to get the output in jupyter (you need to include pybind11/pybind11.h to have that).
More
5 years 3 months ago #1375 by jhauser
I cleaned out the cmake-files of ml-ngs and build/compiled everything anew.
Now it works. I have absolutly no idea why it didn't work before since I haven't changed the code.
Thank you very much for your time! I am sorry, that it took so long.

And thank you for the jupyter suggestion.
Time to create page: 0.138 seconds