search

ArnauMiro / pyLowOrder

High performance parallel Low Order Modelling library by BSC - UPM
MIT License
3 stars 1 forks source link

KeyError: <ElementType.NSIDED: 'nsided'> and <ElementType.NSIDED: 'nfaced'> #35

Open shangwenqiang opened 2 months ago

shangwenqiang commented 2 months ago

Recently, I'm applying pyLowOrder Module to perfoming POD for a large dataset with Ensight Gold format. The ElementType in my data contains the 'nsided' and 'nfaced' types, however, the code can not recognize them. Is there any solution on that? Your response would be greatly appreciated.

ArnauMiro commented 2 months ago

Hello!

Our Ensight reader is legacy and very simple, we are based on the ensight-reader module on python (https://pypi.org/project/ensight-reader/). If they support your data then it can be adapted for pyLowOrder. In fact you can use ensight-reader to read your dataset and then directly call the POD.run with a numpy array containing the data on the points or cells, whatever you work with. Then you could store the output back to ensight for visualization. To adapt to vtkhdf you would need to know the conversion of nsided and nfaced to vtk.

If you can produce a working example for us in this way we can consider adding this functionality to the tool.

Cheers,