Machine learning (ML) code used in the paper "Predicting stress, strain and deformation fields in materials and structures with graph neural networks" published in Scientific Reports.
December, 2022
Each folder contains the ML code for the corresponding mechanics example reported in the paper. The code is reported as developed initially by the authors; hence, the user should use his/her own dataset and modify the code accordingly. The code of the ML models can be instead readily used. The U-Net model used as benchmark for the wrinkling problem is also reported for completeness.
If you use or edit our work, please cite the paper:
Maurizi, M., Chao, G., Berto, F. Predicting stress, strain and deformation fields in materials and structures with graph neural networks. Sci. Rep. 12, 21834 (2022).