Dataset for Probabilistic Physics-Informed Neural Networks for Estimating Heterogeneous Elastic Properties from Low-Resolution and Noisy Displacement Data (PIE-PINN)

This repository contains the processed datasets used for training the Probabilistic Physics-Informed Neural Network (PIE-PINN) model for inverse elasticity estimation. The datasets in the Datasets folder were generated by spatially downsampling the original publicly available displacement data (DOI: 10.1073/pnas.2102721118). During training, synthetic measurement noise is added by the provided code before the data are used to train the PIE-PINN model. This repository contains the processed datasets used in the experiments; the original dataset should be obtained from the cited source.