Deep De-Homogenization: synthetic orientation fields

Data - in zipped form - related to the synthetic orientation fields used in:

Elingaard, M. O., Aage, N., Bærentzen, J. A., & Sigmund, O. (2022). De-homogenization using convolutional neural networks. Computer Methods in Applied Mechanics and Engineering, 388, 114197. https://doi.org/10.1016/j.cma.2021.114197

The dataset is split into a training and test set. The training set contains 10.000 orientation fields, while the test set contains 1000 orientation fields. Each orientation field is saved as .npy file, and can be loaded in python using the numpy.load function (https://numpy.org/doc/stable/reference/generated/numpy.load.html).