The purpose of DeepPhysX framework is to provide an interface between deep learning algorithms and numerical simulations. It is a full Python project with two main pipelines, allowing both to train artificial neural networks with simulated data and to use trained neural networks as components of numerical simulations. DeepPhysX manages not only the production of synthetic data with multiple numerical simulations in multiprocessing but also the storage of the produced dataset. Additional tools are provided to visualize numerical simulations and to follow the evolution of training sessions.
Numerical simulation, Deep neural networks, Scientific machine learning, Python
2021 - now