Creating Datasets for Simulation Models with PyMAPDL – Lesson 4

Private: PyMAPDL to create physics-based datasets for ML/AI applications Creating Datasets for Simulation Models with PyMAPDL – Lesson 4

This lesson features Ansys’ PyMAPDL tool, focusing on utilizing the ‘Pools’ class to enhance computational efficiency. With the help of an example, it demonstrates the generation of synthetic dataset creation for ML/AI based applications. For situations involving intense computation, the ‘Pools’ class enables running multiple instances of MAPDL in parallel, distributing the workload across cores for a faster and efficient execution. The lesson discusses two methods for dealing with the Pools namely, pre-creating the input files and defining a Python function encapsulating MAPDL FEM. Lesson explains the script and the procedure for setting up the model, mesh refinement, solution, and post-processing detail.