PySPOD: A Python package for Spectral Proper Orthogonal Decomposition (SPOD)
Gianmarco Mengaldo, Romit Maulik
Abstract
Large unstructured datasets may contain complex coherent patterns that evolve in time and space, and that the human eye cannot grasp. These patterns are frequently essential to unlock our understanding of complex systems that can arise in nature, such as the evolution of the atmosphere in the short (weather prediction) and long term (climate prediction), the behavior of turbulent flows, and the dynamics of plate tectonics, among several others. Identifying these coherent structures can prove crucial to facilitate the construction of modeling tools that can help anticipate scenarios that would not otherwise be predictable.
Topics & Concepts
Python (programming language)Computer scienceComputer graphics (images)MathematicsArtificial intelligenceProgramming languageControl Systems and IdentificationImage and Signal Denoising MethodsModel Reduction and Neural Networks