sanjayg0 / plom Goto Github PK
View Code? Open in Web Editor NEWPLoM is an open source python package that implements the algorithm of Probabilistic Learning on Manifolds with and without constraints (Soize and Ghanem, 2016; Soize and Ghanem, 2019) for generating realizations of a random vector in a finite Euclidean space that are statistically consistent with a given dataset of that vector. The package mainly consists of python modules and invokes a dynamic library for more efficiently computing the gradient of the potential, and can be imported and run on Linux, macOS, and Windows platform. This repository also archives the unit/integration tests and examples of applying the algorithm to practical engineering problems.
License: MIT License