PLoM
PLoM 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.
Documentation
General
Installation
Examples
- Example 0: Simple example in 20 dimensions
- Example 1: Simple example in 2 dimensions with constraints
- Example 2: Surrogating MSA of a 12-story RC frame
- Example 3: Surrogating IDA of a 12-story RC frame
- Example 4: Application in damage and loss assessment
Acknowledgement
This software was developed under support by the National Science Foundation under Grant Nos. 1612843 and 2131111. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation or the Regents of the University of California.
How to cite the software
Please cite the software as
Zhong, K., Gual, J., and Govindjee, S., PLoM python package v1.0, https://github.com/sanjayg0/PLoM (2021).