This project is intended to provide a facility to assess the impact of input uncertainty on analysis outputs via latin hypercube sampling.
Underlying principles: - define input variables and their statistical characteristics in an input yaml file - use these definitions to generate a random latin hypercube sampling scheme with samples weighted by equal area under CDF - run each sample from the LHS dataset and collate results - collate and regress statistical description onto output variables - test output data for confidence level of mean values
Work to date: - Repo setup based on own project template - provides basic project architecture with initial definition of pre-committ hooks and linting, github actions to run unit tests, structure to allow future packaging for distribution, license, readme and requirements. - Define input file read-in functionality and initial associated unit tests - Define functionality to generate LHS sampling scheme - Define functionality to downselect target solver and provide solver definition
Future tasks: - Refine unit tests - Complete doctrings for self documentation - Add remaining functionality - Package and distribute
Under development. Publicly available for code review only.
If you want to contribute to this project, please contact the author
This project is licensed under the GNU GPL3 terms and conditions