Logo of the Physikalisch-Technische Bundesanstalt

PyThia v4 release and paper publication


Originally developed for fast global sensitivity analysis and efficient parameter reconstruction for applications in nano-optical metrology, PyThia provides an all purpose non-intrusive Python package to approximate high dimensional functions.
Based on general polynomial chaos approximation obtained via linear regression, PyThia generates functional surrogates by relying purely on training data pairs of function input and output values.
The software package released a new version (v4), which adheres to state-of-the-art software development criteria, such as automatic testing, an automatically generated online documentation, executable and documented tutorials and automatically generated static snapshots for all versions with citable DOI.
An article presenting PyThia was published in the Journal of Open Source Software to verify the quality standards of the software.


For more information on the software, installation guides, tutorials and a documentation please visit the Opens external link in new windowPyThia Homepage or the Opens external link in new windowGitLab repository.
A short overview of Pythia's key features can also be found Opens external link in new windowhere).

Related publication:
Hegemann et al., (2023).
PyThia: A Python package for Uncertainty Quantification based on non-intrusive polynomial chaos expansions.
Journal of Open Source Software, 8(89), 5489,
Opens external link in new windowhttps://doi.org/10.21105/joss.05489