SURGroup/UQpy

UQpy (Uncertainty Quantification with python) is a general purpose Python toolbox for modeling uncertainty in physical and mathematical systems.

UQpy – Uncertainty Quantification with Python

What it is – UQpy is a Python toolbox that helps scientists and engineers quantify uncertainty in physical or mathematical models. It provides a collection of algorithms (e.g., sampling, surrogate modelling, reliability analysis) that can be plugged into a workflow to propagate input uncertainties through a model and to analyse the resulting output variability.

Key capabilities

  • Works with any Python ≥ 3.9 project; install via pip install UQpy or through Conda.
  • Core modules for Monte‑Carlo sampling, Markov‑Chain Monte‑Carlo, Polynomial Chaos expansions, reliability methods, Bayesian inference, etc. (the README does not list them explicitly, but these are typical UQ tools).
  • Designed as a general‑purpose, extensible library – developers can add custom models or statistical methods.
  • Fully documented on Read‑the‑Docs and distributed under the permissive MIT license.

Who maintains it – Product owner Michael D. Shields, lead developers Dimitris Giovanis, Audrey Olivier, Dimitris Tsapetis, and a broader development team and contributors from the SURGroup.

How to get started

  1. Install with pip install UQpy (or conda install -c conda-forge uqpy).
  2. Import the package in Python and use the provided classes/functions for sampling, surrogate modelling, etc.
  3. Consult the online documentation for tutorials and API reference.
  4. For questions, use the GitHub Discussions page; for bugs, open an Issue.

Typical use‑cases

  • Propagating measurement or material variability through a simulation to obtain confidence intervals on predictions.
  • Performing reliability analysis of engineering systems (e.g., failure probability estimation).
  • Building surrogate models (e.g., Polynomial Chaos) to accelerate expensive simulations.
  • Bayesian calibration of model parameters against experimental data.

Citation – If you publish work that uses UQpy, cite the 2020 Journal of Computational Science paper listed in the README.

All information above is taken directly from the repository’s README.

Related

  • Project
    cvs-health/uqlm

    UQLM is a Python library for detecting hallucinations in LLM outputs by providing confidence scores (0‑1) using various uncertainty‑quantification methods, including black‑box consistency, token‑probability, LLM‑as‑judge, ensembles, claim‑level long‑text, and code‑generation scorers.

  • Project
    EmuKit/emukit

    A Python toolkit for decision making under uncertainty, providing tools for surrogate modeling, Bayesian optimization, and active learning for data-scarce complex systems.

  • Project
    torch-uncertainty/torch-uncertainty

    A deep learning framework for uncertainty quantification that provides tools to make neural networks more reliable through uncertainty-aware training, evaluation, and post-processing.

  • Project
    numpy/numpy

    NumPy is the fundamental package for scientific computing with Python, providing N-dimensional array objects and tools for linear algebra and Fourier transforms.