latin-hypercube-sampling

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

Efficient global optimization toolbox in Rust: bayesian optimization, mixture of gaussian processes, sampling methods

This is an implementation of Deutsch and Deutsch, "Latin hypercube sampling with multidimensional uniformity", Journal of Statistical Planning and Inference 142 (2012) , 763-772

Provides a number of methods for creating and augmenting Latin Hypercube Samples and Orthogonal Array Latin Hypercube Samples

This repo demonstrates how to build a surrogate (proxy) model by multivariate regressing building energy consumption data (univariate and multivariate) and use (1) Bayesian framework, (2) Pyomo package, (3) Genetic algorithm with local search, and (4) Pymoo package to find optimum design parameters and minimum energy consumption.

Latin hypercube sampling code for the lhs pacakge in the statistical software package R (www.r-project.org)

Simple implementation of Latin Hypercube Sampling.

ChemDesign: DWSIM Experiment Toolkit

Design of experiments (DoE) and machine learning packages for the iCFree project

Wrapper for sampling methods in R and approximate star discrepancy calculation

Parallel optimization engine for QuantConnect LEAN — a 679K-point parameter search in under 3 minutes via warm Docker workers, a persistent .NET harness, and smart search (GA / Bayesian / LHS).

Implementation, analysis and benchmarking of optimization algorithms. Developed in Python and results showed in Jupyter Notebook

Quantitative risk infrastructure for Meridian Private Bank. Three Monte Carlo methods - LHS for venture valuation, scenario-based correlated simulation for macro stress testing, and historical simulation for LP reporting. Calibrated against Q1 2026 Iran war oil shock.

A workshop covering the calibration of agent-based models.

This project is a simple implementation of the Latin Hypercube Sampling (LHS) method using Python. Its primary purpose is to demonstrate how LHS can be used to efficiently estimate expected values from random distributions or stochastic processes by ensuring better coverage of the input space compared to traditional random sampling.

Sampling and resampling techniques for random sample generation, estimation, and simulation

A C++ header-only library for Latin Hypercube Sampling

Operational modal analysis of offshore wind monopiles: OpenSeesPy transient simulation → Hankel DMD (MAC > 0.99997 on modes 1-3) + 1,000-sample Latin hypercube uncertainty study.

Hyperparameter tuning using a robust simulation optimization framework

OpenMC-based Bayesian Optimisation of BWR fuel assembly enrichment layouts using LLM-based seeding and orchestration. The LLM is used to generate physics-informed initial designs instead of LHS and to adjust the search bounds at each BO iteration, while the main optimisation is still handled by the standard GP/Expected Improvement loop