expandLHS

expandLHS is a Python module that implements a model-free expansion algorithm for a Latin Hypercube sample set. The Latin Hypercube Sampling (LHS) is a stratified sampling technique that allows to generate \(N\) near-random samples in the \(P\)-dimensional hypercube \([0, 1)^P\). It is a space-filling sampling strategy that ensures the one-dimensional projection property, i.e. the samples are uniformly distributed in each one-dimension projection. This module extends the usage of this technique by implementing an expansion algorithm. Starting from an initial LHS set of size \(N\), expandLHS samples \(M\) additional points in a LHS-like fashion trying to preserve the LHS properties at most.

This algorithm is introduced in

• “LHS in LHS”: a new expansion strategy for Latin hypercube sampling in simulation design. M. Boschini, D. Gerosa, A. Crespi, M. Falcone (to be published)

The code is distributed under version control at

• github.com/m-boschini/expandLHS

The documentation is available at

• m-boschini.github.io/expandLHS

To install the code simply use

pip install expandLHS

An example notebook can be found in the documentation together with a detailed description of the functions.

expandLHS is released under the MIT License.

Change log

• v1.1.0 New feature: now it is possible to initialise a Latin Hypercube when the class is created

• v1.0.0 First public release.

(Third-level versions not explicitly indicated refer to patches for minor typos/bug fixes)