mlmc 1.0.3

Multilevel Monte Carlo method.

MLMC is a Python library implementing the Multilevel Monte Carlo (MLMC) method. It provides tools for sampling, moment estimation, statistical post-processing, and more.

Originally developed as part of the GeoMop project.

Features

• Sample scheduling

• Estimation of generalized moments

• Advanced post-processing with the Quantity structure

• Approximation of probability density functions using the maximum entropy method

• Bootstrap and regression-based variance estimation

• Diagnostic tools (e.g., consistency checks)

Installation

The package is available on PyPI and can be installed with pip:

pip install mlmc

To install the latest development version:

git clone https://github.com/GeoMop/MLMC.git
cd MLMC
pip install -e .

Documentation

Full documentation, including tutorials, is available at: https://mlmc.readthedocs.io/

Topics covered include:

• Basic MLMC workflow and examples

• Definition and composition of Quantity objects

• Moment and covariance estimation

• Probability density function reconstruction

Development

Contributions are welcome! To contribute, please fork the repository and create a pull request.

Before submitting, make sure all tests pass by running tox:

pip install tox
tox

tox creates a clean virtual environment, installs all dependencies, runs unit tests via pytest, and checks that the package installs correctly.

Requirements

MLMC depends on the following Python packages:

• NumPy

• SciPy

• h5py

• attrs

• ruamel.yaml

• gstools

• memoization

• scikit-learn

License

• Free software: GNU General Public License v3.0