MFTT learns monotone triangular transport maps from samples of one or more target distributions. It implements the single-fidelity, hierarchical, and non-hierarchical constructions in the submitted manuscript Multifidelity Formulations for Triangular Transport.
MFTT supports Python 3.10 or newer. From a clone of the repository, create an isolated environment and install the runtime package:
python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install .Verify the installation from the repository root:
python -m pytestThe installation guide includes Windows commands, direct GitHub installation, and the complete standard dependency list. The standard installation is ready to run the tutorials.
The complete documentation is available from the Sphinx documentation index:
- Notation and mathematical foundations
- Single-fidelity guide
- Hierarchical multifidelity guide
- Non-hierarchical multifidelity guide
- Shared operations
- Public API reference
- Single-fidelity notebook
- Hierarchical notebook
- Non-hierarchical notebook
- Developer guide
The manuscript is the authority for the mathematical definitions and method names. This repository's citation page records the reference and maps its Sections 3–4 to the documentation.
MFTT is developed by Owen Davis and Gianluca Geraci and released under the MIT License. The implemented methods are based on the referenced manuscript by Owen Davis, Daniel Sharp, Youssef Marzouk, and Gianluca Geraci.